CAUJ EARTH SCIENCE? JdBRARY THE LIBRARY OF THE UNIVERSITY OF CALIFORNIA PRESENTED BY PROF. CHARLES A. KOFOID AND MRS. PRUDENCE W. KOFOID PALEONTOLOGY LIBRARY / ! GEOLOGICAL MANUAL BY HENRY T. DE LA BECHE, F.R.S., V.P.G.S. MEMB. GEOL. SOC. OF FRANCE, CORK. MEMB. ACAD. NAT. SCI. PHILADELPHIA, ETC. THIRD EDITION, CONSIDERABLY ENLARGED. LONDON: CHARLES KNIGHT, 22 LUDGATE STREET, AND 13 PALL-MALL EAST. 1833. PALEONTOLOGY LIBRARY Gift of w. **. Kofoid PRINTED BY RICHARD TAYI.Ott, RKD MON COURT, FLKET STIIBBT. I 33 fkfeo. PREFACE, it is attempted, as in works of the following description, to sketch the actual state of a particular science, and at the same time to point out a few of the conclusions that may be hazarded from known facts, an author has always great difficulty in avoiding unnecessary and tedious detail on the one hand ; while, on the other, he must notice such facts as may convince a student that he is not wandering in a wilderness of crude hypotheses or unsupported assumptions. The present edition contains so many additions to the greater part of the Work, that it would be tedious, and, indeed, somewhat difficult to enumerate them. The chief alteration consists in removing the various lists of organic remains to the end of the volume, where they can be more readily consulted. The Work has also, at the suggestion of friends, been printed in a larger form, and in a larger type, the small type of the former editions having only been retained for the lists of organic remains. Under the heads of Inferior Stratified, and Unstratified Rocks, calculations have been introduced respecting the substances of which such rocks are chemically composed; and it is hoped that these calculations may be found A 2 IV PREFACE. useful, as also some observations respecting geological maps and sections, and the geological examination of a country. The Author has availed himself largely of the additions made by M. von Dechen, with the assistance of the ce- lebrated Von Buch and other German geologists, to the German translation of this Work, more particularly as respects the geology of Germany and the lists of or- ganic remains. He has not been able to avail himself of any additions to the French translation of the Manual, made under the superintendence of M. Brochant de Vil- liers, as it will not appear much before the present edition ; but the Author is informed that it will contain a further development, by M. Elie de Beaumont, of his theory of the elevation of mountain chains, as also additions to the lists of organic remains. He has not seen a copy of the American edition of this Work, and is therefore not aware that any additions have been made to it. There can be little doubt that, from a strong desire to find similar organic remains in supposed equivalent de- posits, even at great distances, and from an equally strong desire to discover new species, the same organic remains, particularly shells, often figure in our catalogues under two names, while different species are made to appear as one. Notwithstanding these difficulties, it will, how- ever, be evident, from a glance of these catalogues, that a great mass of information has been gradually collected on this subject alone, from which the most important results must follow,, even though the various lists may require very considerable correction. While availing himself of these and similar catalogues, the student should be careful to recollect, that however great and valuable the aid of Zoology and Botany may be in geological investigations, Physics and Chemistry PREFACE. V are of still greater importance ; inasmuch as the former can only be employed with advantage in explanation of a portion of the phenomena observed, while the latter are available to a very great extent in explanation of the whole. The Author has been particularly anxious to point out his various sources of information, even when he himself has visited the same countries ; that, independently of the fundamental principle suum cuique, the student should be enabled more fully to avail himself of the labours of the various authors cited, by referring to their published works for greater detail than could be admitted into a volume of this description. In a rapidly advancing science like Geology, to which new facts are constantly added, and in which the chances of new views by their combination are consequently mul- tiplied, it is almost impossible to avoid hazarding certain general conclusions, when the various known facts pass in review before us. In those which the Author has ventured to bring forward, he has endeavoured always to follow that system of induction which can alone lead to exact know- ledge ; but as truth, and truth alone, is the object of all sci- ence, he can sincerely declare, that if from the discovery of new facts, or from more sound views respecting those already known, his conclusions should not appear tenable, he would not only be most ready to abandon them, but to rejoice that an untenable hypothesis may have been the means of leading to more exact knowledge, if it should have fortunately so happened that it promoted the requi- site inquiry. Essentially it is of little importance whose or what theory may in the end be found most accurate ; so long as we approximate towards the truth, we accomplish all that can be expected ; and it is clear, that the greater the amount of known facts, the greater the chance of accu- Vl PREFACE. racy, not only from the larger mass of information presented to the mind, but also from the frequent checks offered to hasty conclusions. Happily facts have become so multiplied, that Geology is daily emerging from that state when an hypothesis, pro- vided it were brilliant or ingenious, was sure of advocates and temporary success, even when it sinned against the laws of physics and facts themselves. It is not difficult to foresee, that this science, essentially one of observation, instead of being, as formerly, loaded with ingenious specu- lations, will be divided into different branches, each in- vestigated by those whose particular acquirements may render them most competent to do so ; the various com- binations of inorganic matter being examined by the Natural Philosopher, while the Natural Historian will find ample occupation in the remains of the various ani- mals and vegetables which have lived at different periods on the surface of the earth. TABLE OF CONTENTS. SECTION I. Page. Figure of the Earth 1 Density of the Earth ib. Superficial Distribution of Land and Water 2 Saltness and Specific Gravity of the Sea 3 Temperature of the Earth 6 Temperature of Springs 14 Temperature of the Sea and of Lakes 21 Temperature of the Atmosphere 27 Valleys 29 Changes on the Surface of the Globe 34 Classification of Rocks 35 SECTION II. Degradation of Land 43 Rivers 50 Glaciers 65 Delivery of Detritus into the Sea 67 Action of the Sea on Coasts 77 Shingle Beaches 79 Sandy Beaches 84 Tides 92 Currents 99 Transporting Power of Tides 110 Transporting Power of Currents 112 Active Volcanos 115 Extinct Volcanos 134 Mineral Volcanic Products 137 Volcanic Dykes, &c 138 Earthquakes 140 Hurricanes 149 ABBREVIATIONS OF AUTHORS' NAMES THE LISTS OF ORGANIC REMAINS. Ag- Agassiz. Hoen. Bast. Basterot. Jag. Beaum. Elie de Beaumont. Lam. Blain. Blainville. Lam*. Blum. Blumenbach. Linn. Bobl. Boblaye. Lons. Broc. Brocchi. L. & H. Al. Brong. Alex. Brongniart. Mant. Ad. Brong. Brug. Adolphe Brongniart. Bruguiere. Munst. Murcb. Buckl. Buckland. M. de S. Conyb. Cuv. Conybeare. Cuvier. Nils. Park. DcC.,orDcCau De Caumont. Pas. Defr. Defrance. Phil. De la B. De la Beche. Raf. Desh. Deshayes. Rein. DesM. Des Moulins. Sauv. Desm. Desmarest. Schlot. Desn. Desnoyers. Sedg. Dufr. Dufrenoy. Sow. Dum. Dumont. Sternb. Fuuj. de St. F. Faujas de St. Fond. Thir. Fleni. Fleming. Thur. Goldf. Goldfuss. Y. & B. G. T. German Tran si. of Manual. Wahl. Her. Herault. Weav. His. Hisinger. Hceninghaus. Jager. Lamarck. Lamouroux. Linnaeus. Lonsdale. Lindley and Hutton. Mantell. Munster. Murchison. Marcel de Serres. Nilsson. Parkinson. Passy. Phillips. Rafinesque. Reinecke. Sauveur. Schlotheim. Sedgwick. Sowerby. Sternberg. Thirria. Thurman. Young and Bird. Wahlenberg. Weaver. ERRATA. Page 63, line 6, for adn read and 141, 15, for hwich read which 161, 9, for Planobris read Planorbis 193, 34, for from places read from their places 320, 15, for von Decken read von Dechen 350, 15, for is read as 351, 24, for beds deposited read beds were deposited GEOLOGICAL MANUAL. SECTION I. figure of the Earth. IT has been concluded, both from astronomical and geo- desical observations, that the figure of the earth is a spheroid. This spheroid has been considered as one of rotation, or such a figure as a fluid body would assume if possessed of rotatory motion in space. The amount of the flattening of the poles, or the difference of the diameter of the earth from pole to pole, and its diameter at the equator, has been variously estimated ; but it is com- monly received that the polar axis is to the equatorial diameter as 304? to 305, the compression of the earth, or flattening at the poles, being thus considered as = 7 j. The equatorial diameter about = 7924 miles.* The polar axis = 7898 Difference 26 Density of the Earth. Various opinions have been entertained on this subject ; but it appears certain that the internal density is greater than * Considering the flattening of the poles as = -3-^-5-, M. Daubuisson has made the following calculations Radius at the equator . Semi-terrestrial axis Diff. or flattening of poles Radius in lat. 45 . . A degree at same lat. . A degree of long, in same lat. Surface of our earth 5098857 square myriametres. The volume 1082634000 cubic myriametres. Traite de Geognosie, ed. 2mc, torn. i. B 6376851 metres. 6355943 20908 6366407 111115 78828 2 Supetfaial Distribution of Land and Water. the solid superficial density. Daubuisson infers from the ob- servations of Maskelyne, Play fair, and Cavendish, that " the mean density of the earth is about five times greater than that of water, and consequently, about double that of the mineral crust of our globe*." Laplace considered the mean density of our spheroid as = 1*55, the solid surface being I. Accord- ing to Baily, the density of the earth is 3-9826 times greater than that of the sun, and is to that of water as 11 to 2 f . Superficial Distribution of Land and Water. The relative proportion of dry land to the ocean, as it at present exists, is such, that nearly three-fourths of the whole surface of the globe may be assigned to the latter. Of the former, the configuration is very various, presenting the greatest surface in the Northern hemisphere. Although the land sometimes rises high above the level of the sea, according to our general ideas on such subjects, it is, in reality, but slightly removed above that level, when considered, as it should be, with reference to the radius of the earth J. The superficies of the Pacific Ocean alone is estimated as some- what greater than that of the whole dry land with which we are acquainted. Dry land can only be considered as so much of the rough surface of our globe as may happen, for the time, to be above the level of the waters, beneath which it may again disappear, as it has done at different previous periods. La- place calculated that the mean depth of the ocean was a small fraction of twenty-five miles, the difference produced in the diameters of the earth by the flattening of the poles. It has been variously estimated at between two and three miles. The mean height of the dry land above the ocean-level does not exceed two miles, but probably falls far short of it ; therefore, assuming two miles for the mean depth of the ocean, the waters occupying three-fourths of the earth's surface, the present dry land might be distributed over the bottom of the ocean, in such a manner that the surface of the globe would present a mass of waters ; an important possibility, for, with it at command, every va- riety of the superficial distribution of land and water may be imagined, and consequently every variety of organic life, each suited to the various situations and climates under which it would be placed. The surface of the globe's, solid crust is so uneven, that the ocean, preserving a general level, enters among the dry land * Trait6 de Gognosie, ed. 2me, torn. i. p. 18. f Baily, Astronomical Tables. J See the diagram in my Sections and Views illustrative of Geological Phsenomena, pi. 40. Saltness and Specific Gravity of the Sea. 3 in various directions, forming what are commonly termed in- land seas ; such as the Baltic, Red, and Mediterranean Seas, in which geological changes may be effected different from those in the open ocean. Masses of salt water are sometimes included in the dry land, which have been termed Caspians, from the Caspian Sea, the largest of them. These have no communication with the main ocean ; indeed the level of the Caspian is much lower than that of the Black or Mediterranean Seas, the former body of salt water occupying, with lake Aral and other minor lakes, the lower part of an extensive depression in Western Asia, (from 200 to 300 feet under the general ocean-level,) which receives the waters of the Volga and other rivers. These bodies of salt water have been variously accounted for ; some supposing that they have been left isolated by a change in the relative level of land and water, while others imagine their saltness to arise from their occurrence in countries impreg- nated with saline matter. It is stated, in support of the latter opinion, that the Caspian, and the lakes Aral, Baikal, &c. are situated where salt springs abound. Whatever may be their origin, it will be obvious, that if the fresh water they receive be not equal to their evaporation, they will become gradually more saline, until, the water being saturated, the surplus salt will be deposited at the bottom, and strata of it will be formed of a size and depth proportioned to those of the lake or sea. It would be out of place to attempt a general description of all the various combinations of land and water, with which all must be more or less familiar ; but it may be useful to notice that fresh-water lakes cover very considerable spaces, and that thus very extensive deposits may now take place, which can only envelope the remains of terrestrial or fresh -water animals and vegetables. Saltness and Specific Gravity of the Sea. The whole body of the ocean is composed of salt water, which does not vary very materially in composition, as far as we can judge from the experiments made on it. From evaporation and the fall of rain, the sea will be less salt at the surface than at some little depth beneath it. According to Dr. Murray, sea-water collected from the Firth of Forth contained, in \ 0,000 parts, Common salt .... 220-01 Sulphate of soda . . . 33-16 Muriate of magnesia . . 42-08 Muriate of lime . 7'84 303-09 B 2 4. Saltness and Specific Gravity of the Sea. Dr. Marcet states that 500 grains of sea- water, taken from the middle of the North Atlantic, contained, Muriate of soda . . .13-3 Sulphate of soda . . . 2-33 Muriate of lime . . . 0-995 Muriate of magnesia . . 4-955 21-580 According to the experiments of Dr. Fyfe (Edin. Phil. Journal, vol. i.), the waters of the ocean between 61 52' N. and 78 35' N. do not differ much in their saline contents, these being between 3'27 and 3'91 per cent. The waters were obtained by Scoresby. M. Eichwald informs us, that the waters of the Caspian Sea contain much sulphate of magnesia, in addition to the various other salts held in solution by them. Dr. Marcet instituted a series of experiments on the spe- cific gravity of water, of which the following are the results : Sp. Gr. Sp. Gr. Arctic Ocean .... 1-02664 Northern Hemisphere . 1-02829 Equator 1-02777 Southern Hemisphere . 1-02882 Yellow Sea 1-02291 Mediterranean. . 1-0293 Sea of Marmora . . 1-01915 Black Sea .... 1 01418 White Sea . . . . 1-01901 Baltic ..... 1-01 523 Ice-Sea Water . . 1-00057 Lake Ourmia . . 1-16507 The same author concluded from his observations, " 1. That the Southern Ocean contains more salt than the Northern Ocean in the ratio of 1-02919 to 1-02757- " 2. That the mean specific gravity of sea-water near the equator is 1-02777, intermediate between that of the Northern and Southern hemispheres. " 3. That there is no notable difference in sea-water under different meridians. " 4?. That there is no satisfactory evidence that the sea at great depths is more salt than at the surface *. " 5. That the sea, in general, contains more salt where it is deepest and most remote from land ; and that its saltness is always diminished in the vicinity of large masses of ice. * The author of the abstract of Dr. Marcet's observations in the Edin. Phil, Journal, cites the following observations of Mr. Scoresby in support of this conclusion. Sp. Gr. f Surface . . . 1-0261 Lat. 7616'N.<> At 738 feet LAt 1380 feet f Surface . . I At 120 feet Lat. 7634'N.{ At 240 feet | At 360 feet LAt 600 feet 1-0270 1-0269 1-0265 1-0264 1-0266 1-0268 1-0267 Salt ness and Specific Gravity of the Sea. 5 " 6. That small inland seas, though communicating with the ocean, are much less salt than the ocean. " 7. The Mediterranean contains rather larger proportions of salt than the ocean *." M. Lenz, who accompanied Kotzebue's expedition, inferred from numerous experiments that, 1. The Atlantic Ocean is salter than the South Sea; and the Indian Ocean, being the transition from the one to the other, is salter towards the Atlantic, on the west, than towards the South Sea, on the east. 2. In each of these three great oceans there exists a maxi- mum of saltness towards the north, and another towards the south ; the first being further from the equator than the se- cond. The minimum between these two points is a few de- grees south of the equator in the Atlantic, and probably also in the Pacific, though M. Lenz's observations did not extend sufficiently low in the Pacific. 3. In the Atlantic the western portion is more salt than the eastern. In the Pacific the saltness does not appear to alter with the longitude. 4. In proceeding north from the northern maximum, the specific gravity of the water diminishes constantly as the lati- tude increases. The same author considers that, from the equator to 4-5 N., the water of the sea, to the depth of 1000 fathoms, possesses the same degree of saltness f. The saltness of the sea, particularly that of its surface, would seem greatly to depend on the proximity of nearly permanent ice," and of large or numerous rivers. Thus, as is seen above, the Baltic, White, Black, and Yellow Seas are less salt than the main ocean, because they are supplied with comparatively large quantities of fresh water. From the small proportion of salt contained in the Black Sea and Sea of Azof, the bays of the former frequently contain ice, and the latter is stated to be frozen over during four months in the year. The superior saltness of the Mediterranean, though an ii> land sea, is attributed to the evaporation of its surface, which is supposed greater than the quantity of fresh water with which it is supplied. In consequence, two great currents, one from the Black Sea and the other from the Atlantic, flow into it to supply the waste caused by evaporation. The saline contents of the sea are important, as all chemical changes or deposits, taking place in it, will be more or less af- fected by them. The gravity and pressure of the sea are of * Phil. Trans. 1819 ; and Edin. Phil. Journal, vol. ii. f Edin. Journ. of Science, April 1832. 6 Temperature of the Earth. still greater consequence ; for, as the pressure increases with the depth, effects, which would be possible at one depth, would be impossible at another. Thus, it is obvious from the inge- nious experiments of Sir James Hall, that carbonate of lime may be fused by heat without the loss of its carbonic acid, if subjected to great pressure, such as exists at the bottom of the deep sea. The pressure of the sea must also have consider- able influence on the kind of animal and vegetable life found at different depths ; and we may infer that beneath very deep seas such life does not exist, great pressure and the absence of the necessary light being as destructive to it as the cold and the rarity of the air are in the higher regions of the atmo- sphere. The compressibility of water, which was for a long time doubted, has been proved by experiment. According to the observations of M. CErsted, corrected for the pressure of part of the apparatus employed, this compressibility amounts to 46*65 millionths of its volume for a pressure equal to each at- mosphere. The experiments of MM. Colladon and Sturm, corrected in the same manner, make the compressibility of water, not deprived of air, equal to 4-7'85 millionths for each atmosphere ; while that of water, deprived of air, is equal to 49*65 millionths under the same pressure *. M. Poisson es- timates that it would require a pressure equal to 1100 atmo- spheres to reduce water six hundredths of its volume f . Water containing salts in solution is found to be somewhat less compressible. It follows, that at great depths, and be- neath a great pressure of the ocean, a given quantity of water will occupy a less space than on the surface, and will, conse- quently, by this circumstance alone, have its specific gravity increased. Temperature of the Earth. The superficial temperature of our planet is certainly very materially influenced by, if it may not be entirely due to, solar heat. That the difference of seasons, and of the climates of various latitudes, originates in the greater or less exposure to the sun, is obvious. That local circumstances cause great va- riations of superficial temperature, is also well known ; yet the principle seems to prevail, that under equal circumstances, the temperature decreases from the tropics to the poles. It would be useless to increase the size of this volume with a detail of the various temperatures that have been observed * Pouillet, Elemens de Phys. Experimentales, 2me ed. t. ii. p. 65. I Poisson, Nouvelle Theorie de 1'Action Capillaire, p. 277. Temperature of the Earth. 7 in different situations, or of the modifications arising from local causes ; this will be found in various works devoted to the subject, more particularly in Humboldt's Treatise on Iso- thermal Lines. Respecting the temperature of our globe, M. Arago has made the following remarks : " 1st, In no part of the earth on land, and in no season, will a thermometer raised from two to three metres above the ground, and protected from all re- verberation, attain the 46th centigrade degree : 2ndly, In the open sea, the temperature of the air, whatever be the place and season, never attains the 31st centigrade degree: Srdly, The greatest degree of cold which has ever been observed upon our globe, with the thermometer suspended in the air, is 50 centigrade degrees below zero : 4-thly, The temperature of the water of the sea, in no latitude, and in no season, rises above + SO centigrade degrees *." Geologists have discovered that the superficial temperature of the earth has not always remained the same, and that there is evidence of a very considerable decrease. This evidence will be found scattered over such parts of the following pages as treat of organic remains, and therefore need not be adduced here. It may, however, be right to remark, that it rests on the discovery of vegetable and animal remains entombed in situations, where, from the want of a congenial temperature, such animals or vegetables would now be unable to exist. Undoubtedly this inference rests on the supposed analogy be- tween animals and vegetables now existing, and those ot a si- milar general structure found in various rocks, and at various depths beneath the earth's surface : but as we now find every animal and vegetable suited to the situations proper for them, we have a right to infer design at all periods, and under every possible state of our earth's surface; and therefore to consider, that similarly constituted animals and vegetables have, in ge- neral, had similar habitats. This decrease in surface-temperature may arise either from external, superficial, or internal causes. External Iiifluence. Heat, derived from the sun, producing such great effects at present, it has been supposed that a dif- ference in the relative position of our planet and our great lu- minary would cause a corresponding change in the surface- temperature of the globe. Theories have been invented which suppose such a change in the earth's axis as would render the present poles parts of the equator, and thus capable of having once supported a tropical vegetation, which has gradually disappeared, and been replaced by such plants as can exist * Ann. clc Phys. et dc Chim. torn, xxvii. ; and Edin. Phil. Journ. 1825. 8 Temperature of the Earth. amid masses of ice and snow. Mr. Herschel, viewing this subject with the eye of an astronomer, considers that a di- minution of the surface-temperature might arise from a change in the ellipticity of the earth's orbit, which, though slowly, gradually becomes more circular. No calculations having yet been made as to the probable amount of decreased tempera- ture from this cause, it can at present be only considered as a possible explanation of those geological phaenomena which point to considerable alterations in climates *. Superficial Influence. A decrease of temperature may arise from such a variation in the relative position of land and water, and in the elevation and form of land, as may cause the cli- mate, in any given position on the earth's surface, so to change, that a greater heat may precede a less heat, and the land be capable of supporting the vegetables and animals of hot cli- mates at one time, and be incapable of doing so at another. For this ingenious theory we are indebted to Mr. Lyell f. It supposes a combination of external and internal causes ; the latter raising or depressing the land in the proper situations, the former supplying the necessary heat. It also supposes the possible recurrence of a warm climate, so that the same situations might alternately be placed under the influence of a raised and a depressed temperature. We have so few data for estimating the value of this theory, that it can only be con- sidered as a possible explanation of a diminished temperature. It must, however, be admitted, that, in every state of the earth's surface, the relative disposition of land and water, and the form or elevation of the land, would always have had, as they now have, very considerable influence on climate. Internal Iiifluence. From the earliest times an opinion has existed among philosophers that a central heat exists ; an opinion naturally arising from the phaenomena of volcanos and hot springs. But, notwithstanding this opinion, it was not until a comparatively late period that direct experiments were instituted, for the purpose of determining whether the tem- perature does, or does not, increase with the depth, or from the surface downwards. Various observations have been made on the temperature of mines in Great Britain, France, Saxony, Switzerland, and even Mexico. All those made previous to 1827 were collected, arranged, and commented on by M. CordierJ. Experiments on the temperature of mines have been made in various ways ; * Herschel, Geol. Trans., 2nd series, vol. iii. t Principles of Geology. J Essai sur la Temperature de 1'Intcrieur dc la Terrc : Mem. 66 cent." The air in these galleries was estimated at 12,560 cubic metres. Another source of error arises from the circulation of air in mines, and its introduction from the surface. This will vary according to the local distribution of the galleries in a mine ; but there will always be a tendency to replace expanded and heated air by that which is more dense and cold ; consequently, from whatever cause the heat of a mine may be derived, if the air in it be, as usually happens, warmer than that of the sur- face, the cold air will always strive to get into the mine, and the heated air to escape from it. It follows, that the entrance of air from the exterior surface tends to lower the tempera- ture of the mine, and in some measure to check the heat caused by the workings. M. Cordier * observes, on this sub- ject, that the mean temperature of the mass of 'air -, introduced into a mine during a year, is lower than the mean temperature of the country for the same year, and estimates the difference * Essai sur la Temperature de 1'Inte'rieur de la Terre. It has been supposed, the air in mines being under a greater pressure than that at the sud'aee, and undergoing this change in a short time, that heat 10 Temperature of the Earth. between them at between 2 arid 3 cent, for the greater part of the mines in our climate. The waters in mines may either give too high or too low a temperature, as they may be either derived from beneath or above. If waters descend from the surface into a mine, they will carry with them their original temperature, modified by the heat of the substances through which they pass ; so that their difference of temperature in the mine and on the surface will depend on their abundance or scarcity, and on their slowness or rapidity of motion. Moreover they will constantly tend to reduce the surfaces of rock through which they per- colate to their own temperature. The same remarks apply to water derived from a lower level. The temperature observed in the rock itself will be more or less affected, according to circumstances, by that of the water or air near it. So that the sides of a mine, to certain distances, might possess a heat not common to the mass of rock at the same level. From these various sources of error, to which others might be added, the observations made under circumstances that might be influenced by them, can only be considered as ap- proximations towards an estimate of the value of this mode of inquiry. To render each set of observations available for what they may be worth, M. Cordier has classed those made under different circumstances under different heads. His tables, thus formed, have also the great advantage of being reduced to common measures of heat and depth. From these the fol- lowing have been selected as, perhaps, least liable to error*. would be evolved sufficient to cause the appearance of an increase of tem- perature corresponding with an increased depth. But as the cold air will become expanded by the heated air of the workings, and as the change of rsure cannot be very sudden, this does not appear sufficient to account the phenomena observed. According to Mr. Ivory (Phil. Mag. and Annals, of Phil. vol. i. p. 94), one degree of heat, of Fahrenheit's scale, will be extricated from air when it undergoes condensation = ^fa. ; and if a mass of air were suddenly reduced to half its bulk, the heat evolved would be = 90. * The temperature in these tables is marked in degrees of the centigrade thermometer. When we consider the simplicity of this scale, and the faci- lities with which calculations can be made with it, it seems strange that its use should not be generally adopted in this country, where we continue to employ, from habit, the least philosophical of the tbree scales. The centi- grade scale can easily be reduced to that of Fahrenheit, by considering that the latter is to the former, between the freezing and boiling points of water, as 180 to 100, or as 9 to 5. The degrees of Reaumur's scale are to those of Fahrenheit's as 4 to 9. As the zero of Fahrenheit's scale is 32 of that scale below the zero in the others, it is always necessary to make a proper allowance for it. Temperature of the Earth. 1 1 Table of Observations made on the Springs in Mines. Names, Authors, and Dates. Mines. Depth. Tempe of the Springs. rature mean of the Country. Saxony. Daubu- isson. End of^ winter, 1802. Brittany. Daiibu- isson. 5thSept.,^ 1805. Cornwall. Fox. 7 Publ. 1821. J Mexico. Humboldt * Lead and Silver of Junghohe-Birke . Metres. 78 217 256 224 39 75 140 60 80 120 230 439 522 Deg. 9-4 12-5 13-8 14-4 11-9 11-9 14-6 12-2 15- 15' 19'7 27-8 35-8 Deg. 8- 8- 8- 8- 11-5 11-5 11-5 11- 11- 11- 11- 10- 16- Beschert Gliick . Himmelfahrt . Poullaouen . . Huelgoet Dolcoath Copper . Guanaxuato Silver I Tables of the Temperature of the Rock in Mines. Thermometer placed in a niche cut in the rock, distant from the principal workings : the bulb in the rock ; the rest in a glass tube; the whole covered by a glass door, closing the niche, and only opened for observation. Saxony. De Trebra. ) ( Mine 1805, 1806, 1807 Depth. Metres, of Beschert) 180 Temperature of Rock, of Country. 11-25 8' Gliick ; lead & sil. J 260 15- 8-75 12-81 15- 18-75 8- {71-9 168-2 268-2 379-54 II. Thermometer plunged in the earthy matters at the bottom of galleries, which had been inundated two days*. C p 'n W f, Fw I United Mines.. ' 348 30 ' 8 10 ' Published 1821. .. J \366 31-1 III. Thermometer fixed in the rock of a gallery, for eighteen months at a yard deep. Cornwall. Fox. ) ^ , ,-, Published 1822. > Dolcoath 421 24 ' 2 | Dolcoath 421 10- * M. Cordier remarks on the error that may, in this case, arise from the mixed temperature of the galleries, before inundation, produced by the usual causes in mines at work, and of the waters during inundation. On this sub- ject he cites some observations of his own at Ravin, near Carmeaux, which show that the differences of temperature between the rubbish on the floor of the galleries, and that proper to the level, amounted to 2-6, 2*8, and even 3*-l centigrade. 12 Temperature of the Earth. Table of the Temperatures of the Rock observed in the Coal- mines at Carmeaux, Littry, and Decise. Depth. Temperature. Metres. Deg. Water of the well V6riac ......... 6'2 12-9 Water of the well Bigorre ....... 11 '5 13-15 Rock at the bottom of Ravin Mine . . . 181'9 17'1 Rock at the bottom of Castellan Mine . . 192* 19*5 Littry. Surface ................ 0- 11- Rock at the bottom of St. Charles Mine j 99 . j g. 135 mean of 2 obs .......... / Decise. Water of the well Pelisson ....... 8'8 11 '4 Water of the Puits des Pavilions .... 16'9 11 '67 Rock in the Jacobe Mine ........ 107' 17'78 -- ....... 171' 22-1 These observations were made with great care ; " the ther- mometer was loosely rolled in seven turns of silk paper, closed at bottom, and tied by a string a little beneath the other ex- tremity of the instrument, so that so much of the tube might be withdrawn as might be necessary for an observation of the scale, without fearing the contact of the air : the whole con- tained in a tin case." This was introduced into a hole from 4 to 26 inches in depth and 1^- in diameter, inclined at an angle of 10 or 15; so that the air once entered into the holes could not be renewed, because it became cooler, and consequently heavier, than that of the galleries. The thermometer was kept as nearly as possible at the tempera- ture of the rock, by plunging it among pieces of rock or coal freshly broken off, and by holding it a few instants at the mouth of the hole, into which it was afterwards shut, a strong stopper of paper closing the aperture. The thermo- meter generally remained in this hole about an hour *. Temperature of Water in Artesian Wells, and in neglected Mines. Artesian wells are well known as borings, by which water, at different distances from the surface, rises to, and even above, that surface, from its endeavour to escape. According to the * Where the investigation of the increase or decrease of temperature, be- neath such a depth as may be out of atmospheric influences, is so easy, with a few necessary precautions, it is surprising, that in the British collieries, which are so numerous, and many of which are very deep, so few direct ex- periments should have been made on the temperature of the rock itself. Temperature of Artesian Wells. 1 3 observations of M. Arago, the greater the depth of these wells, the higher is the temperature of the waters that flow from them. From experiments made by M. Fleuriau de Bellevue, in an Artesian well on the sea-side near Rochelle, the temperature increases with the depth. The well, at the time of the first experiment, was 3 inches in diameter, and 316 feet deep, and in it a column of brackish and stagnant water rose to the height of 294? feet. On February 14-th, 1830, he found the temperature at the bottom, after the thermometer had re- mained there 24 hours, to be = 16*25 centigrade; the ex- ternal air being=10*6. At 11 feet beneath the surface of the water the temperature was found = l3*12 cent, after the instrument had remained 17 hours. Common wells, varying in depth from 22 to 28 feet, afforded at the same time a mean temperature of 8*75. On March 22nd, MM. Emy and Gon made further experiments on the same well, which was then sunk to the depth of 125-16 metres, or 369i metrical feet. They found the temperature at the bottom, after the thermo- meter had remained there 25 hours, = 18- 12 cent. Fearful of some inaccuracy in this experiment, they repeated it the next day, when, after the instrument had remained at the bottom for 15 hours, they obtained exactly the same result. M. Fleu- riau de Bellevue estimates the mean temperature of the country at ll-87 cent.* These experiments were conducted with great care, and seem highly illustrative of an increase of heat from the surface to the interior; for the column of water being subject to the usual laws, it would equalize its temperature by the descent of the cooler, and the ascent of the warmer water, if a constant source of comparatively considerable heat did not exist at the bottom. In the waters of neglected mines also there are numerous observations tending to show that the waters do not follow the laws of their greatest specific gravity in such situations, but that the temperatures greatly increase with their depth. Cer- tainly, in many situations, such as in recently flooded mines, the water would be heated by the galleries in which work had been carried on ; but such influence could not continue for a long period, and there are numerous observations which show an increase of temperature in neglected mines. On a subject of this kind, however, great caution is necessary in obtaining the true temperature, and it is very desirable that many of the experiments should be repeated f. * Fleuriau de Bellevue, Journal de Geologic, torn. i. f A cold spring percolating rapidly from the surface to the deep waters of a neglected mine would tend to cool the waters at such depths. 14- Temperature of Springs. Temperature of Springs. The temperature of surface-springs has been supposed to give nearly, if not altogether, the mean temperature of the countries in which they appear. Their value in this respect would depend on whether the waters which supply them be derived from above or beneath, that is, whether they percolate from the surface through porous strata until thrown out by im- pervious beds, or are forced by some means from comparatively greater depths upwards. Many springs, we well know, come within the first class; but many, we are also certain, come within the second, for their temperatures are greatly above what they could have acquired by mere percolation down- wards. At Paris, the oscillations of the temperature of the earth do not quite cease at 28 metres. Professor Kupffer considers that 25 metres from the surface will afford a depth beneath which springs rise with a uniform temperature throughout the year, being sufficiently removed from atmospheric influences. Admitting this, it is clear that if surface- springs be small, and rise slowly, they may have their temperatures somewhat changed during their passage through the 25 metres ; while if they rise quickly, and their waters be copious, they will suffer little change in their traverse through that thickness. The question, however, of whence the waters may have been de- rived, remains the same. Professor Kupffer has constructed the following Table, principally from Von Buch's Treatise on the Temperature of Springs, and from Humboldt's Treatise on Isothermal Lines, with the view of corroborating the observations of Wahlen- berg, that the temperature of springs in high latitudes is greater than that of the air, and of those of Von Hum- boldt and Von Buch, who found that in low latitudes the tem- perature of springs was lower than that of the air; showing "that the temperature of the earth is sometimes very different from the mean temperature of the air, and that its distribution follows different laws*." * Kupffer on the Mean Temperature of the Atmosphere and of the Earth in some Parts of Russia : Edin. New Phil. Journ. vol. viii. : and Poggen- dorfs Annalen, 1829. Temperature of Springs. 15 Places. Lati- tude. Height above sea. Temp, of Earth. Fahr. Temp, of Air. Fahr. Observers. 9 S. 10iN 15 18 23 28 28| 30 39 40 43 46 49 52 53 54 54i- 54| 56 56i 60 Metres. 45 0? 160 300? 350 75 40 500 72-95 78-12 76-10 79-02 74-30 73-85 64-40 72-5 54-27 54-95 55-40 52-02 57-70 50-22 49-32 47-75 48-65 46-62 47-75 47-30 43-70 37-17 34-25 78-12 82-40 77-00 80-60 78-12 77-00 70-92 72-5 53-82 54-27 57-87 49-32 51-57 46-40 40-10 46-16 47-97 43-25 47-75 47-30 42-12 33-35 25-25 Smith. Humboldt. Hamilton. Hunter. Ferrier. Hamilton. Von Buck. Nouet. Mansfield. Warden. Cordier. Saussure. Bouvard. Kir wan. Dalton. Erman. Playfair. Wahlenberg. Cumana St. Jago (Cape Verde Isles) Rock Fort (Jamaica) . . . Havannah Teneriffe Cairo . . Philadelphia Paris Berlin Dublin Ki'iidal Konijjsberg ... Edinburgh Upsal .... 64 66 To this should be added Professor Kupffer's own observa- tions in Russia. Places. Lati- tude. Height. Temp, of Earth. Temp, of Air. 54i Metres. 300 o 39-87 H-7 Kasan 56 30 43-25 37-4 Nishney-ta^ilsk 58 200 37-17 Jl'Stl Werchoturie 59 9QO 36-27 Qn-49 Bo^oslowsk ..... 60 9QO 35-37 20-30 The above tables, if correct, are sufficient -to show that, though the terrestrial temperature, as deduced from springs, decreases from the equator to the poles, it does not decrease according to the mean temperature of the air above it. This seems to point out that there is some modifying cause in action independent of solar influence. Wahlenberg has no- ticed that many deep-rooted plants and trees only flourish because the temperature of the earth exceeds the mean tem- perature of the air; and Professor Kupffer remarks that he has often had occasion to confirm this observation in the northern Urals. 16 Temperature of Springs. At the contact of the atmosphere and earth, we should ex- pect, if they possessed different sources of temperature, that they would mutually act on each other, and that therefore the equal mean temperature of different parts of the earth's sur- face would, to a certain extent, correspond with equal terres- trial temperatures, as deduced from moderate depths. This may perhaps account for Professor Kupffer's conclusion, that " if we draw lines through all the points which have the same terrestrial temperature, these isogeothermal lines resemble the isothermal, as they are parallel to the equator, but diverge from it in several points*." The temperature of the surface, as deduced from spripgs, is undoubtedly liable to many errors, as it rests on the assump- tion that they take the temperature of the earth at moderate depths. Those springs which percolate through porous strata, until thrown out, may take this temperature ; but those which seem to come from beneath cannot be supposed, though cooled in their passage upwards, to do so. The evidence that many springs rise from considerable depths, and possess a temperature independent of solar influ- ence, rests on their great heat, which varies from the boiling point of water downwards to ordinary temperatures. It is impossible to account for this, otherwise than by supposing such heat communicated to the water in parts of the earth far beneath the surface, and removed from atmospheric influence. The source of the heat in thermal waters has occupied the at- tention of Berzelius, Von HoffJ Keferstein, Bischoff, Daubeny, and others. Theformer remarks on those thermal springs which are charged with various salts of soda and carbonic acid, and attributes their origin to the percolation of atmospheric waters to volcanic regions, after which they are forced up to the sur- face, charged with the substances with which they have become combined in those situations. Von Hoff opposes the theory of a mere volcanic point supplying the necessary heat, and considers it much more probable that this is due to those pro- cesses in the interior of our globe which produce volcanos and earthquakes. Keferstein considers that hot vapours and springs are due to volcanic agency, which may be very deeply seated, even below the oldest formations. Bischoff, who de- tails these various opinionsf, does not appear to have adopted any decided one of his own on the subject, but directs atten- tion to the possible increase of temperature in the waters by the internal heat of the earth at great depths, independent of * Kupffer (memoir cited above). f Uberdie Vulchanischen Mineralquellen Deutschland und Frankrcichs : and Edin. New Phil. Journal, 1830. Temperature of Springs. 17 volcanic fires, and observes that if the channels through which the waters flow upwards become once heated, their walls would conduct little heat outwards, for rocks are bad conduc- tors of heat, as is well shown in the case of lava streams, on the outside of which the hand may sometimes be placed, while the melted rock is still flowing inside*. In support of the opinion that thermal waters may have their high temperature caused by a general internal heat, and not by mere volcanic points on the earth's surface, it may be remarked that thermal springs occur in almost all situations, some of which are far removed from any volcanic points on the surface. The immediate connexion of the Geysers and the volcanos of Iceland is so obvious that few will be found to doubt it ; yet when hot springs have been found traversing cracks in strata not volcanic, theories have been invented to explain their ori- gin by chemical combinations at small depths. The salts, however, usually held in solution in these waters do not afford support to this view, and Berzelius has shown it to be unte- nable with respect to the Carlsbad waters. To show the various rocks among which thermal springs occur, we will select a few examples. In ranges of mountains they would appear to be far from uncommon, a circumstance which, supposing the ranges to have been elevated by a force acting from beneath, lends additional probability to a general heat beneath the surface. They have been observed in various places in the range of the Himalaya. Captain Hodgson no- tices them in the course of the Jumna river, so hot that the hand could not be kept in them many moments, and the tem- perature was too great to be measured by the short scaled thermometer usually employed to ascertain atmospheric heat. Again, at Jumnotri, very copious thermal springs rise through crevices in the granite. The heat was estimated at nearly the boiling point; the finger could not be kept in it two seconds. As the height of Jumnotri is estimated at 10,483 feet above the sea, the water would have the appearance of boiling at a lower temperature than in the plains below : moreover, the springs seem to evolve gas, for they rise with great ebullition; still, however, the temperature of the waters would appear to be very considerable f. In the range of the Alps, there are also many thermal springs, as has been already remarked by Bakewell. The thermal waters of Bad-Gastein in the Salzburg country are well known. * Monticelli and Covelli. f Hodgson, Asiatic Researches, vol. xiv. : and Edin.Phil.Journ vol.viii. c 18 Temperature of Springs. The following are Alpine warm springs noticed by Bake- well*: Naters, HautValais; temperature = 86 Fahr. Leuk, Haut Valais, twelve springs; temperature varying from 117 to 126. Bagnes, in the valley of the same name; the baths, village, and one hundred and twenty inhabitants de- stroyed by the fall of part of a mountain in the year 1545 ; temperature unknown. Thermal springs in the valley of Chamonix; temperature unknown. St. Gervais, near the Mont Blanc ; temperature from 94? to 98. Aix les Baines, Savoy; two springs; temperature from 112 to 117. Mon tiers, Savoy; temperature not noticed. Brida, Savoy ; temperature 93 to 97. Saute de Pucelle, Savoy; tempe- rature not noticed. Thermal springs at Cormayeur and St. Didier, on the Italian side of the Pennine Alps ; temperature 94- . Warm springs in the Alps near Grenoble. Many of these thermal waters are of recent discovery, al- though those of Aix were known to the Romans ; therefore there may be many in other parts of the Alps which remain unnoticed. There are also warm springs in the Caucasus, to the N. W. of the fortress of Constantinohor, with a temperature of from 110 to 114? F. ; and there are, no doubt, numerous other thermal waters in great mountain ranges, with which we are as yet unacquainted. In the Pyrenees, we have the two celebrated thermal waters of Barege and Bagneres ; the former having a temperature of 120, at the hottest spring, and the latter of 138, also at the hottest spring. The thermal springs at both these places are numerous, At the latter place there are no less than thirty of them, the temperature of the least hot of which is = 83} F. There are also thermal waters in the valley of Barege, at St. Sauveur, = 98^ ; as also several springs at Cautieres not far from the latter place, of which the temperatures vary from 98 to 131. At Caberu, three leagues from Bagneres, there is a spring = 80. It would be tedious to give a long list of thermal springs ; they occur in all parts of the world, as well remote from, as in the vicinity of, active volcanos. A great burst of hot springs takes place near the base of the south-eastern slope of the Ozark mountains, North America, and about six miles north from the Washita, from which they take their name. They are about seventy in number, and occur in a ravine between two slate hills. James states the temperature of these waters at 160 Fahr. Major Long gives that of several of them, as respectively, 122, 104, 106, 126, 94, 92, 128, 132, * On the Thermal Waters of the Alps, Phil. Mag. and Annals, 1828. Temperature of Springs. 19 151, 148, 132, 124, 119, 108, 122, 126, 128, ISO , 136, 14-0. He also states that, "not only confervas and other vegetables grow in and about the hottest springs, but great numbers of little insects are seen constantly sporting about the bottom and sides*." Another example of the existence of animals and vegetables in thermal springs is to be found at Gastein, where the Ulva thermally and a fresh-water shell, the Limneus pereger Drap., are found in waters at a temperature of 117 Fahr. A very copious discharge of hot water takes place in an alluvial plain in a granitic district at Yom-Mack, about twenty miles from Macao, China. Three large springs have respec- tively the temperatures of 132, 150, and 186 Fahr. That with the temperature of 150 is described as in a state of active ebullition, about thirty feet in diameter, and discharging at least fifteen gallons in a minutef. The temperature of the waters of Carlsbad is also consi- derable, being, according to Berzelius, 165 Fahr. Those of Aix-la-Chapelle are = 143; and at Borset, near Aix-la- Chapelle, there are two springs, of which the temperatures are respectively 158 and 127 Fahr. At Balarue, depart- ment of Herault, there is one = 128 Fahr. The waters of Baden Baden are = 153 F.; of Bude, near Chemnitz, = 158 F.; of Ems, Nassau, = 122F. ; of Ofen, Hungary (in granite), = 147 F.; of Plombieres, Vosges, = 140 F. ; of Burtscheid, Prussian Provinces, = 171 F. Numerous ther- mal springs, varying from 1 18 to 122 F., rise amid a calca- reous district near the Euphrates, about five wersts from Fort Diadine, in Armenia. The thermal springs of our own coun- try are not very remarkable for their elevated temperature ; for with the exception of those of Bath J, which are at 116 Fahr., the others can only be considered as tepid, the waters at Buxton being at 82, those of the Hotwells, Bristol, at 74, and those of Matlock 68 . In the volcanic districts of Italy, thermal springs, as might be expected, are numerous. The waters of the Bagni di Lucca are however sufficiently removed from a volcano to be here noticed : they rise on the sides of a hill, composed of a sandstone, the macigno of the Italians. The district is one of sandstone and limestone, and the hottest spring has a tempe- rature of 131 F. * James, Expedition to the Rocky Mountains. f Livingstone, Edin. Phil. Journal, vol. vi. t These rise through lias, traversing probably red sandstone, carboniferous limestone, &c. The thermal springs of the Hotwells, Matlock, and Buxton, appear among carboniferous limestone. C % 20 Temperature of Springs. It may not be altogether out of place to notice the thermal waters of Bath, St. Thomas in the East, Jamaica, to show how widely distributed these heated springs are. They rise at the base of the Blue Mountains, in a valley composed of trap, limestone, and slate. I observed their temperature to be = 127 F.* The hot and cold springs of La Trinchera, three leagues from Valencia (America), may be cited to show, how dif- ferently derived waters may be, which make their appearance close to each other. According to Humboldt, there are two springs, only 4-0 feet asunder, the one cold, the other hot, the thermal waters having the great temperature of 90*3 centi- grade (194'5 Fahr.). At Cannea, in Ceylon, a thermal spring is stated to exist which does not preserve a constant temperature, but varies from 38 to 41 cent. (100'4 to 105'8 F.). Hot springs are common to the volcanic districts of dif- ferent parts of the world, as also amid extinct volcanos, such as those of Central France; to enumerate them would be useless ; but those of Iceland are so remarkable, that a short notice may not be unacceptable to the reader, particularly as they are the most extraordinary thermal springs with which we are acquainted. Hot springs are numerous in Iceland, but those named the Geysers are the most singular. They are alternately in a state of rest and of violent activity, discharging, at intervals, immense quantities of hot water and steam. Sir G. Mackenzie states that an eruption of the Great Geyser, which he witnessed, commenced with a sound resem- bling the distant discharge of a piece of ordnance. " The sound was repeated irregularly and rapidly ; and I had just," observes this author, "given the alarm to my companionsf, who were at a little distance, when the water, after heaving several times, suddenly rose in a large column, accompanied by clouds of steam, from the middle of the basin, to the height of ten or twelve feet. The column seemed as if it burst, and sinking down it produced a wave, which caused the water to overflow the basin in considerable quantity. After the first propulsion, the water was thrown up again to the height of about fifteen feet. There was now a succession of jets to the number of eighteen, none of which appeared to me to exceed fifty feet in height; they lasted about five minutes. Though the wind blew strongly, yet the clouds of vapour * Although no active volcanos exist in Jamaica, there are the remains of an extinct one on the north side of the island ; and earthquakes are, as is well known, sufficiently common. f Dr. Bright and Dr. Holland. Temperature of Lakes. 21 were so dense, that after the first two jets I could only see the highest part of the spray, and some of it that was occa- sionally thrown out sideways. After the last jet, which was the most furious, the water suddenly left the basin, and sunk into the pipe in the centre*." The water sunk in the pipe to the depth of ten feet, but afterwards rose gradually; when sufficiently high, its temperature was observed, and found = 209 F. A subsequent eruption of the same Geyser is thus described by the same author. After an alarm given of its approaching activity, "in an instant," he says, "we were within sight of the Geyser; the discharges continuing, being more frequent and louder than before, and resembling the distant firing of artillery from a ship at sea It raged furiously, and threw up a succession of magnificent jets, the highest of which was at least ninety feetf." One of the other fountains, which was formerly an insig- nificant spring, and now known as the New Geyser, alter- nates in like manner. The eruption commences, as at the Great Geyser, by short jets, which increase in size. When a considerable mass of water is thrown out, the steam rushes forth furiously, accompanied by a loud thundering noise, car- rying the water, when Sir G. Mackenzie observed it, to at least seventy feet. He describes it as continuing in this mag- nificent play for more than half an hour. " When stones are dropped into this pipe, while the steam is rushing out, they are immediately thrown up, and are commonly broken into fragments, some of which are projected to an astonishing height^" There are other alternating hot springs in Iceland, which are, however, of greatly inferior magnitude to the Geysers. The springs of Reikum, with a temperature of 212 Fahr., rise and fall, and dash up spray to the height of twenty or thirty feet. In the valley of Reikholt, there is a singular alternation of two boiling jets, one throwing the water up twelve feet, the other five. Temperature of the Sea and of Lakes. This temperature will probably be in part derived from that of the atmosphere, and partly from the earth ; but water being, under certain circumstances, able to communicate heat * Mackenzie's Travels in Iceland. f Ibid. J Travels in Iceland : where views of these fountains in full operation will be found. Waters actually at the boiling point seem exceedingly rare. The thermal waters of Urijino, in Japan, are stated to have a temperature of 212 Fahr., but it docs not appear among what rocks they occur. 22 Temperature of Lakes. with great rapidity, the temperature will be more speedily equalized in it, than in the solid earth beneath. Water, more- over, at a given temperature possesses a greater specific gra- vity than when that temperature is either increased or dimi- nished, and will consequently, at that given temperature, sink to the lowest depths. Even if it should be heated there, on the presumption of an internal heat in the earth, the water will still obey the same laws, the newly heated water will ascend, and be replaced by that which is cooler and of greater specific gravity. For, in order that the water should sink to these depths in the first instance, it must be of such a tempe- rature, or specific gravity, as shall enable it to do so, and any change in that temperature, if it be that of the maximum density of water, will cause it to rise. According to Dr. Hope, the maximum density of fresh water is at a temperature between 39|. and 40 Fahr.*, and this determination has been confirmed by Professor Moll. According to the experiments of Professor Hallstrbm, the maximum density of water occurs at the temperature of 4*108 centigrade (39'394 Fahr.). It has been considered that the temperature of the maximum density of sea-water is not far removed from that of fresh water. On this head we have no very satisfactory experi- ments, but it may be supposed that the saline contents of sea- water would have considerable influence on its relative gra- vity at different temperatures. In the years 1819 and 1820 I made numerous experiments, with great care, on the temperature of the Swiss Lakes at various depths, which are often considerable. The results of more than one hundred observations on the Lake of Geneva, in September and October 1819, were, that between the sur- face and a depth of 40 fathoms the temperature varied consi- derably. From 67 to 64 Fahr. was a common heat from one to five fathoms, and there was a general diminution of temperature downwards to the depth of 40 fathoms, whatever the surface-heat might be ; in other words, there was a gene- ral increase of specific gravity downwards. From 40 fathoms to 90 fathoms the temperature was always 44, with one ex- ception near Ouchy, where 45 were observed at a depth of 40 fathoms. From 90 fathoms to the greatest depths, which amounted to 164 fathoms, between Evian and Ouchy, the temperature was invariably = 43*5 Fahr. It will be ob- served, that in these experiments, made with a register ther- mometer constructed for the purpose, the water arranged itself according to the temperatures that would be expected, * Trans. Ro^. Soc. Edinburgh. Temperature oj the Sea. 23 on the supposition of the maximum density of water being between 39 and 40*. After the severe winter of 1819, I made some further ex- periments, and found that the temperature of the lake still followed the same law. In May, 1820, 1 tried the temperature of the lakes of Thun and Zug, and obtained the following results-]-. Lake of Thun. Surface 60 At 15 fathoms 42 At 50 fathoms 41-5 At 105 fathoms 41-5 Lake of Zug. Surface 58 At 15 fathoms 42 At 25 fathoms 41 At 38 fathoms... .... 41 In these experiments also, the results are in accordance with the maximum density of water being between 39 and 40, as was also the case in some which I made in the Lake of Neufchatel, during very cold weather, so cold indeed, that the water froze on the oars of the boat, when the temperature increased towards the supposed maximum density of water. We now turn to the experiments that have been made by different navigators on the temperature of the sea at various depths. The following observations by Scoresby show an in- crease of temperature from the surface downwards in cold latitudes. Deg. of Situation. Depth. Temp. {Surface 29-0 13 fathoms ...31-0 37 fathoms ....33-8 57 fathoms ....34-5 100 fathoms... 36-0 400 fathoms... 36-0 Deg. of Situation. Depth. Temp, f Surface 28'8 T nt 7fi Ifi' N J 50 fathoms... 3 1-8 Lat. 76 ^-< 123 f at homs. .33-8 (_230 fathoms. .33-3 T A Tn if AT f Surface 29 Lat. 79" 4' N. { 730 fathoms> . 37 Again, in lat. 78 2' N. and long. 10' W. the same scien- tific navigator obtained 38 at 761 fathoms, the surface-water being 32. In one situation, indeed, in lat. 76 34' N. the same observer obtained a temperature of 34 at 60 fathoms, and 34-7 at 100 fathoms, after having had 35 at 40 fathoms. The experiments of Capt. Ross are, indeed, opposed to this view, for they give a decrease down to 25 at 660 fathoms, from 30 at 100 fathoms, 29 at 200 fathoms, and 28 at 400 fathoms ; in lat. 60 44' N. and long. 59 20' W. According also to Dr. Marcet, the maximum density of sea-water is not at 40 Fahr. He states that this water increases in weight to the freezing point, until actually congealed. In four experi- ments Dr. Marcet cooled sea-water down to between 18 and * A detailed account of these experiments, with a chart of soundings in the lake, were inserted in the Bibliotheque Universelle for 1819; from whence it was copied, in part, into the Edin. Phil. Journal, vol. ii. t See also Bibliotheque Universelle for 1820. 24 Temperature of the Sea. 19 Fahr., and found that it decreased in bulk till it reached 22, after which it expanded a little, and continued to do so till the fluid was reduced to between 19 and 18; when it suddenly expanded, and became ice with a temperature of 28. According to M. Erman, salt water of the specific gravity of 1*027 diminishes in bulk down to 25 F., and does not reach its maximum density before congelation. It should always be recollected that a saturated solution of common salt does not become solid, or converted into ice at a less temperature than 4 Fahr. ; and therefore if the sea should be, as is sometimes supposed, more saline at great depths, and as it appears to be in the Mediterranean from the experiments of Dr. Wol- laston, ice could not be formed there at the same temperature as it could nearer the surface. Kotzebue, in lat. 36 9' N. and long. 148 9' W. found the surface-water = 71'9, the air being 73; at 25 fathoms the water was at 57'l ; at 100 fathoms, 52 8; and at 30 fathoms, 44 : showing a decrease of temperature towards 39 or 40. In lat. 23 & N. and 181 56' W. Krusenstern obtained, at the surface, 78; at 25 fathoms, 75; at 50 fathoms, 70'5; and at 125 fathoms, 61'5. In latitudes south of the tropics, Kotzebue observed a tem- perature of 49'5 at 35 fathoms, the surface being at 67, the air at 68, in lat. 30 39' S. The same navigator found the temperature at 196 fathoms to be = 38'S, in lat. 44 17' S. and long. 57 31' W. ; the surface-water being 54'9, and the air at 57'6. The following are among the temperatures obtained by Captain Beechey* at various depths and situations. In lat. 47 18' S., and long. 53 30' W., the surface-water being at 49'S, he found 44'7 at 270 fathoms, 39'2 at 603 fathoms, 40-1 at 733 fathoms, and 59'4 at 854 fathoms. In lat. 55 58' S-, and long. 72 10' W., the surface-water being at 43'5, he obtained 42-5 at 100 fathoms, 42-5 at 230 fathoms, 42-5 at 330 fathoms, and 41'6 at 430 fathoms. In the South Pacific, he found in lat. 28 40' S , and long. 96 W., 71 at 100 fathoms, 53 at 200 fathoms, 49 at 300 fathoms, and 45 at 400 fathoms, the surface-water being at 74. Among the observations made by the same navigator in the North Pacific are the following: in lat. 61 10' N., and long. 183 28' W., in July 1827, at" 5 fathoms 41'5', at 10 fathoms 38, at 20 fathoms 29'5, at 20 fathoms 30*5, (this is apparently a se- cond observation at the same depth), at 30 fathoms 30*5, at 52 fathoms 32'5, at 100 fathoms 32'5, and at 200 fathoms 32'5, the surface-water being at 43- 5 and the air at 45. * Becclicy, Voyage to the Pacific, &c. Temperature of the Sea. 25 Observations have been made at considerable depths in the tropics. Capt. Sabine found in lat. 20 30' N., and long. 83 30' W., a temperature of 45*5 at 1000 fathoms, the surface- water being at 83. Capt. Wauchope obtained in lat. 10 N. and long. 25 W., a temperature of 5J at 966 fathoms, the surface-water being at 80: and the same observer also found, in lat. 3 20' S. and 7 39' E., a temperature of 42 at 1300 fathoms, the surface-water being at 73. M. Lenz, in lat. 21 14' N., and long. 196 I' W., found a temperature of 61-4 at 150 fathoms; of 37'7 at 440 fathoms ; of 37'2 at 709 fathoms; and of 36'5 at 976 fathoms ; the surface-water being at 79 0> 5 *. Other observations within the tropics, at inferior depths, show the same decrease of temperature downwards. Thus Kotze- bue in lat. 9 21' N. obtained 77 at 250 fathoms, the surface- water being at 83 and the air at 84 ; and under the equator, in long. 177 & W., 55 at a depth of 300 fathoms, the surface- water being at 82*5 and the air at 83. M. Berard found, at a depth of 1200 fathoms, (without reaching bottom,) between the Balearic Isles and the coast of Algiers, a temperature of 53'4, the surface-water being at 69' 8, and the air at 75*2 F. From other observations in the western part of the Mediterranean, at the respective depths of 600 and 750 fathoms, and another not stated, it was found that the water was still at 55'4, though the temperature of the surface-water varied materially. M. D'Urville remarks that these experiments accord with some made by himself, also in the western Mediterranean, at 300, 200, 250, 600, and 300 fathoms, when he obtained the respective temperatures of 54-5, 54-l, 57'3, 54'6, and 54-8 F. He hence infers that the waters of the western Mediterranean, beneath a depth of 200 fathoms, rest at a temperature of about 55 F.f It will be observed, from what has been stated above, that the waters of lakes arrange themselves according to certain temperatures, which show that experiments made in the cabi- net, and which fix the maximum density of fresh water at a temperature of between 39 and 40 Fahr., are correct. With respect to the waters of the ocean, sea water evidently arranges itself, in the warmer climates of the globe, according to its density, supposing its maximum density to be at a temperature approaching that of its congelation ; but in the colder regions there would appear to be some disturbing forces in action, which in many situations cause the temperature to increase with the depth, thus interfering with the densities, if it be true, * The same observer obtained, in lat. 32 6' N., and in long. 136 48' W., a temperature of 56 at 96 fathoms; of 437, at 228 ; of 38-7, at 480; and of 35;9, at 632 ; the surface-water being at 70-6 F. t D'Urville, Bui. de la Soc. de Geographic, t. xvii. p. 82. 26 Temperature of the Sea* as it seems to be, that the maximum density of sea-water is as above stated. The probability of a central heat would appear to rest, first, on the experiments made in mines, which, notwithstanding their liability to error from various sources, still seem to show, particularly those made in the rock itself, an increase of tem- perature from the surface downwards ; secondly, on thermal springs, which are not only abundant among active and ex- tinct volcanos, but also among all varieties of rocks, in various parts of the world ; thirdly, on the presence of volcanos them- selves, which are distributed over the globe, and present such a general resemblance to each other, that they may be con- sidered as produced by a common cause, and that cause pro- bably deep-seated ; and fourthly, on the terrestrial temperature at comparatively small depths, which does not coincide with the mean temperature of the air above it. The temperature at the bottom of seas and lakes is not at variance with this probability, as the waters endeavour to ar- range themselves according to their greatest specific gravity. Indeed, so far from the temperature found at various depths in the ocean being at variance with the hypothesis of a central heat, they rather accord with it ; more particularly when we regard the increase of temperature with the depth in many places in high latitudes. We might assume that such increase of temperature, interfering with the densities, was due to heal, in the bottom beneath ; which, though sufficient to produce a visible effect upon waters so closely approaching their maxi- mum density as those in high latitudes, was yet insufficient to be apparent beneath warm climates, though it would still cause an increase in the temperature of the waters in low latitudes. Neither is the probability of internal heat at variance with the figure of the earth or observed geological phaenomena. The figure of our planet being that which a fluid body would assume if revolving in space, it is as probable that this fluidity should be igneous as aqueous. Geological phaenomena attest the eruptions of igneous matter from the interior at all periods ; as also elevations of mountains and great dislocations of the earth's surface, caused by forces acting from beneath ; and, finally, a great decrease of surface temperature. Should we be inclined to build a theory on the probability of a central heat, we may suppose, as has often been done, that our world is a mass of igneous matter in the act of cooling. Baron Fourier considered it as proved, from the form of our spheroid, the disposition of the internal strata (shown by experiments with the pendulum) to increase in density with their depth, and from other considerations, that a very in- tense heat formerly penetrated all parts of our globe. He Temperature of the Atmosphere. 27 concluded that this temperature was dissipated into the surrounding planetary spaces, the temperature of which he considered, from the laws of radiant heat, to be =50 cent. ( 58 Fahr.). He moreover inferred that the earth had nearly reached its limit of cooling. The original heat contained in a spheroidal mass equal in magnitude to our globe, would diminish more rapidly at the surface than at great depths, where the elevated temperature would remain for a great length of time. He further inferred from these circumstances, and from the temperature of mines and springs, that there is an internal source of heat, raising the temperature of the surface above that which the action of the sun could alone give it*. Temperature of the Atmosphere. The gaseous compound termed the Atmosphere, which sur- rounds the earth, has been calculated, from its powers of re- fraction, to extend upwards about forty-five miles. Dr. Wol- laston considered, from the laws of the expansions of gases, that it might reach to at least forty miles, with its properties uninjured by rarefaction. On this head Dr. Turner observes, " that the tension or elasticity of gaseous matter is lessened by two causes, diminution of pressure, and reduction of tempe- rature." And he further remarks, that the former alone has been taken into account by Dr. Wollaston, while it appears to him that the extreme cold at great heights would also be sufficient to limit the extent of the atmosphere t. Though no part of the solid earth is so elevated above the general surface as to be exposed to a very considerable depres- sion of temperature, yet numerous mountains are of sufficient height to be covered, at their summits, with what has been termed perpetual snow, the prolific parent of innumerable rivers, without which many regions would be uninhabitable. * M. Svanberg, calculating what might possibly be the temperature of the planetary spaces, proceeds upon another principle than that of the radiation of heat. He supposes that the planetary spaces never undergo any change of temperature, but that the capacity for elevation of temperature, above that which constantly reigns in the ethereal regions, exists only within the limits of the planetary atmosphere. He obtains for the result of his calcu- lations a temperature = 49-85 cent. Observing this near approach towards Baron Fourier's supposed temperature, he had the curiosity to cal- culate the temperature according to Lambert's statements, respecting the absorption which takes place in a ray of light passing from the zenith through the whole atmosphere, and found that he obtained 50-35 for the result. A curious coincidence between the results of the three modes of calculation. Berzelius. Annual Progress of Chemical and Physical Science. Edin. Journ. of Science, vol. iii. New Series. t Turner, Elements of Chemistry, p. 221. 28 Temperature of the Atmosphere. The line of perpetual snow differs generally according to the latitude, and is liable to very great variations from local causes. Some of these variations will be observed in the following Table, by Humboldt*, of the snow-line in certain mountain chains. ,, . . Latitude. Height in Mountains. Deg Deg English feet. Cordillera of Quito to 1 S 15,730 Cordillera of Bolivia 16 to 17f S 17,070 Cordillera of Mexico 19tol9iN 15,020 Himalaya : Northern Flank 30|to31 N 16,620 Southern Flank 12,470 Pyrenees 42to43 N 8,950 Caucasus 42to43 N 10,870 Alps 45Jto46N 8,760 Carpathians 49 to49|N 8,500 Altai 49 toSlN 6,400 Norway : Interior 61 to 62 N 5,400 Interior 67 to 67 N 3,800 Interior 70 to 70^ N 3,500 Coast 71to7lN 2,340 Among other variations from the theoretical line of perpetual snow which have been produced by a combination of physical circumstances, it will be observed that there is a difference be- tween the northern and southern flanks of the Himalaya of more than 4000 feet in favour of the former, by which means a large surface is inhabited which would otherwise be unfit to support animal and vegetable life. It has been supposed that the temperature of the atmosphere diminished equally upwards in different latitudes; but the fol- lowing Table, also by Humboldt, will show that this is not the case, and that, on the contrary, the decrease is much more rapid in the temperate than in the equatorial zone. Height in Equatorial zone; from to 10. Temperate zone ; from 45 3 to 47. English feet. Mean Differ. Mean Differ- Temp. ence. Temp. ence. 81-5 53-6 3,195 71-2 10-3 41-0 12-6 6,392 65-1 6-1 31-6 9-4 9,587 57-7 7-4 23-4 8-2 12,792 44-6 13-1 15,965 34-7 9-9 The curve, representing the line of perpetual snow, will not * Fragmens Asiatiques, p. 549. Valleys. 29 be equal in the northern and southern hemispheres, as the latter is found to be colder than the former. From the variable height at which perpetual snow com- mences, it follows, all other circumstances being the same, that the extent of dry land capable of sustaining animal and vege- table life, will decrease from the equator to the poles, and, con- sequently, that there is a greater probability of an abundance of terrestrial remains being entombed in any deposit now taking place in the tropics, than in similar deposits in high latitudes*. Valleys. A classification of valleys cannot well be accomplished with out some violence, as the various depressions of land, to which the term valley has been much too generally applied, pass into each other in such a manner as to produce compounds of no easy arrangement. No great value is therefore attached to the following sketch. Mountain Valleys. These are both longitudinal and trans- verse ; ranging either in the direction of the mountain chain, or across that direction. Their sides are generally rugged, crowned by lofty pinnacles and broken masses, and are, for the most part, steep. Atmospheric agents, far from producing a milder outline, generally add to their broken appearance. The melting of ice and snow, and the drain of rain-waters furrow their sides, bringing down detritus to the rivers, which, when levels are favourable, deposit it in situations well suited to ve- getation ; so that in mountain regions patches of verdure occur amid the wildest scenes, presenting a singular contrast to the broken forms of the surrounding heights. When levels are unfavourable, or the fallen blocks large, the masses accumulate in the water-courses, and produce innumerable cascades, add- ing to the desolate character of such regions. Lowland Valleys. These differ from the preceding in their rounded form, which would render a section of them an un- dulating line, the undulations varying in the proximity of the higher parts and in depth, so that the more elevated portions may even be many miles asunder, and the depth inconsiderable. From the comparatively gentle slopes of these valleys, atmo- spheric agents, though still able to decompose the rocks be- * If we consider that animal and vegetable life decreases in proportion as the atmosphere becomes colder and less dense, and that marine life is less abundant as the pressure of the sea increases, and the necessary light dimi- nishes, we obtain, if I may so express myself, two series of zones, one rising above the ocean-level, the other descending beneath it, the terms of the two series, all other things remaining the same, affording the greater amount of animal and vegetable life, as they respectively approach the ocean-level. 30 Valleys. neath, do not transport the detritus to any considerable distance, except in climates and situations where heavy tor- rents of rain descend on land unfavourable to vegetation ; yet, even in this case, the general rounded outlines of the hills are not very considerably impaired, though deep furrows are made in their sides. Ravines and Gorges. These are bounded by more or less perpendicular walls of rock, and are common both among mountain and lowland valleys, but more particularly the former. They frequently communicate between more open spaces, and their edges may often be approached without any suspicion that they exist, the country appearing as one conti- nuous slope or level. Broad Flat-bottomed Valleys. Level plains of greater or less extent, bounded by hills or mountains on either side; such as the great valley of the Rhine below Basle, bounded on one side by the Swartzwald, and on the other by the Vosges. Such a diversity of form would seem to suggest a diversity of origin. The mountain valleys for the most part resemble large cracks, produced when the strata were suddenly elevated and contorted, while the lowland valleys appear as if a large body of water had passed over them, rounding the inequalities, and acting on masses of strata in proportion to their power of resistance. The gorges or ravines would seem due to the cutting power of running waters, or to rifts in the rocks pro- duced by violent convulsions. The flat-bottomed valleys have the character of drained lakes, or situations where the rivers or floods, not having any great velocity, deposit considerable quantities of sediment over a flat surface. As we may suppose hill and dale, mountain and valley, to have existed from the earliest geological periods, and that strata were by no means deposited in one even plane surface, we have now a very complicated system of depressions ; though as a general fact it may be stated, that the superior stratified rocks have filled up and covered over numerous inequalities of the inferior stratified rocks, as is the case in Normandy, where the oolite group covers over the uneven surface of slates, limestones and grauwacke, the latter rocks here and there protruding through the stratification of the former, and becoming visible where rivers cut the superincumbent beds. If we can imagine a violent disruption of strata, contorting or throwing them on their edges, large rents and fractures would be the natural consequences, producing longitudinal and transverse fissures ; but these would merely gape, and their origin would appear clear, if not modified' by some subsequent action. If we suppose, with the advocates for no greater ef- fects than we daily witness, that mountains have been raised Valleys. 31 gradually by a multitude of earthquakes acting always in the same line, we shall have great difficulty in explaining the po- sition of strata in high ranges, more particularly those (such are by no means uncommon in the calcareous Alps,) where whole mountains are contorted, and even appear as if thrown over, as at the Righi. Whereas, if we suppose that the eleva- tions have been more violent, these difficulties would appear to vanish, and the upturned, overthrown, and contorted strata, the longitudinal and transverse cracks or valleys, would be more in harmony with each other. If we should suppose a violent disruption of strata to take place beneath the waters of an ocean, these waters would be greatly agitated and react upon the land) rushing into the cracks ; sweeping away pinnacles ; driving blocks and loosely aggregated strata before them ; rounding off angles ; and ac- cumulating detritus at the bottom of hollows. Should such a sudden elevation be effected, partly in the ocean, and partly out of it, the reaction of the sea would only reach the lower portion of the upraised strata, and these only would present rounded forms. Should the strata be elevated only in the atmosphere, the modification of the original cracks would be effected by atmospheric agency alone. Although lowland valleys generally present rounded forms, the strata composing such districts are often far from undis- turbed ; on the contrary, they are often upturned, contorted, and fractured, the lines of valleys being frequently the same with those of the faults or fractures. Often, however, no appearances of fracture are visible in the hills, though these are traversed by faults in various directions. Of this fact the neighbourhood of Weymouth, in our own country, may be cited as affording good examples. Valleys of Elevation are those which seem to have originated in a fracture of the strata, and a movement of the fractured part upwards, so that the strata dip from the valley on either side. Probably a very large proportion of mountain valleys might be arranged under this head ; but at present geologists seem to have confined the application of the term to those which are bounded by hills of moderate height. Prof. Buckland (in 1825) noticed valleys of this kind at New Kingsclere, Bower Chalk, near Shaftesbury, and Pox- well near Weymouth. The annexed diagram is a section of that of Kingsclere. Fig. 1. 32 Valleys. V. Valley of Kingsclere : a a, chalk with flints : b b, chalk without flints : c c, green sand. It will at once be observed, that the strata on either side were once continuous, and that they have been upheaved, producing a fracture, which, by subsequent denudation, has been formed into the valley we now see. Subsequently to the observations of Prof. Buckland, similar valleys in Germany have occupied the attention of M. Hoffman, who endeavours also to show that they are connected with springs impregnated with carbonic acid gas. In support of this opinion he cites the valley of Pyrmont, of which he gives the following section, which will be seen closely to correspond, in its general characters, with that of Kingsclere. M, the Muhlberg, 1107 feet: B, the Bomberg 1136 feet: P, Pyrmont, the bottom of the valley being 250 feet : a a, keuper (red or variegated marl) : b b, muschelkalk : c c, gres bigarre, broken into fragments at d, through which the acidu- lous waters are forced out. As at Kingsclere, the strata have not been forced up to equal heights on either side. The gres bigarre rises to 850 feet on the Bomberg or north side ; while on the Muhlberg or south side it only reaches 540 feet, with an inferior dip. The theoretical opinions connected with these appearances will be noticed in the sequel ; at present it is only necessary to point out the existence of such valleys. M. Hoffman also notices similar appearances, with acidu- lous springs, in the valley of Dribourg, on the left bank of the Weser, and several other combinations of the like kind*. Valleys of Denudation. Although the valleys of elevation above noticed, may also be termed valleys of denudation, this name seems given, in preference, to those valleys where the strata are not far removed from an horizontal position on either side, and of which also the former continuity cannot be doubted. Of these, the following section of the valley of Charmouth will afford an example. Journal de Geologic, torn. i. Valleys. 33 , summits of the hills composed of angular flint and chert gravel, the remains of former superincumbent chalk and green sand which have been partially dissolved in place : b b, green sand, with an uneven upper surface resulting from the causes that have produced the gravel : c c, lias, in which the lower part of the valley has been excavated : d, small river Char flowing at the bottom. If proportions had been strictly at- tended to, the stream would have been invisible. On the sides of the hills, from a to d, much chert and flint gravel is distributed over the rocks b and c, and it may be questionable how much of it has, during a great lapse of time, descended from the heights, as has occurred on the slopes of similarly rounded hills, in the South Hams in Devon, and how much may have been left at the original formation of the valley. The advocates, indeed, of such excavations by no greater powers than those we daily witness, would consider this valley formed by the insignificant streamlet which now flows through it, aided by the rain-waters. This valley is, however, the sole channel of drainage for a district many miles in extent, in which the actual river, with every assistance from floods, has only effected a cut, varying from four to fifteen feet deep, bounded by perpendicular walls ; these walls not composed, for the most part, of lias, but of gravel and drifted materials, such as are strewed over the valley of all heights, from the bed of the river to the tops of the hills. Such valleys are common in various parts of the world, and not unfrequently are without running waters in them, so that these could not have caused them. Even in Jamaica, where heavy tropical rains are sufficiently common, there are valleys, in which the waters are swallowed up by subterraneous cavities, or sink- holes, and no continuous streams are formed. In England we have examples of dry valleys, in our chalk districts, in the oolite of Yorkshire, and among the slates of the South Hams, Devon * ; a covering of vegetation or turf most commonly protecting the surface from removal, even during heavy rains. On the west coast of Peru, where rain never falls, there are also some remarkable examples of dry valleys, which, judging from sketches, resemble many a lowland valley with rounded sides in Europe. The form of these valleys is also opposed to their production by running waters, for they are rounded and not bounded by perpendicular walls. Sometimes the upper part of a hill being composed of harder materials than the lower portion, it advances with a somewhat bold escarpment. * These latter are due to the highly inclined position of the strata, between the fissures of which the rain-water, after having been received in a porous superficial gravel, percolates. D 34 Changes on the Surface of the Globe. The general form of these valleys would seem to suggest a mode of formation somewhat different from that of the moun- tain valleys ; one which has permitted a very general removal of bold projecting points. There is scarcely a district of any considerable extent, composed of these valleys, which does not contain fissures, or faults, even when the strata are, as a mass, not far removed from an horizontal position. In other situations the strata are upheaved, contorted, and intruded by trap rocks, yet the general forms of the valleys are not consi- derably altered ; the same rounded forms still prevail. From this general prevalence of the same character, it may not be unreasonable to conclude that some similar cause has produced them. They appear as if scooped out by large moving masses of waters ; the least resisting parts first giving way. We might imagine this to have been effected by great disturbances beneath an ocean ; such as would be caused by the elevation of long ranges of mountains near them, or a disruption of the strata of which they were actually composed ; in fact, by submarine earthquakes of much greater force than those which we now witness. Earthquakes of the present day frequently produce violent waves, which discharged on a coast ravage all within their reach. The sudden elevation of mountains to the height of several thousand feet would be accompanied by violent disturbance of the land, causing the waters of neigh- bouring seas to rise considerably, and overwhelm land within their reach ; and these discharges of masses of water would have great scooping powers, more particularly if they acted on fractured strata, or small previously existing depressions. These valleys may also have been formed beneath agitated waters, in which currents moving with great velocity were produced ; the land having been afterwards protruded above the level of the sea. These observations on the origin of low- land valleys should be considered as mere speculations, which future investigations may show to be either probable or im- probable. One argument in their favour, in preference to the supposition that they have been scooped out by rivers, is, that in many instances the rivers quit the valleys, which would appear continuations of their natural channels, and pass through gorges or ravines cut in lands of considerable eleva- tion on one side ; the barrier to their natural passage onwards being merely a gentle rise of a few feet at the bottom of the valley, not easily observed. Changes on the Surface of the Globe. The present condition of our planet's surface is far from stable; on the contrary, if time enough could be allowed, a Classification of Rocks. 35 great change in the relations of land and water would be effected. This process is undoubtedly slow, but it is never- theless certain, and so apparent, that many persons have been inclined to refer all geological phaenomena to a continuance of those effects of existing causes which we daily witness. As far as we can judge from known facts, this opinion seems to have been somewhat hastily adopted, and not altogether in accordance with all those geological phenomena with which we are at present acquainted. As the student may, however, be supposed not to possess a knowledge of these phaenomena, the consideration of their relative value must be waived until he becomes more familiar with the subject. After geologists had ceased to amuse themselves by fabri- cating theories, without being at the trouble of examining the surface structure of that world which they made, modified, and broke to pieces at their own good-will and pleasure, and when it was thought that a knowledge of facts was somewhat necessary to a knowledge of the subject, it was soon observed that considerable changes had taken place on the world's surface. Facts being still few, hypotheses were easily formed, and were more or less plausible according to the knowledge of the day. These will bs found in the various works which treat of the history of geology, and therefore need not be produced here ; it will be sufficient to observe that the two prevailing theories of the present time are, 1st, That which attributes all geological phenomena to such effects of existing causes as we now witness ; and, 2ndly, That which considers them referable to series of catastrophes or sudden revolutions. The difference in the two theories is in reality not very great ; the question being merely one of intensity of forces, so that, probably, by uniting the two, we should approximate nearer to the truth. Classification of Rocks. The term Rock is applied by geologists, not only to the hard substances to which this name is commonly given, but also to those various sands, gravels, shales, marls, or clays, which form beds, strata, or masses*. Rocks were first divided into two classes, Primitive and Secondary, it being considered that they originated under different circumstances; the latter only containing organic remains. To this Werner added a third class, which he named Transition, considering that it exhibited a passage from the primary into the secondary. Subsequently, from obser- * For the terms used in geology, see Appendix A. D 2 36 Classification of Rocks. vations made by MM. Cuvier and Brongniart on the country round Paris, a fourth class was instituted, and called Tertiary, because the strata composing it occurred above the chalk, a rock considered as the highest of the secondary class. These divisions or classes are more or less in use at the present time, though it seems somewhat generally admitted that they are insufficient, and not in accordance with the present state of science. Numerous modifications and divisions have been proposed, which, though preferable to the preceding, have not been adopted, the force of habit, possibly, having prevailed. To propose in the present state of geological science any classification of rocks which should pretend to more than temporary utility, would be to assume a more intimate ac- quaintance with the earth's crust than we possess. Our know- ledge of this structure is far from extensive, and principally confined to certain portions of Europe. Still, however, a mass of information has gradually been collected, particularly as respects this quarter of the world, tending to certain ge- neral and important conclusions ; among which the principal are, that rocks may be divided into two great classes, the stratified and the unstratified ; that of the former some con- tain organic remains, and others do not ; and that the non- fossiliferous stratified rocks, as a mass, occupy an inferior place to the fossiliferous * strata also taken as a mass. The next important conclusion is, that among the stratified fossili- ferous rocks there is a certain order of superposition, apparently marked by peculiar general accumulations of organic remains, though the mineralogical character varies materially. It has even been supposed that in the divisions termed formations, there are found certain species of shells, &c. characteristic of each. Of this supposition, extended observation can alone prove the truth ; but it must not be supposed, as some now do, that in any accumulation of ten or twenty beds, characterized by the presence of distinct fossils in a given district, the or- ganic remains will be found equally characteristic of the same part of the series at remote distances. To suppose that all the formations, into which it has been thought advisable to divide European rocks, can be detected by the same organic remains in various distant points of the globe, is to assume that the vegetables and animals distributed over the surface of the world were always the same at the same time, and that they were all destroyed at the same moment, to be replaced by a new creation, differing specifi- cally, if not generically, from that which immediately preceded * The term fossiliferous is here confined to organic remains. Classification of Rocks. 37 it. From this theory it would also be inferred that the whole surface of the world possessed an uniform temperature at the same given epoch. It has been considered, but has not yet been sufficiently proved, that the lowest rocks in which organic remains are found entombed, show a general uniformity in their organic contents at points on the surface considerably distant from each other, and that this general uniformity gradually disap- peared, until animal and vegetable life became as different in different latitudes, and even under various meridians, as it now is. How far this opinion may, or may not, be correct, can only be seen when geological facts shall have been suffi- ciently multiplied; but it is one which demands considerable attention, as the classification of fossiliferous rocks greatly depends upon it. Should it eventually be found to a certain degree correct, it would not be at variance with the theory of a central heat, which having diminished, permitted solar heat gradually to acquire an influence on the earth's surface. Classifications of rocks should be convenient, suited to the state of science, and as free as possible from a leading theory. The usual divisions of Primitive, Transition, Secondary, and Tertiary, may perhaps be convenient, but they certainly can- not lay claim to either equality with the state of science, or freedom from theory. In the accompanying Table, (pp. 38, 39,) rocks are first divided into Stratified and Un stratified, a natural division, or at all events one convenient for practical purposes, independent of the theoretical opinions that may be connected with either of these two great classes of rocks. The same may, perhaps, also be said of the next great division ; namely, that of the stratified rocks into Superior or Fossiliferous, and Inferior or Non-fossiliferous. The superior stratified or fossiliferous rocks are divided into groups. We are yet well acquainted with so small a portion of the real structure of the earth's exposed surface, that all general classifications seem prema- ture; and it appears useless to attempt others than those which are calculated for temporary purposes, and of such a nature as not to impede, by an assumption of more knowledge than we possess, the general advancement of geology. Stratified Rocks. Group 1. (Modern) seems at first sight natural and easily determined ; but in practice it is often very difficult to say where it commences. When we take into con- sideration the great depth of many ravines and gorges, which appear to originate in the cutting power of existing rivers, the cliffs even of the hardest rocks which more or less bound any extent of coast, and the immense accumulations of compara- tively modern land, such as those which constitute the deltas 38 Classification of Hocks. SUPERIOR STRA- TIFIED, or Fos- SILIFEROUS. 1. Modern Group 2. Erratic Block Group Detritus of various kinds pro- duced by causes now in action Coral islands ; Travertin, &c ro-~| m; > 3. Supracretaceous Group , f Transported boulders and"J J blocks ; gravels on hills and I < plains, apparently produced by > I greater forces than those now in I (. action. (A provisional group.) J {Various deposits above the chalk, such as in England, the Crag, Isle of Wight beds, Lon- don and Plastic clays. In France, the freshwater and ma- rine rocks of Paris, &c. 4. Cretaceous Group.^ 5. Oolitic Group 1. Chalk. 2. Upper green- sand. 3. Gault. 4. Lower green- sand. To which may be added, for convenience, 1. Weald clay. 2. Hastings sands. 3. Purbeck beds. The rocks usually known as the Oolite formation, including the Lias. r 1. Variegated or Red marl. 6. Red Sandstone J 2. Muschelkalk. 3. Red sand- Group j stone. 4. Zechstein. 5. Red L conglomerate. 7. Carboniferous Group r 1. Coal measures. 2. Carbo- s niferous limestone. L 3. Old red sandstone 8. Grauwacke Group. Grauwacke, thick-bedded and "] schistose, sometimes red; Grau- wacke limestones; Grauwacke | clay slates, &c. Inferior beds, V frequently mixed with stratified I compounds resembling those of .the unstratified rocks J r f~ Various schistose rocks, and"J INFERIOR brRA- J jj o ^terminate order J many crystalline stratified com- I TiFiED,orJNoN-^ of superposition...) pounds, such as Gneiss, Proto- f ^^ Mica g^ &( . ............. J - FOSSILIFEROUS. UNSTRATIFIED ROCKS. Volcanic, Trappean, Serpentinous, and Granitic rocks Ancient and modern Lava, " Trachyte, Basalt, Greenstone, Corneans, Augite, and Horn- blende Porphyries, Serpentine, Diallage rock, Sienite, Quartzi- . ferous Porphyry, Granite, &c. w Classification of Rocks. 39 improved Werner Ian. Conybeare. Omalius d'Halloy, 1830. Brongniart, 1829. Alluvion )iluvium : Ancient Alluvion. ^ertiary. Alluvial and } Lysian > * , rocks. ) Clysmian rocks, Superior > Tertiary rocks. Order. Secondary. Superme- dial Order. Ammonean rocks. ( [ Medial Or- der. m Transition. ) L Submedial " Order. > Hemilysian rocks. Primitive, or Primary. ( Inferior j Order. rimordial. Arranged among the stratified rocks, ac- cording to the order in which they are supposed to occur. The same as } the improv- I Pyroidal and Agaly- ed Werneri- f sian rocks. Izcmian rocks. Hemilysian rocks. kAgalysian rocks. Modern volcanic rocks, classed as Pyrogene- ous ; igneous rocks of an old- er date, as Ty- phonian. 40 Classification of Rocks. of great rivers, and the great flats, such as those on the eastern side of South America, there is a difficulty in referring these phenomena to the duration of a comparatively short period of time. Geologically speaking, the epoch is recent ; but according to our ideas of time, it appears to reach back far beyond the dates commonly assigned to the present order of things. Group 2. (Erratic Block] is exceedingly difficult to cha- racterize, and should only be regarded as provisional. It may be considered, merely for convenience, as comprising various superficial gravels, breccias, and transported materials. The most extraordinary feature of this group is the distribution of those enormous blocks or boulders found so singularly perched on mountains, or scattered over plains, far distant from the rocks from whence they appear to have been broken. Group 3. (Supracretaceous] comprises the rocks usually termed tertiary ; they are exceedingly various, and contain an immense accumulation of organic remains, terrestrial, fresh- water, and marine. This group has lately been shown to approach, more closely than was supposed^ to the existing order of things on the one side, and to the following group on the other. Group 4. (Cretaceous] contains the rocks which in England and the North of France are characterized by chalk in the upper part, and sands and sandstones in the lower. The term * cretaceous* is perhaps an indifferent one ; for probably, the mineralogical character of the upper portion, whence the name is derived, is local, that is, confined to particular por- tions of Europe, and may be represented elsewhere by dark compact limestones and even sandstones. As, however, geo- logists are perfectly agreed as to what rock is meant when we speak of the ' chalk,' there seems no objection to retain it for the present. The Wealden rocks have been arranged, for the present, in this group, though their organic remains show a different origin, because they may be conveniently studied in connexion with it. Group 5. (Oolitic] comprises the various members of the oolite or Jura limestone formation, including lias. The term ' oolitic' has been retained upon the same principle as that of 4 cretaceous.' In point of fact, this mineralogical character is found only in an insignificant part of the rocks known as the oolite formation in England and France ; and moreover it is not confined to the rocks in question, but is common to many others. In the Alps and in Italy the oolite formation seems replaced by dark and compact marble limestones, so that its mineralogical character is of little value. Classification of Rocks. 4-1 Group 6. (Hed Stfndstone] contains the red or variegated marls (marnes irisees, keuper), the muschelkalk, the new red or variegated sandstone (gres bigarre, hunter sandstein), the zechsteiri or magnesian limestone, and the red conglomerate (rothe todte liegende, gres rouge). The whole is considered as a mass of conglomerates, sandstones, and marls, generally of a red colour, but most frequently variegated on the upper parts. The limestones may be considered subordinate ; some- times only one occurs, sometimes the other, and sometimes both are wanting. There seems no good reason for supposing that other limestones may not be developed in this group in other parts of the world. Group 7. (Carboniferous.) Coal-measures, carboniferous limestone, and old red sandstone of the English. The former would appear in the greater number of instances to be natu- rally divided from the group (6) above it; but the latter, though disconnected from the group (8) beneath in the North of England, is apparently so united with it in many other situations, that the old red sandstone may be considered as little else than the upper part of the grauwacke series in those places. Group 8. (Gm'iiwacke.) This may be considered as a mass of sandstones, slates, and conglomerates, in which limestones are occasionally developed. Sandstones which mineralogically resemble the old red sandstone of the English, not only oc- cupy the upper part, but frequently also other situations in the series. In the lower portion of this group stratified com- pounds, resembling some of the unstratified rocks, are by no means unfrequent. Inferior or Non-Fossiliferous Stratified Rocks, comprising slates of different kinds, such as mica slate, chlorite slate, tal- cose slate, hornblende slate, &c., and various crystalline com- pounds arranged in strata, such as saccharine marble, in which other minerals may or may not be imbedded, gneiss, protogine, &c. From various circumstances, many rocks in the previous division so assume the mineralogical characters of those in this, as to be undistinguishable from them, except by geological situation ; but it may be assumed, that, as a mass, the strata in this division are far more crystalline than in those of the superior stratified rocks, the origin of which seems chiefly mechanical. Unstratified Rocks. This great natural division is one of considerable importance in the history of our globe, as the rocks composing it seem to have caused, jointly with the forces that ejected them, very considerable changes on the earth's surface. They are very generally admitted to be of igneous origin; some of them indeed, those produced by active vol- 42 Classification of Rocks. canoes, never could have been doubted. Their great charac- teristic is a tendency to a crystalline structure, though, in many, this cannot be traced. Every gradation from the cry- stalline to the non-crystalline structure can frequently be ob- served in the same mass. The minerals, felspar, quartz, horn- blende, mica, augite, diallage, and serpentine, enter largely into the composition of these rocks, more particularly the former. In proposing this classification, I am fully aware that many objections may be made to it ; but it pretends to little beyond convenience : and if geologists could be induced to use some- thing of this kind, or any other that would better answer the purpose of relieving us from the old theoretical terms, I can- not but imagine that the science would derive benefit from the change. In the following part of this volume, geological pheeno- mena will be noticed in accordance with this classification. But to enable those who prefer other arrangements to avail themselves of any facts that may be brought forward, the equivalents of the divisions or groups above noticed are given in the Table (pages 38, 39), where the classifications of Cony- beare, Brongniart, and Omalius D'Halloy, as also the im- proved Wernerian, are placed in parallel columns with it. Degradation of Land. 43 SECTION II. MODERN GROUP. Degradation of Land. THERE is a constant tendency in all decomposed or disinte- grated substances to be removed, by the agency of rains and superficial waters, to a lower level than they previously occu- pied, and finally to be transported into the sea. There is no rock, even the hardest, that does not bear some marks of what has been termed weathering, or of the action of the atmosphere upon it. The amount of surface-change, so produced, is ex- ceedingly variable, depending much on local causes. Thus, a rock may undergo complete disintegration in one situation, though composed of nearly the same materials as that in another, of which the change has been comparatively trifling. When we contemplate the present surface of our continents and islands, we cannot but be struck with the great effects that have been produced upon them by the agents commonly known as existing causes ; and among these, the weathering and degradation of land are very remarkable, attesting a lapse of time far beyond the usual calculations. The tors of Dart- moor, Devon, may be referred to as excellent examples of the weathering of a hard rock. These are composed of granite, which, as Dr. MacCulloch has observed, are divided into masses of a cubical or prismatic shape. " By degrees, sur- faces which were in contact become separated to a certain di- stance, which goes on to augment indefinitely. As the wear- ing continues to proceed more rapidly near the parts which are most external, and therefore most exposed, the masses which were originally prismatic acquire an irregular curvili- near boundary, and the stone assumes an appearance resem- bling the Cheese- wring (Cornwall). If the centre of gravity of the mass chances to be high and far removed from the per- pendicular of its fulcrum, the stone falls from its elevation, and becomes constantly rounder by the continuance of decom- position, till it assumes one of the spheroidal figures which the granite boulders so often exhibit. A different disposition of that centre will cause it to preserve its position for a greater length of time, or, in favourable circumstances, may produce a logging stone *." The weathering of these tors is so ex- * MacCulloch. Geol. Trans. 1st series, vol. ii. ; where there are views of Vixen Tor, the Cheese-wring, and Logan Rock : as also in Sections and Views illustrative of Geological Phenomena, pi. 20. 44 Degradation of Land. ceedingly slow, that the life of man will scarcely permit him to observe a change ; therefore the period requisite to produce their present appearances must have been very considerable. The surface of the whole country round these districts attests the same great lapse of time. Whatever may be the nature of the rock, it is disintegrated to considerable depths ; porphy- ries, slates, compact sandstones, trap rocks, all have suffered; but the valleys appear to have previously existed, and the ge- neral form of the land to have been much the same as it now is. The following section will explain this decomposition of surface. Fig. 4. a , decomposition of the rock b b following a line of pre- vious elevation and depression, the accumulation being great- est at the bottom of the valley c, frequently cut through by a river or rivulet, and sometimes exposing a stratified appear- ance, as if the disintegrated substances of the hill-sides had slipped over each other to the bottom of the valley. The maximum quantity of detritus so brought down to the bottom of a valley, sometimes amounts to 25 or 30 feet. This detritus, which is often very loosely aggregated, is now indeed protected from removal, at least to a great extent, by grass and general cultivation. The various appearances of this detritus are sin- gular ; for often larger pieces, perhaps of twenty or thirty pounds weight, are included among small fragments and even sand. Of this the following section, exhibited on the sea- shore at Black Pool, Dartmouth, affords an example. a a, detritus from the grauwacke slates b b, more thickly accumulated at ef. c c, a high beach of small quartz shingles, defending the bottom of the valley d (which is much lower than the crown of the beach) and the cliffs on either side. The drainage of the valley escapes in a serpentine manner by a rivulet at e. At e and^ many large fragments are mixed with the smaller. The slates in the South Hams, Devon, are frequently sur- mounted by a superficial covering of fragments, which, at their union with the undecomposed rock, appear as if some force had been exercised at the commencement; the slates being broken and turned buck in the manner represented beneath. Degradation of Land. 45 a, vegetable soil: &, small frag- Fig. 6. ments of slate resting in various di- rections : c> portions of laminae, turn- ed backwards, sometimes without fracture. If we proceed to the eastward from the South Hams, the same appearances present themselves, whatever may be the nature of the rock, though they become somewhat more com- plicated upon Haldon Hill, and on the coast of Sidmouth and Lyme Regis, as this decomposition of the surface seems mixed with a disintegration effected previous to the deposit of the supracretaceous rocks. A deep disintegration of surface, conforming to the undulations of the country, is well observed in Normandy, where it has been described by M. de Caumont and M. de Magneville, and seems due to the action of the same causes which have produced the decomposition of sur- face in the South of England. This destruction of the surface is common to most countries; and if the rock so weathered be limestone, there is, not unfre- quently, a reconsolidation of the parts by means of calcareous matter deposited by the water that percolates through the fragments, and which dissolves a portion of them. At Nice, the fractured surface thus reunited is so hard, that, if it occur on a line of road, it must be blasted by gunpowder for removal. There are some fine examples of this reconsolidation upon the limestone hills of Jamaica ; as for example near Rock Fort, and at the cliffs to the eastward of the Milk River's mouth. The felspar contained in granite is often easily decomposed, and when this is effected the surface frequently presents a quartzose gravel. D'Aubuisson mentions that in a hollow way, which had been only six years blasted through granite, the rock was entirely decomposed to the depth of three inches. He also states that the granite country of Auvergne, the Vi- varrais and the eastern Pyrenees, is frequently so much de- composed, that the traveller may imagine himself on large tracts of gravel *. Some trap-rocks, from the presence of the same mineral, are so liable to decomposition that there is frequently much difficulty in obtaining a specimen. The depth to which some rocks of this nature are disintegrated in Jamaica is often very considerable. This decomposition is attributed to the chemical as well as mechanical action of the atmosphere. The oxygen of the at- mosphere produces considerable alteration in rocks, more particularly observed in those containing iron, which are thus * Traite de Geognosie. 46 Degradation of Land. often reduced from a hard to a soft substance. With the slow and quiet changes effected by electricity on the surface we are very imperfectly acquainted; but all are familiar with the effects of a discharge from a thunder-storm, shivering rocks, and hurling fragments from the heights into the valleys beneath. In these electrical discharges the lightning often fuses the surface of rocks. Thus, De Saussure found a compound rock, on Mont Blanc, fused on the surface, white bubbles being on the felspar, and black bubbles on the hornblende. Similar observations have been made by other geologists in other parts of the world. The effects of lightning upon loose sand were beautifully exhibited in the drifted sand hills between the sea and the em- bouchure of the Irt, near Drigg, in Cumberland. The sand consists of quartzose grains intermingled with a few grains of hornstone porphyry, and a few fragments of shells, and rests, at the depth of twenty-nine feet, upon a bed of pebbles. Two feet beneath these pebbles is a bed of wet sand containing small pebbles. Upon a single hillock, about forty feet above the sea, and within an area of fifteen yards, three hollow tubes were discovered, about an inch and a half in diameter. They consisted of the matter of the sand fused, and rendered vi- treous. These tubes descended in a vertical manner to more than thirty feet, branching downwards in their descent. One tube coming in contact, at the depth of twenty-nine feet, with a fragment of hornstone porphyry, glanced off at an angle of 45, and afterwards resumed its vertical course. As might have been expected, the tubes were very irregular in their passage into the wet sand. The annexed sketch repre- sents the termination of one of these vitreous tubes upon a granite pebble, and exhibits the manner in which they branched off in their de- scent, thus marking the course of the electrical discharge *. At Peninis Point, St. Mary's, Scilly Islands, there is a curious example of that decomposition of granite, which antiquaries have termed rock- basins^ and considered the work of the Druids. The Kettle and Pans, as these depressions are there named, occur in the large blocks of granite on the top of this promontory ; they are gene- rally three feet in diameter and about two feet deep ; they are mostly circular and concave, but there are others much indented at the sides. " Some have perpendicular sides and flat bottoms, some are * Geol. Trans, vols. ii. and v. pi. 39. Degradation of Land. 47 of an oval form, and others of no regular figure. Many of the blocks are six or seven yards high, eight or nine yards square, and several of them have four, five, six or more of these cavities in them. A large rock near the extremity of this group has two basins, of an immense size, besides several smaller ones. The upper and larger one appears to have been formed by the junction of three or more large basins. It is irregularly shaped, and about eighteen feet in circumference and six feet deep. When the water in this basin has attained the height of three feet, it discharges itself by a lip into a lower basin, more regularly formed, the back of which is about five feet high, but which is incapable of containing more than a depth of two feet of water, owing to the declivity of the sur- face of the rock *." As a proof that similar decomposition sometimes takes place on the sides of a block, the author above cited mentions an oval cavity, six feet long, five wide, and nearly four feet deep, thus situated. The following wood-cut will afford an idea of the Kettle and Panst. Fig. 8. There is scarcely a substance, which having been exposed to the action of the atmosphere for a considerable time, does not exhibit marks of weathering. It will even be observed on the hardest siliceous rocks. The action of the atmosphere on cliffs of sandstone, in which the cement varies in induration or otherwise, produces the most grotesque forms, which must be more or less familiar to the least observing. Variations in temperature much assist the chemical decomposing power of the air. Water may be considered as the principal mechanical agent in the great work of atmospheric destruction, uniting at the same time the character of a chemical agent. By infiltration it tends to separate the particles of which the rocks are com- posed, uniting chemically with the cementing matter in some cases, and in others forcing it away mechanically ; in both instances leaving the particles not previously acted upon, more easily disturbed by a continuation of infiltration. In * Rev. G. Woodley ; View of the present State of the Scilly Islands, 1822. t From a sketch by Mr. Holland. 48 Degradation of Land. those situations where the temperature descends sufficiently low to produce frost, the mechanical action of the atmosphe- ric water becomes much more considerable. Having entered into the interstices of rocks when liquid, it assumes a greater volume when it becomes solid from a sufficiently diminished temperature, felt at greater or less depths in proportion to the amount of decreased heat of the climates where the rocks may be situate. Portions of rock are thus forced asunder, and fine particles so separated, that the mere return of the water to a liquid state, assisted by gravity, is sufficient to re- move them. The large masses have their centres of gravity often so altered relatively to rocks on which they rest, that when no longer cemented by the ice, they fall from their si- tuations to a lower level. The fall of rocks occasioned by this means is common in lofty mountains, where considerable heights are exposed to the alternations of frost and thaw. By percolation through porous rocks the water attains strata which are not so, such as clays. The water thus stopped in its course downwards, escapes as it best can to the sides of hills and other situations, producing springs. At the places where this discharge of water takes place, there is also a me- chanical destruction of the parts through which the water de- livers itself. Rocks are affected by this action of the water in proportion to their composition ; which, though not porous, may still be acted on by the water. An argillaceous substra- tum will get gradually moist at the surface, and in favourable situations may become a wet clay. The stability of the mass above will depend upon the relative position of the strata. Thus in the wood-cut annexed, if on the mountain a, water perco- late through the porous strata b to the impervious clay bed c c, the surface of the latter would become slippery, and the mass above be launched into the valley d. Now this is precisely what hap- pened in the case of the Ruffiberg in Switzerland. This mountain, also known as the Rossberg, is 5196 feet above the level of the sea, and rises opposite the well-known Righi. Its upper part is composed of beds of a compound rock formed from the debris of the Alps at a previous geological epoch. These are to a certain extent porous, and the water percolates through them to a clay stratum on which they rest ; the whole dipping at a considerable angle (about 45). The clay becoming soft by the action of the water, and the thick superincumbent beds losing their support, the latter were launched over the slippery and inclined surface beneath, and the valley below was covered with their ruin. This slide took place on the 2nd of September 1806, and Degradation of Land. 49 covered a beautiful valley with rocks and mud. The villages of Goldau and Busingen, the hamlet of Huelloch, a large part of the village of Lowertz, the farms of Unter- and Ober- Rothen, and many scattered houses in the valley, were over- whelmed by the ruin. Goldau was crushed by masses of rocks, and Lowertz invaded by a stream of mud. The torrent of rubbish and mud which rushed into the lake of Lowertz produced such a motion of the waters, that the village of Seven, situated at the other extremity, was in- undated, and in great danger of being destroyed, two houses having been washed away. Live fish were found in the vil- lage of Steinen, thrown there by the flood. The lives lost were calculated from 800 to 900. Several travellers perished. It appears that there are traditionary accounts of former, though smaller, slides from the Rouffi or Rossberg*. Large falls from mountains take place from the percolation of water to certain portions, which they mechanically loosen or chemically destroy without sliding over an inclined plane, as in the case of Rossberg, though the force of gravity still causes the fall. The Alps have afforded many examples of this fact, among others that of the great fall from the Dia- blerets in 1749. Nothing is so common in mountainous regions as a talus of detritus brought to the foot of a cliff; this detritus com- posed of fragments of the rocks above, detached by decom- position from their surface, and brought down directly by their own gravity, or by the union of their own gravity and the force of surface-water, the latter derived from rains and the melting of snows. Avalanches of snow are great trans- porters of such fragments ; and in the places where they fall there are always large accumulations of them, often borne from the greatest heights by the irresistible fury of the de- scending snow. The under cliffs at Pinhay, near Lyme Regis, may be taken as an example of the destruction of a cliff' by means of land springs, greater than that which is produced by the ac- tion of the sea in the same place. * For a view of this fall, taken four days after the catastrophe, see Sec- tions and Views illustrative of Geological Phenomena, pi. 33. E SO Degradation of Land. a, gravel : b 9 chalk : c, green sand, both porous rocks through which the water percolates to the clay bed d 9 com- posed of the lower part of the green sand beds c and the up- per part of the\ lias beds e ; being arrested in its progress downwards, the water escapes by the easiest road, which is that presented by the cliff originally formed by the sea. It here gradually carries away the clay, first rendering it moist. The chalk and green sand lose their support, give way, and fall over into the sea. The lias e does not give way so fast before the sea at the cliff g, as the superincumbent mass, af- fected by the land springs ; therefore the latter retreats until it has formed a great talus at f\ but this talus tends con- stantly to move forward both by the destruction of the lias cliff at g, and by the tendency of the land springs to loosen its base, and to propel it into the sea. The chalk and green sand containing hard substances, often of considerable size, great protection is afforded to the cliff g, by their fall over its top, the fury of the breakers being greatly spent upon these masses. Rivers. These most frequently, though not always, take their rise among hills and mountains, and are supplied either by the melting of snows or glaciers, by the draining of rain- waters, or by springs. They transport the detritus formed either by the atmospheric agents, previously noticed, or by themselves. The power of this transport depends upon their velocities. Now, the velocity of a river current is greatest in the centre, and least on the sides and bottom, being retarded by friction ; consequently, the transporting power of a river is least where it comes in contact with the substances to be transported. These substances are generally angular if de- tached from simple rocks for the first time, such as pieces of limestone, granite, &c., and at the commencement present great obstacles to transportation ; for the velocity of a current must be sufficient to move these angular fragments before they can suffer attrition. Rocks composed of fragments which have been previously rounded, such as conglomerates, will, if they decompose easily, contribute ready-formed gravel to the river, which might thus be able to carry them forward, while its velocity was insufficient to transport angular frag- ments of equal weight. The transport of sandstones will depend on their state of induration, and be easy where the particles are slightly aggregated ; difficult, when so compact as to form angular fragments. When the velocity of a river is sufficient to produce attrition of the substances which it has either torn up, collected by un- dermining its banks, or which have fallen into it, such sub- stances gradually become more easy of transport, and would, Degradation of Land. 51 if the force of the current continued always the same, be forced forward until the river delivered itself into the sea ; but. as the velocity of a current greatly depends on the fall of the river from one level to another, the transport is regulated by the inclina- tion of the river's bed. Now it is well known that this inclina- tion varies materially, even in the same river; so that it may be able to carry detritus to one situation, but may be unable to transport it further, under ordinary circumstances, in con- sequence of diminished velocity. But this may be, and often is, so much increased further down, that its original transport- ing power may be, in a great measure, restored. It can now, however, only carry forward such detritus as it can receive or tear up in its course, and the pebbles which were left behind at the place of its first diminished velocity can only be brought within its power by floods, or, in other words, by extraordinary circumstances. As a general fact, it may be stated that rivers, where their courses are short and rapid, bear down pebbles into the seas near them, as is the case in the Maritime Alps, &c. ; but that when their courses are long, and changed from rapid to slow, they deposit the pebbles where the force of the stream diminishes, and finally transport mere sand or mud to their mouths, as is the case with the Rhine, Rhone, Po, Danube, Ganges, &c. It will follow that the form and weight of the detritus carried to the sea will depend upon the length and velocities of rivers, all other circumstances being the same. If in its course the form of the land be such that lakes are produced, the detritus borne down by a river will be deposited in their beds, which have thus a tendency to be gradually filled up, the quality of the detritus depending on the velocity of the river. Such inequalities, producing small lakes, are common in mountain valleys, and have evidently been once much more so. The velocity of the stream issuing from the lake will greatly depend upon the fall of land over which it flows. The stream will endeavour to cut down the barrier which produced the lake, but if the velocity be trifling or the rocks hard, it will effect little ; while if the stream be rapid or the rocks easily cut, it will traverse the natural bar, drain the lake, and permit the river to flow in an uninterrupted course. Should the lake, while it existed, have been partially filled up by the detritus from above, the river will cut through this also, and the part thus cut away will be transported to a lower level. The following diagram may assist the reader. Fig. 11. E 2 52 Degradation of Land. a b, course of a river flowing into the lake b h c, which is filled with water to the level b c, the surplus falling over the slope c d, and continuing its course in the direction d g : ef^ deposit of detritus derived from the river a b, at the bottom of the lake c h b : b d, bed of the river formed by cutting through the barrier e c d, and part of the detritus e /if, so as to form a continuous course with a b on the one side and d g on the other. When lakes are large, such for instance as those of Geneva and Constance, an immense lapse of time will be required to fill them with detritus, so that, eventually, a continuous river may traverse land occupying a space once filled by the water. Lakes of this magnitude oppose great obstacles to the transport of pebbles. The progress of a large proportion of detritus from the Alps is arrested by lakes on their north and south sides. Thus, on the north, the Rhine deposits its mountain detritus in the lake of Constance, and the Rhone its trans- ported pebbles and sands in the lake of Geneva. Between these two great lakes, those of Zurich, Lucerne, &c., receive the gravels of other Alpine rivers. On the south, the Lago Maggiore receives the Alpine detritus of the Ticino; the lake of Como, that of the Adda ; and the lakes of Garda, &c. perform the same office to other rivers. From these circum- stances it will be evident, that the detritus of a large portion of the Alps cannot travel, by the rivers, either into the ocean or the Mediterranean. The Po receives the waters of a large portion of the Alps, and carries sand and silt into the sea ; but the pebbles are arrested before it receives the Ticino, which, though it transports rounded stones, does not bring them directly from the Alps, but from its banks, after quitting the Lago Maggiore, which banks contain the rounded Alpine fragments of a previous epoch. The same with the Rhone near Geneva, in which Alpine pebbles occur, and which could not, in the actual state of things, be derived from the Alps, because they would have been stopped in the lake of Geneva. They are derived from its banks and bed immediately on quitting the lake. Geological students, in examining river- courses, should be very careful in distinguishing between pebbles from the immediate banks of rivers, and those which might be derived from a distance, but to the transport of which, by the rivers, physical obstacles oppose themselves. From a want of attention to this circumstance, many errors have arisen. It has been considered that the mode in which a river discharges itself into a lake, and pushes forward its detritus, would be such that the deposit would assume a nearly horizontal stratification. The angle of deposit must, however, depend upon the depth of the lake and the quality of the Degradation of Land. 53 detritus discharged into it. Thus, if the detritus be composed of sand and mud, it will be propelled further into the body of the lake than if it consist of pebbles. Examples of both cases will be found in the lake of Geneva. The ordinary deposit from the Rhone is sandy and muddy, which sinks in clouds, from its greater specific gravity, beneath the clear waters of the lake ; yet the initial velocity is sufficient to transport a part of it about a league and a quarter, for I found a portion of it at the depth of 90 fathoms, raising the bottom of the lake between St. Gingolph and Vevey *. This would give a very slight dip from the embouchure of the Rhone. Off' the mouth of the Drance, a torrent rushing into the lake near Ripaille, the pebbles, forced down, must arrange themselves at a much more considerable angle ; for 80 fathoms are obtained at a short distance from the shore. The same variations in dip will also be observed in the lake of Como, where the turbid waters of the Adda have deposited a considerable quantity of sand and mud, which slopes gradually at a gentle angle; while the torrent-borne detritus at Bellano, Mandello, Ab- badia, and other places, arranges itself with a much more considerable inclination. It would seem to follow that the stratification of lake deposits derived from the land around them, would not be uniform, but would depend on local cir- cumstances, rivers or torrents propelling detritus before them, which would be as various as the rocks they respectively tra- versed ; each collection would have a mode of deposit of its own, independent of the others, and they would tend to ap- proach, and finally to unite with each other. The higher part of the lake of Como is nearly filled up by the detritus transported by the Adda and Meraf . The former has divided the lake into two ; the smaller portion (known by the name of the Lago di Mesola,) being so shallow from the united deposit of the two rivers and some torrents, that aquatic plants grow through the water on the eastern part ; while on the western, in which there is a greater depth, the process of filling up is hastened by means of stones, detached in such numbers, in particular seasons of the year, from the heights on that side, that a passage in a boat beneath the cliffs be- comes exceedingly hazardous. Considering the many thousand revolutions of our planet round the sun, that must have taken place since the land assumed its present general form, we should expect to find the barriers even of considerable lakes cut through under favourable circumstances, and accordingly we do discover appearances which would seem to warrant this conclusion. * For a map and sections of this lake, see BibliothequeUniversellefor 1819. t See Sections and Views illustrative of Geological Phenomena, pi- 31, 54- Degradation of Land. It is by no means uncommon to find plains of greater or less extent, bounded on all sides by high land, and through which a principal river meanders, entering at one end by a valley, and passing out through a gorge at the other, aug- mented by tributary streams from the surrounding hills: sometimes these plains have no principal river passing through them; but many small streams, descending from the moun- tains, unite in the plain and pass out also through a gorge. In such cases the plain presents the appearance of a drained lake, such as we may suppose would be exhibited in many now existing, if passages for the waters were cut or broken through any part of the basins holding them. The gorge at Narni seems to have let out the waters of a lake supplied by the Nera, which now flows through the plain of Terni, the former bottom of the lake. The great fertile plain of Florence seems once to have been the bed of a lake, the drainage of which was effected by a cut through the high land that bounds it on the west. If this outlet were again closed, the waters of the Arno would again cover the plain and convert it into the bed of a lake. The period at which the break in the Jura was formed at the Fort de 1' Eel use, may perhaps be ques- tionable ; but if closed, it would stop the course of the Rhone, and convert the lake of Geneva into a much larger body of water. These appearances are not confined to one part of the world ; they would appear, from the descriptions of intelligent travellers, to exist very commonly. I hare myself observed examples in Jamaica. The district named St. Thomas in the Vale is a marked one. Here we have low land bounded on all sides by hills, which would form the banks of a lake, were not the waters let out by the gorge through which the Rio Cobre flows. It would therefore appear, though large lakes collect moun- tain detritus, which is distributed over a large surface, en- veloping, probably, animal and vegetable remains, that the barriers of the lakes may be cut or broken through, and the rivers again act on a portion of the previous deposit. The probability that many gorges originate from the cutting power of rivers discharged from lakes, is rendered stronger by examining those natural basins which are drained by sub- terraneous channels, and where gorges are not found. Thus Luidas Vale, in the island of Jamaica, is a district surrounded on all sides by high land, and would form a lake, were not the waters, derived from heavy tropical rains, carried off by sink-holes in the low grounds. A body of water, brought to turn the water-wheel of an estate's works, is swallowed up close to these works. A cavern, out of which water some- times issues, near another estate, is speedily engulfed in a cave Degradation of* Land. 55 not far distant. In consequence of this escape of the waters, a gorge is not formed by means of a discharging river flowing over the lowest lip of the high land, as appears to have hap- pened in the case of St. Thomas in the Vale, which adjoins Luidas Vale. It is stated, " that a velocity of three inches per second at the bottom will just begin to work upon fine clay fit for pot- tery, and however firm and compact it may be, it will tear it up ; yet no beds are more stable than clay when the velocities do not exceed this ; for the water soon takes away the impal- pable particles of the superficial clay, leaving the particles of sand sticking by their lower half in the rest of the clay, which they now protect, making a very permanent bottom, if the stream does not bring down gravel or coarse sand, which will rub off this very thin crust, and allow another layer to be worn oiF. A velocity of six inches will lift fine sand ; eight inches will lift sand as coarse as linseed ; twelve inches will sweep away fine gravel ; twenty-four inches will roll along rounded pebbles an inch in diameter ; and it requires three feet per second at the bottom to sweep along shivery angular stones of the size of an egg*." The destructive power of rivers on solid rocks appears to act both chemically and mechanically. Chemically, by the affinity of water and of the air which it holds in solution for the various substances it encounters ; and mechanically, by the friction of the detritus, independent of that of the water, upon the bottom and sides, but principally on the former. They may have thus effected a passage through the lake bar- riers previously noticed, and by these means they destroy the obstacles opposed to their courses. When a bank, a small hill, or the foot of a mountain, opposes their progress, they assail it, and form cliffs, the materials of which, if soft, fall into the stream, or make under cliffs, which are removed, and the work of destruction is slowly continued (Fig. 12. a.) ; or when Fig. 12. the cliff, thus formed, is of harder materials, blocks are accu- mulated in a talus at its base, and the cliff is secured, in a great measure, from attack, until this protecting mass is removed * Encyclopaedia Britannica, art. River. 56 Degradation of Land. (Fig. 12. b.). There is scarcely a river of any considerable length which does not afford examples of cliff's thus produced ; very frequently they overhang flat or gently sloping land, on which the river has flowed while employed in cutting the cliff". It is not a little curious to trace, in countries where rivers wind considerably, the various obstacles which have deter- mined the course of the stream, causing it to attack the ori- ginal more or less rounded forms of the bases of moderately elevated hills. Rivers appear to be constantly striving to arrange their beds in such a manner that they should suffer the least resist- ance in their courses, cutting down obstacles and filling up depressions which checked them. But the constant addition of new detritus from the neighbouring highlands embarrasses this operation, causing accumulations in one situation which direct the waters in another. Thus the fall of a considerable quantity of rocks on one side will throw the stream upon the opposite bank, which might previously have been little at- tacked. This again forces the current in a direction that it did not previously follow ; the bottom becomes torn up by the new line of the principal stream, and the effect of such a fall is felt far down the course of the river. In consequence of this endeavour to avoid a new obstacle, continual changes in a river's bed take place, as also from the destruction of an old obstacle, which permits a new course in a direction that the river has been striving to follow. D'Aubuisson observed two rocks at the falls of the Rhine, near Schafhausen, isolated at the head of the precipice over which the waters leap ; these were observed corroded at their bases by the action of the pent-up current between them. By gradually diminishing their support, the rocks would finally be forced over the cataract, and the waters, having overcome this obstacle,, would fall in a different manner on the bottom beneath, producing a different effect from that which they had previously caused. As all rivers must vary in their cutting power, according to velocity, volume of water, and amount and quality of detritus in the act of transport, it becomes exceedingly difficult to ge- neralize on the subject; but as barriers of even the hardest rocks have suffered, and as the destructive power of the same rivers on the same obstacles is so exceedingly small as to be scarcely perceptible during the life of man, it seems fair to infer that this also tends to confirm the opinion of the great age of the present general state of the world. Mr. Lyell indeed produces, as an example of the compara- tively quick cutting power of a river, a gorge in a lava-current at the foot of Etna, formed by the erosion of the Simeto. The Degradation of Land. 7 lava is considered modern, and Gamellaro is cited as sup- posing it thrown out in 1603. The lava is described as not porous or scoriaceous, but as a compact homogeneous rock, lighter than common basalt, and containing crystals of olivine and glassy felspar. Though there are two waterfalls, each about six feet, the general fall of the river's bed is stated as not considerable. The gorge is cut in some places to the depth of forty or fifty feet, and its breadth varies from fifty to several hundred feet*. It is therefore inferred that this is a good example of the speedy formation of gorges by running water ; and this inference cannot be denied, if the date of the lava-current be correctly ascertained. It may be remarked that the present fall in the bed of the Simeto does not give that of the river during the great cutting operation. It must once have occupied a different level, or else the gorge could not have been commenced ; and there must always have been a rapid fall, or, in other words, a cascade into the low land off the lava, equal to the height of the lava-current ; the waters being raised to the top of the lava, at this place, by the forma- tion of a lake behind, produced by the bar of Java. It would therefore follow, that the gorge in the lava-current has been principally formed by the cutting back of rapids or a cataract. Though this circumstance would facilitate the progress of destruction, and render it less remarkable than if the Simeto, with its present fall, had cut the gorge, it yet leaves this a good example of a ravine formed in hard rock during the course of two centuries, it being always understood that no doubt exists of the period when the lava-current was ejected, and crossed the previously existing valley. The dates obtained by the well-known examples of the Auvergne rivers are only relative ; but they are sufficient to show that a valley existed, through which a river kept its course, conveying detritus in the usual way, and that the pro- gress of the river was barred by a lava-current (as in the in- stance just cited), which descending from a neighbouring vol- cano traversed the valley, and formed a lake. This lake, when full, discharged itself over the lower lip of its basin, which happened to be in the direction of the valley, and over the lava-current. This, by erosion, is cut down, not only to its original bed, but through it into the rock which constituted the bottom of the original valley. Notwithstanding appearances, there are numerous gorges or ravines through which rivers flow, which could not have been cut out by them, at least during the existence of the present general disposition of land ; for the relative levels are such, Principles of Geology. 58 Degradation of Land. that the rivers must be supposed to have run over land of much greater elevation towards their embouchures than they flowed over from their sources ; in other words, such rivers must be supposed to have run up hill, if they be considered the agents which have formed these gorges. As a striking example of this fact, we may cite the course of the Meuse previous to, and during its traverse through the Ardennes. M. Boblaye in- forms us, that previous to its passage through these moun- tains, the Meuse is only separated from the great basin of the Seine by hills or low cols, not more than thirty or forty yards above the present bed of the river; while the Ardennes, through which it actually passes, rise to the height of several hundred feet above the same level. Now, if all rivers had really cut the beds or valleys through which they actually flow, the Meuse must have run up hill, and have cut a narrow channel about three hundred yards deep ; while nothing pre- vented its flowing in the opposite direction into the Paris basin, when it had effected a rise of not much more than a tenth part of that height*. At Clifton, near Bristol, we have also a striking example of the same fact. The Avon here runs through a gorge or ravine, which if closed would form a lake behind it ; but this lake would exert no action on the range of hill through which the present channel passes ; on the contrary, the lowest lip of the basin, and consequently the drainage, would be found in the direction of Nailsea, to the sea beyond which the Avon would continue its course from Bristol. The real rise of land between high water at Bristol and the sea beyond Nailsea is trifling, and is bounded on the north by the high ridge through which the Avon now finds its passage to the Severn. M. v. Dechen remarks that these facts can nowhere be better observed than at the confluence of the Nahe and Rhine near Bingen. The Nahe flows from Kreutznach to Miinster and Budesheim through a broad valley, bounded on the right by low limestone hills. It then traverses the slate mountains through a narrow defile, while the broad valley continues south-east from the Rochusberg by Sponsheim and Okenheim to the great valley of the Rhine at Gaulsheim. Nothing can be more evident than that the Nahe was unable to cut this defile between the Rochusberg and the heights of Weiler, while such a broad and deep channel into the Rhine presented itself so immediately to the passage of the watersf- Other examples might easily be cited, but these are suffi- cient to point out the fact. There are many gorges through which rivers pass, the formation of which remains question - Boblaye, Ann. des Sci. Nat. t. xvii. p. 37. f Von Dechen, German Transl, of Manual. Degradation of Land. 59 able from our ignorance of the relative levels in their vicinity, and thus it becomes difficult to assign them any particular origin. They may be either due to the same causes which have produced the ravines of the Meuse in the Ardennes, and of the Avon near Bristol, or to the cutting power of rivers discharging the surplus waters of lakes. Under this head may be enumerated the celebrated Vale of Tempe, in Thes- saly; the tortuous course of the Wye, between Monmouth and Chepstow ; the famous Rheingau ; the ravine by which the Potomack traverses the Blue Mountains in the United States ; the Gates of Iron, through which the Danube escapes into Wallachia; &c. The Falls of Niagara may be adduced as an example of a river discharging the surplus waters of a lake, and now cutting back a gorge to that lake, which may eventually be drained by it. This celebrated cataract is situated between the lakes Ontario and Erie. For some distance above the embouchure of the river into the former, the country is flat, and apparently alluvial, when suddenly a plateau rises above it and continues to lake Erie. Over this plateau the surplus waters of the latter lake have taken their course, and appear to have ori- ginally fallen over the face of the plateau fronting lake Ontario. By degrees they have cut back their passage about seven miles, leaving about eighteen more to be worn away by future ages. When this shall have been accomplished, .the gorge or ravine will be similar to those previously noticed. The man- ner in which the river cuts its passage is singular, and perhaps somewhat different from what, at first sight, might have been expected. It will be best explained by the following diagram. a b 9 original level of the plateau : a h, river flowing over the plateau, and falling over to the abyss c 9 forming the cascade h c 9 after which the waters take their course in the direction eg: d, beds of limestone resting on beds of shale e, both being surmounted, in the neighbouring flat country, by a mass of transported substances, varying from ten to one hun- dred and forty feet in depth, and containing large blocks. The rush of waters from h to c occasions violent gusts of wind, charged with water, to be driven against the shale e at/*. The continued action of these water-charged whirlwinds displaces the shale, and throws it down in a talus at k. From the re- moval of this shale, the superincumbent limestone loses its support, falls from the combined gravity of itself and the water above, is dashed into the abyss beneath, and thus the falls are 60 Degradation of Land. cut back so rapidly that they have considerably receded within the memory of man. The same operations are again renewed, and again the same results follow. So that unless some extra- ordinary circumstance should arrest their retreat, these falls will discharge the waters of lake Erie; but not suddenly, as is sometimes supposed, so as to produce a violent deluge over the lower country further down the river, but much more gradually ; for the lake waters will only be lowered in propor- tion to the depth of the draining channel, as may be illustrated by the annexed wood-cut, in which a b represents the level of /' g h! i< k 1 m 1 n 1 the lake and of the plateau, rising but little above it : h e, the slope (exaggerated) of the lake bed from h, the spot where the surplus waters are delivered over the plateau, f 1 n 1 the level of the river below the falls. Supposing^ g to represent the falls which have approached the lake by gradually cutting back the channel fromff to gg 1 , it will appear that the same kind of retreat may be effected to h h 1 without discharging more water than now passes down the river. But the falls being once at h h' 9 the retreat of every succeeding yard will occasion more water to pass over them, by draining the waters of the lake down to the point which now becomes its lowest lip ; so that when the falls have cut their way back to i z', the surface of the lake will sink to the horizontal line i c, and the mass of water above the new level will have passed over the falls in addition to the usual drainage. Such an addition must add greatly to the velocity and cutting power of the falls, which will now re- treat more rapidly and effect their passage to kh', reducing the level to k d in less time than it reduced it from a b to i c. After a certain time the water forced over the falls would be- come less, because the superficies of the lake would be smaller, if it were not for the diminished evaporation of the waters sup- plying lake Erie. It has been remarked * that as the surface of the lake became less, the waters carried off by evaporation would also decrease, and that consequently a greater body of water would be carried over the falls of Niagara, thus accele- rating their retrograde movement. Such would no doubt be the effect of the union of the streams, now supplying lake Erie, into one considerable river, without being exposed to that great evaporation which they now suffer in the lake ; but this would by no means assist in producing a debacle or deluge * Lyell, Principles of Geology, 2nd ed. vol. i. p. 209. Degradation of Land. 61 from the supposed sudden discharge of the waters of lake Erie. The addition to the waters passing over the falls of Niagara could only take place as the superficies of the lake became less from the gradual drainage of lake Erie above no- ticed, and consequently the additional force could only come into action and increase in intensity in proportion as the possibility of a sudden discharge of waters would become less. Finally, from the operation of these various causes, a river, fed by a prolongation of the streams now discharging them- selves into lake Erie, would traverse the ancient bed of the lake, and run through the ravine, cut by the falls of Niagara, into lake Ontario. The waters of a lake with a rocky barrier can only be sud- denly let out, and produce a debacle, when the hard barrier separating it from the land at a lower level presents a perpen- dicular face to the whole depth of the lake, which, even then, must be suddenly thrown down, in its whole height, to pro- duce the effect required. Such rocky barriers must be ex- ceedingly rare; and it must be still more rare, that where they existed they were not cut down, to a certain extent, by degrees. The common character of lakes, as respects the inclination from their bottoms to the discharging outlet, varies materially, but in general the slope is very gradual, particu- larly in lakes of considerable magnitude. The often cited debacle caused by the bursting of a lake in the Val de Bagnes was produced from a very different state of things from that attending the drainage of a lake existing in a depression of land, with a rocky barrier. The Val de Bagnes, in the Vallais, is drained by the Dranse, which, when unobstructed, is joined by the waters from the valley of Entremont, leading to the Grand St. Ber- nard, and runs into the great valley of the Rhone, near Martigny. In a part of the valley near the bridge of Mau- voisin, the channel is precipitous and much contracted. Mont Pleureur and Mont Getroz rise^near this spot on the north, and Mont Mauvoisin on the south. Between the two former there is a ravine communicating with the Val de Bagnes, hav- ing a considerable glacier at its upper extremity. Through this ravine blocks of ice and avalanches of snow descend into the Val de Bagnes, and more or less obstruct the channel of the Dranse, which is able, under ordinary circumstances, to remove the greater part, if not the whole, of such ob- structions. When however the blocks of ice are nume- rous, and the avalanches are heavy, the force of the tor- rent is unable to contend with them, and they accumulate. " For several years previous to 1818," says M. Escher de la Linth, " the progress of the Dranse had begun to be obstructed by the blocks of ice and avalanches of snow that 62. Degradation of Land. descended from the glacier of Getroz ; and as soon as this accumulation was able to resist the heats of summer, it ac- quired new magnitude during every succeeding winter, till it became an homogeneous mass of ice of a conical form. The waters of the Dranse, however, still found their way beneath the icy cone till the month of April, when they were observed to have been dammed up, and to have formed a lake about half a league in length*. The danger that threatened now became apparent, and ac- cordingly the gradual drainage of the lake was attempted by means of a gallery through the ice. This reduced its contents from about 800,000,000 cubic feet to 530,000,000 cubic feet. Finally, the discharging waters attacked the debris at the foot of Mauvoisin, and excavating a passage between the rocks and the ice, rushed furiously out, carrying houses, trees, large blocks of rock, &c. before it. Escaping from the narrow valley it desolated a large portion of Martigny f, and passed with gradually diminished velocity down the Rhone into the lake of Geneva. As might be expected, the velocity of the torrent varied materially in different parts of its course. M. Escher de la Linth calculates that from the glacier to Le Chable, a distance of 70,000 feet, the velocity was 33 feet per second ; from Le Chable to Martigny, 60,000 feet, at the rate of 18 feet; from Martigny to St. Maurice, 30,000 feet, at llf; and from St. Maurice to the lake of Geneva, 80,000 feet, with a diminished velocity of 6 feet per second {. The lake was drained in half an hour. As has been noticed by Mr. Yates , lakes are produced in mountainous countries by the fall of rocky masses across nar- row valleys, the waters being thus arrested in their progress down such valleys. Mr. Yates cites the Oschenen-see in the canton of Berne, as a good example of lakes thus formed J ; and M. De Gasparin mentions a recent example (November, 1829) of the formation of such a lake, in the department of the Drome, by the fall of a mountain mass across the river Oule near Lamothe Chalancon. The lake produced in the latter case was 500 or 600 yards long, 60 broad, and 3 or 4? yards deep || . It will be obvious that the possibility of the sudden discharge of waters, thus pent up, will depend upon the nature of the materials composing the dam or barrier : if * Edin. Phil. Journ. vol. i. p, 188. f Among the dehris transported to Martigny were many trees, resting upright on their roots, the attached gravel and soil having kept them in a position with the branches upwards. J Edin. Phil. Journ. vol. i. p. 191. Yates, Remarks on Alluvial Deposits, Edin. New Phil. Journal, July, 1831. II De Gasparin, Ann. des Sci. Nat., Avril 1830. Degradation of Land. 63 these be of such a form, quality, and magnitude, that the body of water is unable to overcome their resistance, and that they do not give way before the cutting power of the surplus waters discharged over the lower lip of the dam, the barrier will remain, acln be clothed with wood and other vegetation, as that of the Oschenen-see now is. Should, however, the dam be composed of soft materials, which might either sud- denly give way before the force of the pent-up water, or be rapidly cut down when a discharge of the surplus water took place, a debacle somewhat analogous to that of the Val de Bagnes might be produced, the effects of which would de- pend upon the body of waters let out, the suddenness with which this was accomplished, and other obvious circum- stances *. Lakes may be suddenly drained, if but a thin perpendicular partition divides them from an inferior level ; for this barrier maybe rendered soft by the percolation of water, and suddenly give way ; but such cases must be of very rare occurrence ; and the lakes are not likely to be of such magnitude as to cause appearances, by their sudden discharge, that may be equal to those producible by the passage of a more general mass of waters over land. Mr. Strangways notices the bursting, or sudden consider- able drainage, of the lake Souvando, on the north of St. Pe- tersburg. Previous to 1818 this lake was separated from that of Ladoga by the little isthmus of Taipala. The lake discharged its waters into the Voxa at Keognemy, and so passed into the Ladoga at Kexholm. In the spring of 1818, the water broke down the isthmus and changed the direction of the discharging waters, by presenting a lower lip in an- other direction. The water has been lowered considerably, and continues to run through its new channel into the lake of Ladoga, having deserted the Voxa f. The same author describes the falls or rapids of Imatra, about six wersts below the point where the surplus waters of the lake Saima first drain off by the Voxa. This river sud- denly contracts itself above the rapids, over which it runs, with great noise and impetuosity, through a gorge that it has evidently cut for itself. According to Mr. Strangways, we * The same observations apply to those cases also noticed by Mr. Yates, in the memoir above cited, where from various circumstances a toirentmay bring with it from a transverse or tributary valley such a mass of detritus into the main valley, as to arrest the progress of water flowing down it. In these cases, however, the barrier, from the nature of things, is not likely to be permanent, but, on the contrary, to be removeable with greater or less rapidity by the main river or torrent. t Strangways, Geol. Trans. First Series, vol. v. p. 344. 64 Degradation of Land. may consider the water to have originally passed over a plat- form between two ranges of hills, forming the bottom of a valley. The platform is composed of gneiss, in very highly inclined strata ; and into this the river has cut a channel. " The surface of this platform is apparently now about fifty feet above the level of the water, at the lower extremity of the rapids. Its surface is in many parts quite bare and deeply channelled in a direction parallel to the river. It is covered with heaps of pebbles and boulders of great size, some of which are hollowed and scooped into the most fanciful shapes. One of the largest of the blocks now left dry, standing nearly in the middle of the elevated platform, is worn through perpendicu- larly with a cylindrical hole*." It is stated that the level of the lake Saima and its discharging river fall gradually. Freshets. These take place, more or less, in all rivers, greatly augmenting their velocities and transporting power, carrying forward substances that could not have been moved under ordinary circumstances. They are also important, as they surprise terrestrial animals in low situations, hurry them on with trees and other matters into the sea, where they may be entombed entire with estuary and marine animals in mud and silt. It has been observed that, during freshets, a river tends chiefly to widen its bed, " without greatly deepening it : for the aquatic plants, which have been growing and thriving during the peaceable state of the river, are now laid along, but not swept away, by the freshes, and protect the bottom from their at- tacks ; and the stones and gravel, which must have been left bare in a course of years, working on the soil, will also col- lect in the bottom, and greatly augment its power of resist- ancef." During these freshes, low lands on the sides of the river are frequently under water, and a deposit takes place ; but notwithstanding all checks, a large quantity of detritus passes onwards to the sea. We should be careful, in our estimates of the effects of a flood in a cultivated country, not only to separate the loss of lives and the destruction of property, which may affect the feelings, from the real physical change produced in the country; but also to remember, that the works of man greatly aid the destructive power of a flood. Instead of a body of water rushing into a plain, where from its diffusion over a more considerable space its velocity and transporting power are both diminished, all cross hedges and bridges, though they may check the waters for the moment, are the means of Strangways, Geol. Trans. First Series, vol. v. p. 341. f Encyc. Brit. art. River. Degradation of Land. 65 producing innumerable debacles, when they give way before the pressure exerted upon them. Suppose a bridge arrests the progress of the flood downwards, and, as very frequently happens on small plains, a causeway connects the bridge with the hills on either side, the waters will accumulate, and will finally burst through the least resisting part of the barrier, which will most probably be the bridge. Having once found a vent, the pent-up waters will issue forth with a velocity pro- portioned to the difference of the level and the mass of water, and a debacle will be produced, whose transporting power will be much greater than that of the general force of the flood if no such barrier had existed. It must also be recol- lected, that man, by his contrivances of ditches and drains, prevents the rain-water from remaining the time that it would otherwise do on the slopes of hills, conducting it as he does by numerous free channels into the valleys below ; so that, in a given time, a much greater body of water is collected than could happen in an uncultivated country. He moreover, by dams and banks, often confines a body of river water within narrower channels than it would naturally take; and thus its dispersion over a larger surface being prevented during a freshet, its ordinary velocity is greatly increased, and with this its transporting power. Glaciers. These are large bodies of ice or indurated snow, formed upon land in the cold regions of the atmosphere, which descend into the valleys of mountainous countries; thus frequently presenting the singular appearance of deso- lation amid fertility, of ice amid vegetation. The levels to which glaciers descend depend greatly on the latitude of the place. Thus, in the arctic regions, where the line of perpe- tual snow approaches very nearly to the level of the sea, gla- ciers are produced in lower hills than could be the case in the Alps, where the line of perpetual congelation is much more elevated. So again in the Himalaya range the line of perpetual congelation being higher than in the Alps, the gla- ciers form at higher levels. Glaciers are instruments of the degradation of land, inasmuch as they drive before them and transport such substances as they may have the power to move. In front of glaciers there is usually a pile of rubbish composed of pieces of rock, earth, and trees, which they have forced forward, known in Switzerland by the name of moraine. If there be a line of moraine some distance from the front of the glacier, it is considered that the glacier has retreated to the amount of that distance ; but if there be no other than that which the glacier immediately drives before it, it is considered to be on the increase. Glaciers assist the degradation of land by transporting blocks, often of very large dimensions, into 66 Degradation of Land. lower regions than they could otherwise attain in so short a time. Many glaciers, particularly where they pass beneath precipices, are charged with fallen rubbish, which, as the ice constantly advances, are carried on with it ; and should a pre- cipice occur in the front of the moving mass, they are hurled over with it into the ravines beneath. Such falls are common in the high regions of the Alps, producing, with the rents suddenly formed in the glacier itself, the few interruptions to the dead silence which reigns in those lofty and wild regions. The velocity with which a glacier advances depends on the angle that it makes with the horizon, of course increasing with the steepness of the declivity. A ladder, left by M. de Saussure at the upper end of a glacier, when he first visited the Col du Geant, has lately been discovered in the Mer de Glace, the continuation of the same glacier, and nearly opposite the aiguille named Le Moine. It must therefore have advanced about three leagues since the year 1787*. From some experiments by Chamonix guides, mentioned by Capt. Sherwill, we learn that this rapid pro- gress ceases, as might have been expected, where the decli- vity becomes less in the Mer de Glace itself; for it was there found that a block of rock advanced about two hundred yards in a twelvemonth f. No better proofs could be afforded of the advance of a glacier, the amount of which corresponds with the declivity. It hence appears to follow, that as the declivity remains nearly the same for a long period, the ad- vance or retreat of the lower part of a glacier will correspond with the local variations in climate, which shall produce more or less ice in the higher, or destroy more or less of the glacier in the lower regions. Almost all glacier waters are charged with detritus, the larger portions of which are deposited near the ice, but the lighter particles are transported to considerable distances ; as is, for example, the case with the Arve, which having depo- sited its heavier burden in the valley of Chamonix, carries the lighter parts to its junction with the Rhone, near Geneva. Not unfrequently the turbid glacier waters are carried on, and deposit the detritus in some lake, as is the case with the Rhone, which transports silt, mud, and occasionally pebbles, into the lake of Geneva. The grinding of the glacier against the bottom over which it passes, may perhaps mechanically assist in the work of destruction. In the northern regions glaciers have sometimes such a short distance to pass over before they reach the sea, that they project into it, as has been observed by northern navi- Phil. Mag. and Ann. of Philosophy, Jan. 1831, f H>M> Delivery of Detritus into the Sea. 67 gators. The mass so forced into the sea will have a constant tendency to float, from its inferior specific gravity, and there- fore when detached by any force from the glacier behind, it will be carried away ; thus, forming those icebergs, so well known and so dangerous, in the Northern Atlantic Ocean. Delivery of Detritus into the Sea. We have seen above, that from the action of the atmo- sphere, the melting of snows and glaciers, landslips, and the cutting power of rivers, considerable destruction of dry land is effected. Local circumstances arrest a considerable portion of this detritus; lakes are filled up, and again cut through; low lands are occasionally flooded, and considerable deposits left upon them ; the velocity of the streams diminishes, and with it the power of transport; so that, as previously observed, rivers when short and rapid may carry a large portion of their detritus forward, while, when long, they leave a considerable part of it in their courses. In favourable situations, such as in plains, they will raise their beds, if confined within bounds, that do not either permit a change of course, or a deposit in a new channel. This fact is well observed in Italy, where many plains have been under cultivation for a long period, during which it was always necessary to restrain the rivers within artificial banks, to prevent their range over the culti- vated land, which would otherwise have been devastated by them; so that, in travelling in that country, the road fre- quently passes up hill, over high artificial ridges, upon which the rivers hold their course at a higher level than that of the surrounding country. These artificial ridges are particularly striking on the little plain of Nice, which has been under cul- tivation since the country was settled from the Phocaean co- lony of Marseilles. The height of the latter elevated river- courses is not only due to their antiquity, but to the loose nature of the conglomerate hills behind, which permits an easy transport of the pebbles. The annexed diagram will illustrate this fact : a b, the level of the country, now cultivated, upon which the arti- ficial banks have been gradually raised to c d, in order to protect the cultivated lands from being in- vaded by the detritus of the river or torrent e, which is thus accumulated from f to e. There is a very general system of endeavouring to check this accumulation, and consequent rise of bed, by throwing, when the waters are low, the transported detritus out of the bed e, upon the protecting banks c d. The Po affords a well-known example of this rise of bed, F 2 68 Delivery of Detritus into the Sea. so that it becomes higher than the houses in the city of Fer- rara. In Holland also the same phenomenon is observable, though not on so great a scale; and may always be expected where artificial banks prevent detritus-bearing rivers from changing their beds on plains. Although rivers, in certain situations, raise their beds, in others they deepen them. This arises from two or more streams uniting into one river, when the water does not expose a surface equal to the two previous surfaces, but one very considerably less, the action of the united waters being to deepen their channel ; so that even with a diminished general inclination of the bed, the velocity continues the same, or is even increased. This deepening of beds by the union of rivers is well ex- hibited by the following facts observed in the Po : " About the year 1600, the waters of the Panaro, a very con- siderable river, were added to the Po Grande ; and although it brings along with it in its freshes a vast quantity of sand and mud, it has greatly deepened the whole Tronco di Venezia from the confluence to the sea. This point was clearly ascer- tained by Manfredi about the year 1720, when the inhabitants of the valleys adjacent were alarmed by the project of bring- ing in the waters of the Rheno, which then ran through the Ferrarese. Their fears were overcome, and the Po Grande continues to deepen its channel every day with a prodigious advantage to the navigations ; and there are several extensive marshes which now drain off by it, after having been for ages under water : and it is to be particularly remarked, that the Rheno is the foulest river in its freshes of any river in that country*." It might be supposed that all rivers would, by means of freshes, propel pebbles into the sea. They certainly accom- plish by these means a greater transport than could be effected in the same channels under ordinary circumstances; but during freshes rivers can only be considered as of greater magnitude, and are therefore still subject to the general laws of rivers; a greater body of water tending to deepen the channel ; the velocities, inclinations of beds, and the power of transport still being in proportion to each other. In the beds of torrents, dry, or nearly dry, for the greater part of the year, we see examples of the deepening of river beds in proportion to the volume of water which passes through them, to the inclination of the beds, and to the resisting power of the bottoms and sides. The transport of detritus will also be observed greater or less in proportion to these circum- * Encyc. Brit., art. River. Delivery of Detritus into the Sea. 69 stances : the finer particles being more easy of transport, there are few rivers which, during freshes, do not convey a great quantity of such detritus into the sea : other kinds of detritus will be also transported, if levels permit ; if not, they remain in the interior. Consequently, according to the circumstances already noticed will be the nature of the detritus conveyed to the mouths of rivers. But as circumstances vary in the same river, a deposit of such detritus in these situations also varies, and there may be alternations of clay or marl, and of sand or gravel. If the mouths of rivers be tidal, the river detritus is com- mitted to the charge of the estuary tides, and is dealt with according to the laws by which these are governed. If they be tideless, the whole mass of transported matter will be pro- pelled without check into the seas at the embouchures. Be- tween the extremes of great resistance and non-resistance the variations are so great and depend so much on local circum- stances, as to be of exceedingly difficult classification. The principal variations are produced by the difference in the volume of the discharging rivers, their velocities, and the quantity and quality of the substances they may transport. As a general fact, however, it may be stated that rivers tend to form deltas in tideless, or nearly tideless, seas, or where they can overcome the resistance of tides, currents, and the destructive action of the breakers ; thus increasing the land by their deposit, and splitting into several channels; the su- perficial increase being in proportion to the depth of water into which the rivers discharge themselves. In calculations of the advance of deltas, care has not always been taken to show the general depth of water into which they may have been protruded ; so that a less quantity of transported detritus might expose a larger surface when thrown on a shallow bottom, than a larger quantity in deeper water. The Nile, Danube, Volga, Rhone, and Po, afford us ex- amples of deltas thrown forward into seas, which may, in common terms, be called tideless. As the Nile receives little atmospheric water from Egypt, on which rain seldom fails, the detritus which it brings down must be principally derived from above. This river begins to rise in June, attains its maximum of height namely, twenty-four or twenty-eight feet in August, and then falls till the next May. During a succession of ages, the Nile has transported a great mass of detritus into the Mediterranean, which has accumulated in a delta at the mouth, and is constantly on the increase. It has been calculated, that, as the sea deepens at the rate of a fathom in a mile, and supposing that the deposit is the same as in the 70 Delivery of Detritus into the Sea. Thebais, the addition would amount to a mile and a quarter since the time of Herodotus. According to Girard, the Nile has raised the surface of Upper Egypt about six feet four inches since the commencement of the Christian aera. The quantity of water discharged per annum by this river is esti- mated at 250 times that of the Thames*. The delta is traversed by two main streams, which separate a few miles be- low Cairo ; one descending to Rosetta, the other to Damietta. The present position of the latter city has led to very exag- gerated ideas respecting the rapid increase of this delta. It was supposed that the present town was the same with that which during the first crusade of St. Louis was situated on the sea. Now, as Damietta is two leagues from the sea, it was calculated that this distance had been produced by de- posits from the Nile wilhin about 600 years. It now, how- ever, appears, from the labours of M. Renaud, that after the departure of St. Louis, the Egyptian Emirs, wishing to pre- vent a new invasion on the same side, destroyed Damietta, and founded a new city in the interior, the present Damiettaf . From the effect of the waves and currents, banks are thrown up on the outer edge of the delta, forming lakes, of which those of Menzalen, Bourlos, and that behind Alexandria, are the largest. The delta of the Po advances at a rapid rate, in consequence of the shallow sea into which it is protruded. We are indebted to M. Prony for a very interesting collection of facts, which authorize him to conclude, " First, that at some ancient period, the precise date of which cannot now be ascertained, the waves of the Adriatic washed the walls of Adria. Secondly, that in the twelfth century, before a passage had been opened for the Po at Ficarrolo, on its left or northern bank, the shore had already been removed to the distance of nine or ten thousand metres from Adria. Thirdly, that the extremities of the pro- montories formed by the two principal branches of the Po, before the excavation of the Taglio di Porto Viro, had ex- tended by the year 1600, or in four hundred years, to a medium distance of 18,500 metres beyond Adria ; giving from the year 1200 an average yearly increase of the alluvial land of 25 metres. Fourthly, that the extreme point of the present single promontory, formed by the alluvions of the existing branches, is advanced to between thirty-two and thirty-three thousand metres beyond Adria; whence the average yearly pro- gress is about seventy metres during the last two hundred years, being a greatly more rapid proportion than in former times ;f." * Supplement to Encyc. Brit., art. Physical Geography. f Extraits ties Histcriens Arabes relatifs aux Guerrcs des Croisades. J Prony, as quoted by Cuvier. Dis. sur les Rev. du Globe. Delivery of Detritus into the Sea. 71 The Mississippi, the great drain of so large a portion of North America, may be considered as delivering its waters into a nearly tideless sea. Its delta is very considerable, and little raised above the level of the ocean. During the greatest heights of flood, the fall of the river from New Orleans to the sea, a distance of about one hundred miles, has been calculated at only one inch and a half in a mile. When the waters are low, the fall is scarcely perceptible, the level of the sea being then nearly that of the river at New Orleans*. This river affords a good example of a flood being higher at a distance from the embouchure of a river than at the mouth itself; for the rise of water, during the great freshets, is fifty feet at Natchez, three hundred and eighty miles inland, while at New Orleans it is only thirteen f. Darby has furnished us with a mass of information respect- ing a large portion of the Mississippi's course, and of its delta, from whence very important geological information may be obtained J. It would appear that the Atchafalaya, which now, at a distance of about two hundred and fifty miles from the sea, conducts a large part of the Mississippi's waters into the Gulf of Mexico, did not always form a drain from that river, but that it once constituted a continuation of the Red River, which now flows into the Mississippi. During the autumns of 1807, 1808, 1809, Mr. Darby had frequent opportunities of examining the bed of the Atchafalaya, the waters in which were then at a low state. He found that " the upper stratum invariably consisted of a blueish clay common to the banks of the Mississippi. This is usually followed by a stratum of red ochreous earth peculiar to the Red River, under which the blue clay of the Mississippi was again to be perceived ." From this we may infer, not only that the Red River flowed through the channel of the Atchafalaya, previous to the present course of the Mississippi, but that the latter river preceded the former, and that there have been alternations. From the form of the Mississippi, where the Atchafalaya de- taches itself, an immense quantity of trees brought down by the former are thrown into the latter. About fifty-two years since, these trees began to accumulate and form the "raft." " This mass of timber rises and falls with the water in the river, and at all seasons maintains an equal elevation above the surface. The tales that have been narrated respecting this phenomenon, its having timber of large size, and in many places being compact enough for horses to pass, are entirely * Hall's Travels in North America. f Ibid. % Darby's Geographical Description of the State of Louisiana. Ibid. 72 Delivery of Detritus into the Sea. void of truth. The raft is, in fact, subject to continual change of position, which, superadding its recent formation, ren- ders either the solidity of its structure, or the growth of large timber, impossible. Some small willows and other aquatic bushes are frequently seen among the trees, but are too often destroyed by the shifting of the mass to acquire any con- siderable size. In the fall season, when the waters are low, the surface of the raft is perfectly covered by the most beau- tiful flora, whose varied dyes, and the hum of the honey-bee, seen in thousands, compensate to the traveller for the deep silence and lonely appearance of nature at this remote spot*. Mr. Darby estimated the -cubic contents of the raft, from observations made in 1808, at 286,784,000 cubic feet, consi- dering the breadth of the river = 220 yards, the length of the raft =10 miles, and the depth = 8 feet. The distance be- tween the extremities of the raft was actually more than twenty miles; but, as the whole distance was not filled up by timber, he assumed ten miles as near the truth. Rafts of this description, but of less size, occur in other parts of the Mississippi or its great tributaries. The banks are destroyed by the currents, and large collections of trees are suddenly hurled into the stream. Captain Hall was present when a large mass of earth, loaded with trees, suddenly fell into the Missouri, and a larger mass had been detached a short time previous to his arrival*!'. There are few rivers whose course is more instructive than the Mississippi, as man has not yet effected many changes on its banks ; and we thus contemplate great natural operations, such as cannot be so well observed in those which have been more or less under his dominion for a series of ages. Its course is so long, and through such various climates, that the freshets or floods produced in one tributary are over before they com- mence in another : hence arise those frequent deposits of de- tritus at the mouths of the tributaries. These latter have their waters forced back, and rendered, to a certain distance, stag- nant by the rush of the flood across their embouchures, and the consequence is a deposit, which remains until the annual floods in the tributary remove itj. When the Ohio is in flood, it stagnates the waters of the Mississippi for many leagues; when the Mississippi is in flood, it dams up the waters of the Ohio for seventy miles . Darby remarks that the Mississippi, in its long course from * Darby's Geographical Description of the State of Louisiana, p. 65. f Hall's Travels in North America. J James, Exp. to Rocky Mountains. Hall's Travels in North America, vol. iii. p. 370. The same author notices the curious mixture of the Missouri waters with those of the Missis- sippi, the former charged with detritus and wood, the latter beautifully clear. Delivery of Detritus into the Sea. 73 the embouchure of the Ohio to Baton Rogue, washes the eastern bluffs, which it tends to carry away and destroy, and that, even to the sea, it does not come in contact with the western side of the valley through which it flows. He at- tributes this, with great probability, to the deposits brought down by the great tributaries, which all enter the Mississippi from the west, and thus accumulate detritus on that side. Notwithstanding the general tendency of the river to the eastward, innumerable smaller changes of channel take place. Thus winding courses shorten themselves, by cutting through isthmuses, the tendency of the winding currents being to de- stroy the barriers between them, as may be observed in nume- rous rivers flowing through plains. New obstacles present themselves; new sinuosities of channel are produced; trees growing upon old alluvial deposits of the river are carried away ; and new vegetation springs up upon the recent alluvium, to be again removed by a new change of channel. During these various minor changes of bed, the degradation of the higher lands supplies a great abundance of detritus, which not only tends to raise the general level of the valley, by deposits over the low lands at floods, but is carried forward towards the sea, and forms an immense delta, composed of clay, mud, and sn% mixed with a large proportion of drifted trees and other vegetable substances. The delta is divided into innumerable lakes, marshes, and streams, inhabited by a multitude of alligators. The main stream of the Mississippi will be observed to project forward, on all good maps, in a singular manner. The detritus brought down by it produces constant alterations, which require all the attention of the pilots. According to Captain Hall, mil- lions of logs, or trunks of trees, are brought down during freshets, and carried several miles into the sea, so that it is difficult to navigate among them. When not carried to sea, these logs are bound together by a kind of cane, which retards the river and collects mud. The same author considers " that a belt of uninhabitable country, from fifty to one hundred miles in width, fringes the edge of the whole of that part of the coast*." It has been supposed that the mud of the Mississippi, sink- ing through one foot of water in an hour, would be carried by the Gulf Stream a distance of fifteen hundred miles before it fell to the depth of five hundred feet, estimating the velocity of the current at three miles per hourf . This has been employed as an argument in favour of the great extent over which river de- * Hall's Travels in North America, vol. iii. p. 340. t Babbage, Economy of Manufactures, 2nd edition, p. 51. 74? Delivery of Detritus into the Sea. tritus may be transported ; and there could be little doubt of the correctness of the calculation if the data on which it is founded were also correct. There is, however, every reason to conclude that the detritus of the Mississippi never finds its way into the Gulf Stream, but that it is deposited on the western shores of Florida, and in the northern portion of the Gulf of Mexico, generally. There is in fact no current to carry the detritus of the Mississippi into the Gulf Stream, as is well known to those who navigate the northern portion of the Mexican Gulf. The waters, moreover, of the Gulf Stream, even so far southward as the Florida reefs, are among the clearest of sea waters yet discovered ; indeed, the depth at which objects can be seen through them has always been re- marked with surprise by those who have visited that part of the world for the first time. This great clearness continues to the Bahama Bank and Islands. It might also be shown that three miles per hour is an over-estimate of the mean an- nual velocity of the Gulf Stream. The mouth of the Ganges will afford us an example of the power of rivers to force forward deltas where no violent cur- rents run across their embouchures, and where the body of water, particularly during freshets, is very considerable, even when such rivers are opposed to considerable tides. Major Rennel described this delta in 1781, so that probably, since his account was written, very material changes have been effected; yet as all these changes are likely to have been made in the same manner, Major Rennel's description will always be valuable, as showing the mode in which they have been carried on. The delta of the Ganges commences about two hundred and twenty miles from the sea in a direct line ; or nearly three hundred, if the distance be reckoned along the windings of the river. The Ganges makes frequent windings, like many other rivers, and thus considerable changes of its bed take place, the opposing bends cutting through the isthmus between them, as in the Mississippi. During the eleven years which Major Rennel remained in India, the head of the Jel- linghy river was gradually removed three-quarters of a mile further down. He also states, that " there are not wanting instances of a total change of course in some of the Bengal rivers. The Cosa (equal to the Rhine) once ran by Purneah, and joined the Ganges opposite Rajenal. Its junction is now nearly forty-five miles higher up. Gour, the ancient capital of Bengal, once stood on the Ganges." It seems probable that the Ganges once ran in the line now occupied by the lakes and morasses between Nattore and Jaffiergunge*. * Rennel, Phil. Trans. 1781. Delivery of Detritus into the Sea. 75 This delta is constantly on the increase. The quantity of detritus must be abundant, for the sea into which it is borne is by no means shallow, the depths being considerable. The amount of detritus held in mechanical suspension by the Ganges has, however, been greatly exaggerated ; for instead of being equal to one fourth of the volume of water discharged, as was supposed by Major Kennel, it appears that 2^ per cent, is the utmost that can be allowed.* The usual checks are produced by the tide, but during the freshets the ebb and flow are little felt, except near the sea. During these times, therefore, the advance of the delta is most considerable, the quantity of transported detritus being then greatest, and the resistance of the sea at its minimum. The sea may ravage the new lands, and apparently remove them for a time ; but even- tually they must gain, even from the accumulative power of the breakers themselves, which also equalize the depths, by conveying the detritus to a short distance : thus rendering the sea more shallow, and consequently more easily filled up by river-borne detritus. Coarse gravel transported by the Ganges does not approach the sea within four hundred miles, and consequently does not occur within one hundred and eighty miles of the commence- ment of the delta ; therefore it would appear that during the present order of things the Ganges has not transported coarse gravel into the sea at its present relative level. A great por- tion of the periodical inundations, represented as flowing on the level lands at the rate of half a mile per hour, has been attributed to the rains which fall on the low lands of India, as it has a blackish tint, from being long almost stagnant among vegetables of different kinds. Small obstacles accu- mulate, as might be expected, very considerable banks and islands ; a large tree arrested in its progress downwards, of even a sunken boat, being sufficient for the purpose. As these islands are quickly formed, so are they easily swept away by any change in the mighty current. At the junction of the Ganges and Burrampooter below Luckipoor, there is a large gulf in which the water is scarcely brackish, even at the extremity of the islands, some of which are described by Major Rennel as equalling the Isle of Wight in size and fertility. The sea is represented as per- fectly fresh to the distance of several leagues from this place during the rainy season. * Gleanings of Science, vol. iii. p. 185 ; Calcutta 1831. On the other hand, it appears that the average discharge of water from the Ganges is much greater than was estimated by Major Rennel, being about 500,000 cubic feet per second. Ibid. 76 Delivery of Detritus into the Sea. It will be seen that deltas not only occur in situations where there is neither tide nor considerable current to prevent a great accumulation of new land, as at the embouchures of the Nile and Po, but also where the tides are small (Mississippi), and even where they are considerable (Ganges). The deltas thus produced are no doubt large, and the amount of animal and vegetable matter which they may entomb very consider- able ; but we must not be led away by measurements and comparisons with the length, breadth, or superficies of districts with which we may be familiar, and which we may, from habit, consider important. They should be regarded with reference to their relative importance as portions of dry land, when it will be seen that they do not expose so considerable a surface as might at first be supposed. The augmentation of deltas will correspond with the detritus carried forward to the embouchures of rivers; and it will be obvious that the facility of the transport will depend, all other circumstances being the same, on the length and fall of the channel. Now the course will be shortest and the declivity greatest at the commence- ment of the delta, and therefore it might be concluded that deltas would accumulate heavier materials, and increase most rapidly at the first periods of their formation, and that this increase would gradually diminish as the fall of the river channel became less, and its length increased ; without reckon- ing on the innumerable checks given to the stream by the in- creasing divisions in the delta. It may also be supposed that the detritus from the high lands would become gradually less, from the equalization of levels, and the fewer asperities that meteoric agents have to act on. Should these remarks, made under the supposition of the non-interference of man, be correct, it will follow that the increase of deltas would gradu- ally diminish if these were the only circumstances which regulated them. But it must be admitted that heavy rains, more particularly in tropical countries, would tend to cut up and destroy the delta itself, (still accumulating at its highest parts,) and force the detritus into the sea. The dense aquatic vegetation, common at the extremities of deltas, would render this transport difficult, yet still some detritus would escape. The amount of such additions to the outskirts of the new land would not, perhaps, be considerable, but it would corre- spond with the size of the delta, and consequently the larger this was, the greater would be the increase thus derived. Between those rivers, such as the Ganges, which obtrude deltas into tidal seas, and those which have large open em- bouchures, such as the Maranon, St. Lawrence, Tagus, and Thames, there are such variations, produced by local causes, Action of the Sea on Coasts. 77 that it would be exceedingly difficult, even if useful, to classify them. In the delivery of their detritus, therefore, such rivers will either produce deltas or estuaries at their embouchures, as they either partake of the characters of the Ganges or the St. Lawrence ; if of the latter, the detritus will be dealt with according to the mode of deposit or transport in estuaries. Action of the Sea on Coasts. Breakers, or the waves falling on sea beaches or coasts, are continual and powerful agents of destruction in some situations; while in others they pile up barriers against themselves. Their destructive influence is principally felt when the rocks on which they are discharged are composed of soft materials, and rise somewhat abruptly above the level of the sea. Their pro- tecting influence is most commonly experienced in front of low level lands, and across the mouths of valleys, on each side of which a hard rocky point supports the ends of a beach. The destruction of coasts of equal hardness almost always bears a proportion to the extent of open sea to which such coasts are exposed, all other circumstances being the same. The configuration of most coasts will be seen to be deter- mined by the hardness of the rocks composing them ; the softer strata giving way before the battering power of the breakers, while the harder rocks preserve their places for a greater length of time. If the rocks forming a coast be stra- tified, much depends on the dip of the strata relatively to the breakers. Thus, in many situations on the southern coasts of Devon and Cornwall, the slaty rocks dip in such a manner towards the sea, that the waves have never effected more than the removal of some loose superficial matter, the same that covers all the hills in the vicinity. In fact, a skilful engineer could not have protected the coast better than has been accom- plished by the dip of the strata. The destructive power in other situations is well known; and of this, the eastern coast of our island presents abundant proof, where very considerable encroachments of the sea have been recorded within the lapse of a few centuries. The substances so forced away by the action of the breakers will be acted on according to their weight, form, and solidity. The tides or currents will remove so much of them as they are able to transport, and the rest will remain on the shore within the immediate influence of the breakers, which constantly tend to grind them down into smaller portions, and finally into sand. In the destruction of a cliffof unequal hardness, it not un- frequently happens, that the harder portions, when large, such 78 Action of the Sea on Coasts. as many concretions in sandstones and marls, or blocks of in- durated strata, remain at the base of the cliff, and in a great measure protect it from the more powerful effects of the breakers, as will be seen in the annexed figure. a, a defence of blocks, derived from the hard strata b, and the Fig. 16. concretions c. Among the un stratified rocks, great variety of hardness prevails, so that they frequently present an uneven front to the sea, resulting from the quicker decomposition and destruction of some parts than of others. Veins of one sub- stance, or rock, traversing another are generally of different textures and solidity from that which they cut, and con- sequently nothing is more frequent, on sea snores, than to ob- serve them either standing out in relief or hollowed into coves. When a shingle or sandy beach, but more particularly the former, is partly torn up and held in temporary mechanical suspension by the breakers during a heavy gale, the action of the waves is very considerable, even on the hardest rocks, so as to scoop them out near the ordinary level of the sea. In exposed situations, the hardest rocks are often drilled into holes or caverns, from the force of the broken wave being driven, by local circumstances, more in one direction than another, or from the inferior hardness of different portions of the rock. The most beautiful of ocean caverns, Fingal's Cave in Staffa, owes its existence to the circumstance of the basaltic columns being jointed in that place, while the general character is to be without divisions in the columns*. After the sea has formed a cavern, the vault of which does not rise above high water, it sometimes works its way upwards at the inmost extremity, partly by means of the compressed air held between each wave as it rolls into the cave. Of this kind of cavern Bosheston Mere in South Wales is an example on the large scale. It is formed through strata of carboni- ferous limestone, and the noise caused by the blast of com- pressed air and sea water upwards is heard at a considerable distance. The protecting influence of breakers is shown in long lines of shingle and sandy beaches, which often defend low and marshy land, particularly at the mouths of valleys, from the destructive power of the sea. MacCulloch, Western Islands of Scotland. Shingle Beaches. 79 Shingle Beaches. In the case of shingle beaches, it will be observed, that during a heavy gale every breaker is more or less charged with the materials composing the beach; the shingles are forced forward as far as the broken wave can reach, and in their shock against the beach drive others before them that were not held in momentary mechanical suspension by the breaker. By these means, and particularly at the greatest height of the tide, the shingles are projected on the land beyond the reach of retiring waves. Heavy gales and high tides combined seem to produce the highest beaches ; they do indeed sometimes cause breaches in the rampart they have raised against themselves, but they quickly repair them. The great accumulation of beach upon the land being effected at high water, the ebb tide, it is clear, cannot deprive the land of what it has gained. In moderate weather, and during neap tides, various little lines of beach are formed, which are swept away by a heavy gale ; and when these little beaches are so obliterated, it might be supposed, by a casual observer, that the sea was diminishing the beach ; but attention will show that the shingles of the lines, so apparently swept away, are but accumulated elsewhere. These remarks do not apply to situations where the sea, during gales, has access to cliffs or piers, from whence there might be a retiring wave carrying all before it ; but to such situations and they are abundant where the breakers meet with no resistance of that kind, and strike nothing but the more or less inclined plane of a shingle beach. Even in cases where the waves in heavy gales and high tides do reach cliffs, and for the time remove shingle beaches, it is curious to see how soon these latter are restored, when the weather moderates, and when the breakers, in con- sequence of a diminished projecting force, cease to recoil from the cliff behind. Shingle beaches travel in the direction of the prevalent winds, or those which produce the greatest breakers : of this there are abundant examples on our own southern coast, where the prevalent winds being W. or S.W., the beaches travel eastward until arrested by some projecting land, when the sea forms a barrier against itself, and not unfrequently leaves a space between it and the cliff' which it formerly cut : this space, under favourable circumstances, is covered by ve- getation, suited to such a situation, even the cliff being some- times studded with sea-side plants, when they can find root. Works are sometimes constructed to arrest beaches, either to protect land behind or to prevent their passage round pier 80 Shingle Beaches. heads into artificial harbours; and thus engineers are prac- tically aware of their travelling power in the direction of cer- tain winds. This progressive march of beaches is far from rapid, and can only be in proportion to the greater power or duration of one wind to another ; moreover, the pebbles be- come comminuted in their passage, and thus the harder can only travel to considerable distances. The Chesil Bank, connecting the Isle of Portland with the main land, is about sixteen miles long, and, as a general fact, it may be stated that the pebbles increase in size from west to east. It protects land which has, evidently, never been ex- posed to the destructive power of the Atlantic swell and seas, which break with great fury against the bank ; for the land behind is composed of soft and easily disintegrated strata, which would speedily give way before such a power. Perhaps a gradual sinking of the land might produce the present ap- pearances ; for though the sea would have attacked the land when the relative levels were different, the form of the bay, and the projection of the Isle of Portland, would soon cause a beach to be formed, which would rise as the land sunk, so that, finally, no traces of a back cliff could be observed. Under this hypothesis, Portland would not have formed an island, but merely the projecting point of a bay, which, with its exposure, would soon have accumulated the beach required. It may be remarked, that this supposed gradual sinking of the land is in accordance with appearances more westward on the same coast, where the facts presented seem to require this explanation. The sea separates the Chesil Bank from the land for about half its length, so that, for about eight miles, it forms a shingle ridge in the sea. The effects of the waves, however, on either side are very unequal; on the western side the propelling and piling influence is consider- able, while on the eastern, or that part between the bank and the main land, it is of trifling importance. The following is a section. Fig. 17. a, the Chesil Bank : 6, the water called the Fleet : c, small cliffs formed by the waves of the Fleet and land springs : d, various soft rocks of the oolite formation, protected from de- struction by the Chesil bank a : e, the open sea. Another curious example of land protected by a shingle bank occurs on the southern coast of Devon, and is remark- able, as it shows that the sea, at its present relative level with Shingle Beaches. 8 1 the land, has never reached the land behind the beach, a fact that will admit of the same explanation as that previously given for the Chesil Bank. At the bottom of Start Bay, and for the distance of about five or six miles, a considerable bank, principally composed of small quartz pebbles, has been thrown up by the sea. The line of coast faces the east. Between Tor Cross and Beeson Cellar, a point of land comes within the reach of the breakers ; but here, as well as elsewhere behind the bank, the land has evidently gained on the sea, or, in other words, the latter has piled up a barrier which prevents its reaching the c\ifi\ as it once did, even during heavy gales. This bank, generally known as Slapton Sands, though composed wholly of small pebbles, protects and blocks up the mouths of five valleys. Between Slapton Sands (properly so called) there is a fresh-water lake, divided into two at Slapton Bridge, where the waters of the northern lake drain into the southern. The northern portion is nearly silted up by the detritus borne down by a river that drains a few miles of country, and is nearly covered by bulrushes and other aquatic plants. The southern and larger portion is open, and of many acres in extent. The waters are supplied by the rivers behind, and commonly percolate through the pebbles into the sea. When, however, the tides are high, and the waters kept up by heavy gales, it sometimes happens that, the relative levels being altered, the sea-water passes through the shingles into the lake, and renders it to a certain extent brackish. This usually happens in winter; but, generally speaking, the relative levels are such, that the lake drains into the sea and remains perfectly fresh. It contains a great abundance of trout, perch, pike, roach, and flounders. The presence of the latter, a marine or estuary fish, shows that it can be gradually accustomed to fresh water. The percolation of the sea through the pebbles, during heavy gales, does not seem to injure the fresh-water fish ; but when a breach was made through this beach during the gale of November 1824, they were nearly all killed by the sudden influx of the sea. Those which escaped up the streams were sufficient, in five years, again to stock the lake abundantly. The breach made through Slapton Sands continued open for nearly a year, becoming gradually smaller. The complete restoration of the sands was hastened by throwing a few bags, filled with shingles, into the gap, upon which two or three gales soon piled up a heavy beach. The old bank must have remained undisturbed for a long period ; for vegetation had become active upon it, as we see by those portions which remain uninjured, where turf and even furze-bushes have established themselves upon the shingles. 82 Shingle Beaches. The above exhibits a section of the beach and lake. a, the sea which throws up the beach b : c, the fresh-water lake be- hind the beach : d, several feet in depth of pieces of slate and sand derived from the slate-rocks e. This diagram shows that the sea could not have acted upon the hill d e since the accumulation of the loose substances d, which it would have instantly removed. The great size of rock fragments moved by the action of the breakers attests their power. During heavy gales, blocks of many tons in weight have been forced from their places ; and others, even squared and bolted together in the form of piers and jetties, have been torn asunder by the battering power of the waves. During the gale of November 1824, which ravaged a considerable part of the southern coast of England, a square block, from a ton and a half to two tons in weight, strongly trenailed down, was torn away from a jetty at Cyme Regis, and tossed upwards by the force of a breaker. Mr. Harris, of Plymouth, informs me that, during the same severe gales, and at the commencement of 1829, blocks of limestone and granite, from 2 to 5 tons in weight were washed about on the Breakwater like pebbles ; about 300 tons, in blocks of these dimensions, being carried a distance of 200 feet, and up the inclined plane of the Breakwater. These blocks were thrown over on the other side, where they re- mained, after the gale, scattered in various directions. A block of limestone, weighing 7 tons, was washed round the western extremity of the Breakwater, and carried 150 feet. Two or three blocks of this size were washed about. At the Pier in Bovey Sand Bay, on the east side of Plymouth Sound, a piece of masonry may be now seen, which was washed back about 10 feet, being, at the time it was struck, 16 feet above the level of an 18-feet spring tide. This piece of masonry weighs about 7 tons, and consists of a few blocks of limestone cemented together and covered by a large block of granite. The mass was dovetailed into, and formed part of, a parapet facing the sea. At the Scilly Islands the blocks of granite that fall from the cliffs are ground by attrition into great boulders, which become the sport of the heavy Atlantic seas in tempestuous weather. The effect produced by a heavy sea must depend consider- ably on the form of the block on which the sea acts. Thus, a flat front would present the greatest resistance to the shock, and the mass so struck would have a tendency to be more Shingle Beaches. 83 easily moved than a rounded mass, if it were not that the re- sistance to removal offered at its base, is very considerably greater than in a rounded mass. The wedging power of the breakers is also very consider- able where heavy blocks of difficult removal are mixed with smaller stones easily transported. A beach of this nature sometimes acquires much solidity, as the smaller pieces are often forced among the larger so tightly as to require very great force, and even fracture, before they can be taken out. It would appear that, though shingle beaches, or those composed partly of pebbles and partly of larger masses, may be moved in the direction of the predominating and heaviest breakers, we have no evidence of their being transported outwards, or into the depths of the ocean, but that, on the contrary, the waves of the sea strive to throw them upon the land ; and this, not only in the case of substances derived from the land, but also in that of corals, shells, and marine plants which have been produced in the sea itself. In tropical countries it is found that many coral reefs and islands are defended on their windward sides by beaches of coral shingles, and even large fragments of coral. Lieut.-Col. Hamilton Smith informs me, that, during a hurricane which he witnessed at Cura^oa in September 1807, large pieces of coral were torn up from a depth often fathoms, and thrown on the bank uniting Punta Brava with the land. Beaches composed wholly or entirely of comminuted marine shells are not un- common, and will be noticed in the sequel. The seaward front of most shingle beaches, particularly when they defend tracts of flat country, is bounded by a line along the edge of the beach ; above this line the beach gene- rally makes a considerable angle with the sands, in cases of sandy flats. In cases where shingle beaches are not entirely quitted by the tide, sandy, shelly, or very fine gravel sound- ings are commonly obtained at a short distance from the shore, unless the bottom be rocky. It would appear that, if the present continents or islands were elevated above, or depressed beneath, the present ocean-level, shingle beaches would be found to fringe the land, but not to extend far seaward*. * We should be careful, when we obtain shingles in various soundings, to consider that the probability is as great of finding pebbles at the bottom of the sea as on the dry land ; and that their presence there, is no proof that they have been transported by existing currents, unless it can be shown that the velocity of the existing current is sufficient to transport such detritus, and that the direction of the current is that which would carry the fragments from the known place of the parent rock. Without attention to this circumstance, it might be supposed that the small shingles, covering the bottom of the newly discovered bank off the north-west coast of Ireland, were carried there by the present currents, when they are quite as likely to have been otherwise cj 2 84? Saudi/ Beaches. Sandy Beaches. The observations made respecting shingle beaches apply, in a great measure, to those composed of sand. The sand is derived either from the detritus borne down by rivers, from the attrition of sea-shore shingles against each other, or im- mediately from the sand and sandstones of the land. The breakers have the same tendency to force sand upon the land, as was observed in the case of shingles ; but, being so much lighter than the latter, sand can be transported by coast tides or currents whose velocity would be insufficient to move shingles. On the other hand, however, smaller forces and bodies of water can throw sand on the shore. The spray that could not transport a pebble can carry sand, and thus this substance can be, and is, conveyed far beyond situations where the reflux of a wave can be felt. When the tide is low, or the sea less agitated, sand, dried by the sun or winds, is transported by the latter to great distances, so that whole districts of once fertile land have been overwhelmed by it. Such transported sand, when sufficient to form hills, is known by the name of dunes, more or less common behind sandy shores or beaches over the globe. A striking example of the progress of such drifted sand inland, is to be found in the Bay of Biscay, on the eastern shore of which the sands have overwhelmed and are continuing to cover large tracts of country. Cuvier states the advance of these dunes as per- fectly irresistible, forcing lakes of fresh water before them, derived from the rains which cannot find a passage into the sea. Forests, cultivated lands, and houses disappear beneath them. Many villages noticed in the middle ages have been covered, and in the department of the Landes alone, ten are now threatened with destruction. " One of these villages, named Mimisan, has been striving for twenty years against them ; and one sand-hill, more than sixty feet high, may be said to be seen advancing. In 1802, the lakes invaded five fine farms belonging to Saint Julien ; they have long since covered a Roman causeway which led from Bourdeaux to Bayonne, and which was seen, about forty years since, when the waters were low. The Adour, which was once known to flow by Vieux Boucaut, and fall into the sea at Cap Breton, is now turned aside more than a thousand toises *." produced. That they are not now rolled about to any extent, is evident from the serpulae and other marine productions attached to some of them brought up by Captain Vidal, during his survey, by the arming of the sounding lead. * Cuvier, Dis. sur les Rev. du Globe. Sandy Beaches* 85 M. Bremontier calculated that these dunes advance at the rate of sixty, and even seventy-two, feet per annum. Under favourable circumstances, sands, transported from a beach into the interior, become consolidated : of this a good example is found on the north coast of Cornwall, where the matter thrown up is formed from comminuted sea-shells, and the consolidation is principally effected by means of oxide of iron. From the drift having taken place at different times, this recent calcareous sandstone is stratified, with occasionally interposed vegetable remains. Houses have been overwhelmed, and human remains entombed where churchyards have ex- isted. Mr. Carne describes a pot of old coins dug out of it. The induration of this rock is so considerable, that holes are drilled in it at New Kay, for the purpose of securing vessels to the cliff. It is also used for architectural purposes, and according to Dr. Paris the church of Crantock is built with it. The same author states that the high cliffs of this recent rock, which extend several miles in Fistrel Bay, are occa- sionally intersected with veins of breccia. " In the cavities, calcareous stalactites of rude appearance, opaque, and of a gray colour, hang suspended." " The beach is covered with disjointed fragments, which have been detached from the cliff above, many of which weigh two or three tons*." Indurated dunes occur in various parts of the world : they have been noticed by Peron in New Holland ; and the rock in which the human remains of Guadaloupe have been found would appear to be similar. These latter are discovered at the Port du Moule, in an indurated beach composed of com- minuted shells and corals. The specimen in the British Museum is formed of coral and small pieces of compact lime- stone, and in it Mr. Konig has observed Millepora miniacea, madrepores, and shells referred to Helix acuta and Turbo Pica. According to Cuvier, the specimen in the Jardin du Roi, at Paris, exhibits a gangue of travertin containing shells * Paris, Geol. Trans, of Cornwall. Not only sands but shingle beaches are sometimes indurated. Captain Beaufort describes a plain several miles in length, near Selinty, coast of Karamania, as bounded by a gravel beach, which has become consolidated from the top of the crest to some distance into the sea; the consolidation extending to the depth of from one to two feet, and being generally covered with loose sand and gravel, so that it is not easily observed. The pebbles are cemented by a calcareous paste, and the whole is so hard, that a blow " more frequently fractures even the quartz pebbles than dislodges them from their bed." Other beaches of the like kind, but on a smaller scale, were observed on other parts of the coasts of Asia Minor and of Greece. Rocky ledges of a similar nature occur to the westward of Side, partly above and partly under the water. They con- tain broken tiles, shells, bits of wood, and other rubbish. They are very hard, and are cemented by calcareous matter, probably derived from some calcareous slate in the vicinity. Beaufort's Karamania, pp. 182 and 185. 86 Sandy Beaches. of the neighbouring sea, and terrestrial shells, especially the Bulimus gnadaloupensis of Ferussac. Near Messina, loose sand becomes consolidated on the beach, and is used for build- ing. It is stated that the cavities thus made are again filled up by sand, which becomes consolidated and used in its turn. Von Buch notices a limestone and sandstone formation of a similar kind on the coast of the Great Canary. The violent north wind of summer raises the light fragments of broken shells, and little rounded grains of trachyte and basalt, (be- tween the town and Isleta,) and drives them over the narrow tongue of land of Guanateme; thus forming dunes about 30 or 40 feet in height. The waters behind these dunes act on the sand, uniting it into a compact mass, which is broken away during the ebb. These waters are, during the greater part of the year, at a temperature of 77 F., which greatly promotes their action on the calcareous particles. The sandstone is produced on the Confidal shore of the Isleta, but not at Catalina, which is exposed to the N.E. wind. It is a real oolite, most of the grains being round and calcareous, sur- rounding a nucleus of basalt, trachyte, or the fragment of a shell. Similar rocks containing shells, resembling those still living on the coasts, occur at the height of 300 or 400 feet above the level of the sea, showing that a difference of level to that amount has taken place since the formation of these rocks *. Dr. Clarke Abel describes a large bank, rising from the sea to the height of about a hundred feet, to the eastward of Simon's Town, Cape of Good Hope, formed of shell and sand, thrown up by the S.E. wind. In this he discovered singular cylindrical bodies, which resembled bones bleached by the air. " On a closer examination, many of them are found to be branched; and others are discovered rising through the soil, and ramifying from a stem beneath, thicker than themselves. Their vegetable origin immediately sug- gests itself!, and is confirmed by a further inquiry. They are seldom solid, their centres being either hollow or filled with a blackish granular substance, which in many specimens, except in colour, resembles the substance called roestone by minera- logists. Their outer crust is chiefly composed of a large pro- portion of sand and a small proportion of calcareous matter, and in many specimens contains fragments of ironstone and quartz an inch square. That they are really incrustations formed on vegetables which have afterwards decayed, is proved by the different degrees of change which the internal parts of different specimens have undergone. In some the organiza- German Transl. of Manual. Sandy Beaches. 87 tion of the plant sufficiently remains to leave its nature un- equivocal ; and near the sea the very commencement of the process of incrustation may be witnessed on the large Fuci which strew the shore*." Peron's previous description of the change undergone by vegetable substances in similar situations on the coast of Aus- tralia, is nearly the same. He considers that the shells un- dergo decomposition, and form a cement with the sand ; and that the vegetables become altered and finally replaced by this sandstone, leaving nothing to show its origin but its general form. On our coasts the sands thrown on shore by the ac- tion of the sea, and afterwards drifted by the winds, are often comparatively considerable. Mr. Ritchie describes a di- strict of ten square miles in Morayshire, once termed the Gra- nary of Moray, as having been overwhelmed. " This barren waste may be considered as hilly; the accumulation of sand composing these hills frequently varying in their height, and changing their situation f ." The following account by Mr. Macgillivray affords an ad- ditional example of the tendency of coast-seas to throw even the substances formed in them upon the land. " The bottom of the sea, along the whole west coast of the Outer Hebrides, from Barray Head to the Butt of the Lewis, appears to consist of sand. Along the shores of these islands this sand appears here and there in patches of several miles, separated by in- tervals of rock of equal or greater extent. In some places the sandy shores are flat, or very gently sloping, forming what are here called Fords ; in others, behind the beach, there is an accumulation of sand to the height of from twenty to sixty feet, formed into hillocks. This sand is constantly drifting ; and in some places islands have been formed by the removal of isthmi. The parts immediately behind the beach are also liable to be inundated by the sand ; and in this manner most of the islands have suffered very considerable damage The sand consists almost entirely of comminuted shells, ap- parently of the species which are found in the neighbouring seas. It is rather coarse in the grain ; but during high winds, by the rubbing of its particles on each other, a sort of dust is formed, which at a distance resembles smoke, and which, in the island of Berneray, I have seen driven into the sea to the distance of upwards of two miles, appearing like a thin white fog f." It would be useless to accumulate notices of these various * Clarke Abel, Voyage to China, p. 308. t Notes appended to Cuvier's Theory of the Earth, by Jameson. 1 Ibid. 88 Sandy Reaches. sand drifts, which often contain seams of vegetable matter that have been successively covered up, and of which sections are afforded*. The action of the waves round coasts tends to disturb the bottom at certain depths, and to move the shells, sands, and other substances, of which this bottom is com- posed, towards the land. The exact depth to which the mov- ing action of waves extends, seems never to have been very accurately estimated ; indeed, when we consider that the power of the wave is continually varying, such an estimate becomes exceedingly difficult. Ninety feet, or fifteen fathoms, has been sometimes considered as the limit, in depth, to which this disturbing power extends; but this requires confirmation. Around coasts and on shores which do riot much exceed ten or twelve fathoms, the action of the waves is very apparent in the discoloration of the water during heavy gales. This tur- bid character of the sea is due to the moving power of the waves on the bottom, and becomes more marked as the water becomes more shallow, either in approaching the land or over shoals. The transporting power of the waves will therefore be in proportion to the depth of water beneath them, the trans- port being greatest in the shallowest places. The waves will tend to throw substances on coasts, because the off-shore wind produces smaller waves than the wind blowing upon the land. On shoals distant from the land, the effect will be somewhat different, and the piling or propelling power will be greatest on the side of the prevalent or more violent winds. Shoals will be also liable to shift, as the turbid waters on the crown of a shoal will be forced over on the lee side. Accordingly, we do find that shoals shift, more particularly when near the surface, unless there be an equal counteracting effect in a cur- rent or tide. We may, in some measure, learn the effects of waves at different depths, from the form of the outer talus of the Digue, or Breakwater, at Cherbourg, where they have, to a certain extent, arranged the stones, four-fifths of which * Not only are sand-hills thrown up by the sea, but also by the waves of extensive fresh-water lakes. Dr. Bigsby (Journal of Science, vol. xviii.) and Capt. Bayfield (Trans, of Lit. and Hist. Soc. of Quebec, vol. i.) both remark the beaches thrown up in the bays of Lake Superior. The latter author notices some curious lines of ancient beaches rising one above the other, like the seats of an amphitheatre, in valleys at some distance from the shores of the present lake, and hence infers that the level of Lake Su- perior has fallen. Similar beaches are observable on other lakes of North America. Capt. Bayfield noticed seven ridges of shingle, rising above each other, near Cabot's Head, Lake Huron : the highest was overgrown with spruce firs; the second had bushes or smaller trees of the same kind; the third, shrubs and flowers ; the fourth, lichens and mosses ; the rest being bare of vegetation. Dr. Bigsby and Capt. Bayfield also notice the sand- hills thrown up on the shores of Lake Superior by the prevailing N.VV. winds. Sandy Beaches. 89 are small, in the manner best fitted to resist themselves. Ac- cording to M. Cachin, there are four kinds of taluses, ar- ranged one beneath the other. The upper line of talus, being only touched by the higher break of the waves, presents a height proportioned to its base, as 100 is to 185. The se- cond line, comprising the whole distance between the line of high and low water at the equinoxes, and thus exposed to the battering power of the breakers during the whole flood and ebb, is consequently the most inclined, and its height is to its. base as 100 to 540. The third line, being below the lowest water at the equinoxes, is only acted upon during the first of the flood or the last of the ebb. Its height is to its base as 100 to 302. The fourth line, or the base of all, not being acted on by the waves, maintains a talus, of which the height is to the base as 100 to 125*. The action of waves on coasts is not only exhibited by pi- ling up detritus in the direction of their greatest force on the shore, by which embouchures of rivers are turned on one side, but also by heaping up bars, as they are termed, even at their mouths, rendering their navigation dangerous, and in many instances preventing it altogether ; though, behind these bar- riers, the rivers may have considerable depth and breadth. In some situations these bars are partially dry at low water, at others they are never uncovered, though rendered visible by the breaking of a furious surf. To produce examples would be useless, as they are common in all parts of the world. In many cases, the bars are liable to shift, particularly after a gale of wind, so that vessels are frequently lost by keeping the direction of the old channels ; and it requires the constant at- tention of pilots to be aware of the exact position of the new passages. When the rivers are small, the force of the waves frequently blocks up their embouchures, and artificial means are neces- sary to permit the escape of the pent-up waters, that would otherwise form a lake in the low country behind. If the dam be a shingle beach, the water usually percolates through it; but if composed of sand, the water will accumulate until its level enables it to cut a passage through the barrier and es- cape. This done, the breach will be again repaired, and an- other accumulation of water take place behind, and so on. But, in the mean time, the level of the low land would rise, first, by deposition from the river waters; and, secondly, from the sand blown over the bank. In such an alluvial land there would probably be found remains of terrestrial, fresh- water, and even marine shells, the latter worn or broken. * Mem. dc 1' Academic, torn, vii, p. 413. 90 Sandy Beaches* Rivers are deflected from their courses into the sea by beaches extending from one side, and produced by the winds and breakers ; both forcing detritus before them, if it be com- posed of sand or comminuted shells, while the latter acts upon the shingles alone, except when light pebbles are caught up in the heavier spray, and are thus driven by the wind. Ex- amples of this deflection may be seen in many situations, and the harbour of Shoreham, on our southern coast, is a marked one Rivers, when thus deflected from their courses by beaches, generally escape into the sea by the sides of cliffs, which seem to give them such support that they can cut channels. In tropical countries the breakers commonly throw up bar- riers against the advance of the mangrove trees, either from a deep bay or creek, or at the mouths of rivers, if they come within their influence. Capt. Tuckey remarks, that " the peninsula of Cape Padron and Shark Point, which forms the south side of the estuary (of the Zaire), has been evidently formed by the combined depositions of the sea and river, the external or sea shore being formed of quartzy sand constitu- ting a steep beach ; the internal or river side, by a deposit of mud overgrown by mangroves ; and both sides of the river to- wards its mouth are of similar formation, intersected by nu- merous creeks (apparently forming islands), in which the water is perfectly torpid." This mangrove tract appears to extend inland, on both banks, about seven or eight miles, and is represented as impenetrable. Did not the sea pile up a bar- rier against it, and thus afford it protection from its own at- tacks, it would be destroyed f. Similar phaenomena, though on a much smaller scale, are seen at the mouths of the Rio Minho, and other rivers in Jamaica. Beaches are accumu- lated in front of mangrove trees, under somewhat similar cir- cumstances, in the same island, on the south side of which, particularly near Albion estate, lakes are formed on the inside of a shingle beach thrown up by the sea. The lake near Al- bion has a small opening in the protecting bank, permitting the surplus water to escape ; this water being apparently de- rived from the drain of the mountains behind, and the splash of the sea during gales. The mountain drainage has carried much mud into the lake, upon which mangrove trees have * See Geological Notes, pi. 1. fig. 2.; and Phil. Mag. and Annals of Philosophy, N. S. vol. vii. pi. 11. fig. 2. f Expedition to the Zaire or Congo, p. 85. This author further remarks, that " small islands have in many places been formed by the current (of the river) ; and doubtless in the rainy season, when the stream is at its maxi- mum, these islands may be entirely separated from its banks, and the en- twined roots keeping the trees together, they will float down the river, and merit the name of floating islands." Sandy Beaches. 91 established themselves. These by their roots entangle various substances, and form land, the accumulation being a com- pound of mineral, vegetable and animal substances*. A much larger lake of the same description is found under Yallah's Mountain, the most projecting part of the beach forming Yallah's Point +. The bank called the Palisades, at the end of which stands Port RoyaJ, Jamaica, seems thrown up by the action of the prevalent breakers, caused by the sea breezes, or winds from the east and south-east, which propel the materials of the beach from east to west. This bank is between eight and nine miles long, of little elevation above the sea, having a beach on the seaward front, with mangrove trees on many parts of the in- ward side. If the passage between the western end of this bank and the land opposite to it should be barred up by a continua- tion of the bank, a large lake would be inclosed, into which the Rio Cobre would discharge itself. The mangrove trees would assist in the formation of new land, in which a mixture of marine, fresh-water, and terrestrial remains might be en- tombed. Mangrove trees afford support to beaches thrown up by the sea ; and if such a beach originated from a shoal, there is al- ways a tendency to increase land to leeward by their agency. Protection being once afforded, the mangrove trees establish themselves, and accumulate silt, mud, and drift-rubbish about their roots. Thus, support is afforded to the original bank, and new materials are piled upon it to windward by the action of the breakers, additional consolidation being produced by the tropical sea-side creepers. Meanwhile the advance to lee- ward continues, until the land immediately against the beach becoming too dry for the support of the mangrove trees, others, more suited to the new land, establish themselves ; and, finally, a grove of cocoa-nut trees may gradually appear \. Tides and Currents. The principal motions in the waters of seas and oceans are * For a section of this lake, see Sections and Views illustrative of Geo- logical Phenomena, pi. 35. fig. 6. f These waters contain a multitude of alligators. \ For a section of such an island near Jamaica, see Sections and Views illustrative of Geological Phenomena, pi. 36. fig. 2. According to M. Gutsmuth, the great band of alluvial matter, deposited by the sea for a distance of 200 miles between the Maranon and Oronoco, is increased by the mangrove trees, which, when the deposits still continue submerged, advance into the shallow sea and soon form woods. (Hcrtha, vol. ix. 1827.) In this and similar cases we may consider that, from the shallowness of the sea, heavy breakers cannot reach the mangrove trees, and therefore a beach is not thrown up against them. 92 Tides. produced by tides and currents ; the former due to the action of the sun and moon, the latter probably caused by the winds and the motion of the earth. The streams of water caused by tides are chiefly felt on coasts, while the currents produced by winds are more or less experienced over the whole surface of the ocean. It must frequently happen that the direction of a tide and a current being the same, they add mutually to the velocity of each other, while the contrary arises with opposed courses. The streams of water produced by tides and currents are geologically important, as they may be the means of distribut- ing the detritus derived from the land over spaces at a greater or less distance from the shore ; their power of affecting this being proportioned to their velocity and depth. Tides. The velocity of a stream of tide depends on the obstacles it encounters. These obstacles generally present themselves in the form of projecting headlands, a gradually diminishing channel, or a group of islands and shoals. In the former case the velocity of the tide is considerably increased round the op- posing capes, gradually diminishing to its usual rate at a short distance on either side, or in the offing. The English Chan- nel will present us with many examples, more or less striking, according to circumstances. Round the Start and the Bill of Portland the tides run exceedingly strong, causing dangerous Races when opposed to the winds. But these considerable streams of tide are merely local ; for in the bays, and at a short distance out at sea, the velocity of the tides does not exceed a mile and a half or two miles ; while at the headlands above noticed, it frequently flows at the rate of four or five miles *. Generally speaking, the increased velocity of the tidal stream round capes is in proportion to the body of water forced into the bays of which they form the extreme points. The greatest obstacle opposed to the tidal wave flowing up the English Channel, is the great bight on the west of Cap la Hague, where we find innumerable islands and rocks, of which the principal are Guernsey, Jersey, and Alderney. The stream of flood being completely opposed to the line of coast, and pent-up by the islands and rocks, it rises to a very consi- derable height, and escapes through the Race of Alderney, be- tween the island of the same name and the main land, with a velocity of seven miles an hour. It continues to run with great rapidity round Cap Barfleur, gradually decreasing in * All the miles mentioned in the following notice of tides and currents are nautical, sixty being equal to one degree. Tides. 93' strength until the general level is restored. Some idea may be formed of the variation in the Channel level, caused by this obstacle, by the differences in the rise of tide observed be- tween the mouth of the Channel and the Straits of Dover. The perpendicular rise of tide on each side of the mouth of the Channel is nearly the same, being twenty-one feet at Ushant, and twenty feet at the Land's End. In the great bight or bay west of Cap la Hague, the tide rises forty-five feet between Jersey and St. Maloes, and thirty-five feet at Guernsey. At Cherbourg this great elevation of the level is diminished ; the tide there rising about twenty-one feet. On the opposite side of the Channel, on the English coast, the perpendicular rise of the tidal wave is comparatively trifling, being thirteen feet at Lyme Regis, seven feet in Portland Road, fifteen feet at Cowes, and eighteen feet at Beachy Head. Therefore, the elevated level of the Guernsey and Jersey waters produces no perceptible effect on the English coast op- posite. Between Beachy Head and Dover, there is a rise of twenty-four feet on the west of Dungeness, and twenty feet at Folkestone. On the opposite coast there is a rise of twenty feet at Havre, nineteen feet at Dieppe, and nineteen feet at Boulogne. The tides are twenty feet at Dover, and nineteen feet at Calais. The Bristol Channel is a familiar example of a high rise of tide caused by a gradually contracted channel, at the end of which there is no outlet. At St. Ives, Cornwall, the perpen- dicular rise of the spring tides is eighteen feet, of the neap tides fourteen feet*. At Padstow the tide rises twenty-four feet; at Lundy Island, thirty ; at Minehead, thirty-six ; at King Road, near Bristol, from forty-six to fifty; and at Chepstow, about the same. The difference of level, produced by obstacles to the tide, is remarkably exhibited on each side of the isthmus separating Nova Scotia from the main land of North America. In the Bay of Fundy, on the south side, the tides have a very consi- derable rise, amounting, according to Des Barres, to sixty and seventy feet at the equinoxes; while on the northern side, in Baie Verte, they rise and fall only eight feet. The tidal stream is, as might be expected, very rapid in these gradually dimi- nished channels, particularly where the rise and fall is most considerable. This unusual rapidity ceases by degrees as we approach the mouths of such channels, and arrive at the more common levels. From the great diversity in the line of coasts, innumerable modifications are effected in tidal streams, causing them to * The rise of tide at St. Ives is sometimes stated at twenty-two feet. 94- Tides. flow with augmented or diminished velocity. As such streams are only visible on coasts, it. seems fair to infer that the effects produced by them do not extend to any considerable distance beyond the land. The tide in the offing, and the tide along shore, do not ex- actly correspond, the flood tide continuing in the offing some time after the ebb has commenced on shore ; the ebb tide the same. It has been stated that " the length of time between the changes of the tide on the shore and the stream in the offing, is in proportion to the strength of the current and the distance from the land ; that is, the stronger the current, and the greater the distance that the current is from the land, the longer it will run after the change on the shore*." Among the small islands of the Pacific Ocean the tide rises about two feet, there being no great range of coast near them to produce a greater elevation. At the islands of the Atlantic Ocean the rise is greater, being at the Azores from six to seven feet; at Madeira, eight or nine; among the Canaries, eight or ten ; at the Cape Verde Islands, from four to six ; at the Ber- mudas, five or six ; at St. Helena, three ; at Fernando No- ronho, six ; and at Tristan da Cunha, eight or ten feet. The stream of tide along a coast is greatly increased at the time of full and new moon, so that at spring tides the current often runs at double the rate experienced at neap tides. The transporting power of tidal streams is therefore perpetually changing, independent of the variations produced by winds upon them. From various circumstances the tides of flood and ebb are sometimes unequal. Thus, at the Land's End the flood runs nine hours to the north, and the ebb three to the south. In the expedition under Captains Parry and Lyon, it was found that in the higher part of Davis's Straits the flood tide set from the north at the rate of three miles an hour for nine hours, the tide of ebb making only three hours. A current setting into the Straits of Malacca, during part of the year, causes the tide to run nine hours one way arid three hours the other. The tides are irregular through the Straits of Banca, with an easterly wind. The ebb sets to the northward for sixteen hours, while the flood only lasts eight * Purdy, Atlantic Memoir, 1829. In the same work it is stated that " the time which the flood-stream runs in the middle of the English Channel after the time of high water on shore, is, westward of the meridian of Port- land, about three hours; but to the eastward, off Beachy Head, only one hour and three quarters. In the offing, between the meridians of Dunge- ness and Folkestone, the North Sea and Channel tides seem to meet ; and the ebb of the one uniting with the flood of the other, set in an easterly di- rection off the French coast, more than four hours after high water on the western shore of Dungeness." p. 88. Tides. 95 Hours. In common tides there are two floods and two ebbs in twenty-eight hours in these straits, the duration of which is in some sort regulated by the winds : the flood lasts six hours, and the ebb eight hours ; or there are five hours flood, and nine hours ebb. The tides are very trifling and irregular in the West Indies, perhaps owing to the accumulation of water pent up by the equatorial current and trade winds. At Vera Cruz there is only one tide in twenty-four hours, and that irregular. Among these islands the tide varies in perpendicular rise from a few inches to two feet or two feet and a half. The stream or cur- rent produced by them must consequently be very trifling. Theoretically, all bodies of water, even large fresh-water lakes, have tides ; but they are so insignificant that inland seas, such as the Mediterranean and Black Seas, are generally termed tideless. The current setting into the Mediterranean from the At- lantic is somewhat modified by the tides. In the middle of the Straits of Gibraltar the current sets eastward ; on each side, however, the flood tide sets to the westward. " On the European side, west of the island of Tarifa, it is high water at ll h , but the stream without continues to run until 2 h . On the opposite shore of Africa, it is high water at 10 h , and the stream without continues to run until one o'clock ; after which periods it changes on either side, and runs eastward with the general current. Near the shore are many changes, counter currents and whirlpools, caused by and varying with the winds. Near Malaga the stream runs along shore about eight hours each way. The flood sets to the westward *." The strongest tides of which I can find mention, occur among the Orkney and Shetland Isles, and through the Pent- land Frith, between the main land of Scotland and the former. The flood comes from the north-west, and is not of unusual strength until it encounters the obstacles of the islands and main land. The tides change near the shores sooner than at a di- stance from them. The difference of time varies according to si- tuation, amounting in some places to two or three hours. The velocity of the tide through Stronsa Frith is about five miles an hour during spring tides, and a mile or a mile and a half at neaps. In North Ronaldsha Frith, the springs run at five miles an hour ; the neap tides at one mile and a half. The flood divides near the shore at Fair Isle, forming a large eddy on the east side. The springs here run six miles an hour, the neaps two. These tides increase in velocity when sup- ported by the winds. The most rapid stream of tide occurs * Purely, Atlantic Memoir, p. 90. The tide rises three feet at Malaga. 96 Tides. at the Pentl'ind Frith, its velocity being nine miles an hour during the springs, though it runs only three miles an hour at neap tides. Tides in Rivers and Estuaries. These are necessarily much modified by circumstances; but, generally speaking, the tide of ebb is stronger than the flood, from the body of fresh water being pent up by the flood, to which the rivers must always present a certain resistance, proportioned to their velocity and abundance of water; the greatest resistance to the flood, and increased velocity of the ebb, being during freshets, or when the rivers have a surcharge of water produced by rains in the interior. When the flood tide takes place in rivers of sufficient depth, the first operation of the tide appears to be that of a wedge, elevating the fresh water from its inferior specific gravity to a higher level. The flood gradually opposes greater resistance to the outflow of the river, and in the end succeeds in damming it up. I have found many fishermen aware of this " creep- ing," as they have termed it, of the salt water beneath the fresh at the commencement of the flood, and have seen a rise of five or six feet caused in water in the higher parts of tidal rivers, while the water so raised has continued perfectly fresh at the surface. At the ebb, if the fresh or river waters be abundant, they will, after the salt water has been discharged 5 flow over the salt water to greater or less distances from the shore accord- ing to circumstances. After the rains, a strong freshet sets down the Senegal, and a powerful current of fresh water runs some distance out at sea. Masters of vessels crossing this stream have been surprised by the sudden increased draught of their ships, caused by their entrance into a fluid of inferior specific gravity. Captain Sabine states, that while proceeding in his voyage faom Maranham to Trinidad, on September 10, 1822, the general current running at the great rate of ninety-nine miles in twenty-four hours (more than four miles per hour), they crossed discoloured water in 5 08' N. lat., and 50 28' W. long. He considers this water as that of the river Amazons or Maranon, which had preserved its original impulse three hundred miles from its embouchure, having flowed over the waters of the ocean, from its less specific gravity. The line between the ocean water and discoloured water was very distinct, and great numbers of gelatinous marine animals were floating on the edge of the river water. The temperature of the ocean water is stated as = 81 0i l, and that of the supposed river water = 8l*8, both near the division line: *' the specific gravity of the former was 1 0262, and of the latter 1 -0204-." From experiments made, the depth of the discoloured water Tides. 97 was superficial, and did not amount to 126 feet. There was no bottom at 105 fathoms. In this discoloured water the ship was set N. 38 W., sixty-eight miles in twenty-four hours, or rather less than three miles per hour. The western side of the fresh water was gradually lost in that of the sea. Captain Sabine attributes the unusual velocity of the ocean-current of ninety-nine miles per day, to the obstacle which this fresh- water current opposes to it*. In the river St. Lawrence we have a striking example of the superior velocity of the ebb tide to the flood, " At the Isle of Coudre, in spring tides, the ebb runs at the rate of two knots. The next strongest tide is between Apple and Basque Isles ; the ebb of the river Saguenay uniting here, it runs full seven knots in spring tides; yet, although the ebb is so strong, the flood is scarcely perceptible ; and below the Isle of Bic there is no appearance of a flood tidef." The great difference in the ebb and flood of river tides must depend on many local causes, but be principally in proportion to the perpendicular rise of tide on the one side, and the mass of fresh water on the other. The flood tide sets up many rivers so suddenly, as to cause a wave of greater or less mag- nitude, according to circumstances, called the bore, appearing * Experiments to determine the Figure of the Earth. We have other accounts of discoloured waters in the Atlantic, which would render it necessary that the specific gravity and relative freshness of simply discoloured water should always be ascertained, as was done by Captain Sabine, before we can be certain that waters even flowing in the necessary direction were derived from rivers. Captain Cosme de Churruca states, that 128 leagues to the eastward of St. Lucia, and l. r >0 to the N.E. of the Orinoco, there is always discoloured water as if on soundings, but there is no bottom at 120 fathoms. The same appearances are observed about seventy or eighty leagues to the eastward of Barbadoes. Humboldt notices a place in the la- titude of Dominica at about 55 W. longitude, where the sea is constantly milky, although it is very deep ; and seems to think that there may possibly be a volcano beneath it. Captain Tuckey observed the same kind of milki- ness upon entering the Gulf of Guinea ; but considered it due to multitudes of Crustacea which were caught, and which produced great luminosity at night. ^ Sir Gore Ouseley mentions that on February 12, 1811, when off the Ara- bian shore, a partial line of green ^a/er, such as generally indicates shallows, and perfectly different from the blue of a deep sea, was perceived extending considerably. It appeared eight or nine miles from the land. The change from the blue to the green waters was sudden, so that the ship was in green and blue waters at the same time. Having entered the green water they sounded, and found bottom at seventy-nine fathoms; proving that the change of colour was not due to a shoal; for previous to entering this water they sounded in the blue water, and found sixty- three fathoms, so that the blue was more shallow than the green water. This was observed not far from the Persian Gulf. Sir Gore Ouseley, Travels, vol. i. In this case there was no great river near to produce the difference of colour. " Green Sea " is the name given to the Persian Gulf by Eastern geographers. t Purdy, Atlantic Memoir, p. 91. u 98 Tides. as if the flood suddenly overcame the resistance of the ebb. The bore of the Ganges is very considerable. According to Major Rennel, it " commences at Hughly Point, below Fulta, the place where the river first contracts itself, and is percep- tible above Hughly Town ; and so quick is its motion, that it hardly employs four hours in travelling from one to the other, although the distance is near seventy miles. At Calcutta, it sometimes causes an instantaneous rise of five feet ; and both here and in every other part of its track, the boats on its ap- proach immediately quit the shore, and make for safety to the middle of the river *." According to Romme, there is a considerable bore at the mouth of the Amazons or Maranon during three days at the equinoxes. It is observed between Maraca and the North Cape, and opposite the mouth of the Arouary. A wave of twelve or fifteen feet in height is suddenly formed, and is followed by three or four others. The advance of this bore is exceedingly rapid, and the noise caused by it is stated to be heard at the distance of two leagues. It occupies the whole breadth of the river, and in its progress carries all before it until it has passed the banks into deeper and wider water, where it ceases. M. De la Condamine has described this phenomenon, and has ob- served that there are two opposing currents during the flood, one superficial, the other deep. There are also two super- ficial currents, one setting by the shore on each side, while a central but retarded current descends. Tides are stated to be felt two hundred leagues up the Amazons, so that there are several in the river at the same time, and the surface of the water for that distance forms an undulating line. The most curious bore which I find recorded, was observed by Monach, Port Commandant at Cayenne : he states, that " the sea rises forty feet in less than five minutes in the Turury Channel, river Arouary; that this suddenly elevated water constitutes the whole rise of tide, the ebb immediately taking place, and running with great velocityf ." In the Zaire or Congo we have an example of the compa- ratively small effect of the tide upon a large body of fresh water discharged with sufficient velocity. Notwithstanding the aid of Massey's machine, bottom was not found in Tuckey's ex- pedition at 113 fathoms in mid-channel and at the mouth, and the stream ran at the rate of four and five miles an hourj. This stream became checked but not overcome in mid-channel, and the tide onty produced counter currents near the shore. The rise of water is felt between thirty and forty miles up the * Phil. Trans. f* Romme : Vents, Marges et Courants clu Globe, torn. ii. p. 302. J It has been since supposed that this stream had greater velocity. Currents. 99 river. Alluvial land is continually forming into flat islands, which are covered by mangrove trees and papyrus, and are often partially or wholly carried by the river into the ocean*. Professor Smith describes a floating isle of this kind which he saw further north off the coast of Africa; it was " about 120 feet long, and consisted of reeds resembling the Donax, and a species of Agrostis ? among which were still growing some branches of Justicia\" Currents. Currents are sometimes classed as constant, periodical, and temporary. The great current which flows from the Indian Ocean round the Cape of Good Hope, up the coast of Africa to the equa- torial regions, whence it strikes across the Atlantic to the West Indies, is considered a constant current, produced by the tropical or trade winds, assisted by the motion of the earth. The current having driven, by these means, a body of water to the continent of America, through which it cannot escape, passes up through the channel offered it at the Straits of Florida, flows considerably to the northward, and then bends to the eastward, and south-east, taking its course to the west coast of Europe and the upper part of Africa. It is considered that the latter division of the current again unites with the northern portion of the equatorial current, and again traverses the Atlantic. Between Cape Bassas in Africa and the Laccadives or Lakdivas, there is a constant current to the westward, mostly to the S.W. or W.S.W. Its rate is supposed to be from eight to twelve miles per day. The current south of the equator, in the Indian Sea, runs to the west. During the N.E. monsoon the currents of the Mosambique Channel run to the south along the African coast, and even in the offing ; their usual velocity being about seven or eight leagues in twenty-four hours. On the coast of Madagascar the currents take an opposite direction, and set towards the north. At the southern extremity of Africa, the currents set round the bank of Agulhas, or Lagullas as it is more commonly termed, a bank of considerable extent, the soundings in which are described as mud to the westward of Cape Lagullas, and sand to the eastward, the latter containing numerous small shells. Rennel informs us that this current is strongest during the winter, and that the outer verge of the stream runs into 39 S. before it turns to the northward, after which it proceeds slowly along the western coast of Africa to, and even beyond, * Tuckey's Expedition to the Zaire or Congo. f Ibid- P- 259. H 2 100 Currents. the equator*. The general velocity of the current round the bank is not stated ; but it appears that one vessel was carried by it one hundred and sixty miles in five days, or thirty-two miles per dayf. Beyond St. Helena, the current above noticed unites with the equatorial current of the Atlantic, and sets across from the Ethiopic sea to the West Indies. The velocity of this current has not been well ascertained, but is generally con- sidered as about one mile and a half per hour, increasing as it proceeds westward, and setting off the coast of Guyana at the rate of two or three miles per hour. Captain Sabine states that, sailing from Maranham in 1822, and entering the cur- rent, he estimated it as running at the rate of ninety-nine miles in twenty-four hours, or a little more than four miles per hour. The central direction of this current is W.N.W. " On the Colombian coast, from Trinidad to Cape la Vela, the currents sweep the frontier islands, inclining something to the south, according to the strait they come from, and running attout a mile and a half an hour with little difference. Between the islands and the coast, and particularly in the proximity of the latter, it has been remarked, that the current at times runs to the west, and at others to the east. From Cape la Vela, the principal part of the current runs W.N.W. ; and, as it spreads, its velocity diminishes : there is, however, a branch which runs with the velocity of a mile an hour, directing itself towards the coast about Cartagena. From this point, and in the space of sea comprehended between 14 of latitude * Captain Tuckey in his expedition to the Zaire or Congo, found a current setting to the N.N.W. after making St. Thomas off* the African coast. Its velocity was thirty-three miles in twenty-four hours. f As the current round the Lagullas Bank evidently conforms to the bank, we may, perhaps, consider that it there has considerable depth, that is, a depth equal to about sixty or seventy fathoms. But of this we cannot be quite certain, for we do not know to what distance water thrown off by the bank at lesser depths may be carried round it. There is an easterly or counter current which sets to the south of this main current. Capt. Horsburgh mentions having been carried by it at the rate of 20 to 30 miles in 24 hours ; and, in two instances, at the rate of 60 miles in the same time. My friend, Mr. Becher of the Hydrographical Office, informs me that there is also an easterly or counter current close to the shore, so that the main or western current flows between two eastern or counter currents. The survey of the coast of Africa, to the east of the Cape of Good Hope, was made by Capt. Owen with the assistance of this current against the full strength of the trade wind. It also appears to exist more northward of the Cape, in the Atlantic. Respecting the depths to which surface currents descend, we learn from Capt. Belcher, that he found a current running with nearly the same velocity at 40 fathoms as on the surface, off the west coast of Africa, in lat. 1 5 27' 9" N., and long. 17 31' 50" W., the rate being 0-75 per hour, and the direction south. Jour. Geog. Soc., vol. ii. Currents. 101 and the coast, it has however been observed, that in the dry season the current runs to the westward, and in the season of rains to the eastward*." It is asserted, that there is a constant stream entering the Mexican Gulf by the western side of the channel of Yucatan ; and that there is commonly a re-flow on the eastern side of the same channel around Cape Antonio f. On the northern coasts of St. Domingo and Cuba, in the windward passages, at Jamaica, and in the Bahama passages, the currents appear variable, their greatest observed velocity being about two miles per hour. The accumulation of water in the Caribbean and Mexican seas does not raise the level of those seas so much as was, perhaps, once supposed. The difference of level observed by Mr. Lloyd, in his researches on the Isthmus of Panama, be- tween the Mexican Sea and Pacific Ocean, was in favour of the -greater height of the Pacific Ocean by 3'52 feet, an unexpected result ; but the measurements were conducted with such care, that we can scarcely doubt it. The high- water mark at Panama is 13' 55 feet above high- water mark of the Atlantic at Chagres ; but from the difference in the tides on each side the isthmus, the Pacific is lower than the Atlantic at low water by 6*51 feetf. If \ve consider the body of water pent up by the effects of currents over so large a space as the Mexican Sea at eight feet, or even less, above the Atlantic Ocean, we need not be surprised at the velocity of the current produced by its escape through the Straits of Florida. If the temperature of the waters, heated in the Gulf of Mexico and Caribbean Sea, be greater, as we know it is, than that of the waters north of the tropics through which the Gulf Stream flows, the specific gravity of the former waters will be less, and consequently they will flow onwards over the colder waters or those of greater specific gravity, precisely as river- water flows out to sea over that of the ocean, and will continue to do so until their progress be gradually checked and finally stopped. From a mass of information that has been collected, it ap- pears that the Gulf Stream varies considerably in breadth, length, and velocity. It has been found that winds much affect the current, diminishing its breadth and augmenting its velocit} r , or augmenting its breadth and diminishing its ve- locity. In mid-channel, on the meridian of the Havanna, the di- rection is E.N.E., and the velocity about two miles and a half * Purdy, Atlantic Memoir, translated from the " Dcrrotero de las An- tillas." f Ibid. I Phil. Trans. 1830. 102 Currents. per hour. Off' the most southern parts of Florida, and at about one third over from the Florida Reefs, it runs at the rate of about four miles per hour. Between Cape Florida and the Bernini Isles it runs to the N. by E., with a velocity of more than four miles an hour. The stream is weak on the Cuba side, arid sets to the eastward. A re- flow or counter current sets down by the Florida Reefs and Kays to the S. W., and W., and by its aid many small vessels have made their passages from the northward*. To the northward of Cape Canaveral there is no stream of tide, along the southern coast of the United States, further from the shore than in ten or twelve fathoms of water ; from that depth to the edge of soundings, a current sets to the southward at the rate of a mile an hour ; out of soundings, the Gulf Stream is found setting to the northward f. It is also stated that there is a re-flow or counter current on the eastward of the stream. Capt. Sabine remarks, that in the latter part of 1822 the velocity of the current after passing Cape Hatteras was seventy- seven miles per dayf. Rennel, considering the force of the stream as determined at different points, calculates that the water requires about eleven weeks to run in the summer, when its rapidity is greatest, from the Gulf of Mexico to the Azores, a distance of about 3000 miles. Capt. Livingston, however, observes, that the calculations of the velocity of the Gulf Stream are not to be depended on. He found it setting at the rate of five knots and upwards on the 16th and 17th of August 1817. On the 19th and 20th of February 1819, it seemed to be almost imperceptible. In September 1819, it set at about the rate described in the charts . Lieut. Hare has found in the meridian of 57 W., that the stream ranges to 4?2f N. in the summer, and even to 4?2 N. in the winter. It would appear, that the waters after issuing through the Straits of Florida, run off from the eastern edge of the stream to the eastward, as might be expected from their tendency to equalize their level, particularly in those parts not carried forward with considerable velocity. A strong current sets from the Polar Seas, and through Hudson's Bay and Davis' s Strait, commonly denominated the Polar or Greenland current. It sets southerly down the coast * Purely, Atlantic Memoir. t Ibid. I Capt. Livingston observes that the current set him, off Cape Hatteras, 1 8' to the northward of his dead reckoning ; this he ascertained by stellar and solar observations. Atlantic Memoir. Purdy, Atlantic Memoir. These observations appear to have reference to the stream between Cape Florida and the Bernini Isles. Currents. 103 of America to Newfoundland, bringing down large icebergs beyond the Great Bank. Captains Ross and Parry found the velocity of the current from three to four miles per hour in Baffin's Bay and Davis's Strait. A current from the polar regions sets into the North Atlantic between America and Europe : it produced such a drift of the ice to the south in Capt. Parry's attempt to reach the North Pole over the ice, that the expedition was finally abandoned in consequence of it. The Polar current coming from Davis's Strait, may be said to unite with the Gulf Stream, and then to set eastward, di- recting its course to the coasts of Europe and Africa. Off the coast of Newfoundland, the current sometimes runs at the rate of two miles an hour, but is much modified by winds. About five degrees to the westward of Cape Finisterre the current has a velocity of thirty miles in twenty-four hours. Between Cape Finisterre and the Azores there is a tendency of the surface waters to the S.E., being variable in winter. Lieutenant Hare, in September 1823, found a current setting E.S.E. with a velocity of a mile and a half per hour between N. latitude 45 20' and 4-3 40', and W. longitude 22 30' to 16. Rennel remarks, respecting the currents between Cape Finisterre and the Canary Islands, that " it may be taken for granted, that the whole surface of that part of the Atlantic from the parallel of 30 to 45 at least, and to 100 or 130 leagues offshore, is in motion towards the Straits of Gibraltar." " Near the coasts of Spain and Portugal, commonly called The Wall, the current is always very much southerly (as it is more easterly towards Cape Finisterre), and continues as far as the parallel of 25, and is, moreover, felt beyond Madeira westward ; that is, at least 1 30 leagues from the coast of Africa ; beyond which a S.W. current takes place, owing, doubtless, to the operation of the N.E. trade wind." The same author observes, that the velocity of the current varies considerably, being from twelve to twenty, or more, miles in twenty-four hours. He considers sixteen as below the mean rate. A current sets along the coast of Africa from the Canaries to the Gulf of Guinea, running westerly out of the Bight of Biafra. The rainy seasons, and Harmattan wind, interrupt this stream. From Cape Bojador and the Isles de Los, the velocity of the current has neverbeen found toexceed a mile and a half per hour on the coast and on the outer edge of the bank. Its more common rate is less than a mile. At the distance of four leagues from the coast it becomes half a mile, and even less. In the meridian of 11 W. the current runs twenty-five miles to the E.S.E. in twenty-four hours. Off Cape Palmas it sets to the E. at forty miles; off Cape Three Points, and 104 Currents. thence to the Bight of Benin, at from fifteen to thirty miles. It then decreases in strength, runs to the southward, turns to the S. W. between 6 and 8 S., and thence flows N. W. to the Cape Verde Islands. It is considered that the portion flowing eastward into the Gulf of Guinea, is not altogether continuous with that which comes from Cape Bojador to the south. A current is described to pass round Cape Horn and Terra del Fuego, from the Pacific into the Atlantic, during the greater part of the year*. From the Straits of Magellan to the equator, a current sets northward along the western coast of South America. At eighty leagues from the coast, between 15 S. latitude and the equator, arid even to 15 N. latitude, the currents generally run westward. Captain Hall found a constant current setting off the Galapagos, to the N.N.W. At Guayaquil a strong current sets out of the Gulf at the rate of forty miles in twenty- four hours. Between Panama and Acapulco, and at about 180 miles from the latter place, Cap- tain Hall met with a steady current running E. by S. at rates varying from seven to thirty-seven miles per day. Great quan- tities of wood are drifted from the continent of America to Easter Island by the force of a current setting in that direction. Currents have been found at Juan Fernandez, and 300 leagues to the westward of it, running W.S.W. at sixteen miles per day. At the Marquesas they flow with a velocity of twenty-six miles in twenty-four hours. Between the Marquesas and the Sandwich Islands they have been found to run westward at the rate of thirty miles a day, in April and May. A southerly current has been observed at California ; and a northerly cur- rent along the N.W. coast of America, from Cape Orford, the latter having a velocity of a mile and a half per hour. A northerly current sets through Behring's Straits f, and is supposed to run along the north coast of America, and deliver itself through Baffin's Bay and Hudson's Straits, into the Atlantic. King found a current setting N.E. near the Japanese Is- * Captain Hall states, that he did not meet with any current round Cape Horn. A naval officer, however, assures me that a current runs out of the Pacific into the Atlantic during nine months ; and this is rendered probable from the prevalence of strong westerly winds during the greater part of the year, which would drive the waters before them. Kotzebue found a cur- rent which turned rapidly to E.N.E. near Staaten Land, having had another direction (S.W.) off Cape St. John. f Kotzebue describes this current as setting through the Straits with a velocity of three miles per hour to the N.E. At Anchorage, near East Cape, the current was found to set at the rate of one mile per hour ; but shortly afterwards, notwithstanding a brisk wind, the expedition under Kotzebue made but little way against it, though going, by the log, at the rate of seven miles per hour. Currents. 1 05 lands, the velocity five miles per hour ; but he also found it to vary considerably in direction and strength. Among the Philippine Islands a current comes from the N.E., and runs with considerable force among the passages dividing the islands; it has been found with a strength of twenty miles a day near these isles. This current varies. Cook found a southerly current in August, flowing ten or fifteen miles a day, between Botany Bay and 24- S. On the same side of Australia a vessel was set forty miles to the southward in twenty-four hours, in the month of March ; and in July another vessel was carried thirty miles in two days in the same direction. A constant current sets eastward into the Mediterranean, with a velocity of about eleven miles in twenty-four hours. It has been considered that there is an under or counter current setting westward, and carrying out the dense water, rendered more than usually saline from evaporation within the Straits of Gibraltar ; but this has lately been controverted. It was remarked by Dr. Wollaston, that the salt carried into the Jlediterranean by the current from the Atlantic must remain there after the evaporation of the water which held it in solu- tion, unless it could escape by some means. He inferred its escape to be by an under current, usually thought to exist, and this he considered proved by experiment; for water brought up from the depth of 670 fathoms about fifty miles within the Straits, by Captain Smyth, was found to contain about four times the usual quantity of saline matter. Water taken from depths of 450 and 400 fathoms, at 680 and 450 miles within the Straits, did not exceed in its saline con- tents many ordinary examples of sea- water. He further ob- served, that if the under current moved only with one fourth the velocity of the upper current, and was of the same depth and breadth as it, the former would convey out as much salt as the latter brought in *. Mr. Lyell infers that this dense water cannot pass out, because the bottom of the sea rises be- tween Capes Spartel and Trafalgar, and has only 220 fathoms of water upon it; and therefore, if the under and more saline water be as deep as is supposed, it would be impossible for it to escape, and it would deposit great quantities of salt in the bed of the Mediterranean f . It is much to be regretted that we do not possess better information on this subject, and that direct experiments have not been made on this supposed under current. That this has not been done is the more remarkable, when we consider the numerous opportunities afforded by the * Wollaston, Phil. Trans. 1829. t Lyell, Principles of Geology, vol. i. 106 Currents. continual passage of ships, and the proximity of such esta- blishments as those of Gibraltar. Mr. Ly ell's theory of a great deposit of salt at the bottom of the Mediterranean, though very ingenious, can scarcely be true ; for, supposing it to be so, the sea would, as the depth increased, be more and more charged with saline matter, until it finally became mere salt, the density increasing at the same time. This being the case, we should bring up salt with the sounding-lead, and little else. But the fact is, that the deep soundings, as shown by Captain Smyth, are mud, sand and shells. Sand and shells form the bottom, beneath 980 fathoms of water, a little east of the me- ridian of Gibraltar; and the same bottom is found in the Straits beneath 700 fathoms of water. Now these places are near where the sea-water, so highly charged with saline mat- ter, was brought up ; and where, according "to the theory, there should be a bottom of salt. The same may be said of other situations*. The current entering the Mediterranean passes along the southern shores of that sea, and is felt at Tripoli and the Is- land of Galitta. At Alexandria there is a stream flowing east, as well as between the coast of Egypt and Candia : arrived on the coast of Syria, it runs north, and then advances between Cyprus and the coast of Karamania. A strong current flows from the Black Sea into the Mediterranean, through the Dar- danelles. A constant current flows out of the Baltic, through the Sound and Cattegat, into the German Ocean. Its velocity in the narrowest part of the Sound is about three miles per hour ; but the ordinary rate, in fine weather, is about one mile and a half or two miles. The currents out of the Sound and two Belts are directed towards the Scau or Skagen, and flowing thence, turn N.E. towards Marstrand, at the rate of about two miles per hour. It is not impossible that a counter and under current setting into the Baltic from the ocean may exist ; for Captain Patton observed, when at anchor a few miles from * In all our remarks on the changes that may be supposed to occur at the bottom of the Mediterranean, we should be careful to remember that this bottom is divided into two great basins (See Smyth's Charts) by a winding shoal, which connects Sicily with the coast of Africa. This shoal, known as the Skerki, has the following line of soundings upon it, proceeding from the African to the Sicilian coast ; namely, 34, 48, 50, 38, 74, 20, 70, 52, 91, 16, 15, 32, 7, 32, 48, 34, 54, 70, 72, 38, 55, and 13 fathoms, from whence an idea of its inequalities may be formed. There are soundings in 140, 157, and 260 fathoms, on either side, as also places where 190 and 230 fathoms of line have been run out, without finding bottom. It may be here remarked, that, at the entrance of the Dardanelles into the Mediter- ranean, there are only thirty-seven fathoms of water; so that the quantity of solid matter requisite to bar the communication between the Black and Mediterranean seas, would not be very considerable. Currents. 107 Elsinore, in an upper current setting at the rate of four miles an hour outwards, that in sounding in fourteen fathoms, he found the line continue perpendicular to his hand, when the lead was raised a little from the ground. Hence he con- cluded that there was an under current that prevented the line from being carried away. In the Indian and Chinese seas we have good examples of periodical currents, evidently referable to the periodical winds or monsoons. From St. John's Point to Cape Cormorin there is a nearly constant current in the direction of the coast from N.N.W. to S.S.E. ; except that between Cape Cormorin and Cochin, it flows from S.E. to N. W. from October to the end of January. The current sets from the ocean into the Red Sea from Oc- tober to May, and runs out of that sea from May to October. A current commonly sets from the Gulf of Persia towards the ocean during the whole time that the current flows into the Red Sea, and runs into the Gulf from May to October. In the Gulf of Manar, between Ceylon and Cape Cormo^ rin, the current flows northward from May to October, set- ting the remaining six months to the S.W. and S.S. W. From Pedro Point on the north of Ceylon, to, Pointe de Galle on the south, the current runs S.E,, S.S.E. , S., S.W., and W., according to the nature of the coast, uniting at the Pointe de Galle with the current that comes out of the Gulf of Ma- nar. The ordinary velocity of the stream on the south coast of Ceylon is about a league an hour. The Ceylon currents are weak in June and November. In the Bay of Bengal the cur- rents run with the wind towards the N.E. during the S.W. or W. monsoon, and slacken in September. On the coast of Orissa, about eight days before the equinox their direction is towards the south, and they become strong at the end of the month. During the N.E. and E. monsoon, the currents are, as before, with the wind, and strong in proportion to it. In the S.W. monsoon the current between the coast of Malabar and the Lakdivas sets to the S.S.E. with a velocity of twenty, twenty- four, or twenty-six miles in twenty-four hours. Between the Lakdivas its direction is to the S.S.W. and S. W., its rate being from eighteen to twenty-two miles in twenty-four hours. The current, setting W. or W.S.W. to the westward of these islands, varies in velocity from eight to eleven miles per day. There is a strong current among the Maldives : among the southern isles the direction is generally to the E.N.E. in March and April ; in May it sets to the eastward ; in June and July the currents often run to the W.N.W., particularly to the south of the equator. Between these isles and Ceylon they frequently set strongly to the west- 108 Currents. ward during the months of October, November, and De* cember. The currents in the China seas, at a distance from shore, generally flow, more or less, towards the N.E. from the mid- dle of May to the middle of August, and have a contrary di- rection from the middle of October to March or April. The velocity of the currents from the N.E. is usually greater in Oc- tober, November, and December, along the adjacent shores, than that of the opposite set in May, June, and July. Their strength is most felt among the islands and shoals near the coast. The strongest currents of these seas are experienced along the coasts of Cambodia, during the end of November. They run with a velocity of from fifty to seventy miles to the south- ward, in twenty-four hours, between Avarella and Poolo Cecir da Terra. Some part of the stream setting into the Straits of Malacca, causes the tide to run nine hours one way and three hours the other. The currents to the northward commence running in April through the Straits of Banca, past the Straits of Malacca, and along the west coast of the Gulf of Siam, setting along the north-east side of the same Gulf to the E.S.E. until to the eastward of Point Ooby ; they then bend to the N.E. running along the coasts of Cambodia, Cochin-China, and China, till September, when the oppo- site monsoon and currents prevail from the N.E., and conti- nue to March or April. Periodical currents occur, according to M. Lartigue, along the west coast of South America, from Cape Horn to latitude 19. The S. and E.S.E. winds produce a current, setting to the N. W., off 1 the coast of Peru, of which the maximum velocity is fifteen miles in twenty-four hours, and the mean velocity about nine or ten miles. Between this current and the coast, there is a counter current flowing to the S.E. During the prevalence of the wind from N. to W. the current flows S.E., but is only sensible near the land*. Temporary currents are innumerable, every severe gale of any duration producing one. Nothing is more common than these partial currents, which are more particularly felt along coasts and through channels. The direction arid velocity of the currents above enumerated may be considered rather as approximations to the truth, than as the truth itself, for the determination of currents is liable to many errors. The usual manner of ascertaining them is by comparing the true place of a ship, determined by means of chronometers and astronomical observations, with the position * Lartigue, Description dc la Cote du Pcrou. Currents. of the same ship as deduced by dead reckoning. The latter is a calculation of the vessel's way through the water in a given direction. The rate of the vessel's way is estimated by means of a contrivance called a log-line, or a line at the end of which there is a float. According to the quantity of line run out in a given time, with allowances for the agitation of the sea, &c., is the rate of the vessel's way calculated. This operation is liable to numerous errors; and even with the line and glasses in the highest order, requires a nicety of execution seldom practised. The direction of the vessel's course is estimated by the compass, with allowances for magnetic variation. Here we have a most fruitful source of error, for until lately no al- lowance whatever was attempted for the local attraction of the ship. It is now well known that the disposition of iron in a vessel is such, that no two ships will be found to have the same local attraction ; consequently no rules can be adopted for correcting the error of aberration by means of placing the magnets in any particular situation, though some situations have been found more favourable for true observations than others. It was not until Mr. Barlow invented his plate of iron for counteracting the effect of aberration, that the error arising from it could be fully known. Now nearly all the preceding observations, as to the direction and velocity of currents, were made before this great source of error was un- derstood ; consequently many of them are erroneous, and re- quire that re-examination which the advance of science has rendered necessary. It is clear, that if a vessel is steering one course, and those on board consider they are taking another, the position deduced from dead reckoning must wander from the truth in proportion to the amount of aberration, even sup- posing the rate of way through the water and other necessary observations correct. If, in the annexed diagram, a vessel, without any allowance for aberration, be supposed to hold her course from a to b, while in reality her course, with c \^ lg ' 9 ' proper allowance for aberration, is from a to . \^^ r, the distance from b to c will, according to ~~ the usual practice, be referred to current, after an observation shall have shown that her true place is at c. It will be clear that in this case no such current exists, and that the difference between the true and calculated situations of the ship arises solely from want of attention to local attrac- tion. Another great source of error in estimating the value of currents has been noticed by Captain Basil Hall. This au- thor observes, that the usual method of laying down ships' tracks by two lines, one representing the course as estimated 110 Transporting Power of Tides. from the dead reckoning, and the other as deduced from chronometers and lunar observations, leads to no information as " to where the current began, or where it ceased, or what was its set, or its velocity." He proposes instead of this, that the position of the ship tbund at each good observation should form the point of departure, both for the line representing the distance and direction to the next observed true position, and for that representing the ship's course as estimated by dead reckoning*. A very superior plan, and one that should super- sede the old method. Although these causes of error render the exact velocity and course of currents heretofore observed vague and uncertain, so that many minor streams may be found imaginary, and that the. navigator may be exposed to great danger from im- plicitly depending upon them ; yet to the geologist, perhaps, they may not be so formidable ; as, probably, the general ve- locity of currents will not be found greatly altered ; and as it is with their velocity and consequent transporting power that he is principally concerned. Transporting Power of Tides. The stream caused by tides varies much in strength, but a common velocity appears to be one mile and a half per hour, when head-lands, shallow banks, and other obstacles are not opposed to it; and therefore, even supposing the superficial velocity to extend to the bottom, which would not be the case except in comparatively shallow seas, the general transporting power of such tides would appear, judging from the effects we witness near shores, to be but small. This the unchanged character of soundings for a great length of time, though prin- cipally composed of mud and sand, seems to attest. Where obstacles are opposed to the tides, the transporting power will be increased, and the changes produced more rapid. The tide through the Pentland Firth having a velocity of nine miles per hour, would scour out pebbles of considerable size from its channel ; but its power to do this would cease at each extremity, where the tides flow at the rate of two or three miles per hour, and the local cause would merely produce a local effect. The same with the Race of Alderney, and other similar places. Changes in the shape of sand-banks frequently take place when they approach the surface ; but as they then come within the influence of another cause, the action of the waves, the transporting power of which is very considerable, too much must not be attributed to the mere force of a tidal stream. * Edinburgh Philosophical Journal, vol. ii. Transporting Power of Tides. Ill The transporting power of tidal rivers outwards, or into the waters of the sea, is considerable, more particularly during the time of freshets or floods. As has been seen, the tide of ebb in rivers is always greater than the flood ; therefore, al- though estuary waters are very turbid, and a great proportion of them merely carried backwards and forwards, detritus will escape into the open sea in proportion to the difference of ve- locity between the ebb and flood. It should be remarked, that all estuaries have a tendency to be filled up by deposition of the matters held mechanically in suspension by their waters. The heads of estuaries are very frequently alluvial plains, formed of the same kind of mud and silt as are at present brought down by the rivers ; and it often appears as if the tides had flowed up to much greater distances than they now do, the higher parts having been gradually silted up*. These appearances are so common, that it is useless to insist upon them ; but the extent of flat lands, evidently accumulated in this way on the sides and heads of estuaries, is often very re- markable, and would seem to have required a long lapse of ages for their formation; more particularly when the present deposits of the same estuaries are considered. Notwithstanding this deposit in the estuary itself, and the bars and banks accumulated at the mouths of so many tidal rivers, above noticed, mud and silt escape into the sea, and are transported by the tides to greater or less distances from the rivers ; as may often be seen at low water, on coasts where tidal rivers discharge themselves. The transporting power of tides and currents being propor- tioned to their velocity, and this being greatest when obstacles are opposed to either, it is in these situations that we should look for the greatest transporting power. The difference between the velocity of tides on the surface and at moderate depths must be very considerable, otherwise the previously noticed power of water to tear up different kinds of substances at given velocities must be incorrect ; for if the velocities were nearly as great at moderate depths as on the surface, tidal streams would be little else than a mass of turbid waters. The discoloration of the sea to greater or less distances from the shore, according to the depth, is well known to be effected during heavy gales, and is due to the action of the waves, and not to that of the tide merely passing over sand or mud with a certain strength, and therefore must not be con- founded with it. * If we could always give implicit confidence to old maps and charts, great deposits of this nature would seem to have taken place within historical times. 1 1 2 Transporting Power of Currents. To take an example of tidal waters running over a certain bottom : At the Shambles, a well-known bank near the island of Portland, the tides run at the rapid rate of about three nau- tical miles per hour, over soundings of gravel which do not alter. Now, if the calculations above noticed were correct, and the inferior velocity not very considerably different from that on the surface, stones, the size of eggs, could be torn up by water with a velocity of three feet in a second, or 3600 yards in an hour; consequently the pebbles on the bank would be carried away, and nothing but bare rock or masses of stone would be left; but the soundings on the Shambles are the same at present as they are represented to have been, by the charts, many years ago. The preservation of the same kind of bottoms or soundings, over which tides or currents pass with considerable velocity without their being altered, is familiar to most mariners; and it would seem that we are far from being acquainted with the respective velocities required to tear up mud, sand, and pebbles at various depths in the sea. Tidal streams flow over mud banks in some estuaries at the rate of a mile and a half or two miles per hour, without removing them ; though, if the above- noticed calculations were always applicable, the current would be sufficiently strong to remove pebbles of some size. The same remark applies to innumerable sand-banks*. Transporting Power of Currents. In estimating the transporting power of currents, we should consider the causes which produce them, and the nature of the fluid in which they are produced. The motion of the earth, * While on the subject of soundings, it may be noticed that the British Islands are in reality united to the continent beneath the sea by banks of various kinds, at greater or less depths ; the principal soundings on which are mud or sand. The whole is more or less known by the name of sound- ings, because bottom can be easily obtained by a line of eighty or ninety fathoms in length. The boundary of these soundings is traced on all good charts, and is seen to commence at the bottom of the Bay of Biscay, then to run round the British Isles, and to communicate with the shallows of the German Ocean. The bed of the sea in these soundings can only be considered as so much of the continent, which happens to be at no great depth beneath the ocean level. The upper part of the bottom, tenanted by various animals whose exuviae are daily left in it, is probably in a great measure derived from the detritus of the British Islands and such parts of the continent of Europe as are either bathed by, or discharge their waters into, these seas. The depths being comparatively inconsiderable, the tides, currents, and waves are pro- bably enabled to act, according to circumstances, in the distribution of the detritus. The course of the tides round the British Islands is represented in Dr. Young's Natural Philosophy, vol. i. pi. 38. fig. 521 ; see also Lubbock on Tides, Phil. Trans. 1831. Transporting Power of Currents. 1 1 3 although it would seem to give a certain general movement to the waters of our globe, does not appear capable, taken by itself, of producing currents of geological importance. The great cause of ocean-currents seems to be prevalent winds ; and accordingly we find that in the equatorial regions of the world, over which the more or less easterly winds, commonly called the Trade Winds, prevail, there is a tendency of the waters to flow westward in the Pacific Ocean, in the Atlantic, and in those parts of the Indian seas free from the monsoons. That the winds are the great cause of ocean-currents, is a fact suffi- ciently proved by the velocity and direction of such currents in the Indian and Chinese seas, varying with the force and di- rection of the monsoons. On this subject Major Rennel ob- serves, " It is well known how easily a current may be induced by the action of the wind, and how a strong S.W., a N.W., or even a N.E., wind on our own coasts raises the tide to an extraordinary height in the English Channel, the river Thames, the ease coast of Britain, &c., as those winds respec- tively prevail. The late ingenious Mr. Smeaton ascertained, by experiment, that in a canal of four miles in length, the water was kept up four inches higher at one end than at the other, merely by the action of the wind along the canal. The Baltic is kept up two feet at least by a strong N.W. wind of any continuance; and the Caspian Sea is higher by several feet, at either end, as a strong northerly or southerly wind prevails. It is likewise known that a large piece of water, ten miles broad, and generally only three feet deep, has, by a strong wind, had its waters driven to one side, and sustained so as to become six feet deep, while the windward side was left dry. Therefore, as water pent up so that it cannot escape acquires a higher level, in a place where it can escape the same ope- ration produces a current, and this current will extend to a greater or less distance according to the force by which it is produced or kept up*." It is also considered that the moon exercises an influence on the waters of the tropical regions, increasing their velocity by drawing them from E. to W. The current setting six hours one way and six hours the other through the Straits of Messina, though there is no rise or fall of water with it, is at- tributed to the influence of the moon, and may be considered as a tide. Capt. Livingston observes, that " when the sun's declination is N., the N.E. trade \\ind blows fresher, and ex- tends further to the northward than when the sun's declina- tion is S., thus farcing a greater body of water into the Ca- ribbean Seat." The current setting into the Mediterranean through the * On the Channel Current. \ Purdy's Atlantic Memoir. 1 1 4 Transporting Power of Currents. Straits of Gibraltar is commonly attributed to the evaporation of that sea, which also receives a large supply of water from the Black Sea through the Dardanelles. The easterly in- draught from the Atlantic is stated to commence nearly one hundred leagues to the westward of the Straits of Gibraltar. It has been supposed that an under and counter current sets outwards; but this, as has been above noticed, has been lately controverted*. That under currents do, however, occur in the Mediterranean, Capt. Beaufort affords us sufficient proof. After remarking that from Syria to the Archipelago there is a constant current to the westward, slightly felt at sea, al- though very perceptible on shore, amounting to three miles per hour, between Adratchan Cape and the opposite island, he observes, " The counter currents, or those which return beneath the surface of the water, are also very remarkable : in some parts of the Archipelago they are sometimes so strong as to prevent the steering of the ship ; and in one instance, on sinking the lead, when the sea was calm and clear, with shreds of bunting, of various colours, attached at every yard of the line, they pointed in different directions all round the compassf." These observations of Capt. Beaufort are of the highest importance when we consider the transporting power of cur- rents, because they seem to show that we cannot judge of the direction of under currents from those known to flow on the surface. The winds being, generally speaking, the cause of the great ocean-currents, and effects being only in proportion to their causes, the streams of water thus produced will not extend deeper than the propelling power of the winds can be felt. Now, as the ocean varies in density according to its depth, the cause sufficient to move waters on the surface, and to certain depths beneath it, will constantly meet with opposi- tion, at an increasing ratio ; until finally, the moving power and the resistance being equal, no effect whatever is pro- duced ; and all water beneath a certain depth would be, as far as respects surface causes, immoveable, and consequently would have no transporting power. Hence it would appear that the transporting power of cur- rents will depend on the depth of the sea, all other things being equal, and that the smaller the depth the greater the transporting power. Consequently, coasts are the situations where we may look for this power. If the current entering into the Mediterranean from the Atlantic be due to the evaporation of the former, this also is a superficial cause, and its effects will gradually become less, until, in deep water, it ceases altogether. * Lyell's Principles of Geology. } Beaufort's Karamanin. Active Volcanos. 115 We have seen that tides as well as currents have their great- est velocity in shallow water, across headlands, or in con- tracted channels; consequently, their greatest transporting power exists in the same situations, and will be local. Tides commonly exert an equal transporting power in two direc- tions, for the most part opposite to each other, except in the case of rivers, where this power is greater on the ebb than at the flood. Unless the rivers be very considerable, . the detri- tus brought through their embouchures by the superior velo- city of the ebb, enters into the power of the coast-tides, and is carried backwards and forwards by them until deposited. But in the case of great rivers, such as the Maranon, St. Law- rence, and Orinoco, the unchecked detritus is borne forward, until stopped and turned by the ocean-currents. Large ad- ditions are daily made to the coast of South America by the deposit from the waters of the Maranon, which are carried toward the shore by the prevailing current *. Upon a review of what has been stated respecting the streams of water caused by tides and currents, it would ap- pear that their geological importance will depend upon the re- lative depth of water which they traverse, and their proximity to land, by which their velocity is increased. Round coasts they have a transporting power, which varies according to circumstances; being greatest, all other things remaining the same, nearest the land. In great depths we have no reason to suppose that this transporting power exists; or if it does, the causes must be different from those which produce motion on the surface. It does not appear that we are acquainted with the velocities which could tear up mud, sand, or gravel ; for currents pass over the bottom in shallow water, composed of mud and sand, without mixing them, with a considerable surface velocity. The changes produced on the bottom are scarcely perceptible, within the periods we should consider long, unless in shallow water, and near the mouths of great rivers, the deposits from which must gradually accumulate, and diminish the depth of the water. In the soundings round coasts, we do not generally find any great inequalities; but in the ocean these must exist to a very great extent, as is shown by the rocks, shoals, and small islands scattered over it, the tops of mountains emerging from the water, which is generally of great depth close to them. Active Volcanos. The surface of the earth is irregularly marked by orifices, * The water upon this coast i$ so shallow, that the land is dangerous to approach without great care, the only harhours being the mouths of rivers. 116 Active Volcanos. through which various gases, cinders, ashes, stones, and streams of. red-hot melted rocks are projected. From this continued propulsion of matter through a vent or vents, a co- nical mass is accumulated, to which the name of volcano is applied. Volcanos differ materially in the quantity of matter ejected, but agree in such a general resemblance to each other, that they seem all referable to the operations of the same causes. Various theories have been formed for the explanation of volcanic phaenomena ; but it must be confessed, that they are all more or less defective, and that the real causes of such phaenomena are mere subjects of conjecture. With some of the effects we are familiar; though with the districts most ravaged by erupted matter, we are far from being well acquainted ; our principal knowledge of volcanos being derived from the two largest active vents of Europe, Etna and Vesuvius, but principally from the latter. Etna certainly covers a consider- able surface, but Vesuvius sinks into insignificance before some of the great volcanos of the world. From their general proximity to, or occurrence in, the sea, it has been supposed that the active state of volcanos is pro- duced by the percolation of sea-water to certain metallic bases of the earths or alkalies at various depths beneath the surface ; which metallic bases being thus inflamed, cause the phaeno- mena observed in volcanic eruptions. The volcanos in the interior of Mexico, as also the volcanos of Tartary, have been accounted for by the advocates of this theory: the former, by supposing a connexion between the vents of Colima, Jorullo, Pococatepetl, and Orizaba, all situated on the same line ; the latter, by considering that the waters of salt lakes may perco- late to their foci. Recent researches would, however, appear to render this hypothesis untenable ; for according to MM. Klaproth, Abel Remusat, and Humboldt, there is a volcanic region, with an area of about 2500 square geographical miles in central Asia, between 300 and 400 leagues distant from the sea. The principal seat of volcanic action is the range of the Thian chan, in which are the two volcanos of Pe chan and Ho tcheou, distant about 105 miles in an east and west di- rection from each other, the former being about 225 leagues from Lake Aral *. Recent observations also, in the central chain of the Andes, by MM. Roulin and Boussingault, show that volcanic eruptions take place at a considerable distance from the sea. The Peak of Tolima (according to Humboldt in lat. 4 46' N. and long. 77 56' W. from Paris,) was seen to be in eruption by M. Roulin in 1826 ; and M. Boussin- * Humboldt, Fragmens Asiatiques. Active Volcanos. 117 gault observed a volcano of the same district to be in activity in 1829. M. Roulin discovered from an ancient document that there had been a great eruption of Tolima in March 1595 *. As the first chemical operation, if the theory of a percola- tion of sea-water to the metallic bases of earths or alkalies M-ere true, would be the union of the oxygen with the metal- lic base, and the escape of an immense quantity of hydrogen, M. Gay Lussac has objected to it, that pure hydrogen gas is not evolved from volcanos ; and as a proof of it, observes, that if it were present, it would be inflamed by the red-hot matter ejected from the craters. Dr. Daubeny endeavours to meet this objection, by supposing the hydrogen " to have combined in its nascent state with sulphur, and the two bodies to have been evolved in the form of sulphuretted hydrogen gas." He also considers that the presence of large quantities of muriatic acid would destroy the inflammability of the hy- drogen f. According to the same author, the gases evolved from vol- canos consist of muriatic acid gas, sulphur combined with oxygen or hydrogen, carbonic acid gas, and nitrogen ; to which must be added a great quantity of aqueous vapour J. Sir Humphry Davy found the sublimations of Vesuvius to consist of a common salt (one specimen containing a minute quantity of muriate of cobalt), chloride of iron, sulphate of soda, muriate and sulphate of potassa, and a small quantity of oxide of copper^. Volcanic eruptions are usually preceded by detonations in the mountain, and agitations of the earth, or earthquakes in the vicinity, after which the mountain vomits forth an abun- dance of ashes, cinders, and stones ; and streams of melted lava flow from apertures made in the side of the cone, the resistance of which becomes unequal to the pressure of the melted mass within. The lava very rarely seems to proceed from the lip of the crater. The following is a summary, from various authorities, of the heat and appearances of a lava-current. " Lava, when observed as near as possible to the point from whence it issues, is for the most part a semifluid mass of the consistence of honey, but sometimes so liquid as to penetrate the fibre of wood. It soon cools externally, and therefore exhibits a rough unequal surface ; but as it is a bad conductor of heat, the in- ternal mass remains liquid long after the portion exposed to the air has become solidified. The temperature at which it continues fluid is considerable enough to melt glass and silver, and has been found to render a mass of lead fluid in four * Humboldt, Frugmens Asiatiques. f Description of Volcanos, p. 377. J Ibid. p. 376. Phil, Trans., 1828. 118 Active Volcanos. minutes ; when the same mass, placed on red-hot iron, re- quired double that time to enter into fusion." The heat does not, however, appear to be always equal ; for it is stated, that when bell-metal was thrown into lava (of 1794), the zinc was melted and the copper remained unfused *. The volcanic eruption which produced the greatest quantity of lava known to have been thrown out at one time, is that re- corded as having proceeded in 1783 from the low country near Shaptar Jokul, in Iceland. The lava burst out, accord- ing to Sir G. Mackenzie, at three different points, about eight or nine miles from each other, and spread in some places to the breadth of several miles f. The whole of Iceland may be considered as little else than a volcanic mass, in which there are many apertures through which lava, ashes, and other products have been ejected. The igneous matter struggles to escape in various places, and, con- sequently, many single eruptions from different points have been recorded since historical times; nevertheless, volcanic discharges have taken place at various times through the same apertures. Thus, there have been twenty-two eruptions from Hecla since the year 1004; seven from Kattlagiau Jokul since 900 ; and four from Krabla since 1 724. As might be expected in such a region as that of Iceland, the eruptions are not confined to the immediate dry land, but have pierced through the sea in the vicinity. In January 1783, a volcanic eruption, described as flame, rose through the sea, about thirty miles from Cape Reikianes ; several islands were observed, as if raised from beneath, and a reef of rocks exists where these appearances occurred. " The flames lasted se- veral months, during which, vast quantities of pumice and light slags were washed on shore. In the beginning of June, earthquakes shook the whole of Iceland ; the flames in the sea disappeared; and the dreadful eruption commenced from the Shaptar Jokul, which is nearly two hundred miles distant from the spot where the marine eruption took place J." Another submarine eruption occurred near the same island, on June 13th, 1830. An island was produced, and consequent eruptions were fearedin the interior, as in the case above cited ||. An example of a volcano forcing its way from beneath the sea into the atmosphere was observed off St. Michael's, Azores, in 1811. It was first seen above the sea on June 13th. On the 17th it was observed by Captain Tillard and some other gentlemen from the nearest cliff of St. Michael's. The appearances were exceedingly beautiful, the volcano * Daubeny. Description of Volcanos, p. 381. t Sir George Mackenzie, Travels in Iceland, 2nd edit. J Ibid. \\ Journal de Geologic, torn. i. Active Volcanos. \ \ 9 shooting up columns of the blackest cinders to the height of between 700 and 800 feet above the surface of the water. When not ejecting ashes, an immense body of vapour or smoke revolved almost horizontally on the sea. The bursts are described as accompanied by explosions resembling a mixed discharge of cannon and musketry, and by a great abundance of lightning *. By the 4th of July a complete island was formed, described by Capt. Tillard (who landed upon it,) as nearly a mile in circumference, almost circular, and about 300 feet in height. In the centre there was a crater, then full of hot water, which discharged itself through an opening facing St. Michael's. To this island, which after- wards disappeared, Capt. Tillard gave the name of Sabrina, from that of the frigate which he commanded. By reference to the manuscript journals of the Royal So- ciety of London, I find that a volcanic island was thrown up among the Western Islands about the middle of the seven- teenth century. Sir H. Sheres is described, in the account of the meeting of the Royal Society, on January 7th, 169091, as having informed those assembled, " that his father passing by the Western Islands went on shore on an island that had then been newly thrown up by a volcano, but that in a month or less it dissolved, and sunk into the sea, and is now no more to be found f." The volcano which rose through the sea, between the island of Pantellaria and the coast of Sicily, in July 1831, af- fords us a recent example of the propulsion of igneous and other rocks through the sea into the atmosphere, forming an island. The water was observed, by Neapolitan vessels, to be heaved up and agitated, and smoke to be evolved over the spot in the early part of July. Intelligence of the circum- stance having been received at Malta, vessels were dispatched to ascertain the exact position of the new volcano, and to warn other ships of the danger. On the 18th and 19th of July, Capt. Swinburne estimated the crater, then above the sea, at seventy or eighty yards in external diameter, and twenty feet above the water in the highest place, the agitated and heated water in the crater escaping by an outlet on one side. Dr. Davy visited this volcano on the 5th of August, and has presented us with a detailed description of the various * For a view of this scene, and a plan and elevation of the island, see Sections and Views illustrative of Geological Phsenomena, pi. 34 & 35. t These manuscript and unpublished journals of the Royal Society con- tain a fund of curious information, highly illustrative of the science of the time, the heads of the conversations at each meeting being entered. They moreover afford a valuable insight into the progress of science since the first establishment of the Royal Society. 120 Active Volcano*. phaenomena lie observed. The most common appearance is represented to have been that of a dense white vapour, pro- bably the vapour of water, thrown up to a great height. In violent eruptions, columns of black matter rose to the eleva- tion of three or four thousand feet, spreading out widely even to windward. Vivid lightning, accompanied by thunder, darted amid the atmosphere of eruption, and the sounds thus produced were far more intense than those resulting from the eruptive explosions. Dr. Davy watched in vain for any appearance indicative of the presence of inflammable gas. When in the midst of a dark shower of ashes to the leeward of the volcano, this author remarks, that excepting once or twice, when a slight smell of sulphur was perceived, "no unusual odour, not the slightest bituminous smell, or smell of sulphuretted hydro- gen, or of sulphurous acid, or of any other acid fume, was observable*." The various solid matters examined were im- pregnated with saline matter, resembling that contained in the sea, and afforded slight traces of sulphur. Among the sub- stances ejected were fibres, resembling vegetable fibres, and which emitted a smell like that of burning sea-weed, and it is conjectured that they were portions of weed sucked into the crater. M. C. Prevost, charged by the Academy of Sciences of Paris with the examination of this volcanic island, visited it on the 28th of September. The island was then about 2300 feet in circumference, and from 100 to 230 feet high, having two principal elevations, on different sides of the crater. The crater filled with boiling water. From the various accounts which have been received, it does not appear that any lava-current was ejected. The island seemed merely the result of an accumulation of ashes, cinders, and small fragments, among which were occasionally some of larger size. Among the ejected substances were pieces of dolomite and limestone; one fragment, of some pounds weight, is said to resemble grauwacke. As was to be expect- ed, the island, when the eruptions ceased, was unable to re- sist the action of the breakers ; accordingly it has given way before this force, and is now reduced to a shoal. The bottom whence this island was thrown up is repre- sented to have been at little depth beneath the surface of the sea. Capt. Smyth has, however, clearly shown that this was not the case, but that on the contrary it rose in deep water, unconnected with the Adventure Bank, to the westward of itf. Some miles to the S.E. of this island, which has been named Sciacca, Julia, Hotham, Graham, and Corrao, is the * Dr. Davy, Phil. Trans., 1832. Smyth, Phil. Trans., 1832, pi. 7. Active Vulcanos. 121 Nerita bank, which may perhaps be the subaqueous summit of another volcanic mass. If we consider that the heat of the volcano would destroy the various marine animals, which either lived in or on the sand, mud, or gravel of the previous bottom, such animals would probably be buried beneath the mud and cinders de- rived from the explosions. The volcano would seem to have been in activity some little time beneath the sea before it reared itself above the surface ; for Capt. Swinburne, while passing over the same spot on the 28th of the previous month (June), experienced shocks which were then attributed to an earthquake*. It can only have been since historical times, and by mere accident, that instances of volcanos so forcing themselves from beneath the sea could have been recorded. Now, the power of man to do this is so recent, that we may conclude such oc- currences to have been far from rare ; and that, even in the present day, they may happen in remote regions, into which civilized man rarely, if ever, enters, and therefore they re- main unknown. There are numerous islands in the ocean, composed almost entirely of volcanic matter, and in which active volcanos still exist, that may have been thus formed ; the dome or cone not giving way before the pressure of the water, but gradually accumulating a mass of lava, cinders and ashes, so that the islands have become firm, and even of considerable size. Owhyhee, or Hawaii, is perhaps a magnificent example of such an island. The whole mass, estimated as exposing a surface of 4000 square miles, is composed of lava, or other volcanic matter, which rises in the peaks of Mouna Roa and Mouna Kaah, to the height of between 15,000 and 16,000 feet above the level of the sea. Mr. Ellis describes the crater of Kirauea as situated in a lofty elevated plain, bounded by a precipice fifteen or sixteen miles in circumference, apparently sunk from two hundred to four hundred feet below its original level. " The surface of this plain was uneven, and strewed over with loose stones and volcanic rock ; and in the centre of * It is impossible not to be struck, in the drawings and plans of the islands of Sabrina and Sciacca, with the resemblance they bear to those volcanic islands which have basins in them, into which there is a narrow pas- sage communicating with the sea. Deception Island, New South Shetland, (of which there is a description and a plan in the Journal of the Geogra- phical Society,) affords a good idea of such islands. The interior basin is there five miles in diameter and ninety-seven fathoms deep. Many other examples will readily present themselves to the geographer. The commu- nication between the interior basin and the sea would seem produced, in the cases of Sabrina and Sciacca, by the rush of the waters out of the cra- ter during the explosions. 122 Active Volcano. it was the great crater, at a distance of a mile and a half from the place where we were standing. We walked on to the north end of the ridge, where, the precipice being less steep, a descent to the plain below seemed practicable. After walk- ing some distance over the sunken plain, which in several places sounded hollow under our feet, we at length came to the edge of the great crater, where a spectacle sublime, and even appalling, presented itself before us. Immediately before us yawned an immense gulf, in the form of a crescent, about two miles in length, from N.E. to S.W., nearly a mile in width, and apparently 800 feet deep. The bottom was covered with lava, and the S. W. and northern parts of it were one vast flood of burning matter in a state of terrific ebullition, rolling to and fro its ' fiery surge ' and flaming billows. Fifty-one conical islands of varied form and size, containing so many craters, rose either round the edge, or from the surface of the burning- lake : twenty-two constantly emitted columns of gray smoke, or pyramids of brilliant flame ; and several of these at the same time vomited from their ignited mouths streams of lava, which rolled in blazing torrents down their black indented sides into the boiling mass below." Mr. Ellis concluded, from the ex- istence of these cones, that the mass of boiling lava resulted from the streams poured from the craters into this upper re- servoir, which appeared to vary in its level ; for there were marks on the rocks bounding it, which showed that the great crater had been recently filled up 300 or 400 feet higher to a black ledge, from whence there was a slope to the hot fluid mass *. It will be obvious that this crater by no means resembles those with which we are more familiar. Instead of the more or less rounded orifice usually found, we have a semicircular crack in a level of considerable extent, and, by the descrip- tion, this level does not appear to have been ravaged by lava streams flowing from the crater over it. The depth of water round Owhyhee, and indeed round the Sandwich Islands ge- nerally, is so great, that they are somewhat dangerous to ap- proach in stormy weather, as anchorage cannot be obtained except close to the land ; seeming to show that these volcanic masses rise from considerable depths, and are only partly out of the water. The number of volcanos which fringe the Pacific Ocean, or occur in it, or in that part of the Indian Seas which contains Java and the neighbouring islands, far exceeds that of any * Ellis, Tour through the Sandwich Islands. An interesting account of the state of Kirauea, in 1829, will he found in Stewart's Visit to the South Seas. The general description is not materially different, the changes being principally in the crater. Active Volcanos. 123 other part of the world. From Terra del Fuego they occur northerly through the range of the Andes, often attaining very considerable elevations. In Mexico the northerly line is met by an east and west line, connecting it with the volcanos in the West Indian Islands. In California there are three vol- canos, of which one, Mount St. Elia, is variously estimated from 13,000 to 17,000 feet in height. America is connected with Asia by means of the volcanic vents of the Aleutian Isles. From Kamtschatka southwards we observe volcanos in the Kurule Islands, Japan, the Loo Choo Isles, Formosa, and the Philippines. From the latter islands a range of volcanic vents proceeds to nearly lat. 10 S., ranges westward along this pa- rallel for about twenty-five degrees of longitude, and then turns up N.W. diagonally through about twenty degrees of latitude. This line, which when represented in maps* resembles an enormous fish-hook, passes from the Philippines, by the N.E. point of Celebes, Gilolo, the volcanic isles between New Gui- nea and Timor, Floris, Sumbawa, Java, and Sumatra, to Barren Island. Active volcanos are by no means relatively so abundant in, or on the shores of, the Atlantic. Indeed the shores of this ocean in Europe, Africa, and America, appear free from them, if we except Mexico and the land connecting the main body of North America with the Southern continent, and which may be considered as common both to the Atlantic and Pa- cific Oceans f . Teneriffe affords the greatest volcanic elevation in the At- lantic, the Peak rising 12,216 feet above its surface. Iceland, though its volcanos do not attain any considerable elevation, presents the largest accumulation of volcanic matter above the level of the same mass of waters. We have seen that in Iceland high cones or elevations of land do not always accompany volcanic eruptions, for the lava of ] 783 seems to have flowed from comparatively low aper- tures. Elevations seem more especially formed when the erupted matter consists of cinders, ashes, or stones, which being ejected, arrange themselves in a conical manner around the central aperture, where the amount of melted rock or lava may vary. The escape of this melted rock will, in a great measure, depend on its relative proportion to the cinders, ashes, or stones thrown out. If these be in comparatively * See Von Buch's Canary Islands, pi. 13 ; and a corrected reduction of this in Lyell's principles of Geology, pi. 1. f Mr. Scoresby notices a volcano off the main land of Greenland. This volcano is situated in the island of Jan Mayen, presented marks of recent eruption, and had a crater about 500 feet deep, and 2000 feet in diameter. Edin. Phil. Journal. 124? Active Vulcanos. small quantity, the lava will have the less difficulty to escape, and may easily break down its barrier and rush Forth. But when the proportions are inverted, a large cone may be raised without the escape of any lava-current. Between the two ex- tremes there will be every kind of variation, and lava-currents will flow from various apertures and at various heights. By repeated action a volcano acquires considerable solidity at its base, for the loose erupted matter is, independently of the con- soli lation produced by other causes, bound together by lava radii proceeding from the central aperture. Rents are often produced in the base, particularly when the great vent has accumulated matter to a considerable height, and through these, lava is protruded; the streams so thrown out serving to brace the lower parts of the mountain more firmly toge- ther. The occurrence of such apertures is precisely what we should expect in a volcano, which had accumulated materials upon it nearly equal to the average force of the elastic vapours propelling igneous matter upwards; for the pressure of the elevated column being very considerable, and in proportion to its height, it will always struggle to free itself in the direc- tion of the least resistance. Now the sides of a volcanic moun- tain are not likely to be homogeneous, but to vary much in their resisting powers, being most solid where crossed by lava- currents, and weakest where merely formed of ashes or sub- stances of the like nature. If to these causes of unequal re- sistance to pressure, we add the fractures and rents produced by shocks in the mountain itself, we should always expect to find lateral discharges of lava common, while similar streams from the mouth would be rare. M. von Buch is the author of a theory respecting the ele- vations of volcanos, which has been adopted by many geolo- gists, while it has been combated by others. He observes, that the appearances of many craters are such, that we can scarcely consider them as erupted in the ordinary way ; be- cause they do not seem to present either lava-currents, or such an arrangement in the deposit of other volcanic sub- stances as to justify such a conclusion. To these craters he has given the name of Craters of Elevation (Erhebungskrater}. It has been opposed to this theory, that it presupposes a ho- rizontal accumulation of lava or other volcanic matters, pre- viously to the propulsion of elastic vapours through it, which should elevate the flat mass in a dome or cone, and burst through the highest part, presenting the appearance of a cra- ter of eruption. How far this objection may be valid would seem to depend on the possibility of forming sheets of volcanic matter, which heat might soften and elastic vapours force up, so that the necessary forms should be produced. It may be Active Volcano*. 125 questionable whether, under a great pressure of the sea, there is the same tendency to produce cinders and ashes as in the atmosphere ; and whether the superincumbent weight would not so act upon the solid matter ejected, that it would be forced into fusion, and sheets of melted matter be the result, if the elastic vapours beneath a column of melted rock were suffi- ciently powerful to overcome the resistance both of the column of lava and the superincumbent water. It being by no means probable that the density of sea-water, beneath any depth which we can reasonably assign to the ocean, would be such as to render it of greater specific gravity than liquid lava ejected from a volcanic rent, situated beneath the sea; it would follow, that so long as the lava continued in a state of fusion, it would arrange itself horizontally beneath the fluid of inferior specific gravity. The question then arises, how long a body of lava in fusion would remain fluid be- neath the waters of the sea. The particles of water in contact with the incandescent lava would become greatly heated, and consequently, from their decreased specific gravity, would immediately rise, their places being supplied from above by particles of greater density and less temperature. Thus a cooling process would be established on the upper surface of the lava, rendering it solid. Now as the particles of fluid lava would be prevented from moving upwards by the solid matter above, pressed downwards by its own gravity and the superincumbent water, they would escape laterally, where not only the cooling process would be less rapid, from the well-known difficulty of heated water moving otherwise than perpendicularly upwards, but where also the power of the fluid lava to escape resistance would be greatest. Let a Fig. 20. (Fig. 20) be a volcanic rent, through which liquid lava is propelled upwards in the direction d f. The lava being of greater specific gravity than the water b h e c, it would tend 126 Active J'ola/nos. to arrange itself horizontally in the directions d b, d c. The surface b d c having become solid, the lava would escape from the sides b and c, spreading in a sheet or tabular mass around ; and this effect would continue so long as the propelling power at a was sufficient to overcome the resistance opposed to the progress of the lava, or until the termination of the eruption, if that should first happen. If such would be the state of things beneath a considerable depth of water, the tendency to produce ashes and cinders in a volcanic vent would increase with its approach to the sur- face of the water; and therefore all the phenomena of erup- tions from beneath the surface of the sea would differ but little from those observed in the atmosphere. Another objection to the theory of craters of elevation is, that the stratification of such supposed craters is precisely that of craters of eruption ; and that therefore the inference from this circumstance would be in favour of the latter, because we now have daily examples of such modes of formation, while of the other we have none. Data on this subject are so few, that it seems difficult to esti- mate the value of this objection. The fact, however, that solid rocks can be raised by elastic vapours, is shown in the case of the Little and New Kameni, (Island of Santorino,) where brown trachyte, of a resinous lustre and full of crystals of glassy felspar, was upraised ; the former in 1573, and the latter in 1707 and 1709. The elevation of the Little Kameni was "accompanied by the discharge of large quantities of pu- mice, and a great disengagement of vapour *." By terming this rise an earthquake, we merely seem to be using two names for the same thing. That there were elastic vapours it is clear, and that these vapours were the propelling power may be fairly inferred ; therefore the fact is the same, whether we call it an earthquake or a volcanic elevation, and it would be somewhat difficult to draw fine lines of distinction between the two. The trachyte of New Kameni was observed to have shells upon it when raised, and limestone and marine shells are described as composing a part of these otherwise igneous islands f. These occurrences at Santorino are quite sufficient to show, that volcanic rocks, with shells upon them, may be raised bodily to the surface. Clayslate and limestone ap- pear also to have been forced upwards at some previous pe- riod, as they are seen at Mount Elias, dipping from the in- terior outwards. Langsdorff notices a trachyte rock 3000 feet high, which appeared in 1795 near the island of Unalaschka, and which seemed to have been thrown up as a mass from the Lyell, Principles of Geology, vol. i. p. 386. f Ibid. Active Folcanos. 127 bottom of the sea* . M. Omalius d'Halloy cites M. Rein- wardt as stating, that on the western side of the Isle of Bnnda, a bay was, in 1820, replaced by a promontory formed of huge blocks of basalt, The rise of land is described as having been so gradual, that the inhabitants were not aware of the change until it was nearly completed. It was accompanied by a bubbling and great heat of the seaf. Considering this account as correct, it is a remarkable example of the quiet rise of land above the level of the ocean. Ingenious explanations have been given to account for the large orifices which have been termed craters of elevation. Mr. Lyell considers that the crater resulting from the de- struction of the summit of Etna in 1444-, was as large as those noticed in other places and named craters of elevation ; and supposes that a series of great explosions might so reduce the cone, that finally there would be a circular bay, forty or fifty miles round, in an island seventy or eighty miles in circum- ference, wholly composed of volcanic rocks which should dip outwards. But supposing such appearances to have been produced, the whole base of Etna, a kind of circular island, would still show its lava-currents, sections of which would be observed in the interior bay, or might be exposed outside, and no doubt would remain that it was a crater of eruption. How far the so-called " craters of elevation" may resemble the sup- posed case of Etna remains to be seen ; yet if they should not, as is considered they do not, present traces of lava-currents, radiating from a centre or centres, but largo envelopes of tra- chyte or other fused volcanic rock, they can scarcely be re- ferred to the same origin. There does seem a possibility of producing craters of elevation by the action of heat and elastic vapours on a sheet of lava, therefore the subject should be fairly investigated, without bias, with proper caution, and in the necessary detail. It is supposed that after the craters of elevation were formed, the eruptive action poured forth the usual volcanic substances, which, when it was continued sufficiently long, produced a cone like the Peak of Teneriffe ; but when such eruptive action was small, or the crater comparatively recent, the appearances were such as we now observe at Barren Island in the Bay of Bengal, where a central cone, in activity, in the midst of a basin of water, is surrounded by a circular range of volcanic ground, which, according to the figure given by Mr. Lyell f, rises at an angle of about 45 from the sea. The height of * Daubeny, Description of Volcanos, p. 310. f Omalius d'Halloy, Elements de Geologic, p. 40.5. I Principles of Geology, vol. i. p. 390. 128 Active Folcanos. the central cone is about 1800 feet above the water, and the elevation of the surrounding volcanic circle being nearly the same, the interior is only viewed through a break in it. It would appear that the rocks of this island are extremely hot ; for Capt. Webster, landing upon it in March 1822 or 1823, found the water almost boiling at one hundred yards from the shore; the stones upon the beach, and the rocks exposed by the ebb tide, hissing and steaming, and the water bubbling around them*. Von Buch adduces the Caldera in the Isle of Palma, Cana- ries, as a good example of the craters of elevation. A large precipitous cavity or crater exists in a lofty range sloping outwards, which incloses it on all sides but one, where a gorge forms the only communication from the exterior to it. The sides of this great cavity expose a section of beds of basalt, and conglomerates composed of basaltic fragments, dipping regu- larly outwards. White trachyte, and a rock composed of hornblende and white felspar, are also noticed. Now if the beds be so regular, and not composed of scoriaceous matter or ashes, as it is stated they are not, their formation would seem not to have taken place in the air or beneath a small pressure of water, but under different circumstances, which would permit the basalt to be flattened into tabular masses, not presenting the appearance of lava-currents which have flowed in the atmosphere. Jorullo affords a striking example of the outburst of volcanic action in the interior of dry land, where no active volcanos then existed, though the rocks in the vicinity would seem to indicate their previous presence. Judging from the direction of the vents, a cleft seems to extend east and west across Mexico to the Revillagigedo Isles in the Pacific. Previous to June 1759, the space where the volcano of Jorullo now stands was covered by indigo and sugar-canes, bounded by two brooks, the Cuitimba and San Pedro. In June, hollow subterranean noises were heard, accompanied by earthquakes, which lasted from fifty to sixty days. Tranquillity seemed re-established at the commencement of September, but on the 28th and 29th of this month the subterraneous noises again commenced, and, according to Humboldt, the ground, with a superficies of three or four square miles, rose up like a bladder. The extent of this movement is considered to be now marked by an elevation round its edges of 39 feet, gradually acquiring a height of 524? feet towards the centre of the present vol- canic district. The eruption appears to have been very violent, fragments of rock were hurled to great heights, ashes # Edin. Phil. Journal, vol. viii. Active yolcanos. 129 were thrown up in clouds, and the light emitted was seen at considerable distances. The Cuitimba and San Pedro are described as having precipitated themselves into the volcanic vent, and to have assisted, by the decomposition of theii waters, the fury of the eruption. " Eruptions of mud, and especially of strata of clay, enveloping balls of decomposed basalt in concentric layers, appear to indicate that subterra- neous water had no small share in producing this extraordinary revolution. Thousands of small cones, from 6 to 10 feet in height, called by the natives Hornitos (ovens), issued forth from the Malpays. Each small cone is a fumirole, from which a thick vapour ascends to the height of from 22 to 32 feet. In many of them a subterraneous noise is heard, which ap- pears to announce the proximity of a fluid in ebullition." From amid these cones, six volcanic masses, varying from 300 to 1600 feet in height above the old plain, were ejected from a chasm having a direction N.N.E. and S.S.W. The most elevated mass is named Jorullo, and from its north side a con- siderable quantity of lava, containing fragments of other rocks, has been thrown out. The great eruptions ceased in Febru- ary 1760, and afterwards became gradually less frequent. The opponents to the theory of craters of elevation consider the raising of the ground in the form of a bladder as not alto- gether proved, resting on Indian accounts of appearances, which have been considered with reference to a particular theory. The well-known Monte Nuovo near Naples was thrown up in a day and a night in 1538. This is also described as ejected from a* fissure. The present height of this volcanic elevation is 4-4-0 feet above the sea, and its circumference about a mile and a half. Various descriptions of volcanic eruptions will be found in works dedicated to the subject, and could not be admitted within the necessary limits of this volume. The following ac- count, however, obtained by the exertions of Sir Stamford Raffles, of a great eruption from Tomboro, in the island of Sumbawa, is too important to be omitted. The first explo- sions were heard at various distant places, where they were very generally mistaken for discharges of artillery. They commenced on the 5th of April 1815, and continued more or less until the 10th, when the eruptions became more violent; and such a great discharge of ashes took place that the sky was obscured, and darkness prevailed over considerable di^ stances. It appears that a Malay prow, while at sea on the llth, far from Sumbawa, was enveloped in utter darkness, and that, afterwards passing the Tomboro mountain at the distance of about five miles, the commander observed that the lower part 130 Active Volcanos. appeared in flames, while the upper portion was concealed in clouds. Upon landing, for the purpose of procuring water, he found the ground covered to the depth of three feet by ashes, and " several large prows thrown on shore by the con- cussion of the sea." Quitting Sumbawa, he with difficulty sailed through a quantity of these ashes floating on the sea, which he described as two feet thick, and several miles in ex- tent. This person also stated that the volcano of Carang Assam, in Bali, was convulsed at the same time. The most interesting account is that presented us by the commander of the East India Company's cruiser Benares, which is nearly as follows : At the commencement of the explosions this vessel was at Macasar, and the reports so closely resembled those of cannon, that it was supposed there was an engagement of pi- rates somewhere in the neighbourhood. Troops were conse- quently embarked on board the Benares, and the vessel stood out to sea in search of the supposed pirates. On the 8th of April she returned, without having found any cause for alarm. On the llth the apparent discharges of cannon were again heard, sometimes shaking the ship and Fort Rotterdam. The vessel proceeded southward to ascertain the cause of these explosions. At eight o'clock on the morning of the 12th, " the face of the heavens to the southward and westward had assumed a dark aspect, and it was much darker than when the sun rose ; as it came nearer it assumed a dusky red appear- ance, and spread over every part of the heavens ; by ten it was so dark that a ship could hardly be seen a mile distant ; by eleven the whole of the heavens was obscured, except a small space towards the horizon to the eastward, the quarter from which the wind came. The ashes now began to fall in show- ers, and the appearance was altogether truly awful and alarm- ing. By noon the light that remained in the eastern part of the horizon disappeared, and complete darkness covered the face of day. This continued so profound during the remainder of the day, that I (the commander of the Benares) never saw anything to equal it in the darkest night ; it was impossible to see the hand when held close to the eyes. The ashes fell without intermission throughout the night, and were so light and subtile that, notwithstanding the precaution of spreading awnings fore and aft as much as possible, they pervaded every part of the ship." "At six o'clock the next morning it continued as dark as ever, but began to clear about half-past seven, and about eight o'clock objects could be faintly observed on deck. From this time it began to clear very fast The appearance of the ship when day-light returned was most singular; every part being covered with the falling matter. It had the appearance Active Volcanos. 131 of calcined pumice-stone, nearly the colour of wood-ashes ; it lay in heaps of a foot in depth on many parts of the deck, and several tons weight of it must have been thrown overboard ; for though an impalpable powder or dust when it fell, it was, when compressed, of considerable weight. A pint measure of it weighed twelve ounces and three quarters ; it was perfectly tasteless, and did not affect the eyes with a painful sensation ; had a faint smell, but nothing like sulphur; when mixed with water it formed a tenacious mud difficult to be washed off." The same vessel left Macasar on the 13th, and made Sum- bawa on the 18th. Approaching the coast she encountered an immense quantity of pumice-stone, mixed with numerous trees and logs with a burnt and shivered appearance. When arrived at Bima Bay, the anchorage was found to be altered, as the vessel grounded on a bank where a few months pre- viously there had been six fathoms of water. The shores of the bay were entirely covered with the ashes ejected from Tomboro, which is distant about forty miles. The explosions heard at Bima were described as terrific, and the fall of ashes so heavy as to break in the Resident's house in many places. There was no wind at Bima, but the sea was greatly agitated, the waves rolling on shore, and filling the lower parts of the houses a foot deep. When off the Tomboro mountain, about six miles distant, on the 23rd, the commander of the Benares observed the summit to be enveloped in smoke and ashes, while the sides showed lava-currents, some of which had reached the sea. The explosions were heard at very considerable distances. Not only were they noticed at Macasar, which is 217 nautical miles from Tomboro, but also throughout the Molucca Islands; at a port in Sumatra, distant about 970 nautical miles from Sumbawa; and at Ternate, distant 720 miles. Lieut. Phillips being dispatched to relieve the wants of the inhabitants, who were perishing from famine and disease, learned from the Rajah of Saugar, that about seven o'clock in the morning of the 10th of April there was an appearance of three distinct columns of flame, all within the crater, which united at a great height upwards ; and that, subsequently, the whole mountain appeared like a mass of liquid fire. How far the appearance of flame may be correct, it would be difficult to say, as nothing is so common as deceptive appearances of this kind ; its character, however, would seem remarkable. The Rajah's account proceeds : " The fire and columns of flame continued to rage with unabated fury, until the dark- ness caused by the quantity of falling matter about eight P.M. Stones at this time fell very thick at Saugar, some of them as large as two fists, but generally not larger than walnuts." K 2 132 Active Volcanos. Soon after 10 P.M. a violent whirlwind arose, "which blew clown nearly every house in the village of Saugar, carrying the tops and light parts along with it In the part of Saugar ad- joining Tomboro its effects were much more violent, tearing up by the roots the largest trees, and carrying them into the air, together with men, houses, cattle, and whatever else came within its influence." The sea was agitated, rising twelve feet higher than it was ever known to do before. The water rushed upon the land, sweeping away houses and all within its influence, and destroying the few rice-grounds which pre- viously existed at Saugar. As might have been expected amid such a convulsion, a great destruction of life was effected, and many thousand inhabitants were killed. The vegetation on the north and west sides of the peninsula was completely destroyed, with the exception of a high point of land where the village of Tomboro previously stood, and where a few trees still remained*. The changes produced by such eruptions as that here re- corded, would, independently of the alteration in the shape of the volcano itself, and of the streams of lava which flowed from it, extend to very considerable distances. On the dry land, vegetables and animals would be entombed beneath stones and ashes, the quantity of the covering matter probably increasing with the proximity to the volcano. And if it should chance, as sometimes happens, that the aqueous vapours discharged from the volcanic vent were suddenly condensed, the torrents produced would sweep away not only the looser parts of the volcano, but also the plants and animals which they might encounter, embedding them in a thick mass of alluvial matter. The vegetable and animal substances enveloped by the dis- charged ashes, cinders, and stones falling into the sea, would be both marine and terrestrial ; and a very curious mixture, as far as regarded its organic contents, would be observed : trees, men, cattle, fish, corals, and a great variety of marine re- mains, would be encased, and it might so happen that both on the land and in the sea a bed of lava might cover such ac- cumulations. In the case of the great discharge of lava in Iceland, in the year 1 783, many terrestrial remains may have been covered by the igneous matter, possibly some in such situations as to preserve their form. Should a similar eruption take place in the sea, where, as before observed, the conditions are favourable for the production of a sheet of lava, sands and clays, perhaps full of marine remains, would be covered over, and very considerable changes might be produced by such a Life of Sir Stamford Rafiles. Active Volcanos. 13S superincumbent mass of heated matter. Upon these, after a certain time, sands and clays, charged with organic remains, might be accumulated, and again covered up by a new erup- tion. Thus producing an alternation of igneous and aqueous rocks. Mr. Henderson notices an alternation of fossil wood, clay, and sandstone in Iceland, surmounted by basalt, tuff, and lava. When this accumulation of vegetable matter was so covered is not so clear ; but if Mr. Henderson be right in considering many of the fossil leaves as those of the poplar, it is not, pro- bably, very recent, for it supposes a change of climate, as poplars do not now grow in Iceland*. During great explosions, volcanos cannot be approached sufficiently near for the purposes of very minute observation ; therefore we can only judge of some of the probable effects from appearances at their calmer periods, and consequently a minor state of activity is very favourable for such examinations. After ineffectual attempts to observe the workings of the fluid mass within the crater of Vesuvius at the commencement of 1829, when that mountain was somewhat active, 1 was fortu- nate enough on the 15th of February to have ascended on a calm day, when the vapours darted majestically upwards as they were propelled from the small cone in the middle of the grand crater f. The incandescent matter in the vent was at times distinctly visible, a rare circumstance; for should there be the slightest movement in the air, every object is obscured by the vapours. After the more continued detonations there was a lull or calm, succeeded by a violent explosion, throwing up stones to a considerable height, mixed with pieces of red- hot lava, the latter falling like lumps of soft paste on the sides of the small cone. When the vapour cleared away, the red- hot mass appeared as if in ebullition from the passage of the gaseous matters through it. The light emitted varied exceed- ingly in intensity, being brightest at the moment of the great explosion, when a great volume of vapour suddenly forced its way through the fiery mass, darting up with great velocity, and carrying all before it. Wishing to profit by my good fortune, I continued many hours on the mountain, until night closed in, hoping that objects might be perceived within the crater not previously observed. In this I was disappointed, appearances being the same, though more distinctly visible. The picturesque effect, however, was greatly heightened ; the solid ejected substances darted upwards like a grand discharge * Henderson's Iceland, vol. ii. p. 115. According to this author, the lig- nite deposit occurs extensively in the N.W. peninsula of Iceland. t For a sketch of the crater at this time, see Sections and Views illus- trative of Geological Phenomena, pi. 22. 134- Extinct Volcanos. of red-hot balls, while the reflection of the incandescent mat- ter within, on the vapour above, was at times extremely brilliant, producing, at a distance, those false appearances of flames so frequently noticed. Flame, that is, the combustion of some inflammable gas, does, however, seem to issue, though very rarely, from volcanic vents. Sir Humphry Davy notices a jet of flame as proceeding from an aperture on the side of Vesuvius facing Torre del Greco, in May 1814: it rose to the height of sixty yards, accompanied by a violent hissing noise. It continued for three weeks, but Sir Humphry Davy was unable to collect any inflammable gas*. The products of active volcanos, though man seems to ex- haust his language in rinding terms to express his horror and dismay at their mode of ejection, do not constitute such an addition to dry land as at first sight would appear probable, for their mass must be regarded relatively to the mass of dry land generally, and not with reference to particular districts. Moreover, cavities corresponding with the quantity of matter thrown out will sometimes occur not far beneath the surface ; and when the weight above shall overcome the resistance be- low, either suddenly from a violent convulsion, or slowly from gradual change, the mass above will fall into the abyss beneath, and matter be, in some measure, restored to its place. Among volcanic changes it is by no means uncommon to hear of hills disappearing, and being converted into lakes. The most me- morable example, perhaps, of the disappearance of a volcano, is that which took place in Java in 1772. The Papandayang, on the south-western part of the island, reputed one of its largest volcanos, was observed at night, between the llth and 12th of August, to be enveloped by a luminous cloud. The inhabitants being alarmed, betook themselves to flight; but before they could all escape, the mountain fell in, accompanied by a sound resembling the discharge of cannon. Great quan- tities of volcanic substances were thrown out, and carried over many miles. The extent of ground thus swallowed up was estimated at fifteen miles by six. Forty villages were engulfed or covered by the substances thrown out, and 2957 persons were reported to have been destroyed f. Extinct Volcanos. From a similarity of appearances, rocks existing under cer- tain circumstances where there are at present no active vents, have been attributed to a volcanic origin. To draw fine lines of distinction between volcanos now in activity and those * Davy, Phil. Trans. 1828. f Horsfield, as quoted by Daubeny. Extinct Volcanos. which appear extinct, would be almost impossible, for there is no certainty that the one may not soon be converted into the other. Of this we probably have a good example in Vesuvius, which after being, as far as we can judge from historical re- cords, for a long period extinct, became convulsed in the year 79, destroyed the higher part of its old cone, part of which now remaining is named Monte Somma, and overwhelmed Herculaneum, Pompeii, and Stabiae, entombing not only men, but theatres, temples, palaces, and innumerable works of art, which have afforded by their disinterment more real know- ledge of the manners and customs of the ancient inhabitants of these beautiful regions of Italy, than all the writings which have escaped destruction. Solfataras, as they are termed, are usually considered as semi-extinct volcanos, emitting only gaseous exhalations and aqueous vapour; but there can be no certainty that they also may not again enter into activity. According to Dr. Daubeny, sulphuretted hydrogen and a small portion of muriatic acid are contained in the steam which rushes out of the fuma- roles at the Solfatara near Naples. The rocks of the crater and vicinity are greatly decomposed by the action of these gaseous exhalations ; and, among other salts thus formed, the muriate of ammonia is the most abundant. Solfataras, vari- ously modified, are by no means rare in volcanic countries. Not only do extinct volcanic vents occur in regions where active volcanos now exist, so that we may imagine a mere change of fiery orifice, but they are also found in districts where all trace of activity has been lost since the earliest hi- storical times, if we except the presence of mineral and thermal springs. In Central France and in Germany such appear- ances are particularly remarkable, and it has been attempted to draw a line of distinction between those volcanos which have existed in a state of activity since the establishment of the present order of things, and those whose activity was previous to this state. The subject is full of difficulty, more especially as respects Central France, where volcanic ejections have taken place at different periods ; so that there is no ready mode of making geological distinctions between the ejections, which would seem little else than productions from new orifices opened for the discharge of volcanic matter in the same region. We may be able to observe the extremes, but to mark striking and easily distinguishable points intermediate between them would be exceedingly difficult. Volcanic ejections were probably continued through nearly the same orifices for a long period of time, during which many and great geological changes were taking place around them, and on the surface of the earth generally. i 136 Extinct Volcano*. It has been attempted to determine the relative ages of vol- canos by the absence or presence of craters ; as also on the sup- position that some have existed prior to the excavation of val- leys, while others have been produced after their formation, their lava-currents having been discharged into them. Such distinctions can scarcely be made ; for craters may be easily obliterated, and relative age, from the excavation of valleys, cannot be very satisfactorily established amid circumstances which could so easily produce changes in this respect. A more direct mode has been to try their relative antiquity by means of the mineral structure of their lavas ; and if this should hold good, it would be the safest guide ; but it may be doubted how far our knowledge of volcanic products authorizes so ge- neral a conclusion. That there is a great difference in the mineral character, generally, between the igneous rocks of the older periods of the world and those at present formed, few- will doubt. We know of no granite or serpentine streams thrown out from modern volcanos ; but when igneous rocks so closely allied in geological dates as those produced by active and extinct volcanos are under consideration, such distinctions should not be too hastily adopted. Dr. Daubeny considers that the more modern volcanic pro- ducts of Auvergne are more cellular, have in general a harsher feel, and possess a more vitreous aspect than the more ancient *. In Auvergne and the Vivarais there are numerous examples of the more modern extinct volcanos, the craters of which are frequently very perfect, or merely broken down by the dis- charge of large lava-currents from them. Details respecting them will be found in works written expressly on the subject, and pictorial representations among the views contained in Mr. Scrope's work on Central France f. In the district of the Eifel, near the Rhine, there are also extinct volcanos which have been considered as comparatively recent, from the situations which they occupy ; having been apparently produced after the formation of the valleys in the neighbouring country. In the volcanic district of Central France the lava-currents have in some places traversed valleys, and dammed up the waters that pass through them. The waters so dammed up accumulated into a lake, which was subsequently drained through a gorge cut in the rocky barrier by means ot the surplus water: which not only effected this, but also cut, by continual erosion, into the rock beneath, forming a part of the original valley. Many other examples of extinct volcanos have been noticed * Description of Volcanos. f Part of one of the most striking of these views is copied in " Sections and Views illustrative of Geological Phenomena," pi. 24. Mineral Volcanic Products. 137 in districts where active volcanos do not now exist. Their relative antiquity is however so little understood, that a gene- ral classification of them cannot be attempted. Mineral Volcanic Products. Various classifications of volcanic substances have been proposed, among which the division into Trachytic and Ba- saltic seems to be that most commonly adopted ; trachyte being considered as essentially composed of felspar, and containing crystals of glassy felspar; while basalt is supposed to be es- sentially composed of felspar, augite, and titaniferous iron. Lavas, however, present such various mixtures of different minerals, that exact classifications of them would appear ex- ceedingly difficult; and when we consider that these different compounds may be infinitely modified by circumstances, such classifications cannot be of much value. These products are of such a compound nature, consisting of felspar, augite, leu- cite, hornblende, mica, olivine, and other minerals, that de- finite names can scarcely be attached to them. Mr. Poulett Scrope has distinguished the rocks termed trachyte, basalt, and graystone (the latter a name proposed by himself,) under the following heads: 1. Compound trachyte, with mica, horn- blende, or augite, sometimes both, and grains of titaniferous iron. 2. Simple trachyte, without any visible ingredient but felspar. 3. Quartziferous trachyte, when containing nume- rous crystals of quartz. 4-. Siliceous trachyte, when apparently much silex has been introduced into its composition. 1. Com" mon graystone, consisting of felspar, augite, or hornblende, and iron. 2. Leucitic graystone, when leucite supplants the felspar. 3. Melilitic graystone, when melilite supplants the felspar, c. 1 . Common basalt, composed of felspar, augite, and iron. 2. Leucitic basalt, when leucite replaces the fel- spar. 3. Olivine basalt, when olivine replaces the felspar. 4-. Hauyine basalt, when hauyine replaces the felspar. 5. Ferruginous basalt, when iron is a predominant ingredient. 6. Augite basalt, when augite composes nearly the whole rock*. As all fused substances will tend to crystallize, or arrange their component parts more compactly, where their liquidity continues the longest, and their loss of temperature is the slowest, we find that lava-currents are always more crystalline or compact in their interior parts, and that dykes cutting vol- canic cones are generally more compact and crystalline than the lavas which flow from them ; such dykes being also more Quarterly Journal of Science, vol. xxi. 1826. 138 Volcanic Dylces, $c. crystalline towards their interior parts than towards their walls or sides. It has been inferred from the appearance and dis- tribution of the ejected matters, that many volcanic rocks have not been formed in the atmosphere, but beneath seas, and that they have been subsequently elevated. The ashes and pumice ejected from volcanos seem merely, if I may so express myself, the frothy part of the great fused and incandescent matter within, produced by the action of elastic vapours, or by the intumescence of that matter under diminished pressure. The force required to eject such light substances is evidently far inferior to that necessary for the propulsion of the more solid lava, and consequently the one is in general more com- mon than the other. As might be expected from the nature of such mineral productions, volcanic substances vary, from the lightest ash to a highly crystalline rock, the intermediate states being vitreous, and of the character of obsidian. The quantity of minerals detected in volcanic products is exceed- ingly great, a circumstance by no means surprising when we consider the various elementary substances acted on by heat in the bowels of a volcano, and striving to combine with each other in various ways*. Not only are fused substances ejected, but also various por- tions of rocks traversed by the volcanic vent; and as this is very variously situated, so are the rocks various which are thrown out. Vesuvius having been under observation for so long a period, its products have received greater attention than falls to the lot of most volcanos ; and it has been observed, though no doubt volcanos vary most materially in this respect, that such ejected substances are far from being either rare or of one kind. The Chevalier Monticelli's invaluable collection of Vesuvian products at Naples contains a great variety of these substances, among which may be seen fragments of the compact limestones of the district, with organic remains in them, seeming to show that the vent traverses the limestones, and that the fiery mass rends off portions of them, as indeed might be expected from the nature of the country. The lime- stones so ejected are often impregnated with magnesia, sup- posed to have been acquired in this great natural crucible. Volcanic Dykes^ fyc. Dykes or fissures in the sides of volcanos, subsequently filled by melted lava, are sufficiently common. M. Necker de Saussure mentions numerous dykes which traverse the beds * Sulphur is exceedingly common, and is often sublimed in such quan- tities as to be carried away for economical purposes. Volcanic Dykes, S?c. 139 of Monte Somma. These veins are nearly all of the same composition, differing somewhat from the lava-beds they cut ; augite being more abundant, while leucite, so common in the beds, occurs rarely in the dykes, with the exception of one vein of Monte Otajano, and another near the foot of the Punte del Nasone, which contain large crystals of leucite. The Java of the dykes also contains minute crystals of felspar (?), with a considerable abundance of a yellow substance, which may be olivine. The rock composing the veins is fine-grained on the sides, and more crystalline in the middle. These veins vary from one to twelve feet in width. One remarkable dyke, different from the rest, occurs at Otajano. It is about ten feet and a half wide, and rises per- pendicularly to the crest of the mountain, having apparently turned up the alternating beds of porous and compact lava which it traverses. Another singular dyke cuts the rocks of the Primo Monte. It rises perpendicularly, and is formed of a slightly greenish gray and homogeneous rock. At its base (that of the mountain) it is only eleven inches wide, and for twelve feet of its height is bordered by a line of vitreous lava, half an inch thick, separating it from the porous volcanic breccia which it cuts. Above the twelve feet, the vitreous lava ceases entirely, the solid rock occupying the whole vein*. Dr. Daubeny notices tuff traversed by dykes of a cellular trachytic lava at Stromboli, and at Vulcanello in the island of Liparif. Dykes, described as resembling greenstone, were noticed by Sir George Mackenzie traversing alternate beds of tuff' and scoriaceous lava in Iceland. Dykes of porphyry traverse the older lavas of Etna. Their formation is by no means difficult of explanation, by suppo- sing fissures, which sometimes have, and sometimes have not, penetrated to the surface, injected with incandescent lava. Of fissures extending to the surface, the cleft twelve miles long and six feet broad, which opened on the flank of Etna, be- tween the plain of St. Lio and a mile from the summit, at the commencement of the great eruption of 1669, is an example |. This fissure gave out a vivid light ; from which Mr. Lyell with great probability concludes that it was filled to a certain height with incandescent lava. After the formation of this, five other fissures were produced, and emitted sounds heard at the distance of forty miles. While on this subject it may be as well to notice the pro- * Necker, Memoire sur le Mont Somma, Mem. de la Soc. de Phys. et d'Hist. Nat. de Geneve, 1828. f Daubeny 's Description of Volcanos, p. 185 187, where there are views of these appearances. % Lyell, Principles of Geology. Ibid. vol. i. p. 364. HO Earthquakes. bable effects of a column of lava passing through stratified rocks, insinuating melted matter among the strata, or through fissures formed in them. Let a b in the annexed diagram represent a column of li- quid lava, traversing horizontal strata. It is obvious that it will strive to overcome the resistance of the sides, and such re- sistance will always be less between the strata than elsewhere. If it obtain an aperture in that direction, it will endea- vour to separate one stratum from another ; and it will the more readily accomplish this, as to the pressure of the column of lava will be added the mechanical action of the wedge; and eventually an injection of liquid lava may be made, and carried laterally, so far as the pressure will permit. Thus, if a separation of the strata can be commenced at d, it will be carried on in the direction d c as far as the pressure of the column a d will permit. If, in- stead of this kind of injection, we consider the strata to have been fractured, as is very likely to be the case near volcanic action, the fissure will be filled, and forced asunder as far as resistances will permit. Thus, if a fracture efbe made, it will be filled by liquid lava as far as can be effected by the pressure of the column a e. The strata have here been sup- posed horizontal, for the sake of illustration, but as they might occur in all modes, the effects would be varied accord- ingly, the principle remaining the same. Earthquakes. The connexion between volcanos and earthquakes is now so generally admitted that it would be useless to enumerate the various circumstances that point to this conclusion. They both seem the effects of some cause as yet unknown to us. The motion of the ground produced by earthquakes is not always the same; sometimes resembling the undulatory movement of a heavy swell at sea, though much quicker, and being at others tremulous, as if some force shook the ground violently in one spot. The former of these is far the most dangerous, as it forces walls and buildings off' their centres of gravity, crushing whatever may be beneath them. It has been considered that earthquakes are presaged by certain atmospheric appearances, but it may be questionable to what extent this supposition is correct. Historians of earth- quakes seem to have been generally desirous of producing ef- fect in their descriptions, adding all that could tend to heighten Earthquakes. 141 the horror of the picture. They have not always, moreover, been anxious or able to separate accidental from essential cir- cumstances. As far as my own experience goes, which is however merely limited to four earthquakes, the atmosphere seemed little affected by the movement of the earth ; though I would be far from denying that it may be so ; for we can scarcely imagine such movements to arise in the earth, with- out some modification or change of its usual state of electri- city which would affect the atmosphere. If animals be gene- rally sensible of an approaching shock, it might arise as well from electrical changes as from the sounds which they may be supposed capable of distinguishing. Earthquakes very frequently precede violent volcanic ex- plosions, even though they may be felt far from a fiery vent. Thus, the great earthquake which ravaged the Caraccas, March 26th, 1812, was followed by the great eruption of the Souffrier in St. Vincents, on April 30th of the same year; when, according to Humboldt, subterranean noises were heard the same day at the Caraccas and on the banks of the Apure. Earthquakes are felt over very considerable spaces ; and of this no better example has yet been recorded than the cele- brated earthquake of Lisbon in 1755, the shock of which was felt over nearly the whole of Europe, and even in the West Indies. The force capable of causing such extensive vibra- tions must have been very considerable ; and, with every al- lowance for the easy transmission of motion and sound late- rally through rocks, must have required considerable depth for its production. Motion seems always to be communicated to water during earthquakes, the vibratory movement being very frequently felt by vessels at sea, and waves of greater or less magnitude, according to the force of the shock, being com- monly driven on shore. The wave produced during the great Lisbon earthquake rose sixty feet high at Cadiz, and eighteen feet at Madeira, causing various movements of the water on the coasts of Great Britain and Ireland. Similar waves, though of proportionally less size, are common during volca- nic eruptions ; motion being produced in the surrounding water, which being unable to rend and crack like the land, communicates the impulse it has received to the waters around, and thus a wave is propagated which will diminish in height in proportion as it recedes from the disturbing cause. In al- most all ports irregularities in the motion of the sea are at times observable, which cannot be reconciled with the tides or motions communicated to water by temporary currents or winds in the offing. The movement is generally a quick flow 142 Earthquakes. or reflow of the water, and is often so trifling as to escape the attention of all but seamen or fishermen, who, constantly engaged with their vessels or boats in harbours, are surprised to find them suddenly floated or left dry, and this sometimes several times repeated. May not these movements be caused by earthquakes beneath the depths of the sea, or be so trifling as to escape observation on land ? If, as it seems reason- able to conclude, earthquakes are propagated laterally through considerable distances, in the same manner as sound is con- veyed through the air, the intensity of the shock will depend on the medium through which it is conveyed ; and if this view should be correct, earthquakes will not be equally felt on every description of rock. I once observed a fact, which, though it struck me much at the time, cannot in itself form the basis of any reasonable hypothesis, but as it may be the means of exciting inquiry it is as well to mention it. While sitting in a house in Jamaica, situated on a hill, near the verge of the white limestone of that island, where the large gravelly, sandy, and clay plain of Vere and Lower Clarendon meets it, I experienced a slight shock of an earthquake. Having occa- sion about half an hour afterwards to descend to some houses at the foot of the hill, and on the gravel plain, I inquired if the inhabitants had felt the earthquake ; they, however, ri- diculed the idea, stating that if any such had occurred they must have known it, as they also had been sitting quietly, and were too much accustomed to shocks not to have observed the earthquake if it had really occurred. I then considered that I had deceived myself, and thought no more of the sub- ject until the evening; when some negroes, who had been employed on their own account, a few miles distant in some mountains composed of the white limestone, reported that they had felt the shock of an earthquake ; and it subsequently ap- peared, that a much more considerable shock had been lelt in the vicinity of Kingston, about forty miles distant. The im- portance of this fact certainly rests on the little apparent sen- sation produced at the lower house ; and therefore, as the shock may have escaped the attention of those then present, this circumstance is in itself of no great value, and is merely stated to promote inquiry. It may, however, be remarked, that gravel would transmit a vibration less readily than com- pact limestone, though it might more easily give way before a vertical movement. Humboldt has remarked that during the Caraccas earthquake of 1812, the Cordilleras were more shaken than the plains. This may have arisen from the more easy transmission of the vibration through the gneiss and mica slate, than through the rocks in the plains; or, as Earthquakes. 14-3 might be also the case in Jamaica, the inferior rocks might be more shaken through their continuity than the superior rocks, being nearer the disturbing cause. It may also be remarked that rocks would transmit sounds unequally from variations in their texture and continuity, and that subterranean noises might be audible, while the shock which produced them could not be distinctly felt. Various sounds are recorded as accompanying earthquakes, but the most general seems a low rumbling noise like that of a waggon passing rapidly along. The first shock I ever expe- rienced was, during a beautiful night, on the north side of Jamaica, when it appeared as if a waggon, rolling rapidly to the house, gave it a smart rap and then passed on. It has been considered, and with much probability, that the very great distances at which volcanic explosions from sur- face-vents have been heard, arises from, the transmission of the sound through the rocks. The great explosion at Sum- bawa above noticed is described as having been heard in Su- matra, a distance of 970 geographical miles, and at Ternate, 720 miles in another direction *. It is also stated that the eruption from the Aringuay, in the island of Lu9on, Philip- pines, in 164-1, was heard in Cochin-Chinaf. Earthquakes produce changes in the level of the land, rais- ing and depressing ground, and causing clefts, slips or faults, and various other modifications of surface. The raising of the surface implies either an expansion of the solid matter be- neath, or a separation of parts, which should form a cavity, filled either by gaseous or liquid substances. We are not aware of anything that could produce the expansion required but heat, so that if the temperature were again diminished, contraction would ensue. If a separation of parts were ef- fected, and the upper portion raised, the gaseous or liquid support could scarcely be considered permanent, unless the injected matter became solid, as might happen with liquid lava, and the hollow produced by such injection be far re- moved from the surface. The best example of the bodily elevation of land with con- siderable surface appears to be that recorded by Mrs. Maria Graham, as having taken place during the Chili earthquake of 1822. The shock extended along the coast for more than a thousand miles, and the land was raised for a length of one hundred miles, with an unknown breadth, but certainly ex- tending to the mountains. The beach was raised about three or four feet, as was also the bottom near the shore ; on the former, shell-fish were still adhering to the rocks on which * Life of Sir S. Raffles. f Chamisso, Kotzebue's Voyage. 144 Earthquakes. they grew. It was also observed that there were other lines of beach, with shells intermixed, above that newly elevated, attaining in parallel lines a height of about fifty feet above the sea ; seeming to show that other elevations of the same land had been effected by previous earthquakes. During this earth- quake the sea flowed and ebbed several times. No visible change in the atmosphere was produced previous to the shocks, but it is supposed that some effect, perhaps electrical, may have been caused by the earthquake, for the country was sub- sequently deluged by storms of rain*. Mr. Lyell has accumulated a considerable mass of evidence to show that such elevations have been the consequence of earthquakes in other places, and that considerable depressions have also occurred f. Thus, during the Cutch earthquake of 1819, the eastern channel of the Indus was altered, the bed of which was in one place deepened about seventeen feet, so that a spot once fordable became impassable. It further ap- pears, from the observations of Lieutenant Barnes, that not only was there considerable subsidence in this case, but also a remarkable elevation of upwards of fifty miles in length, running parallel to the subsidence, across the delta of the Indus from east to west. The greatest observed height of this elevated land, in some places sixteen miles broad, was ten feetj. Various surface-changes were effected during the great earthquake in Calabria in 1783. Of these a summary has been given from various authorities by Mr. Lyell, whose ac- count will be perused with interest, however little we may feel inclined to adopt the theoretical conclusions that have been deduced from it. The earth had a waving motion ; nume- rous and deep rents were formed ; faults were produced, even through buildings ; large land-slips took place ; lakes were formed, one about two miles long by one broad, from the obstruction of two streams ; the usual agitation of the neigh- bouring sea was produced, and heavy waves broke upon the land, sweeping all before them. The great earthquake in Jamaica of 1692, generally described as having swallowed up Port Royal, has been adduced as an example of great derangement. It is a common tradition in that island, that many of the accounts which have appeared respecting this earthquake have been much exaggerated ; nor need this surprise us, when we reflect how difficult it is to obtain a clear account of unusual natural phenomena from those who have been dreadfully alarmed by them. In order Journ. of Science ; Geol. Trans, vol. i. f Principles of Geology. J Ly ell's Principles of Geology, vol. ii. p. 269, Earthquakes. 145 to estimate the changes that may be supposed to have taken place during this earthquake, it becomes necessary to notice the condition of Port Royal and of the neighbouring coast previous to its occurrence. The site of the present town of Port Royal is the same as that of the old town, being at the western extremity of a sand-bank, known as the Palisades, about eight miles long, thrown up apparently by the sea. Immediately seaward are numerous shoals and coral reefs known by the name of the Keys. Now it appears from the evidence of Captain Hals, who went to Jamaica with Penn and Venables in 1655, that the land upon which Port Royal then stood, was joined to the Palisades, distant about a quarter of a mile* by a narrow ridge of sand just appearing above the water ; and it further ap- pears, that when Jackson invaded St. Jago de la Vega, about seventeen years before this time* the same land formed an island ; the narrow ridge resulting from the drift of sand by the prevalent E. or S.E. winds, and the action of the breakers. By a continuance of the same forces the whole space between the Palisades and Port Royal was eventually filled up, aided by the contrivances of the inhabitants, who drove in piles and formed wharfs, close to which the water was so deep that vessels of 700 tons came alongside and unloaded *. Upon this newly formed land the greater part of the town was built* and consisted of heavy brick houses. Now the part of Port Royal, described as having been swallowed up or sunk, was situated upon this new formed land. " The ground gave way as far as the houses stood, and no further, part of the fort and the Palisades at the other end of the houses stand- ingf." Sir Hans Sloane says : " the whole neck of land being sandy (excepting the fort, which was built on a rock and stood) on which the town was built, and the sand kept up by pali- sades and wharfs, under which was deep water, when the sand tumbled, on the shaking of the earth, into the sea, it covered the anchors of ships riding by the wharfs, and the foundations yielding, the greatest part of the town fell, great numbers of the people were lost, and a good part of the neck of land, where the town stood, was three fathoms covered with water." We have next to consider the state of the sea * The variation in the depth of this water would be trifling, for the tides only rise or fall eleven inches or a foot at Port Royal. t Phil. Trans, for 1694. Long, who from his office was so well qualified to obtain the best information* says : " the weight of so many large brick houses was justly imagined to contribute, in a great measure, to their down- fall ; for the land gave way as far as the houses erected on this foundation stood, and no further." L 1 4-6 Earthquakes. during the shocks. The harbour is described as having had " all the appearance of agitation as in a storm ; and the huge waves rolled in with such violence, as to snap the cables of the ships, and drive some from their anchors." Again, we find that the houses near the water sunk at once : " and a heavy rolling sea followed, closing immediately over them." The Swan frigate, which was by the wharf careening, was carried over the tops of the houses by the sea, and some hun- dreds of persons escaped by clinging to her. Some houses sunk or settled perpendicularly, so that they remained from the balcony upwards above water ; but the greater part were rendered a mass of ruins. Finally, the land with the fort on it is reported to have formed an island, as at the time of Jack- son's expedition *. This state of things has not however con- tinued; for the same causes, which once joined the Palisades with the fort, continuing, the whole now forms a continuous piece of land. A plan of Port Royal formed from authentic and existing documents, and representing the extent of that town previous to, and immediately after, the earthquake of 1692, as also about a century afterwards, was published in the Jamaica Al- manac for 1806, having first appeared about ten or twelve years previously. The annexed wood-cut (Fig. 22) is a re- duction of this plan (the lines of streets being omitted), with the addition of the present extent of this point of land, accor- ding to Dr. Miller, of Jamaica, a, #, , , a, and /, the Fig. 22. fT p ort R y a l Harbour. boundary of the town and point of Port Royal previous to the great earthquake. The darkly shaded parts P and C, the portion that remained after the earthquake; C being Fort Charles. The lightly shaded part N, N, N, the extent of the town and point at the end of the eighteenth century; the accumulation of sand having been principally produced by Phil. Trans. 1694; Sloane, Nat. Hist of Jamaica; Long, Hist, of Ja- maica ; and Bryan Edwards, Hist, of the West Indies. Earthquakes. 147 the natural drift of the sand. The spaces /, /, /, and H, the additional land, formed by the drifting of the sand since that time, and constituting with the lightly shaded portions N 9 N, N 9 the present extent of the town and point of Port Royal. The space H, formerly known as Chocolate Hole, now filled up, forms part of the garrison parade. It also appears that the portion of Port Royal which re* mained above water, after the shock, is generally considered to be based upon white limestone, as is the case with Fort Charles. This rock is now known to constitute the base of part, if not a large proportion, of the ridge of land, named the Palisades, which commences close to Port Royal (L, Fig. 22.), and very probably also forms the base of the various coral reefs known as the Port Royal Keys. Upon a review of what has been adduced respecting this earthquake, it does not appear that there is evidence of sub- sidence, that is, the bodily subsidence of a mass of land of great depth ; though I would be far from denying that there may have been something of the kind. The whole may be explained by the settlement of loose sand, charged with the weight of heavy houses, during the violent shocks of an earth- quake, and by the inroad of the sea; for had there been a general subsidence, the rocks would have disappeared with the rest *. The evidence of the ruins of houses commonly stated to be seen beneath the sea, in calm weather, close to the present town, will do for either hypothesis ; for they would be similarly situated either from the settlement of the sand, or by the subsidence of land, in the usual acceptation of this term. While in Jamaica, in 1824, I endeavoured in vain to see what are commonly termed the Ruins of Old Port Royal, and it now appears that they are covered up with sand, forming considerable inequalities beneath the sea, to the W. and N. W. of the present naval hospital. When vessels touch on these inequalities they are said to "ring the bells of Old Port Royal." It however by no means follows that the ruins were not more distinctly visible in 1780, as is stated to have been the case by Sir Charles Hamilton f, and others : on the con- trary we must expect them to have been so, for we have seen that the transport of sand, principally produced by the break- ers, driven forward by a prevalent wind (the trade wind), is * Dr. Miller informs me, that when he resided in Port Royal, prior to the great fire of 1815, there were many old people then living, descend- ants of the early settlers, and it was the traditional opinion among them, that the great damage was produced by the slipping of the sand, an opinion in accordance with that previously noticed. f Lyell, Principles of Geology, vol. ii. p. 269. L 2 148 Earthquakes. very considerable at Port Royal, so that the ruins would be gradually covered up. The earthquake was generally destructive of buildings in Jamaica, and masses of rocks were detached from the heights; no great difficulty in a country abounding with precipices and steep mountains. According to one account, two moun- tains met in the Sixteen Mile Walk ; if they did so, they have since been so complaisant as to separate, for there is no- thing at present existing there to warrant a conclusion that they ever did meet. That heavy fragments of rock, and con- siderable masses of earth, blocked up the passage for the time, is exceedingly probable ; but there is a great difference be- tween such an event and the meeting of mountains. Funnel-shaped, or inverted conical cavities are by no means unfrequent on plains after earthquakes; and are so much alike wherever they occur, that they must have some common cause for their production. Circular apertures were produced in the plains of Calabria by the earthquake of 1783 : they are described as commonly of the size of carriage-wheels, but often larger and smaller ; they were sometimes filled by water, but more frequently by sand. Water seems to have spouted through them*. During the earthquake in Mercia in 1829, nume- rous small circular apertures were produced in a plain near the sea, which threw out black mud, salt water, and marine shells f. After the earthquake at the Cape of Good Hope, in December 1809, the sandy surface of Blauweberg's Valley is described as studded with circular cavities, varying from six inches to three feet in diameter, and from four inches to a foot and a half in depth. Jets of coloured water are stated, by the inhabitants of the valley, to have been thrown out of these holes to the height of six feet during the earthquake J. It seems somewhat difficult to account for these appearances, though the common aqueous discharges through rents or chasms can be more readily understood. During the Chili earthquake, previously noticed, sands were forced up in cones, many of which were truncated with hollows in their centres . The courses of springs are, as would be anticipated, often deranged amid such motions of the ground ; and flashes of light, or bright meteors, are so frequently mentioned that we can scarcely doubt their occurrence, and they may, perhaps, be considered as electrical. If we now withdraw ourselves from the turmoil of volcanos * Lyell, Principles of Geology ; where a view and section of these cu- rious cavities are given : pp. 428, 429. f Ibid. ; and Ferussac's Bulletin, 1829. J Phil. Mag. and Annals, January 1830. Journal of Science. Hurricanes. 14*9 and earthquakes, and cease to measure them by the effects which they have produced upon our imaginations, ^we shall find that the real changes they cause on the earth's surface are comparatively small, and quite irreconcileable with those theories which propose to account for the elevations of vast mountain-ranges, and for enormous and sudden dislocations of strata, by repeated earthquakes acting invariably in the same line, thus raising the mountains by successive starts of five or ten feet at a time, or by catastrophes of no greater im- portance than a modern earthquake. It is useless to appeal to time: time can effect no more than its powers are capable of performing: if a mouse be harnessed to a large piece of ordnance, it will never move it, even if centuries on centuries could be allowed ; but attach the necessary force, and the re- sistance is overcome in a minute. Hurricanes. These are of geological importance, as by the sudden appli- cation, if I may so express myself, of a furious wind and de- luges of rain to the surface of land, very considerable changes are in a short time produced on that surface. It has been considered that the wind, during hurricanes, travels with a velocity of from eighty to one hundred miles per hour; but it must be confessed that we possess no very satisfactory infor- mation on this head. Be, however, the velocity of the wind what it may, its force is sufficient to level forests, throw down buildings, and destroy a large amount of animal life ; in a few- hours converting a beautiful and luxuriant country, studded with villages and towns, into a scene of desolation and mourn- ing. Furious torrents are suddenly formed, which not only sweep away a large proportion of the uprooted trees, and the bodies of numerous terrestrial animals destroyed by the effects of the wind ; but also act most powerfully on all the drainage depressions, producing the maximum effects of running water in such situations. In mountainous regions the land-slips are then also frequently considerable ; and if these fall into the bed of a torrent, they add to its destructive effects, by damming up the waters for a time, which, when they have forced their passage through the obstacle, rush onwards with increased velocity and power. In the hurricane in the West Indies of August 1831, we have a melancholy example of the destruction of animal and vegetable life caused by these scourges of that portion of the world. Not only were buildings of various kinds levelled with the earth, and numbers of persons buried beneath their ruins, but a large amount of animal lite was also destroyed ; and 1 50 Hurricanes. those trees which were not uprooted by the fury of the wind, were deprived of their foliage, many even of their branches, so that the unfortunate island of Barbadoes presented that strange phenomenon, a mass of leafless trees on a tropical island. This hurricane also ravaged the islands of St. Vin- cent and St. Lucie, and was even felt at the eastern end of Ja- maica. The sea is, as might be expected, violently agitated during hurricanes, and causes great destruction, particularly on low coasts. Thus, in the great Jamaica hurricane of 1780, the sea suddenly burst in upon the small town of Savanna la Mar, and swept it, and every thing in it, entirely away. The hur- ricane of August 1831, was sufficiently powerful at Hayti to raise the sea at Aux Cayes to a considerable height, and the swell consequent on it was so great on the coast of Cuba as to throw every vessel on shore at St. lago de Cuba. Hurricanes are often more partial, but thev/ are not the less destructive to the land they traverse on that account. The hurricane of 1815, which traversed Jamaica from North to South, was one of this description ; it took its way across the western portion of the Blue Mountains, and was exceedingly destructive. Not only was the wind furious, but the quantity of rain which fell in a given time was considered quite unexampled even in the tropics. The flood which descended the Yallahs river, swept away all the fish in it, and ten years afterwards it was considered that there were no fresh- water fish in that river. The land-slips in the Port Royal, St. Andrews, and Blue Mountains were very considerable ; and when I visited these mountains several years afterwards, many a bare cliff bore evi- dence to the changes that had been thus produced. When these land- slips descended to the bottom of the ravines, they dammed up the waters for the time, and then giving way, were partially swept onwards. The loss of life was considerable, and many buildings were either washed away or buried be- neath detritus. The land communications between Kingston and the Eastern coast were stopped ; and Mr. Barclay relates, that being thus compelled to pass by sea to Morant, the vessel was obliged to make "a considerable offing to keep clear of the enormous quantity of trees, which literally covered the water to a considerable distance." Though so destructive about the centre of its course, this hurricane was neither felt at St. Jago de la Vega (Spanish Town), forty miles to the west- ward, nor at Morant Keys, fifty miles to the eastward. The force of a hurricane is well shown by the facts observed at Guadaloupe, on the 25th of July 1825. Strongly-built houses were blown down, and many tiles driven through the doors of warehouses. A deal plank, 39 inches long, 9-8 Gaseous Exhalations. 151 inches wide, and *9 inch thick, was transported with such ra- pidity that it traversed the trunk of a palm-tree 17'7 inches in diameter. A piece of wood, 7'8 inches square, lind from four to five yards long, was forced by the wind into a hard, frequented, and stoned road, to the depth of about a yard. The exactitude of these and other facts illustrative of the great power of this hurricane, was verified on the spot by General Baudrand, of the French Engineers *. It will be obvious that during hurricanes a comparatively large amount of terrestrial animals and vegetables may, in ad- dition to the land-detritus, be carried outwards into the seas which bathe the shores of tropical islands, such as those of the West Indies, more particularly when such islands are moun- tainous, as is the case with Cuba, Hayti, Jamaica, and others. Not only men, quadrupeds, birds, and land reptiles, but also fresh-water tortoises and crocodiles, may be surprised and carried out to sea, where they would have a poor chance of escape amid the turmoil of the waves at such times. A large proportion of the creatures thus borne by torrents outwards would, most probably, be devoured by sharks and other vora- cious inhabitants of the sea; but there is still a possibility that the river-detritus, and the sands and mud, stirred up by the action of the waves in shallow seas, would, when tranquillity was restored, envelope various terrestrial, fluviatile, and marine remains : such a deposit would thus, to a certain extent, re- semble one formed in an estuary, but would so far differ from it as probably, in the case here supposed, the remains would exhibit the marks of violent transport. In the immediate vi- cinity of the coast, the breakers would throw a considerable quantity of these remains on shore. Gaseous Exhalations. In several situations removed from any volcanic action, so far as is visible on the surface, natural jets of inflammable gases are seen to issue, affording decisive evidence of chemical changes that are taking place at various depths beneath. Of these, some have served the purpose of the priest to delude mankind, while part of the others have been more usefully employed. Carburetted hydrogen gas is well known to be the " fire- damp" of the coal districts, and to issue from the coal strata ; collecting in the ill-ventilated galleries of collieries, and, when sufficiently mixed with atmospheric air, exploding with great violence if approached incautiously with an unprotected flame, * Pouillet, Siemens de Phys. Experimentale, t. ii. p. 718, 2nde E'dit. 152 Gaseous Exhalations. spreading mourning and misery among the families of the miners. If the genius of Davy had merely produced his safety- lamp, it would alone have entitled him to the applause and thanks of mankind. As carburetted hydrogen is so freely liberated in coal-mines, it would be expected that it should occasionally be detected on the surface, and accordingly it has been so discovered *. Inflammable gas also occurs in other situations, where there is no reason to suspect the presence of coal strata. Of this, the well-known jets of gas in the limestone and serpentine district of the Pietra Mala, between Bologna and Florence, afford an example. Captain Beaufort describes an ignited jet of inflammable gas, named the Yanar, near Deliktash, on the coast of Karamania, which perhaps once figured in some religious rites. He states that, "in the inner corner of a ruined building, the wall is un- dermined, so as to leave an aperture of about three feet in di^ ameter, and shaped like the mouth of an oven : from thence the flame issues, giving out an intense heat, yet producing no smoke on the wall." Though the wall was scarcely disco- loured, small lumps of caked soot were found in the neck of the opening. The hill is composed of crumbly serpentine and loose blocks of limestone. A short distance down the hill there is another aperture, which from its appearance seems once to have given out a similar discharge of gas. The Yanar is supposed to be very ancient, and is possibly the jet described by Pliny f. Colonel Rooke informed Captain Beaufort, that high up on the western mountain at Samos there was an intermittent flame of the same kind ; arid Major Rennell stated that a natural jet of inflammable gas, inclosed in a temple at Chittagong, in Ben- gal, is made use of by the priests, who also cooked with it. The village of Fredonia, in the State of New York, is lighted by a natural discharge of gas, which is collected by means of a pipe into a gasometer. The quantity obtained is about eighty cubic feet in twelve hours. It is carburetted hydrogen, and is supposed to be derived from beds of bitumi- nous coal. The same gas is discharged in much larger quan- tities in the bed of a stream about a mile from the village. According to M. Imbert, gaseous exhalations are employed at Thsee-Lieou-Tsing, in China, to distil saline water obtained from wells in the neighbourhood. " Bamboo pipes carry the * It appears very remarkable that in the coal districts of the British Isles, where such a large amount of carburetted hydrogen is annually produced, means have not been adopted for making an economical use of this gas, both as respects light and heat. f Beaufort's Karamania. Gaseous Exhalations. 153 gas from the spring to the place where it is to be consumed. These tubes are terminated by a tube of pipe-clay, to prevent their being burnt. A single well (of gas) heats more than three hundred kettles. The fire thus produced is exceedingly brisk, and the caldrons are rendered useless in a few months. Other bamboos conduct the gas intended for lighting the streets and great rooms or kitchens *." These wells of inflam- mable gas were, according to M. Imbert, formed for the pur- pose of obtaining salt water, which they in fact first gave out. The water failing, the wells were sunk to a considerable depth in order to find the water ; instead, however, of finding salt water, inflammable gas suddenly rushed forth with consider- able noise f. M. Klaproth notices other jets of inflammable gas in China ; one, now extinguished, is stated to have burnt from the second to the thirteenth century of our era. This Ho tsing, or fiery well, was situated 80 li to the S.W. of Khioung tcheou, and like those above mentioned produced salt water J. This connexion of inflammable gas with saline springs or salt is not confined to China, but has also been observed in America and in Europe. While boring for salt at Rocky Hill, in Ohio, and near Lake Erie, the borer suddenly fell, after they had pierced to a depth of 197 feet. Salt water im- mediately spouted out, and continued to flow for several hours; after which a considerable quantity of inflammable gas burst forth through the same aperture, and, being ignited by a fire in the vicinity, consumed all within its reach . It also appears that M. Rceders, inspector of the salt-mines of Gottesgabe, at Reine in the county of Tecklenberg, has for two or three years used an inflammable gas which issues from these mines, not only as a light, but for all the purposes of cookery. He obtains it from the pits that have been aban- doned, and conveys it by pipes to his house. From one pit alone a continuous stream of this gas has issued for sixty years. * Bibl. Universelle; and Edin. New Phil. Journal, 1830. t Humboldt, Fragmens Asiatiques. J Ibid. In the same work will be found an interesting account of the mode in which the Chinese sink these wells, in search of salt water, to con- siderable depths. This is effected by the constant striking of a piece of steel, of about 300 or 400 pounds weight, against the rock, upon the same principle that a hole is made in the solid rock by an iron or steel bar for the purpose of blasting with gunpowder. In the Chinese work, however, the steel weight is suspended by a cord to one end of a piece of wood, placed over a support in such a manner that a workman by dancing or jumping on the other end, raises the weight about two feet at each motion, and suddenly lets it fall again. By these slow, but somewhat sure means, a round per- pendicular hole is formed, about five or six inches in diameter, very smooth, and, according to M. Imbert, from 1500 to 1800 French feet in depth, Trans. New York Phil. Soc. 154 Gaseous Exhalations. It is supposed to consist of carburetted hydrogen and olefiant gas*. Inflammable gases are also found to proceed from ground charged with petroleum and naphtha. The inhabitants of Baku, a port on the Caspian Sea, are supplied with no other fuel than that derived from the petroleum and naphtha with which the earth in the neighbourhood is strongly impregnated. About ten miles to the N.E. of this town there are many old tem- ples of Guebres, in each of which there is a jet of inflammable gas, rising from apertures in the earth. The flame is pale and clear, and smells strongly of sulphur. Another and a larger jet issues from the side of a hill. The ground is generally flat, and slopes to the sea. If in the circumference of two miles, holes be made in the earth, gas immediately issues, and in- flames when a torch is applied. The inhabitants place hol- low canes into the ground, to convey the gas upwards, when it is employed for the purposes of cookery as well as for a lightf. M. Lenz, describing an eruption of mud and flame near the village of lokmali, fourteen wersts to the west of Baku, would seem to attribute the gaseous exhalations of this district to a volcanic origin, but the facts adduced will scarcely admit of this interpretation. He notices this eruption as having taken place on November 27th, 1827. A column of flame burst out, where no flame had been previously seen, and rose for three hours to a considerable height, then lowered itself to the height of three feet, and burnt for twenty-four hours. After this the mud rushed forth and covered the country over an area of 200 toises by 150, to the depth of two or three feet. There is sufficient evidence that other eruptions of mud or clay had previously taken place from the same, or nearly the same, place. This and other " salses" noticed in the same territory cannot be termed volcanic, in the usual ac- ceptation of the word. Moreover we learn from the observa- tions of the same author, that at the Atech-gah, or the great fires of Baku, the principal jet rises through a calcareous rock, with a dip of 25 to the S.E., the fissures or cracks being ren- dered blue by it J. Carbonic acid gas is evolved abundantly in coal-pits and volcanic regions. Its occurrence in the Grotto del Cane, of which such overcharged descriptions have been given, is well known. MM. Bischof and Noggerath notice a pit, on the side of the lake of Laach, in which they found dead birds, squirrels, bats, frogs, toads and insects, killed by the evolu- tion of carbonic acid gas. M. Bischof estimates that the ex- * Journal of Science. f Edin. Phil. Journal, vol. vi. J Humboldt, Fragmens Asiatiques. Gaseous Exhalations. 155 halations of carbonic acid gas, in the vicinity of the lake of Laach, amount to 600,000 pounds daily, or 219,000,000 pounds in a year. In the Brohlthal on the Rhine, an old volcanic country, there is a considerable evolution of carbonic acid gas, which is employed by M. Bischof in the manufacture of chemical pre- parations on the large scale. Six hundred pounds of this gas are calculated to be discharged from only one of the jets in twenty-four hours, being at the rate of 219,000 pounds in the year *. Carbonic acid gas is so abundantly evolved in a district of mineral springs in Armenia, near Fort Diadine, on the Eu- phrates, that it produces a noise while issuing through the cracks in the limestones of the country, killing the birds which come within its influence f. The Guevo Upas, or Valley of Poison, in Java, would ap- pear to be a cavity filled to a certain height with carbonic acid gas. Mr. Loudon describes it as about half a mile in circumference, and of an oval form. The depth is from thirty to thirty-five feet, the bottom being flat, without vegetation, and covered with skeletons of men and of various animals, such as tigers, hogs, deer, &c., which have perished by their entrance into the gas. The destructive gas did not rise, so as to be dangerous, above eighteen feet from the bottom J. A very copious discharge of carbonic acid gas occurs on the Kyll, nearly opposite Birresborn. The gas rises through fissures of the rock, and traverses a pool of rain-water, resting on it, with such violence that the noise is stated to be heard at the distance of 400 yards. Birds are killed when they approach too close, and persons wishing to drink are driven away by the gas, a stratum of which covers the surrounding turf. In many situations gaseous vapours come to the surface mixed with water or petroleum, with sufficient force to produce "salses" or mud volcanos. Dr. Daubeny considers those of Maculaba in Sicily as independent of volcanic action, but due to the combustion of the sulphur existing among the rocks. Mud eruptions from the discharge of gaseous vapours and water are known in many other places ||. * German Trans, of Manual. f Voskoboinikov, Goinoi Journal, 1829; and Boue's Mem. Geol. et Paleontologiques, t. i. 1832. J Journal of Geographical Society, vol. ii. Bischof and Noggerath, Edin. Phil. Journal. || Those near Modena have long been celebrated. 156 Deposits from Springs. Deposits from Springs. Springs are seldom or ever quite pure, owing to the solvent property of water, which percolating through the earth, al- ways becomes more or less charged with foreign matter. Car- bonate, sulphate, and muriate of lime, muriate of soda, and iron, are frequently present in spring waters. Some are more highly charged with these and other substances, such as car- bonate of magnesia and even silica, than others, and have hence obtained the name of mineral springs. Many are ther- mal, as before noticed, and seem not immediately derived from the waters of the atmosphere ; as may also be the case with many that are cold, their more elevated temperature having been lost in their passage upwards through colder strata. Many thermal springs contain silica, though this substance is of exceedingly difficult solution. The siliceous deposits from the Geysers in Iceland are well known. Sir George Mackenzie describes the leaves of birch and willow converted into stone, every fibre being discernible. Grasses, rushes, and peat are in every state of petrifaction. There are also deposits of clay containing iron pyrites, which decompose and communicate very rich tints to it. The deposits from the Geysers extend to about half a mile in various directions, and their thickness must be more than twelve feet, for that depth is seen in a cleft near the Great Geyser. The finest exhibition of such deposits as yet noticed, occurs in the volcanic district of St. Michael, Azores. Dr. Webster describes the hot springs of Furnas as respectively varying in temperature from 73 to 207 Fahr., and depositing large quantities of clay and siliceous matter, which envelope the grass, leaves, and other vegetable substances that fall within their reach. These they render more or less fossil. The ve- getables may be observed in all stages of petrifaction. He found " branches of the ferns which now flourish in the island completely petrified, preserving the same appearance as when vegetating, excepting the colour, which is now ash-grey. Fragments of wood occur, more or less changed ; and one en- tire bed, from three to five feet in depth, is composed of the reeds so common in the island, completely mineralized, the centre of each joint being filled with delicate crystals of sulphur*." The siliceous deposits are both abundant and various : the most abundant occur in layers from a quarter to half an inch in thickness, accumulated to the depth of a foot and upwards. * Edin, Phil. Journal, vol. ri. Depositsfrom Springs. 157 The strata are nearly always parallel and horizontal, though sometimes slightly undulating. The silex forms stalactites, often two inches in length, in the cavities of the siliceous de- posits, and these are frequently covered with small brilliant quartz crystals. Compact masses of siliceous deposits, broken by various causes, have been re-cemented by silica, and the compound is represented as very beautiful. Some of the ele- vations of this breccia Dr. Webster considers upwards of thirty feet in height. The general deposit appears to be con- siderable, and to form low hills. The colours of the clay and siliceous substances are very various, and even brilliant, white, red, brown, yellow, and purple being the principal tints. Where the acid vapours reach the rocks, they deprive them of their colours. Sulphur is abundant, and the springs occur in a district of lava and trachyte*. According to James f, the thermal springs of the Washita deposit a very copious sediment, composed of silex, lime, and iron. This shows that hot springs, when propelled through a non-volcanic district, may yet contain silica. The same may be said of some of the springs in India. Dr. Turner found that the thermal springs of Pinnarkoon and Loorgootha, in that country, which produced 24- grains of solid matter in a gallon, contained 2 1 '5 per cent of silica, 19 of chloride of so- dium, 19 of sulphate of soda, 19 of carbonate of soda, 5 of pure soda, and 15 -5 of water J. The following is an analysis of the Geyser waters and hot springs of Reikum, Iceland, by Dr. Black. A gallon of each produced: Geyser. Reikum. Soda 5-56 3'00 Alumina 2-80 0-29 Silica 31-50 21-83 Muriate of soda 14-42 16-96 Sulphate of soda 8-57 7-53 These analyses do not show the presence of lime, but Sir G, Mackenzie mentions a calcareous deposit from boiling springs (temp. 212) in the valley of Reikholt, in Iceland, charged with carbonic acid gas. Many thermal and other springs con- tain this gas, which seems very abundant in volcanic regions. To its power of dissolving lime, when passing through calca- reous rocks, those deposits are due, that are so common in some countries, particularly when volcanic, which are known under the general name of Travertin or calcareous tufa. Pro- bably, also, many hot springs may contain carbonic acid gas, which, not meeting with calcareous or magnesian strata, is thrown off when in contact with the atmosphere. * Edin. Phil. Journal, vol. vi. f Expedition to the Rocky Mountains. I Elements of Chemistry. 158 Deposits from Springs. Travertins are of greater geological importance than the si- liceous deposits from modern springs, at least so far as their extent of surface and depth are concerned ; though both these have been greatly exaggerated, from the usual mode of compa- ring such deposits, not with the superficies of the land gene- rally, but with their magnitude relatively to the valleys or plains in which they may occur, and not unfrequently with that of man himself. The deposit from the fountain of St. Allire, near Clermont, formed a bridge which was, in 1754-, one hundred paces long, eight or nine feet thick at its base, and twenty or twenty-four inches in its upper part *. Mr. Lyell notices the calcareous deposits from the baths of San Vignone, and states that one stratum, composed of seve- ral layers, is fifteen feet thick, and that large masses are cut out of it for architectural purposes +. According to Dr. Gosse, the thermal waters which deposit this travertin are sufficiently hot to boil eggs. The thermal waters at the baths of San Filippo, not far from the above, have a temperature of 122 Fahr., one spring being about a degree or two higher. They contain silica, sulphate of lime, carbonate of lime, sulphate of magnesia, and sulphur; and, notwithstanding their elevated temperature, Conferva; flourish in them. The ground around is formed of travertin deposited by the springs. There are many fissures ; one 30 feet deep, and 150 to 200 feet long. In it the water is whitish, and in a state of ebullition, whence its name, II Bollore. It emits copious discharges of steam and sulphurous vapour. There are other fissures in which sulphur is sublimed in the same manner as at the Solfatara near Naples, and the produce was sufficient to constitute a branch of industry, now however abandoned. The surfaces of these fissures are penetrated by sulphuric acid. Dr. Gosse observed the siliceous stalagmites mentioned by Professor Santi, and describes them as covering the surface of the travertin to the depth of one eighth of an inch J. Mr. Lyell notices the spheroidal structure of the tra- vertin deposited, and compares it with the magnesian limestone of Sunderland. What the amount of magnesia may be in the San Filippo travertin is not stated, but according to Dr. Gosse it is combined with sulphuric acid. Sulphate of lime exists in great abundance in these springs; so much so, that before the water is conducted to the places where the well- known medallions are formed, it is allowed to stagnate for the purpose of depositing the sulphate of lime. That the sulphates * Daubuisson, t. i. p. 142. } Principles of Geology, p. 202. + Gosse, Edin. Phil. Journal, vol. ii. Deposits from Springs. 159 should be common, would be expected where so much sulphu- rous vapour is evolved ; and it is even stated that sulphur exists in the travertin, though it is principally composed of carbo- nate of lime. Deposits of travertin are by no means uncommon from cold springs in the Apennines, particularly near the volcanic region of southern Italy. The celebrated Falls of Terni are, as is well known, artificial, and have been formed by cutting through a previous calcareous deposit, to form a channel for the Velino, which now rushes over a precipice into the Nera beneath. Upon the flat land above, a considerable deposit of lime has taken place ; when, it does not so clearly appear, but proba- bly since the establishment of the present order of things. Notwithstanding the velocity of the water, its cutting powers are trifling, and the upper channel preserves all the appear- ance of art. The Velino contains much carbonate of lime, which it deposits after the great leap, even in the bed of the Nera, which does not cut it off, but is obstructed to a certain degree by it, as may be seen at a place called the Bridge, over which I crossed the Nera, by taking one or two leaps at the chasms cut by the latter torrent. At this place there must be a constant struggle between the destructive power of the Nera, and the lapidifying power of the Velino. The country around exhibits abundant examples of calcareous deposits from springs charged with carbonate of lime. The usual explanation of this phenomenon seems very probable. It supposes the carbonic acid to be derived from the volcanic regions beneath, (and they appear not far distant on the surface,) which, passing with the water through the calcareous strata, dissolves as much lime as it can take up, giving off the excess of carbonic acid under di- minished pressure in the atmosphere, and causing the carbonate of lime to be deposited. The carbonic acid found so abundantly in acidulous springs is ascribed by Von Buch, Brongniart, Boue, Von Hoff, and other geologists, to volcanic or igneous action at various depths beneath the surface. M. Hoffman has further shown that, in certain valleys of elevation, mineral springs are frequent, and cites the valley of Pyrmont as a good example, where the waters are charged with carbonic acid gas*. In the marshy meadows of the valley of Istrup (one of elevation), mounds of mud, from fifteen to twenty feet high, and 100 feet in circumference, are produced by currents of carbonic acid gas, and on their surface many small reservoirs of water are * The following are the contents of these waters, according to Bergman, in a wine pint : Carbonic acid, 26 cubic inches; carbonate of magnesia, 10 grains; carbonate of lime, 4*5; sulphate of magnesia, 5-5 ; sulphate of lime, 8-5; chloride of sodium, 1*5; and oxide of iron, 0-6. Henry's Elements, and Turner's Elements. 1 60 Deposits from Springs. kept in a state of ebullition by bubbles of gas of the size of the fist *. After producing other examples of this evolution of car- bonic acid gas, either combined with water, or nearly if not altogether free, M. Hoffman observes, that " the country si- tuated on the left bank of the Weser in the direction from Carlshafen to Vlotho, up to the foot of the Teutoburg-Wald, may be compared to a sieve, whose apertures, as yet unclosed, permit the escape of gas, disengaged from volcanic depths by means unknown f. The travertin of Tivoli, and the famous Lago di Zolfo, near Rome, have been much appealed to by those who ascribe ail geological appearances to such causes only as are now in ope- ration ; but the former is a mere incrustation, considerable it is true in some situations, if measured by our own magnitude, but insignificant if compared with the country in which it oc- curs; and the latter is but a pond of water, dignified somewhat strangely by the name of a lake, and containing, according to Sir H. Davy, a saturated solution of carbonic acid, with a very small quantity of sulphuretted hydrogen. The spring is ther- mal, being about 80 Fahr. ; plants thrive in and about it, and they are encased in stone beneath, while they vegetate above, and thus they may become fossil, their mosl delicate structure preserved, and their ramifications uncompressed. All the examples hitherto produced of deposits, that can fairly be traced to existing springs, are relatively unimportant ; and though they may lead us to understand how great geolo- gical deposits may, chemically, have taken place, as the cabi- net experiments of the chemist teach us the laws which govern nature on a large scale, they no more could have produced the great limestone or siliceous deposits observed on the earth's surface, than the experiments above alluded to could produce the great chemical phaenomena they illustrate, however long continued. Mr. Lyell has presented us with an account of calcareous deposits in Scotland, which are remarkable, not for their ex- tent, but for the circumstances which attend them. It appears that the Bakie Loch, Forfar shire, has produced a marl used in the agriculture of the country. The following is a section of the beds : 1. Peat, containing trees, one to two feet; 2. Shell- marl, containing in parts tufaceous limestone, provincially termed " rock-marl" one to sixteen feet; 3. Quick-sand, with- out pebbles, cemented together in some places by carbonate of lime, two feet; 4. Shell-marl of good quality for agriculture, (almost every trace of shell is often obliterated,) one to two * Hoffman, Journal de Geologic, t. i, ; and Poggendorf s Annalen, 1829. t Ibid. Deposits from Springs. 161 feet; 5. Fine sand, without pebbles, resting on transported detritus, at least nine feet. The rock-marl is limited to the vicinity of the springs, irregularly distributed over the lake. The Bakie shell-marl is white, with a yellow tint. The rock- marl has the same yellow tint, and consists almost wholly of carbonate of lime, compact, and even crystalline. Organic remains of the marl. Horns of stags and bulls ; wild boar tusks. Cypris ornata, Lam. Limn&a peregra, Val- vata fontinalis, Cyclas lacustris, Planobris contortus, Ancylus lacustris, all of Lamarck. Mr. Lyell considers this calcareous rock as not immediately due to the springs, but to have been produced through the agency of the testaceous inhabitants of the lake ; for though the springs do contain lime, it is in such small quantities, that they could not directly produce the marl. He considers that the testaceous animals obtained the lime either from the water or from the Charce which they fed upon, and thai, dying, they left their calcareous exuvire to form, by accumulation, the shell-marl, which was converted into cal- careous rock by the action of the water upon it; the water con- taining carbonic acid, and forming a solution of carbonate of lime, which might produce a crystalline limestone. Seeds of Chartz, or Gyrogonites^ are converted into carbonate of lime, in which the nut is sometimes found within ; but commonly that space is empty, and the integument alone preserved. The Char a here found mineralized is the Char a hispida, a plant, which now abounds in the Bakie Loch, and in the other lakes in Forfarshire. It contains such a proportion of carbonate of lime, as strongly to effervesce with acids when dried. Mr. Lyell, noticing the deposits of marl in the Loch of Kin- nordy, states that it is thickest at that end of the lake where the springs are most common. The shells are the same here as at the Bakie Loch, and are, like them, nearly all young, scarcely one in ten being full-sized. A large skeleton of a stag (Cervus Elaphus) was dug out of the marl, and was remarkable as being found in a vertical position, the points of the horns being nearly at the surface of the marl, while the feet were about two yards below it. The marl is covered by peat, and in this peat were discovered other skeletons of stags, and (in 1820) the remains of an ancient canoe, hollowed out of the solid trunk of an oak *. There is something in the formation of these lakes which reminds us strongly of the epoch of the submarine forests and of the lacustrine deposits of East Yorkshire, which will be no- ticed in the sequel ; like them they seem to have succeeded a considerable transport of detritus, and to have been gradually filled up, being surmounted by peat ; previous to the formation * Lyell, Geol. Trans. 2nd series, vol. ii. M 162 Naphtha and Asphalt urn Springs. of which latter production man certainly was an inhabitant of these islands, as his works are entombed in it : the lakes being then, probably, more or less open spaces of water, or else his boat would have been of little service to him. Naphtha and Asphaltum Springs. These are distributed over various parts of the world, and cannot be considered as rare. According to Dr. Holland, the petroleum springs of Zante are much in the same state as in the time of Herodotus. They are situated on a small marshy flat, bounded by the sea on one side, and by limestone and bi- tuminous shale-hills on the others. The principal pool is about 50 feet in circumference, and a few feet deep : the sides and bot- tom of this and the others are thickly covered with petroleum, which by agitation is brought to the surface of the water, and collected. The amount obtained is estimated at 100 barrels annually*. James states that about 100 miles above Pittsburgh, and near the Alleghany river, there is a spring, on the surface of which float such quantities of petroleum, that a person may collect several gallons in a day. He considers that it may probably be connected with coal strata, as is the case with similar springs in Ohio, Kentucky-}-. The pitch lake of Trinidad, estimated at about three miles in circumference, has long been celebrated. According to Dr. Nugent, the asphaltum is sufficiently hard in wet weather to support heavy weights, but during the heats it approaches fluidity. It is intersected with numerous cracks filled with water; and it appears that these cracks sometimes close up again, leaving marks on the surface of the pitch lake. When slightly covered with soil, as it is in some situations, good crops of tropical productions are obtained. From this cover- ing of soil it is difficult to estimate the exact boundaries of the lake {. Captain Alexander states, that at Pointe la Braye masses of pitch advance into the sea, and have the appearance of black rocks among the trees. At the hamlet of la Braye, the coast is covered with pitch, which runs out to sea and forms a bank under water . * Holland's Travels in the Ionian Isles, Albania, &c. f Expedition to the Rocky Mountains. J Nugent, Geol. Trans, vol. i. Alexander, Jameson's Edin. Phil. Journal, Jan. 1833. The same author notices an assemblage of salses or mud volcanosat Pointe du Cac, forty miles southward from the pitch lake. The largest of the salses is about 150 feet in diameter. Coral Reefs and Islands. 163 Large quantities of naphtha are obtained on the shores of the Caspian. The inhabitants of the town of Baku, a port on that sea, are supplied with no other fuel than that obtained from the naphtha and petroleum, with which the neighbouring country is highly impregnated. In the island of Wetoy and on the peninsula of Apcheron, this substance is very abun- dant, supplying immense quantities which are taken away. Thermal springs are found near those of naphtha *. The naphtha springs at Rangoon, Pegu, appear to be ex- ceedingly abundant. Mr. Coxe estimates their produce at 92,781 tons per annum. In the Indian Islands there are also similar springs. Marsden notices them in Sumatra, at Ipu, and elsewhere. Coral Reefs and Islands. In consequence of the numerous situations where these are observable in the Pacific Ocean and Indian Seas, very exag- gerated ideas have generally been entertained of their relative importance. Large masses, supposed to be the work of my- riads of polypi fers, were considered to have been raised by the labour of these animals from great depths, while immense sheets of coral rock were supposed to cover the bottom of the seas. During Kotzebue's voyage, M. Chamisso enjoyed op- portunities of visiting seme remarkable groups of islands, ar- ranged in a circular or oval manner, with openings among them which permitted the passage of a vessel from the outer ocean into the central basin. These islands seemed merely higher portions of a circular or oval ridge of coral reefs of un- equal heights. M. Chamisso presented a description of what he considered the stages which the coral reef passed through before it became an island habitable for man. This descrip- tion has been so often quoted that it must be familiar to most readers. Subsequently to Kotzebue's voyage, MM. Quoy and Gai- mard, who sailed with the expedition of M. Freycinet, paid particular attention to the coral islands and reefs which they had opportunities of examining; and the result of their obser- vations was, that the geological importance of these islands and reefs had been greatly exaggerated. Far from supposing that the polypifers raise masses from great depths, they con- sider that they merely produce incrustations of a few fathoms in thickness. In those situations where the heat is constantly intense, arid where the land is cut into bays, with shallow and quiet water, the saxigenous polypi increase most considerably, incrusting the rocks beneath. The same authors observe, * Edin. Phil. Journal, vol. v. M 2 164 Coral Reejs and Islands. that the species which constantly formed the most extensive banks belong to the genera Meandrina, CaryophyUia, and ^4strea 9 but especially to the latter ; and that these genera are not found at depths exceeding a few fathoms. It is therefore concluded, that unless we are to suppose these animals en- joying the prerogative of inhabiting all depths, under various pressures of water, and different temperatures, they cannot have produced the masses attributed to them. From these and other considerations they infer, that the appearance of coral reefs and islands depends on the inequalities of the mineral masses beneath, the circular character of some being due to the crests of submarine craters*. This conclusion seems far from improbable, for we know that volcanic vents are common in the same seas ; and that in the West Indies, and the tropical parts of the Atlantic, where corals are suffi- ciently numerous, we do not observe these circular groups of islands, the volcanic vents, though existing, not attaining the importance of those in the Pacific Ocean or Indian Seas. MM. Quoy and Gaimard observe, that, neither with the anchor nor the lead, have they ever brought up fragments of Astre<) alone capable of covering large spaces, except where the water was shallow, about twenty-five or thirty feet in depth, though they found that the branched corals, which do not form solid masses, lived at great depths f . They agree with Forster, that the polypifers may form small isles, when masses of land shelter them, by raising their habitations to the level of the sea: thus exposing a surface on which sands and other matters are heaped and consolidated : a mode of formation in accordance with what I have observed on the coasts of Jamaica. With regard to the great depth of water frequently observed close to the coral reefs, the same authors consider, that they may be accounted for on the supposition that the polypifers have erected their dwellings upon the verge of a steep cliffi such as is commonly observed on the sides of mountains and coasts. In support of this opinion they cite the isle of Rota; where corals, resembling those now found in the neighbouring seas, occur on cliffs. There are, however, certain situations where coral reefs run, as it were, in a line with a coast, but separated from it by deep water, which would seem to require a different explanation. * Quoy et Gaimard, Sur 1'Accroissement des Polypes Lithophytes con- side"r6 geologiquement, Ann. des Sci. Nat. torn. vi. f Sounding off Cape Horn at about 56 S., and in about fifty fathoms of water, they brought up small live branched corals ; and sounding in one hun- dred fathoms on the bank of Laghullas (off the southern point of Africa,) they obtained JRetepora. Coral Reefs and Islands. 165 In situations such as those in which these coral isles and reefs abound, where recent, and comparatively recent, volcanic action is so apparent, we should expect to find evidences of the rise of such reefs above the level of the sea ; and, accord- ingly, navigators have presented us with them. MM. Quoy and Gaimard state, that the shores of Coupang and Timor are formed of coral beds, which induced Peron to consider that the whole island was the work of polypifers. But it ap- pears, that, proceeding towards the heights, vertical beds of slate, traversed by quartz, are met with at about five hundred yards from the town : and upon these and other rocks do the coral beds rest, which MM. Quoy and Gaimard estimate as not exceeding twenty-five or thirty feet in thickness. At the Isle of France a similar bed, more than ten feet thick, occurs between two lava-currents ; and at Wahou, one of the Sand- wich Isles, coral beds extend some little distance into the in- terior. To this we may add, that round the east coast, and on the northern side of Jamaica, there is an extensive bed that merely fringes the land, about twenty feet thick, which has every appearance of a coral bank raised above the waters, and brought within the destructive action of the breakers. In situations like those in the Pacific, where volcanos and coral reefs are both abundant, we should expect to find some curious combinations of volcanic matter with coral banks, and even alternations: even admitting, for the argument, that the principal rock-forming polypifers do not build beneath twenty- five or thirty feet of water; still with the movements of land which may accompany volcanic action, such banks may be de- pressed, and covered by lava-currents, and again raised and brought to view. The example adduced in the Isle of France is sufficient to show, that at least one coral bed may be in- closed between lava-currents. We cannot conclude this sketch without noticing a singular fact, observed, as we have been informed, by Mr. Lloyd while engaged in his survey of the Isthmus of Panama. Seeing some beautiful polypifers on the coast, he detached specimens of them; and, it being inconvenient to take them away at the time, he placed them on some rocks, or other corals, in a shel- tered and shallow pool of water. Returning to remove them a few days afterwards, it was found that they had secreted stony matter, and fixed themselves firmly to the bottom. Now this property must greatly assist in the formation of solid coral banks ; for if pieces of live corals be struck off by the breakers, and thrown over into calm water or holes, they would affix themselves, and add to the solidity of the mass. 166 Submarine Forests. Submarine Forests. At various points round the shores of Great Britain, of the northern parts of France, and of the Baltic, accumulations of wood and plants, which do not appear to differ from those now existing, but on the contrary to be identical with them, occur at levels beneath those of high-water. To these lig- neous and other vegetable remains, which are commonly seen at the retreat of the tide, a temporary removal of the beach, or an encroachment of the sea on tracts of land but slightly raised above it, the name of Submarine Forests has been given. Correa de Serra describes the submarine forest on the coast of Lincolnshire as composed of the roots, trunks, branches, and leaves of trees and shrubs, intermixed with aquatic plants; many of the roots still standing in the position in which they grew, while the trunks were laid prostrate. Birch, fir, and oak were distinguishable, while other trees could not be deter- mined. In general, the wood was decayed and compressed, but sound pieces were occasionally found, and employed for ceconomical purposes by the people of the country. The sub- soil is clay, above which were several inches of compressed leaves, and among them some considered to be those of the Ilex aquifolium, as also the roots of Arundo Phragmites. These appearances are not confined to the coast, but extend considerable distances into the interior, so that the former merely presents a natural section of that which occupies a large area inland. A well sunk at Sutton afforded the follow- ing section: 1 Clay 16 feet. 2. Substances similar to the submarine forest 3 to 4 feet. 3. Substances resembling the scouring of a ditch-bottom, mixed with shells and silt . 20 feet. 4. Marly clay ,?.;,. . . 1 foot. 5. Chalk rock* , . . 1 foot to 2 feet. 6. Clay '.'.''. *. . . 31 feet. 7. Gravel and water Not known. Another boring made inland by Sir Joseph Banks afforded a similar section. This "moor" as Correa de Serra terms it, is considered to extend to Peterborough, more than sixty miles south from Sutton f. Mr. Phillips presents us with very interesting details re- specting some lacustrine deposits in Yorkshire, which are ap- parently ot the age of these submarine forests, and which have This would seem not to be chalk, properly so called, but merely a chalky substance. f Correa de Serra, Phil. Trans. 1799. Submarine Forests. 167 become in some places submerged. He remarks that the fol- lowing may be considered as their general section. 1. Clay, generally of a blue colour and fine texture. 2. Peat, with various roots and plants ; and in large deposits, contain- ing abundance of trees, nuts, horns of deer, bones of oxen, &c. 3. Clay of different colours, with fresh-water Limnace. 4. Peat, as above. 5. Clay, "with fresh-water Cyclades, &c. and blue phosphate of iron. 6. Shaly curled bituminous clay. 7. Sandy coarse laminated clay, filling hollows in the diluvial formation. Mr. Phillips considers the accumulations of peat along the banks of the Humber and its tributaries as of the same epoch as these deposits, of which, he observes, the most constant beds are Nos. 1, 2, and 5. The species of deer enu- merated as found in the peat, are, the great Irish elk, (Cer- vus giganteus), the red deer (C. Elapkus), and the fallow deer (C. Dama). The peat deposit of the marsh-lands is covered by silt and clay, sometimes thirty feet thick, such as is now deposited by the Humber*. The peat is represented as beneath low-water mark. Dr. Fleming describes a submarine forest on the shores of the Frith of Tay, extending in detached portions on each side of Flisk beach, three miles to the westward and seven miles to the eastward. It rests on clay of unknown depth. The clay is similar to the carse ground on the opposite side of the Frith, and to the banks in the channel. " The upper portion of this clay has been penetrated by numerous roots, which are now changed into peat, and some of them even into iron py- rites. The surface of this bed is horizontal, and situated nearly on a level with low-water mark. In this respect, how- ever, it varies a little in different places. The peat bed oc- curs immediately above this clay. It consists of the remains of leaves, stems, and roots of many common plants of the natural orders Equisetacea, Graminea, and Cyperacecc, mixed with roots, leaves, and branches of birch, hazel, and probably also alder. Hazei-nuts destitute of kernel are of frequent oc- currence. All these vegetable remains are much depressed or flattened where they occur in a horizontal position, but when vertical, they retain their original rounded form. The peat may be easily separated into thin layers, the surface of each covered with leaves. The lower portion of this peat is of a browner colour than the superior layers; the texture is like- wise more compact, and the vegetable remains more oblite- ratedf." The same author further observes, that stumps of trees, * Phillips, Illustrations of the Geology of Yorkshire, 1829. f Trans. Royal Soc. of Edinburgh, vol. ix. 168 Submarine Forests. with the roots attached, are observed on the surface of the peat, and no doubt can exist that they are in the positions in which they grew. No alluvial soil stratum was observed above the peat, the surface of which does not occur at a higher level than from four to five feet below high-water mark. Dr. Fleming also describes another submarine forest in the Frith of Forth, at Largo Bay. It rests on a brown clay, into which the roots of the trees have penetrated. The author considers it as lacustrine silt. Over this there is an irregu- larly distributed covering of sand and fine gravel. The peat is composed of land and fresh-water plants, among which are the remains of birch-, hazel-, and alder-trees; hazel-nuts are also seen. Dr. Fleming traced the root of one tree, apparently an alder, more than six feet from the trunk *. If we pass from the main land of Scotland to its isles, we shall observe that the same appearances present them- selves. Mr. Watt notices a submarine forest in the bay of Skaill, on the west coast of the mainland of Orkney. Stems of small fir-trees, ten feet long and five or six inches in diameter, are found partly imbedded in, and partly resting on, the sur- face of an accumulation of vegetable matter principally com- posed of leaves. The stems were still attached to their roots, and the whole was greatly decayed, so as to be easily cut by the spade. Many seeds of the size of a turnip-seed were dis- covered among the vegetable matterf. The Rev. C. Smith describes a submarine forest on the coast of Tiree, one of the Hebrides. Beneath a plain of 1500 acres in extent there would appear to be moss- land, similar to that previously noticed, under twelve or sixteen feet of alluvial covering. The moss-land is seen to bound the plain on the east, and the bay in which it appears is open to the whole force of the Atlantic. The general depth of the peat or moss-land amounts to several feet, but at its appearance on the shore it does not exceed four or five inches. This is firm, and ad- heres strongly to a sandy clay, on which it is based. Besides the remains of trees, which are obvious, there are other and smaller plants, and numerous seeds, which at first looked quite fresh, but afterwards became darker from exposure. " The seeds have the appearance of belonging to some plant of the natural order of Leguminostf ; and Mr. Drummond sug- gests that they may probably be those of Genista anglica\" According to the same author, submarine forests are by no means uncommon on the shores of Coll. He also cites the Kev. H. Maclean as having noticed similar appearances, not observed by himself, in the island of Tiree. * Journal of Science. f Edin. Phil. Journal, vol. iii. p. 100. J Smith, Edin. New Phil. Journal, 1829. Submarine Forests. 169 Returning again to the main land, we find similar appear- ances described by Mr. Stephenson, on the shores of the flat lands between the Mersey and the Dee, on the coast of Che- shire. Stumps of trees, ramifying in all directions, are stated to appear as if cut off about two feet from the ground. The ve- getable matter rests on blueish marl, and is covered by sand *. Mr. Yates notices a submarine forest on the coast of Meri- onethshire and Cardiganshire, divided into two by the estu- ary of the river Dovey. A sandy beach and wall of shingles bound it on the land side, beyond which is marsh-land, par- tially drained by the oozing of the water through the sand and shingle. The Pinus sylvestris, or Scotch Fir, occurs among the other treesf. Mr. Homer describes a submarine forest on the coast of the S. W. part of Somersetshire. It is well seen between Stolford and the mouth of the Parret, where the shore is low ; a high shingle beach, principally composed of lias (the rock of the vicinity), protects the level land behind from the sea. The vegetable remains present themselves here, as in the other places, as a stratum of peat or decayed leaves, containing the trunks, stems, and branches of trees. Among these are twigs, nuts, and a plant, (commonly found entire,) which Mr. Brown considered might be the Zostera oceanica of Linnaeus. Some of the stems of trees were twenty feet long, and the woods were considered to be oak and yew, not generally decayed, but sufficiently hard and tough to be used as timber, and for fuel. Even those trees which were soft when taken out, be- came hard when dried. The brown vegetable matter was generally a foot or eighteen inches thick, and rested on blue From this coast thera is an extensive tract of flat land, which extends a considerable distance inland, and from it the hills rise in promontories, islands, and other forms, precisely as they would rise from a level sea. Mr. Horner cites De Luc as stating, that while new channels were digging between the Brue and the Axe, a bed of peat was found beneath the surface. This stratum, if it may be so called, has been noticed in other parts of the same flats, and even trees have been reported as found in it ; seeming to show that the forest noticed on the shore may be only a section of a large deposit beneath the Bridgewater levels. * Edin. Phil. Journal, vol. xviii. Mr. Smith cites the Liverpool Courier of December 1827, to show that after a heavy gale, trunks and roots of trees were found under the sand below high- water mark, which had all the ap- pearance of having grown where then found. t Yates, Proceedings of the Geol. Soc., November 7, 1832. t Horner, Geol. Trans, vol. iii. p. 380, &c. 170 Submarijie forests. A very important addition to our knowledge of submarine forests has been made by Dr. Boase in his description of that in Mount's Bay, Cornwall. The vegetable bed consists of a brown mass, composed of the bark, twigs, and leaves of trees, which appear to be almost entirely hazel. In this there are numerous branches and trunks of trees. The greater part of this wood is hazel, mixed with alder, elm, and oak. " About a foot below the surface of this bed, the chief part of the mass is composed of leaves, amongst which hazel-nuts are very abundant. In this layer may also be found filaments of mosses, and portions of the stems and seed-vessels of small plants, many of them evidently belonging to the order of Grasses ; together with the fragments of insects, particularly of the elytra and mandibles of the beetle tribe, which still display the most beautiful shining colours when first dug up, but on exposure to the air all these minute objects soon crumble into dust." Beneath this, the vegetable matter becomes closer, and finally earthy and of a lamellar structure. It rests on gra- nitic sand, and this again on clay slate. The vegetable stratum slopes from the interior to the sea at about an angle of two degrees. It is covered by a bed of smoothly polished shingles, composed of hornblende rock, about two or three inches in diameter. The bed is sixteen feet thick, and is crowned by a granitic sand about ten feet thick. The. vegetable bed, by its rise, appears beneath a marsh inland, having passed under its covering of pebbles and sand *. M. De la Fruglaye observed that after a heavy gale in 1811, a beach near Morlaix, which previously seemed to consist of sand, presented, from the sand being washed away, an ap- pearance of a large mass of vegetable matter and trees united together and extending along shore for a considerable di- stance. The leaves were well preserved, but the trunks and branches of trees were rotten. Oak was observed among the wood, and insects with their colours preserved were disco- vered in the mass. A few days- after this event this accumu- lation of vegetable matter was again covered up by sand f. The peat moors which occur on different points of the shores of the Baltic, near Greifswald, near Gnageland, on the S.E. side of the Haff, at its confluence with the Oder near Swine- mlinde, in the island of Usedom, and in the vicinity of Col- berg, agree in many respects with the submarine forests above noticed. They are in many places from ten to fourteen feet below the level of the Baltic, above which they only rise a few feet. They are separated from the sea by a slip of coast of various breadths, by dunes and sandbanks, under which they * Boase, Trans. Geol. Soc. Cornwall. f Journ. des Mines, t. xxx. Submarine Forests. 171 do not extend, but gradually disappear. In these peat moors, the greatest depths of which are in their central portions, land, marsh, and fresh-water plants, with their seeds, are alone discovered, to the exclusion of marine vegetation. Trunks of trees with their roots, among which are oaks and pines (Pinus sylvestris), are found in them, not at the bottom, but at some height above it. The roots occur in their natural positions, often several times above each other, still however beneath the present level of the Baltic to the depth of five feet. In some places the Arundo Phragmites so abounds that the peaty moss seems entirely composed of it. The lower layers contain Ceratophyllum demersum, Potamogeton pusillum, Najas major, Nymphr a more detailed description of these localities, with a view and a of Baussi Raussi cliff, see my paper in the Geological Tran.sa.cti ons K *For section vol. iii. 2nd series. t Dela Marmora, Journal de Geologic, torn. N 178 Raised Beaches and Masses of Shells. Tuscany, of the Roman States, and of Sicily*, the change of level would appear not to have been altogether local. M. Boblaye notices various lines of worn rock on the lime- stones of Greece (similar to the lines now produced by the action of the waves on the coasts of the same country), raised at various heights above the present level of the Mediterranean. He also points out the existence of small horizontal terraces, and lines of holes drilled by perforating shells ; circumstances which M. Boblaye attributes to successive elevations of the land above the level of the sea. A littoral cavern, near Na- poli di Romania, contains a breccia referrible to the present epoch, for it contains fragments of antique pottery. This ca- vern has apparently been raised five or six yards above the present level of the Mediterranean f. It has been previously observed, that on the west coast of South America a beach was raised during the earthquake of 1822, and there were evidences of former beaches having been so elevated. M. Lesson also observed at Conception, more southerly on the same coast, banks of shells, correspond- ing with those of the neighbouring sea, now dry and raised above itj. It is almost impossible not to remark in these raised beaches and sea beds, the action of the same forces which have been noticed under the head of Earthquakes. The land has been liable to rise and fall at various epochs, as will be seen in the sequel; the intensity of the force, producing these changes, varying materially. It is exceedingly difficult to assign dates to the Plymouth raised beach, to the shells at Uddevalla, and to the other similar appearances above noticed ; but we learn from them, that since the establishment of animal life, such as we now observe it, the relative levels of the sea and land have been liable to change, as they have been previously to this period, and, referring to the Temple of Serapis, near Naples, as they have been even since man has erected his temples and other works of art. * M. de la Marmora, carefully distinguishes the sandstone from the rock daily forming at Messina. f Boblaye, Journal de Geologic, torn. iii. J Brongniart, Tab. des Ter. qui composent 1'Ecorce du Globe, p. 92. For a detailed account of the geological appearances connected with the celebrated Temple of Serapis, at Puzzuoli, near Naples, consult Lyell's Principles of Geology, vol. i. The rise and fall of land seem to have been as follows: 1. After the original building of the temple, a sinking of the land, and a covering of the lower part of the columns, so that the boring shell (Lithodomus) only attacked them about twelve feet above their pedes- tals. The height to which the shells have bored is also about twelve feet ; therefore the columns, without being overthrown, were certainly lowered to Organic Remains of the Modern Group. 179 Organic Remains of the Modern Group. These will necessarily consist of existing animals, but may also include some no longer found in a living state. Man not only greatly modifies the present surface of the land, by de- stroying tracts of forests, preventing the inundations of low countries, turning torrents, and directing the surface-water through innumerable channels to satisfy his own wants and conveniences, but he also drives all animals before him which do not suit his purposes ; thus circumscribing the domain of those which are not useful to him, while he covers the country with those that are, and which never could exist in such num- bers but for his care and protection. Consequently all terres- trial remains would correspond with the increasing power of man, and therefore a very different suite of such remains would be now entombed, than when his power was more limited. Over the inhabitants of the waters he would exercise little control, excepting in rivers, small lakes, and round some coasts. One very material difference would be effected in the quan- tity of trees and shrubs transported to the sea, more particu- larly in the temperate and colder regions, where man requires wood, not only for the purposes of various constructions, but also for fuel. We see in the delta of the Mississippi that an abundance of wood is now transported there by the river, but this will daily diminish as man converts the forests, whence it is derived, into pastures and corn-fields. The gigantic animal Cervus giganteus, commonly known as the Irish Elk, was once imagined to have existed only at an epoch anterior to man, but it is now considered that he was co-existent with him ; although this by no means proves that it did not live upon the earth previous also to him, as seems to have been the case. We have no great certainty when the Mastodons of North America ceased to exist; it is commonly supposed that they became extinct previous to the commence- ment of the modern group, but of this we have no good proof. The same may be said of some other animals. The Dodo seems to afford us an example of the extinction of an animal in comparatively recent times ; for it is now al- most certain that this curious bird existed on the isle of Mau- ritius, during the voyages of the early navigators to the East the depth of twenty-four feet above their pedestals in water. 2. Elevation of the temple, still standing, above the level of the sea, or nearly so, for the pavement is not flooded to any considerable depth, not more than about one foot. N 2 180 Organic Remains of the Modern Group. Indies. The relative antiquity, therefore, of animals whose remains are only now found entombed, must not be too hastily inferred. The bone of the wolf is that of an extinct animal, as far as the British islands are concerned. In the darkness of ages many animals may have perished, not a tradition of whose existence remains, not only from the advance of man, and the power which civilization affords him, but also from the destruction caused by predaceous animals, though the latter is not so probable as the former. Erratic Blocks and Gravel. 181 SECTION III. ERRATIC BLOCK GROUP. Erratic Blocks and Gravel. WE must impress upon the geological student the necessity of considering this group as simply one of convenience, formed provisionally for the purpose of presenting certain phaenomena to his attention, which in the present state of science could not so easily be done under any other head. The origin of the various transported gravels, sands, blocks of rocks, and other mineral substances scattered over hills, plains, and on the bottoms of valleys, often referred to one epoch, may be- long to several. In a word, all that transported matter com- monly termed Diluvium, requires severe and detailed exami- nation. At the present time, there would appear to be three principal opinions connected with the subject. One, supposing the transport to have been effected at one and the same period; another, that several catastrophes have produced these su- perficial gravels ; while a third would seem to refer them to a long continuance of the same intensity of natural forces as that which we now witness. Perhaps these various opinions may arise from our present inadequate knowledge of the phenomena on which we attempt to reason, and probably also from premature generalizations of local facts. These different opinions, though they cannot each be correct in explanation of all the observed facts, may each be so in part; and it were to be wished that the phaenomena here arranged under one head solely, as above stated, for convenience, were examined without the control of a preconceived theory. At the close of the last section, a local elevation of land was noticed, of somewhat difficult arrangement in our systems. In order to illustrate the changes which have taken place in the same district, without, however, attempting to consider such appearances as general, I shall continue the description of it. At Oreston quarries, Plymouth, clefts and caverns in limestone rocks have afforded numerous remains of the ele- phant, rhinoceros, bear, ox, horse, deer, &c. buried, more particularly in the case of clefts, beneath considerable angular masses and smaller fragments of limestone. In one instance which I noticed, the animal remains occurred beneath ninety feet of such accumulations, the bones and teeth being confined to a black clay under the fragments. The remains of bears, 182 Erratic Blocks and Gravel. rhinoceroses, hyaenas, and other animals contained in the ce- lebrated Kent's Hole, near Torquay, belong to the same dis- trict. In the 'superficial gravel of this part of the country, the remains of animals, of the same kind as those detected in the caverns, have not yet been discovered ; but if we continue our researches eastward, we shall find them in the valleys of Charmouth and Lyme*, where they occur in situations which would appear anterior to the great weathering, if I may so ex- press myself, of the circumjacent hills : thus apparently giving these remains of elephants and rhinoceroses the same relative antiquity as those beneath fragments in the clefts of rocks near Plymouth, and probably also as those contained in the caverns at the same place, and in Kent's Hole. Now the raised beach in Plymouth Sound seems to afford evidence of a configuration of land not widely different, in that place, from the present, and therefore we may perhaps infer the existence of inequali- ties in the land, or hill and dale, in this district generally, not widely different from those we now observe. It will be re- marked that the animal remains which seem to imply a warmer climate existing at that time than at present, occur in low grounds, fissures, and caves. Upon the former they may have lived, and into the two latter they may have either fallen or been dragged by beasts of prey. The elephants probably browsing on branches and herbage, rhinoceroses preferring low grounds, the bears and hyaenas inhabiting caves, and the deer, the ox, and the horse, ranging through the forest and the plain ; all which supposes land fitted for them, and therefore hill and dale, level plains and rocky escarpments with open ca- verns. Consequently valleys were scooped out previous to the existence of the elephants ; and if a mass of waters acted on the land, destroying these animals, it must have been influenced in its direction by the previously existing inequalities of surface. The next question may be, does this district present evi- dences of the exertion of a greater intensity of natural force than that which we now observe? The answer may be, that it does. The whole district is fractured, or, to use geological terms, so broken into faults, that the spaces in which, with careful examination, they may not be detected, are very in- considerable. Such dislocations may, or may not, have been contemporaneous with the raised beach. Perhaps they were previous to it, for there has evidently been a very considerable dispersion of rock fragments, and this apparently by water, which would have scattered such a beach as that noticed at Plymouth. The following section at the Warren Point, near * The line of coast has been preferred in this description, because the sections are there more clear and less equivocal. Erratic Blocks and Gravel. 183 Fig. 26. Dawlish, is not only a good example of a compound fault, but also of transported gravel upon it. b b, conglomerates, and c c, sandstones of the red sandstone formation, fractured or broken into faults at ff> so that conti- nuous strata are displaced. Upon these fractured strata rests a gravel, a a, com- posed of chalk flints, and green sand chert, mixed with a few pebbles similar to those in the conglomerates b b. It has evidently been deposited subsequent to the fracture, for it rests quietly upon it and is unfrac- tured. The chalk and green sand of this district have once covered very considerable spaces, though the latter is now only seen on Haldon Hills ; near this section, it is true, but separated from it by an intervening valley. There are many other dislocations so covered on the same coast, where these appearances can be observed with the greatest ease, particu- larly at low water. It might be supposed that these flints and pieces of chert were merely the remains of superincumbent masses of chalk and green sand, which have been destroyed by meteoric agents, the harder parts falling down on the top of the fracture. We can scarcely consider this physically probable, if even possible; for it supposes the removal of more than 600 feet of sandstone and conglomerate (for not until that height above this section would the green sand and chalk come on), without scarcely leaving any of the pebbles, or large masses of the red con- glomerate, while the flints and cherts, which belonged to up- per, and consequently first destroyed rocks, remain. Let us now consider another class of appearances. Over the whole district, wherever transported gravel occurs, the surface of the rocks (it being of no importance what they hap- pen to be,) is drilled into cavities and holes, similar to those well known on the chalk of the east of p^ 2 7. England. The following sections will illustrate this: a *^4O>;c Broc., Cancellaria cassidea, Broc., Buccinum corrugatwn^ Broc., Cerithium Li?na, Brug., C. quadrisulcatum, Lam., Murex rugosus, Sow., M. mi- nax, Pyrula Jicoides (Bulla Jicoides^ Broc.), Ostrea virginica, Lam., O. edulina, Sow., Pecten latissimus, Broc., P. medius, Studer, Meleagrina margaritacea, Studer, Area antiquata, Lam., Cardium edulinum, Sow., C. oblongum, Broc., C. semi- granulatum. Sow., C. Mans, Broc., C. clodiense, Broc., C. mul- ticostatum^Yoc^ Tellinatumida,Broc.) Venus Island tea, Lam., Venus rustica, Sow., Astarte excavata, Sow., Cytherea convexa, Brong., Corbula Gallica, Lam., Panopaea Faujasii, Solen Va- gina, Lam., S. strigilatus, Lam. (analogue now living), S. Le- gumen, Linnaeus, Balanus perforatus, Studer*. Prof. Sedgwick and Mr. Murchison, in describing the con- tinuation of these rocks on the flanks of the Salzburgh and Bavarian Alps, mention great alternating masses of conglo- merate, sandstone and marl, north of Gmunden; and still further north, in the higher part of the series, beds of lignite. Detailing the section of the Nesselwang, they state that the lowest Supracretaceous or tertiary strata are of great thickness, and are applied vertically against the Alps. The conglome- rates are mentioned as extremely abundant, the molasse and marl being entirely subordinate to them. According to these authors, there are three or four distinct lines of lignite, sepa- rated from each other by thick sedimentary deposits. Hence they infer that the presence of lignites alone is unimportant, as these occur in very different situations. In a section taken through the hills at the east end of the lake of Constance, the lower part of the Supracretaceous or tertiary system is de- scribed as composed of green micaceous sandstone, in which beds of conglomerate are subordinate, and it is considered identical with the molasse of Switzerland. The conglomerates alternating with greenish sandstone and variously coloured marls are noticed as forming the upper Supracretaceous group, and composing the mass of the mountain ridge extending northwards from Bregenz. Supracretaceous rocks are no- ticed in the valley of the Inn, containing coal, worked for pro- fitable purposes, thirty-four feet thick, near Haring. The * Brongniart, Tableau des Terrains qui composent 1'Ecorce du Globe. Supracretaceous Group. 241 coal is described as accompanied by fetid marls variously in- durated. In the coal and overlying beds there are many ter- restrial and fluviatile shells, and also in the latter beds nume- rous impressions of dicotyledonous and other plants. Several marine shells are discovered in these strata. The authors con- sider that the various sections which they observed, prove the comparatively recent elevation of the neighbouring Alpine chain ; and the more recent supracretaceous deposits noticed by them, bear the same relation to the neighbouring Alps as the Sub- Alpine rocks in Northern Italy do to the high moun- tains near them; whence they infer that the northern and western basins of the Danube, and the supracretaceous basin of the Sub-Alpine and Sub-Apennine regions, have been left dry at the same period*. According to Prof. Sedgwick and Mr. Murchison, the su- pracretaceous rocks of Lower Styria consist, in the ascending series of a section from Eibeswald to Radkersburg, 1. Of micaceous sandstones, grits and conglomerates, derived from the slaty rocks on which they now rest at a highly inclined angle. 2. Of shale and sandstone with coal. At Scheineck, where the coal is extensively worked, it contains bones of Anthracotheria^ and in the shale Gyrogonitcs (Char a tuber cu- lata of the Isle of Wight), flattened stems of arundinaceous plants, Cypris, Paludin< f fish-scales, &c. 3. Of blue marly shale and sand. 4. Of conglomerate, with micaceo-calcareous sand and millstone conglomerate, occupying the whole hilly region of the Sausal. 5. Of coralline limestone and marl. The organic contents of this rock are stated to be, many corals of the genera Astrea and Flustra ; Crustacea ; Balanus crassus, Conus Aldrovandi, Pecten infumatus, P/wlas, Fistulana* &c. The authors refer this rock to the epoch of the Sub- Apennine formations and English crag. 6. Of white and blue marl, calcareous grit, white marlstone, and concretionary white limestone. At Santa Egida, concretionary white lime- stone, alternating with marls, contains Pecten pleuronectes, Ostrea bellovicina. Scalar ia, Cyprtfa, &c. 7. Of calcareous sands and pebble beds, calcareous grits and oolitic limestone. At Radkersburg, where the hills sink into the plains of Hun- gary, the strata are charged with shells, some being identical with living species (Mactra carinata and Cerithium vulgatum). The authors consider this group as similar to the more recent rocks of the Vienna basin. In describing another section, Prof. Sedgwick and Mr. Murchison notice that, at the Poppendorf, the marls, sands * Sedgwick and Murchison, Proceedings of the Geol. Soc. of London, Dec. 4, 1829. 2 4-2 Supracretaceous Group. and conglomerates, are crowned by a micaceo-calcareous sand, containing concretionary masses of a perfect oolite, affording a good example, if any were wanting, of the trifling value of mineral ogical character in determining rocks far distant from each other*. Let us now proceed to those parts of the South of France which border the Mediterranean, observing that M. Elie de Beaumont, when remarking on the period at which he con- siders the Alps to have been thrown up in a direction between Marseille and Zurich, notices numerous situations where the newer supracretaceous strata are characterized by the remains of Oysters, Polypifers, Patelltf, the Balanus crassus (fig. 38), (which M. Deshayes considers may only be a variety of Balanus Tu- lipa\ Patella conica, and other shells. He also identifies these rocks in Provence, Dauphine, and Switzerland. In the molasse of Pont du Beauvoisin, M. Elie de Beaumont discovered shells which M. Deshayes recognised to be Balanus crassus, Patella conica, and a Pecten partaking of the characters of P. Beudanti, P. Jacobccus, and P.Jlabel- According to M. Marcel de Serres, the marine supracre- taceous rocks of the South of France rest on each other in the following descending order: 1. Sands, generally yellow or white, and more or less argillaceous, calcareous, or siliceous, according to circumstances. These sands abound in the re- mains of terrestrial and marine mammalia, reptiles, and fish, mixed with the remains of birds, and some wood. Shells are not common, with the exception of Ostrete and Balani. 2. Yel- low and calcareous marls, of no great thickness, sometimes alternating with stony beds. 3. Beds of limestone, to which the same author has given the name of c alcaire moello?i, usually worked as a building-stone in the South of France. The upper beds generally contain the greater quantity of shells; these and the middle strata also contain the remains of mam- malia, fish, Crustacea, annulata, and zoophytes. Terrestrial mammalia are very rare, consisting principally of a few bones and isolated teeth, which mostly approach those of the Pa- Iceotlierium and Lophiodon. The lower beds contain but few shells. 4. Argillaceous blue marls, well known as the blue Sub-Apennine marls. These marls vary much in their mine- ralogical character, being more or less calcareous, argillaceous, or sandy, according to circumstances. They have nearly the * Sedgwick and Murchison, Proceedings of the Geol. Soc. of London, March 5, 1830. f Elie de Beaumont, Rev. de la Surf, du Globe ; Ann. des Sci. Nat. 1829 et 1830. Supracretaceous Group. 243 same colour, passing from a greenish or blueish gray into a blue of greater or less intensity. Their thickness seems to depend on the inequalities of the surface on which they rest, their depth being sometimes very considerable, while at others it is trifling. They contain a large collection of marine re- mains, principally shells. Terrestrial mammalia and reptiles are exceedingly rare. M. Marcel de Serres only mentions one stag's horn, the bones of a land tortoise, and the vertebrae of a crocodile. Marine mammalia and fish are scarce, as are also the remains of zoophytes*. The following section, by M. Marcel de Serres, of the strata of Banyuls, through which the Tech has cut its bed, will re- mind the geologist of sections to be seen at Nice, and in various parts of Italy ; 1. (upper bed.) Transported substances, named by the author diluvium of the plains, rolled pebbles of primary rocks, cemented by a brownish red gravelly clay ; thickness from one to three yards. 2. Another deposit of transported detritus, named mountain diluvium by the author, stated to be distinctly separated from the above, composed of rolled pieces of granite, mica-slate, gneiss, and quartz, cemented by a slightly red clay, more gravelly than the first. The size of the rolled fragments is considerable, the smallest being equal to that of the head ; thickness, two to three yards. 3. Yel- lowish siliceous sands, indurated in parts, the beds thick, varying from four to six yards. Lower portion contains shells and lignites. 4. Argillo-arenaceous marls, blueish gray, and micaceous ; sometimes alternating with the upper yellow sands. Shells very abundant ; thickness, six to eight yards. 5. Blueish argillaceous and tenacious marls. They contain few shells, and even these become less abundant as the section increases in depth ; thickness not known. These marls are supposed to rest upon micaceous clay-slates, from the struc- ture of the Alberes chain, at the foot of which these beds of Banyuls dels Aspre are found. Nos. 3. and 4. are stated to contain the remains of mastodons, deer, lamantins, land-tor- toises, and sharks, disseminated among the marine shells, but they are represented to be scarce f. There are many lignite deposits in this part of France, of which the relative ages have not been determined so accurately as could be wished. M. Marcel de Serres, however, shows that some of them are inferior to the calcaire moellon, and probably occur at the lower part of the blue marls. The fol- lowing is a section at Saint Paulet, about a league and a half * The organic exuviae discovered in these marls are enumerated in the lists at the end of the volume. t Marcel de Serres, Geognosie des Terrains Tertiaires du Midi de la France. Montpellier, 1829. R2 24-4? Supra cretaceous Group. from Saint Esprit (order descending) : 1. Yellowish calcareo- siliceous sands, containing the remains of marine shells. 2. Thick beds of the calcaire moellon, containing numerous casts of Cijtherea, Venus, and Cerithia. 3. Sands with marine shells resembling No. 1. 4-. Alternation of fresh-water limestone (containing Gyrogonites\ earthy lignite, and sandy marls. 5. Compact limestone, with Cerithia or Potamides and Palu- dince. 6. Thin argillaceous rnarls, with small oysters. 7. Thin earthy lignite. 8. Argillo-arenaceous marls, with traces of lig- nite. 9. Compact fresh-water limestone, with Limneeee and Cyreneff. 10. Thin yellowish and calcareous marls. 11. Ar- gillaceous blue marls, with traces of more or less fibrous lig- nite. 12. Argillo-bituminous marls, containing numerous ma- rine and fluviatile shells. These marls, as well as the lignite which succeeds them, contain small pieces of amber. 1 3. Lig- nite in beds of two or three yards in thickness, preserving the woody structure, even resembling charcoal : contains amber. 14. Argillo-bituminous marls, with marine and fluviatile shells, the same as No. 12. 15. Lignite with the same characters as No. 13. All these beds rest parallel on each other with great regularity, and show that they have been deposited tranquilly and successively*. Many species contained in the rocks above noticed are ana- logous with those now existing in the Mediterranean, pointing to some kind of connexion between the ancient state of that sea and the present. We therefore seem to arrive at some- thing like a probability that the blue marls were deposited in a sea, perhaps somewhat similar to the Mediterranean, but presenting more surface than it. From the recent observations of Colonel Silvertop it would appear, that the Sub- Apennine deposits are also discovered at Malaga, and other parts of the South of Spain, the blue marls occupying the same relative position f. M. de la Marmora shows us that the supracretaceous de- posits of Sardinia correspond with those of the South of France, of the South of Spain, and of a large part of Italy. The fol- lowing is his account of their superposition (in the descending order): 1. A fine-grained white, or yellowish white, lime- stone ; 2. A yellow and very earthy calcaire moellon^ mixed with sand; 3. Calcareous, sandy, and siliceous strata; 4. Blue marls, sometimes whitish; 5. Some very rare strata of calca- reous conglomerates, with traces of lignite, or else trachytic tuffa cemented by carbonate of lime. No. 5. is rare. The characteristic shells of the blue marls are stated to be Pecten * According to M. Dufrnoy, these beds rest tinconformably on strata equivalent to the green sand ; Annales des Mines, 1830, pi. v. t Silvertop, MSS. Supracretaceous Group. 245 pleuronectes and Venus rugosa. They likewise contain nume- rous remains of crabs, but univalves are described as rarely found*. The remains of large mammalia, which have rendered the Upper Val d'Arno so celebrated, would appear to be disco- vered in beds of somewhat contemporaneous origin ; a differ- ence in the circumstances attending the deposit of the superior rocks having produced a difference of the remains detected in them, inasmuch as marine exuviae are absent. M. Bertrand-Geslin distinguishes three basins between the source of the Arno, and Florence; namely, the basins of Casentino, Arrezzo, and Figline; the whole valley of the Arno, for that distance, being bounded by a sandstone named macigno, or by dark-coloured limestone. According to the same author, the following section (in the descending order) may be observed between Arrezzo and Incisa: 1. Thick bed of yellow argillaceous sand. 2. Thick beds of rolled quartzose pebbles, intermixed with coarse sand. 3. Fine gray and mi- caceous sands, many fathoms thick, containing thin beds of blue sandy marl ; these sands being, in the middle and lower parts, exceedingly rich in the bones of mammiferous animals. 4. Very thick argillaceous blue marl, constituting the lowest deposit in the basin, and containing many fossils in its upper part. From his various observations on the Val d'Arno, M. Ber- trand-Geslin concludes; I. That the rolled pebbles are larger and more abundant in proportion as they approach the mountain chain on the north, whence they appear to have been derived : 2. That the coarse sands occupy the central part of the valley, while the finest sands skirt the foot of the mountain range on the south : 3. That the lower sands and blue marls are deposited in horizontal beds : 4. That the bones of mammalia are very abundant towards the central part of the Val, on the right bank of the Arno, and are rare on the left bank : 5. That these bones, in good condition, and sometimes disseminated, are generally deposited in different planes, as if not all at one time : 6. That the yellow sands contain fluviatile shells at Monte Carlo; and 7. That this transported mass contains neither the remains of marine shells, solid stony beds, nor lignites f. The animals whose remains are stated to have been dis- covered in the Upper Val d'Arno are: Elephas primigenws, Hippopotamus major, Rhinoceros, Tapir, Deer, Horse, Ox, Hyaena, Felts, Bear, Cavern Fox, and Porcupine. The pre- sence of these remains would appear to indicate that the de-p * De la Marmora, Journal de Geologic, t. iii. p. 319. t Ann. des Sci. Nat. t. xiv. 1828. 246 Supracretaceous Group. posit containing them was not far removed, as to date, from the transported gravels and sands, mingled with volcanic sub- stances, in Auvergne, and which will be noticed in the sequel. During this state of comparative repose, in which similar mineralogical substances enveloped similar animal remains over a considerable surface, there were some situations in which vegetable matter was more abundantly collected than in others, as might now happen at the embouchures of rivers when the streams possessed no great velocity. After the pro- duction of the blue marl, circumstances became somewhat altered, and this over a considerable surface, for the deposit no longer continued the same; sands, showing a greater velocity or transporting power of water, commonly covering these blue marls in the South of France and Italy. There were, how- ever, modifying circumstances; for sheets of calcareous matter, frequently producing limestones, occur mixed with these sands, enveloping terrestrial, fresh- water, or marine remains, as these came within their influence. M. Elie de Beaumont notices the following section near the Pertuis de Mirabeau ; which, while it shows that the rocks belonging to the cretaceous and oolitic groups of that neigh- bourhood were disturbed and contorted, previous to the de- posit of the supracretaceous rocks which rest upon them, also exhibits the superposition of certain supracretaceous strata of that part of France with which we have been occupied, and which, in the neighbourhood of Aix, presents such a curious approach, in their organic contents, to some of the terrestrial inhabitants of the present country. Fig. 39. b a a b a , rocks of the oolitic group : b b 9 rocks of the cretaceous group, containing Ammonites and Belemnites mucronatus. D, bed of the Durance at the Pertuis de Mirabeau, on both sides of which rest nearly horizontal beds of supracretaceous rocks, c c, on the upturned edges of the older strata. On the side p, that of Peyrolles, the supracretaceous rocks constitute a thick fresh-water deposit, " principally composed of gray compact limestone, penetrated by numerous irregular tubular cavities, and of sandstone, analogous to that which near Aix alternates with the variegated marls of the fresh- water series*." On the other side of the Durance, and near * Elie de Beaumont, R6v. de la Surf, du Globe : Ann. des Sci. Nat. 1829 et 1830. Supracretaceous Group. 247 the chapel of La Magdelaine, o, the supracretaceous rocks are seen resting on the edges of the older strata, and the following beds are observed, in the ascending order: 1. A calcareous sandstone, without shells, in some strata containing calcareous pebbles, and passing into a conglomerate. 2. The above beds, with the remains of marine shells. In these beds M. Elie de Beaumont observed dolomite. 3. A bed containing some limestone pebbles, and a great number of oysters, their hinges elongated, among which are probably the Ostrea virginica of the shelly molasse of Piolene and Narbonne ; also other shells, among which M. Deshayes recognised Anomia ephippium, Balanus crassus, and an undescribed Pecten, resembling the P. Jacobins, P. Beudanti) and P. Jlabelliformis. 4>. A con- siderable thickness of molasse, not very shelly, in one bed of which there are vegetable remains. 5. An oyster-bed, analo- gous to No. 3, covered by a certain thickness of shelly mo- lasse. 6. A thickness of three yards of a yellow sand, covering an alternation of calcareous sandstone, and a compact blueish gray limestone, with irregular tubular cavities, containing ter- restrial and fresh-water shells. M. Elie de Beaumont does not consider this limestone as the same as that noticed on the other side of the Durance, but as forming the upper part of the supracretaceous series at this place ; while the beds near Peyrolles constitute the lower part of the same series. The exact relations of these rocks with the fresh-water de- posit at Aix, remarkable for the insects found entombed in part of it, do not appear to have been yet well determined. According to Messrs. Lyell and Murchison, the following is a section of the beds rising above the level of the town of Aix (in the descending order): 1. White calcareous marls and marlstone, passing gradually into a calcareo-siliceous grit, containing Cyclas gibbosa, Sow. ; Potamidcs Lamarckii, Buli- mus pygmceuS) and an undescribed species of Cypris\ thickness about 150 feet. 2. Marls, with plants and shells. 3. Marls, with fish and plants. 4. Bed with insects, with occasionally Potamides and plants. This bed is described as a brownish green, or light gray calcareous marl, composed of very thin laminae. 5. Gypsum, with plants. 6. Marls. 7. Gypsum, with fish and plants. 8. Marls, with traces of gypsum. 9. Pink limestone, containing Potamidcs^ Cyclas gibbosa, Sow., and Cyclas Aqua Sexticc, Sow. This limestone is often highly contorted, and passes either into a calcareous grit or red sandstone, and, still lower, into compact calcareous breccia; the whole is based on a coarse conglomerate. The lower beds dip N.N.E. at about 25 or 30. From the section accom- panying the memoir of Messrs. Lyell and Murchison, it would appear that these conglomerates rest, beyond Aix, on red marl, 248 Supracretaceous Group. fibrous gypsum, and gray limestone, with Limnceee and Pla- norbes; and these again on the compact limestone, sand, and shale, containing coal at Fuveau, accompanied by the remains of an Unio, Melania scalaris. Sow., Cyclas concinna. Sow., C. cuneata, Sow., and Gyrogonites* . The preservation of the insects is very great, permitting the determination of genera and species. According to M. Marcel de Serres, Arachnides accompany the insects, properly so called ; the latter, however, being far more abundant than the former, two or three genera only of Arachnides having been determined, while sixty-two genera of insects have been ob- served. The most curious circumstance attending these re- mains is, that some are considered identical with those now existing in the country; Brachycercus undatus, Acheta cam- pestris, Forflcula parallela, and Pentotoma grisea, being, ac- cording to M. Marcel de Serres, the more remarkable. It is also worthy of observation, that the greater part of the insects are of those kinds which generally inhabit arid and dry places. Although they occur in various positions, they are sometimes spread out, as if by an entomologist for the purpose of dis- playing their wings. Their colour is generally an uniform tint of brown or black. Some of the fish discovered in the same marls are so small that they do not exceed ten or eleven millimetres in length f. The place in the series of the supracretaceous rocks to which the brown coal formation of Germany should be re- ferred, does not appear to be as yet well determined. This deposit is characterized by an immense quantity of vegetable remains, and is probably of different ages. The brown coal may be traced from the environs of Aix-la-Chapelle to the Rhine. It there occurs in a narrow plateau between the latter and the Erft, and acquires a thickness of above 100 feet between Bonn and Cologne, without any extraneous bed. The brown coal deposit rests on the declivity of the grauwacke mountains on the right bank of the Rhine, and is connected with the trachytic conglomerates and basaltic formations of the Siebengebirge. It rises on the plateau of the grauwacke mountains further south in the vicinity of Linz, (Orsberg, Mendenberg,) and spreads in detached por- tions to the Westerwald, where it is greatly extended, and is interrupted and covered by basalt. The deposit extends to the south side of the grauwacke mountains into the Wetterau. Lyell and Murchison, Edin. New Phil. Journal, 1829. f Marcel de Serres, Geog. des Ter. Tertiaires du Midi de la France, in which some of the insects are figured; as also in the Memoir of Messrs. Lyell and Murchison above noticed, in illustration of the remarks of Curtis on the specimens brought to England. Supracretaceous Group. 24-9 Further separated, but still with similar characters, it occurs in the Habichtswalde near Cassel, at the Meissner with basalt, the latter being more recent than it. It again occurs, not far distant, in the basin of Thuringia, the beds being of consider- able thickness at Artern in the Unstrutthal. At Durrenberg and Halle, where the brown coal (near Langenbogen) occurs in beds nearly fifty feet thick, it is broken into separate basins. The brown coal deposit extends to the country around Leipsic, to the Elbe as far as Torgau, and occurs frequently in the low tracts between Magdeburg and the Hartz. It is found in the level country between the Elbe and the Oder. It is discovered at Bockup near Domitz (Mecklenburg), and near Buckow, Freienwalde on the Oder, and to the east at Zielenzig and Gleissin. The brown coal generally rests upon a compact tenacious clay, and is covered by large masses of sand. The organic remains which abound in the superincumbent rocks in Hesse and Magdeburg, have not been accurately determined ; those which occur in the inferior strata of the Rhine are better known. They consist of Pisces; Cyprinuspapyraceus^^roi\i\ (Geistingerbusch near Siegburg) ; an undetermined species from Mendenberg and Friesdorf near Bonn. Reptilia; Rana diluviana, Goldf. ; Salamandra ogygia / Triton Noachinus ; Ophis dubhis 9 from Orsberg. Crustacea; a small crab from Geistingerbusch. Insecta ; species of the genera Lucamis, Meloe, DytisciiS) Buprestris, Cantharis, Cerambyx^ Parandra, Belostoma, Cercopis, Locusta, Anthrax, and Tabanus, all from Orsberg. Plantae; seeds of Ervum hirsutum, and E. tetra- spermum, from Geistingerbusch; reed and dicotyledonous leaves, resembling those still growing in the country, but not specifically the same. Large trunks of trees are very com- mon *. A large part of the South of France, bounded by the ocean, or rather by the sandy dunes it has thrown up, between the districts of Bordeaux and Bayonne, and extending far into the interior, particularly at the foot of the Pyrenees, is com- posed of supracretaceous rocks ; an exact and detailed account of whose varied relations to each other may still perhaps be considered as wanting, though much has been done respecting them. This superficies comprises, among other districts, that extensive and monotonous region named the Landes, where the traveller finds little to relieve the sameness which sur- rounds him, except the peasants stalking over the country mounted on stilts, for the greater convenience of seeing objects afar off. * German Transl. of Manual. 250 Stipracretaceous Group. M. de Basterot has presented us with a very valuable detail of the fossil shells obtained by him from the districts of Bor- deaux and Dax, which is inserted in the Lists of Organic Re- mains, considering that such lists are of the greatest utility to the geological student; referring him, however, to M. de Bas- terot's memoir for the detailed description of each shell. This author remarks, that out of the 330 species of shells noticed by him in the great sandy deposits of the Landes, forty-five only have existing analogues in the neighbouring seas, com- prising the Mediterranean ; and he further observes, that if the basin of the Gironde be taken as a centre, the shells in similar supra cretaceous basins will the more resemble each other as the distances are less. Thus, out of the 330 species collected in the vicinity of Bordeaux, ninety-one are found in the deposits of Italy, sixty-six in those of the environs of Paris, eighteen in those of Vienna*, and twenty-four in the supracretaceous rocks of England f. If reference has been made to M. de Basterot's list, it will have been observed that, though many shells found in this part of France are also discovered at Paris, there is likewise a very considerable correspondence between them and those of Italy. It would appear, from the mention of the fresh- water limestone at Saucats, that there was a change of the relative level of sea and land in that situation, which permitted the envelopment of fresh-water shells in carbonate of lime; and that after this deposit, a change of level was effected, which enabled marine lithodomous shells to bore extensively into the fresh-water rock, and permitted an accumulation of mineral matter and marine shells above it. The analogues of existing species are forty- five; the living species being re- markable for the diversity of their habitats, some being found in the Atlantic and Pacific Oceans, and the Indian and Me- diterranean Seas, while not a few inhabit the coasts of the Channel and the Bay of Biscay, to which, from the fall of the land, the Bordeaux and Dax deposits seem naturally to belong. When the ocean covered this part of France, it seems necessary to suppose that the mean temperature of the situation was above that which it now is, in order to suit the animals, many of whose analogues exist in warm climates. We now proceed to give a short notice of the supracreta- ceous rocks of the Paris basin, as they long constituted the type to which all deposits of this epoch, wherever found, were * M. de Basterot observes, that this number will probably become in- creased as the Vienna basin shall become better known ; which we may expect it soon will be, from the labours of M. Parsch. f De Basterot, Description Geologique flu Basin Tertiairc du Sud- Quest de la France, lere partie; Mem. de la Soc. d'Hist. Nat. de Paris, t. ii. Supracretaceous Group. 251 referred. However the rocks of this group may be eventually discovered to differ from this type, the labours of MM. Cuvier and Brongniart on the rocks of the Paris basin will not the less retain that place in the annals of Geology, which by com- mon consent has been assigned them. Nor will the zoological discoveries of Cuvier, constituting as they did such a brilliant epoch in the history of geological science, the less claim the gratitude of geologists in succeeding ages. The following is the classification of the Paris rocks, ac- cording to MM. Cuvier and Brongniart (order ascending) : fPlastic clay. 1 . First fresh-water formation . < Lignite. (_First sandstone. 2. First marine formation .... Calcaire grossier. {Siliceous limestone. Gypsum, with bones of animals. Fresh-water marls. i Gypseous marine marls. Upper marine sands and sandstones. Upper marine marls and limestone. Millstone, without shells. Shelly millstone. Upper fresh-water marls. Plastic Clay. So named because it easily receives and pre- serves the forms given to it, and is used in the potteries. It rests on an unequal surface of chalk beneath, which is hol- lowed and furrowed in various ways, so as to present hills, valleys, and outstanding knolls, which sometimes have not been covered by the newer and superincumbent rocks ; at least, if they have covered them, the strata which did so have been removed by denudation *. This clay is variously co- loured, being white, gray, yellow, slate-gray, and red. It differs considerably in thickness, as might be expected from the nature of the surface on which it reposes. Above these beds, to which, strictly speaking, the term " plastic clay " is alone applicable, there is often another clay, separated from the former by a bed of sand ; the latter clay being black, sandy, and sometimes containing organic remains. In it occur lig- nites, amber, and shells (both fresh-water and marine). It is stated, that in this deposit, considered as a mass, the lower parts do not contain organic remains ; that in the central por- tion the remains are commonly those of fresh-water animals; and that in the upper part there is a mixture and even an al- ternation of marine and fresh-water remains, the latter gra- dually becoming more scarce, and the former finally prevail- * A breccia of chalk fragments cemented by clay is found at Meudon, separating the chalk and plastic clay. 252 Supracretaceous Group. ing. The following is a list of the organic remains most com- monly found in the plastic clay. Fresh-water Remains. PLANOKBIS rotundatus, Al. Brong. ; P. incertuS) Defr. ; P. Punctum, Defr. ; P. Prevostinus, Defr. PHYSA antiqua, Defr. LIMNEUS kmgiscatus, Al. Brong. PALUDINA virgula, Defr. ; P. indistincta, Defr. ; P. uni- color, Olivier ; P. Desmarestii, Prevost ; P. conica, Prev. ; P. ambigua, Prev. MELANIA triticea, Defr. MELANOPSIS buccinoidea, Poiret ; M. costata, Olivier. NERITA globula, Defr.; N. Pisiformis, Defr.; N. sobrina, Defr. CYRENA antiqua, Defr. ; C. tellinoides, Defr. ; C. cuneifor- misj Sow. Marine Shells contained in the mixture of the upper part. CERITHJUM furtatum, Sow.; C. melanoides, Sow.; another CeritMum not determined. AMPULLARIA depressa, Lam.? (var. minor) ; Ostrea bello- vaca, Lam. ; O. incerta, Defr. Fossil Vegetables. Exogenites ; Phyllites multinervis , En- dogenites echinatus. Calcaire grossier. This, as its name implies, is composed of a coarse limestone, and is more or less hard, so as to be employed for architectural purposes. It alternates with ar- gillaceous beds, and is remarkable for the constancy of its cha- racter throughout a considerable extent of country. It is often separated from the plastic clay beneath by a bed of sand. The organic remains are stated to be generally the same in the cor- responding beds, presenting rather marked differences when the beds are not identical. The inferior beds are very sandy, often more sandy than calcareous, and almost always contain green earth, disseminated either in powder or grains, which, according to the analysis of M. Berthier, appears to be a sili- cate of iron. These beds are remarkable for the abundance of their organic contents. The following is a list of those fossils which are considered to characterize the different parts of this deposit. In the lower beds. MADREPORA, at least three species. ASTREA, three species at least. Tu RBI NOLI A elliptica, Al. Brong. ; T. crispa, Lam. ; 1\ sulcata, Lam. RETEPORITES digitalia. Lam. LUNULITES radiata. Lam.; L. urceolata, Lam. Fu N GI A Guettardi. NUMMULITES lavigata ; N. scabra ; N. numismalis ; N. rotundata. CERITHIUM giga?iteum. LUCINA lamellosa. Supracretaceous Group. 253 CARDIUM porulosum. VOLUTA cithara. CRASSITELLA lamellota. TURRITELLA multisulcata . OsTREAJt,abeIlula ; O. cymbula. In the central beds*. OVULITES elongata, Lam. ; O. mar- garitula, Deroissy. ALVEOLITES miUum, Bosc. ORBITOLITES plana. TURRITELLA imbricata. TEBEBELLUM convolutum. CALYPTR^EA trochiformis. CARDITA avicularia. PECTUNCULUS pulvinatus. CITHEREA nitidula ; C.elegans. MILIOLITES. CERITHIUM? In the upper beds. MILIOLITES. AMPULLARIA spirata. CERITHIUM tuberculatum ; C. mutabile ; C. lapidum ; C. petricolum. LUCINA Saxorum. CARDIUM Lima. COKBULA anatina ? C. striata\. Vegetable Remains, according to M. Ad. Brongniart, in the Calcaire Grossier of Paris : NAYAD^: Caulinites parisiensis. EQUISETACEJE Equisetum brachyodon. CONIFERS Pinus Defrancii. PALMM Flabellaria pari- siensis. MONOCOTYLEDONS, OF UNCERTAIN FAMILY Culmites nodo- sus ; C. ambiguus. DICOTYLEDONS, OF UNCERTAIN FAMILY Exogenites; Phyl- lites linearis. Ph. nenoides, Ph. mucronata, Ph. remiformis, Ph. retusa. Ph. spathulata. Ph. lancea J. Siliceous Limestone. A limestone, sometimes white and soft, sometimes gray and compact, penetrated by silica, infil- trated in every direction and at all points. It is often cellular, the cells sometimes large and communicating with each other in all directions, the silica lining their sides with mammillary concretions, or with small transparent quartz crystals. Osseous Gypsum (Fresh-water), and Marine Marls. The gypseous rocks consist of an alternation of gypsum and calca- reous and argillaceous marls. Above this alternation there are thick marl beds, sometimes calcareous, at others argilla- ceous. In these latter strata are found abundant remains of Limn&de and Planorbes, and in their lower parts, palms of considerable size are discovered prostrate. The gypseous strata contain the remarkable remains of extinct mammalia and other animals, which the genius of Cuvier may almost be said to have restored to life. Above these beds, which, from the nature of their organic remains, are considered to have * Nearly all the well-known fossils from Grignon are found in these beds, t MM. Cuvier and Brongniart, Desc. Geol. des Envir. de Paris, ed. 1822. J Ad. Brongniart, Prod, d'une Hist, des Veg. Fossiles, 1828. 254 Supracretaceous Group. been deposited in fresh water, there is a succession of marls, considered as deposited in the sea, because they contain ma- rine remains ; the marine and fresh- water systems being se- parated by calcareous or argillaceous marls, often thick. The upper marl beds contain numerous remains of oysters, consi- dered to have certainly lived in the places where now entombed, more particularly, as M. Defrance discovered them at Ro- quencourt attached to rounded pieces of marly limestone, which latter are sometimes pierced by Pholades. Organic Remains in the Gypseous Beds. MAMMALIA: Palceotherium magnum, Cuv. (fig. 40, a.} *; P. medium, Cuv. ; Fig. 4-0. P. crassum, Cuv. ; P. latum, Cuv. ; P. curium, Cuv. ; P. mi- nus, Cuv. (fig. 40. .); P. minimum, Cuv ; Anoplothcrium commune, Cuv. (fig. 40. c.) ; A. secundarium, Cuv. ; A. gra- cile, Cuv. ; A. murinum, Cuv. ; A. obliquum, Cuv. ; Charo- ptamusparisiensis,C\\v.\ Canisparisiensis,Cu\.; Coati ; Di- delphis parisiensis, Cuv.; Sciurus; &c. BIRDS. REPTILES : Crocodile ; Trionyx ; Emys. FISH. Organic Remains of the Fresh-water Marls. MAMMALIA: Palceotherium aurelianense, Cuv. (Orleans) ; Lophiodon major, Cuv. (Soissons, &c.) ; jL minor, Cuv. (Paris) ; L. pygmaus, Cuv. (Paris). BIRDS. FISH. SHELLS: Cydostoma mumia, Lam.; Lim- n&a longiscata ,K\. Brong. ; L. elongata, Al. Brong.; L. acu- minata, Al. Brong. ; L. Ovum, Al. Brong. ; Planorbis Lens, Al. Brong. ; Bulimus pusillus, Brard. * The forms of the animals above represented are such as they are con- sidered to have been by Cuvier, Oss. Foss. t. iii. pi. 66. Supracretaceous Group. 255 In the Marine Marls (Yellow}. Fish bones; Cytherea? convexa; Cytherea ? plana ; Spirorbes; Cerithium plicatum. Yellow Marls separated from the above by Green Marls. Spears and palates of the Ray ; Ampullaria patula ? Ceri- thium plicatum ; C. cinctum ; Cytherea elegans ; C. semisul- cata?? Cardium obliquum ; Nucula margaritacea. Calc. Marls, with large Oysters. Ostrea hippopus; O. Pseudochama ; O. longirostris ; O. canalis. Calc. Marls, with small Oysters, Ostrea cochlearia ; O. cy- athula; O. spatulata; O. linguatula ; Balani; Crabs' feet. Upper Marine Sands and Sandstones. These are composed of irregular beds of siliceous sandstone and sand, the lower portion without organic remains that can be supposed to have existed in the places where now found, these being broken and very rare. In some situations, where the broken shells are more common, millions of small bodies are discovered, to which M. Lamarck has given the name of Discorbiles. These non-fossiliferous sands are in many places covered by a limestone, sandstone, or calcareo-siliceous rock filled with marine shells, of which the following is a list : Oliva mitreola ; Fusus? approaching F. longaevus; Cerithium cristatum ; C. lamellosum ; C. mutabile? Solarium; Melania costellata? Melania ? another species ; Pectunculus pulvinatus ; Crassa- tella compressa? Donax retusa? Cytherea nitidula; C. laevi- gata ; C. elegans ? Corbula rugosa ; Ostrea flabellula. Upper Fresh-water Formation. This rock varies very con- siderably in its mineralogical character, being sometimes com- posed of white friable and calcareous marls, at others of dif- ferent siliceous compounds ; among which are the well-known millstones, sometimes without shells, at others charged with Limnaeae, Planorbes, Potamides, Helices, Gyrogonites (seeds of the Charge), and silicified wood. Organic Remains. ANIMAL. Cyclostoma elegans antiqua ; Potamides Lamarckii; Planorbis rotundatus ; P. Cornii; P. Prevostinus; Limneus corneus ; L. Pabulum ; L. ventrico- sus ; L. inflatus ; Bulimus pygmaeus ; B. Terebra ; Pupa De- francii; Helix Lemani ; Helix Demarestina *. VEGETABLE. Muscites ? squamatus ; Chara medicaginula ; C. helicteres ; Nymphaea Arethusae ; Culmites anomalus; Car- polithes thalictroidesf. As has been often remarked, there is evidence in the va- rious organic remains entombed in the strata above noticed, that the space comprised within what is commonly termed the Paris basin, has not always been exposed to the influence of * Cuvier and Brongniart, Desc. Geol. des Env. de Paris, f Ad. Brongniart, Prod, d'une Hist, des Veg. Fossiles. 256 Supracretaceous Group. the same circumstances since the deposit of the chalk, but that there has been an alternation of three lacustrine or fresh-water deposits, with two which are marine; the former constituting the lower and the upper part of the series. It remains to in- quire the probable cause of these variations. By employing the term basin for this collection of supra- cretaceous rocks, we, as before observed, seem to assume that of which we have no great evidence ; the fresh-water deposits may have been, and probably were, effected in basins, but the marine do not require this form. It would seem reasonable to infer that there may have been here, as has been shown to have happened elsewhere, movements in the land, changing its level relatively with the sea. When we regard the mode in which the various deposits are now arranged, we find that, as a mass, they do not repose horizontally on each other ; but that, according to MM. Cuvier and Brongniart, there were various inequalities at different times, commencing with those of the chalk, presenting hills and valleys. In various parts of this unequal soil the lignite and plastic clay were de- posited, thus to a certain extent filling up some of the inequa- lities. Upon this the calcaire grossier was formed, following more or less the inequalities of the surface beneath. To the calcaire grossier succeeded a gypseous deposit, showing an ab- sence of the sea, and the presence of fresh water, of unequal depth. Then followed a large deposit of sand covering up the pre-existing inequalities, in the upper part of which sand are numerous marine remains; the whole presenting a vast plain. A new state of things followed ; the sea disappeared ; and fresh- water remains became entombed*. The mechanical and chemical circumstances attending these deposits have also curiously varied. We will not stop to in- quire whether the inequalities of the chalk were produced suddenly or slowly, for on this head we possess no very de- cided evidence ; but the deposit of the plastic clay (properly so called) would appear to have been slow, even if the detritus, mechanically suspended, may have resulted from a somewhat violent wash of the inferior rocks. In the sands above this, we have the evidence of a transport by water moving with suf- ficient velocity to carry sand onwards. This is followed by a deposit, to a certain extent quiet, composed of vegetables and amber derived from them. The nature of the other organic remains mingled with them, at first indicates the presence of fresh-water animals ; but finally, some variation in the relative level of the land and sea, apparently occurring gradually rather than suddenly, (for there is no evidence of a rush of waters,) * Cuvier and Brongniart, Env. de Paris. Supracretaceous Group. 257 introduces marine animals, which existed at the same time with many fresh-water animals that have gradually become accustomed to live in the same medium with them. This state of things was destined to disappear, and we have a movement of water sufficient to transport sand. This was succeeded by a calcareous deposition, when carbonate of lime, probably in a great measure derived from the ruin of older rocks, was washed away by water, and deposited over a considerable space. It is obvious, from the structure of these rocks, that the materials of which they consist must have been in a state of fine mecha- nical division, such as to have required no violent rush of waters for their removal : they probably subsided during a period of tranquillity. After the deposit of the calcaire grossier, the production of calcareous rocks, remarkable for their cellular structure, took place. The origin of these cells is unknown ; but they probably arose from the calcareous matter, during the act of subsidence, enveloping foreign matter more soluble or perishable than itself, which has subsequently been removed by the agency of water. It is remarkable that the cavities are now lined by silica in such a manner as scarcely to admit of any other supposition, than that the silica was deposited within the cells from a liquid in which it had been previously dissolved. The osseous gypsum presents us with a decidedly new state of things. Singular animals, of which the very genera are now extinct, must have existed somewhere in the district, the remains of which became in some manner entangled in sul- phate of lime, considerable deposits of which were then in pro- gress. The question will arise, Whence did such a quantity of sulphate of lime proceed? Certainly it is a new ingredient, at least in any abundance, in this district; and there is no evi- dence that it was deposited in a sea, as was the case with the carbonate of lime of the calcaire grossier; on the contrary, as it only contains terrestrial and fresh-water remains, it would seem to have been formed through the medium of fresh water. If so, the previous level of the land and sea had been altered, and the springs of the district, if the gypsum was derived from them, must, .instead of carbonate of lime, have produced an abundance of sulphate of lime. This state of things changed ; the sulphate of lime ceased to be produced or deposited in abundance, the relative level of sea and land again became altered, the result was a formation of marls with marine shells in them ; during which, there were at least some places where rolled pebbles were produced, to which oysters became at- tached, some of the pebbles being pierced by boring shells. These deposits are described as conforming more or less to the surface beneath each, and there is no evidence of any par- 258 Supracretaceous Group. ticular movement of water; but to them succeeds a vast quan- tity of sand, the organic remains in which are broken, and the mass fills up inequalities and forms a plane surface. This ap- pears to show a long continued action of water, with a velocity equal to the transport of sand over a considerable space. At the close of this period the causes, whatever they were, that prevented the envelopment of organic remains, ceased, and marine exuviae became entombed in great abundance. Finally, to crown this curious series, we have a deposit of a very vari- ous mineral ogical character, containing the remains of such animals and vegetables as are only known to exist on dry land, marshy places, or in fresh water. This variety of mi- neralogical structure is what we should consider probable in a shallow lake, into which springs, holding various substances in solution, entered at various parts. That the water was shallow, at least in part, has been considered probable by MM. Cuvier and Brongniart, from the remains of Chart?, so commonly found in this deposit; an opinion exceedingly strengthened by the observations of Mr. Lyell on the Charge of the Bakie Loch, Scotland. To produce the friable calca- reous marls, it is not necessary that the waters should be ther- mal ; but judging from the phenomena of existing springs, this condition would seem requisite for the siliceous deposit; for we do not know of any such formation now in progress, except in such springs. If the millstone and other siliceous substances were thus produced (and it seems difficult to ob- tain their formation in any other manner consistent with ex- isting causes), these thermal waters have disappeared, and si- lica is no longer deposited in this district; seeming to show that very great changes in the solvent powers of water, and in the temperature of springs, may take place in the same district at different epochs. Thus we have a great deposit of carbo- nate of lime at the epoch of the calcaire grossier ; another of sulphate of lime at the period of the osseous marls, and, finally, one of silica at the time of the millstone formation. Supracretaceous Rocks of England. Let us now compare the supracretaceous rocks of England with those of the Paris basin. Those of the former country are commonly known by the names of Plastic Clay, London Clay, Bagshot Sands, the Fresh-water formations of the Isle of Wight, and the Crag formerly noticed. Plastic Clay. Unlike the deposit to which the same name is applied in the environs of Paris, this rock, though occasion- ally containing a considerable abundance of clay, employed for various useful purposes, presents us with pebble beds, irregu- larly alternating with sands and clay ; but, like the strata of the same name at Paris, they rest upon an unequal surface of Supracrctaceous Group. 259 chalk beneath. The organic remains also are not principally terrestrial and fresh-water, but for the most part marine, though the others are intermingled with them. These remains are, according to Mr. Conybeare : UNIVALVES Infundibulum echinatum / Murex latus, M. gradalus, M. rugosus, Cerithium funiculatum, C. intermedium, C. melanoides ; Turritella ; Pla- norbis hemistoma. BIVALVES Ostrea pulchra, O. tener; Pec- tunculus Plwnstediensis ; Cardium Plumstedianum ; Myaplana; Cytherea ; Cyclas cuneiformis, C. depcrdita, C. obovata. In ad- dition to this, traces of lignite and vegetables are observed in several places. The three following sections will convey an idea oi this deposit in the neighbourhood of London, according to Prof. Buckland ; and in the Isle of Wight according to Mr. Webster. Section near Woolwich (series ascending}. Chalk with flints, above which: 1. Green-sand of the Reading oyster-bed, con- taining green coated chalk flints, but no organic remains ; 1 . foot. 2. Light ash-coloured sand, without shells or pebbles ; 35 feet. 3. Greenish sand, with flint pebbles; 1 foot. 4-. Greenish sand, without shells or pebbles ; 8 feet. 5. Iron-shot coarse sand, without shells or pebbles, and containing ochreous concretions disposed in concentric laminae; 9 feet. 6. Blue and brown clay, striped, full of shells, chiefly Cerithia and Q/- thereae ; 9 feet. 7. Clay striped with brown and red, and con- taining a few shells of the above species ; 6 feet. 8. Rolled flints, mixed with a little sand, occasionally containing shells like those of Bromley; e.g. Ostrea, Cerithium, and Cytherea, disseminated in irregular patches; 12 feet. 9. Alluvium*. Section at Loam-Pit Hill, three miles S.W. of Woolwich (order ascending). Chalk with flints, above which: 1. Green sand, identical with the Reading bed, and in every respect re- sembling No. 1. at Woolwich; 1 foot. 2. Ash-coloured sand, slightly micaceous, without pebbles or shells ; 35 feet. 3. Coarse green sand, containing pebbles ; 5 feet. 4. Thick bed of ferruginous sand, containing flint pebbles; 12 feet. 5. Loam and sand, in its upper part cream-coloured, and containing nodules of friable marl; in its lower part sandy and iron-shot; 4 feet. 6. Three thin beds of clay, of which the upper and lower contain Cytherece, and the middle, oysters; 3 feet. 7- Brownish clay, containing Cytherece\ 6 feet. 8. Lead-coloured clay, containing impressions of leaves; 2 feet. 9. Yellow sand; 3 feet. 10. Striped loam and plastic clay, containing a few pyritical casts of shells, and some thin leaves of coaly matter ; 10 feet. 11. Striped sand, yellow, fine and iron-shot ; 10 feet. At a higher level than No. 11. on the same hill, the line of the London clay commencesf. * Buckland, Geol. Trans. 1st series, vol. iv. f Ibid, s 2 260 Supracretaceous Group. Section of the vertical beds in Alum Bay, Isle of Wight (order ascending). Above, or rather next to, the chalk : 1 . Green, red, and yellow sand ; 60 feet. 2. Dark blue clay, containing green earth and nodules of dark limestone, in the latter of which Cytherece, Turritella, and other shells are found ; 200 feet. 3. A succession of variously coloured sands; 321 feet. 4. Beautifully coloured sands, alternating with pipe-clay, co- loured white, yellow, gray, and blackish ; 543 feet. In the central parts of these latter deposits are three beds of lignite, and above them, at some distance, five other lignite beds ; each 1 foot thick. 5. Strata of rolled black flint, contained in a yellow sand. 6. Blackish clay, containing much green earth and septaria ; analogous to London clay*. It will be observed, from these sections, that the transport- ing powers of water have not been precisely similar near Lon- don and at the Isle of Wight. At the former place, there would appear to have been a greater movement than at the latter; the mass of the strata near London containing more pebbles in proportion to its depth than the beds of the Isle of Wight, where there would appear to have been a more calm, as well as a more abundant, deposit. This may perhaps in some mea- sure be accounted for, by supposing the Isle of Wight strata, now thrown into a vertical position, to have been gradually accumulated in a hollow or cavity, more remote from the dis- turbing power of currents or motions in the water, than in shallower depths. At all events, the transporting power of the waters appears to have been irregular ; their velocities varying in such a manner that pebbles are carried forward at one time, while fine particles of detritus are alone moved at another. In the Isle of Wight beds we also see that circumstances have been favourable to the accumulation of vegetable mattery which is not irregularly disseminated, but occurs in beds; the cir- cumstances which attended this deposit being continued at ir- regular intervals, such as might be expected at the mouths of rivers. London Clay. This name has been applied to the great argillaceous deposit which underlies the London district. The clay is mostly blueish or blackish, and composed of argilla- ceous and calcareous matter in variable proportions, the latter rarely attaining a sufficient quantity to constitute marl or im- perfect limestone. Layers of calcareous concretions, known by the name of Septaria, are by no means unfrequent; and it is stated that beds of sandstone are occasionally observed in it. It has been often remarked, that if the description of the Paris rocks had not preceded that of the country round Lon- * Webster, Geol. Trans. 1st series, vol. iv. Supracretaceous Group. 261 don and of the Isle of Wight, it never would have been consi- dered that the, so called, Plastic Clay was separated from the London Clay, but rather that they constituted different terms of the same series. It will have been observed that in the above-noticed section at Alum Bay, in the Isle of Wight, there was nothing to warrant such a separation ; neither does there appear to be any good reason why in the London dis- trict they should not be regarded as upper and lower portions of a deposit formed under nearly similar general circumstances. The deposit of the London Clay would appear to mark a comparatively quiet state of things ; and the clay named Plastic marks a similar state, although it occurs among sands and peb- bles. The whole seems merely to show that the velocities of the transporting waters varied, and that they continued for a longer period of little importance during the deposit of the London clay. This clay varies very considerably in thickness. Thus, one mile east of London it is only 77 feet deep; at a well in St. James's-street, 235 feet; at Wimbledon it was not pierced through at 530 feet; and at High Beech, 700 feet*. Organic Remains. A Crocodile; a Turtle. Fish. Crus- tacea, a great variety, few of which have been noticed ; among these few, Cancer tuberculatus, Konig ; C. Leachii, Desmarest; Inachus Lamarckii, Desm. CONCHIFERA Clavagella coro- nata, Desk., cal. gros., Paris; Fistulana personata, Lam., cal. gros., Paris; Gastrochaena contorta; Pholadomya margarita- cea, Sow.; Solen affinis, Sow.; Panopaea intermedia, Sow. ; Mya subangulata, Sow. ; Lutraria oblata, Sow. ; Crassatella sulcata, Lam., cal. gros., Paris; C. plicata, Sow; C. compressa ; Corbula globosa, Sow. ; C. Pisum, Sow. ; C. revoluta, Sow. ; Sanguinolaria Hollowaysii, Sow. ; S. compressa, Sow. ; Tellina Branderi, Sow. ; T. filosa, Sow. ; T. ambigua, Sow. ; Lucina initis, Sow. ; Astarte rugata, Sow. ; Cytherea nitidula, Lam., cal. gros., Paris, Bourdeaux ; Venus incrassata, Sow. ; V. transversa, Sow. ; V. elegans, Sow. ; V. pectinifera, Sow. ; Venericardia Brongniarti, Sow. ; Ven. planicosta, Lam., cal. gros., Paris, Ghent; Ven. carinata, Sow.; Ven.deltoidea, Sow.; Ven. oblonga, Sow. ,- Ven. globosa, Sow. ; Ven. acuticostata, Lam., cal. gros., Paris ; Cardium nitens, Sow. ; C. semigranu- latum, Sow., molasse, Switzerland; C. turgidum, Sow. ; C. porulosum, Lam., cal. gros., Paris; C. edule, Brander, Bor- deaux, analogous to the existing species ; Cardita margaritacea, Sow. ; Isocardia sulcata, Sow. ; Area duplicata, Sow. ; A. Bran- deri, Sow. ; A. appendiculata, Sow. ; Pectunculus decussatus, * Conybcarc and Phillips's Outlines of the Geology of England and Wales; art, London Clay. 262 Supracretaceous Group. Sow.; P.costatus, Sow.; P.scalaris, Sow.; P. brevirostris, Sow.; P. pulvinatus, Lam., cal. gros., Paris, Bourdeaux, Turin, Traunstein ; Nucula similis, Sow. ; N. trigona, Sow. ; N. mi- nima, Sow. ; N. inflata, Sow. ; N. amygdaloides, Sow. ; Axinus angulatus, Sow. ; Chama squamosa, Sow.; Pinna affinis, Sow.; P. arcuata, Sow. ; Avicula media, Sow. ; Pecten curneus, Sow.; P. carinatus, Sow. ; P. duplicatus, Sow. ; Ostrea gigantea, Sow., Traunstein; O. flabellula, Lam., cal. gros., Paris, Bourdeaux; O. dorsata, Sow. ; O. cymbula, Lam., cal. gros., Paris, Bour- deaux ; O. oblonga, Bt -under ; Lingula tenuis, Sow. MOL- LUSCA Patella striata, Sow. ; Calyptraea trochiformis, Lam., cal. gros., Paris; Infundibulum obliquum, Sow. -, L tubercula- tum, Sow. ; I. spinulosum, Sow. ; Bulla constricta, Sow. ; B. elliptica, Sow. / B. attenuata, Sow. ; B. filosa, Sow. ; B. acu- minata, Sow. ; Auricula turgida, Sow.; Au. simulata, Sow.; Melania sulcata, Sow. ; M. costata, Sow. (Qu. M. costellata, Brander and Lam., cal. gros., Paris?); M. minima, Sow. ; M. truncata, Sow.; Paludina lenta, Sow. ; P. concinna, Sow.; Am- pullaria ambulacrum, Sow.; Am. acuta, Lam., cal. gros., Paris; Am. patula, Lam., cal. gros., Paris ; Am. sigaretina, Lam., cal. gros., Paris; Neritina concava, Sow. ; Nerita globosa, Sow. ; N. aperta, Sow. ; Natica Hantoniensis ; N. similis, Sow.; N. glauci- noides,S Y 0w.,; N. striata, Sow.; Sigaretus canaliculatus, Sow., cal. gros., Paris, Bourdeaux ; Acteon crenatus, Sow. ; A.elongatus, Sow. ; Scalaria acuta, Sow.; S. semicostata, Sow.; S.interrupta, Sow. ; S. undosa, Sow. ; S. reticulata, Sow. ; Solarium patulum, Lam., cal. gros., Paris, Bourdeaux ; Sol. discoideum, Sow. ; Sol. canaliculatum, Sow. ; Sol. plicatum, Lam., cal. gros., Paris; Trochus Benettiae, Sow. ; Piacenza, Turin, Bourdeaux; T. extensus, Sow. ; T. monilifer, Lam., cal. gros., Paris ; Tur- ritella conoidea, Sow. *; Tur.elongata, Sow.; Tur. brevis, Sow.; Tur. edita, Sow. ; Tur. multisulcata, Lam., cal. gros., Paris ; Cerithium dubium, Sow. ; C. Cornucopia?, Sow. ; C. gigan- teum, Lam., cal. gros., Paris ; C. pyramidale, Sow. ; C. gemi- natum, Sow.; C. funatum, Sow.']:; Pleurotoma attenuata, Sow.; P. comma. Sow. ; P. semicolon, Sow. ; P. colon, Sow. ; P. ex- erta, Sow. , P. rostrata, Sow. ; P. acuminata, Sow. ; P. fusifor- mis, Sow. ; P. laevigata, Sow. ; P. brevirostra, Sow. ; P. prisca, Sow. ; Cancellaria quadrata, Sow. ; C. laeviuscula, Sow. ; C. evulsa, Sow. ; Fusus deformis, Konig ; F. longasvus, Lam., cal. gros., Paris ; Fusus rogosus, Lam., cal. gros., Paris, Bour- deaux ; F. acuminatus, Sow. ; F. asper, Sow. ; F. bulbiformis, * According to M. Deshayes, Turritclla conoidea, T. dongata and T. edita, of Sowerby, are the same shells, referable to T. imbricafaria of Lamarck. f It is remarkable that, out of the numerous species of Cerithium found in the calcaire grossier of Paris, the C. gigantcum should be the only one yet noticed in the London clay. Supracretaceous Group. 263 Lam. (4* var.), cal. gros., Paris ; F. ficulneus, Sow. ; F. errans, Sow. ; F. regularis, Sow. / F. Lima, Sow.; F. carinella, Sow.; F. conifer, Sow. ; F. bifasciatus, Sow. ; F. complanatus, Sow.; Pyrula nexilis, Sow. ; P. Greenwoodii, Sow. ; P. laevigata, Lam., cal. gros., Paris, Traunstein ; MurexBartonensis, Sow.; M. fistulosus, Sow. ; M. interruptus, Sow. ; M. argutus, Sow. ; M. tricarinatus, Lam., cal. gros., Paris, Vicentin ; M. bispi- nosus, Sow. ; M. frondosus, Lam,, cal. gros., Paris ; M. de- fossus, Sow.; M. Smithii, Sow. (2 var.); M. trilineatus Sow. ; M. curtus, Sow. ; M. tuberosus, Sow. ; M. minax, Sow. ; Switzerland ; M. cristatus, Sow. ; M. coronatus, Sow. ; Rostel- laria Parkinsoni, Sow. (var.) ; R. lucida, Sow. ; R. rimosa, Sow. ; R. macroptera, Sow. (2 var.); R. Pes-Pelicani (Strom- bus Pes-Pelicani, Linn.), Piacenza, &c., analogous to the ex- isting species; Cassis striata, Sow.; C. carinata, Lam., cal. gros., Paris; Harpa Trimmeri, Parkinson; Buccinum jun- ceum, Sow. ; B. lavatum, Sow. / B. desertum, Sow. ; B. canali- culatum, Sow. ; B. labiatum, Sow. ; Mitra scabra, Sow. ; M. parva, Sow. ; M. pumila, Sow. ; Voluta Luctator, Sow. ; V. spinosa, Lam., cal. gros., Paris ; V. suspensa, Sow. ; V. mon- strosa, Sow. ; V. costata, Sow. ; V. Magorum, Sow. ; V. Ath- leta, Sow. ; V. depauperata, Sow. ; V. ambigua, Sow. ; V. no- dosa, Sow.; V. Lima, Sow.; V. geminata, Sow.; V. bicorona, Lam., cal. gros., Paris ; Volvaria acutiuscula, Sow. ; Cypraea oviformis, Sow. ; Terebellum fusiforme, Sow. ; T. convolutum, Al. Brong., cal. gros., Paris ; Ancellaria canalifera, Lam., cal. gros., Paris, Bourdeaux ; A. aveniformis, Sow. ; A. Turritella, Sow. ; A. subulata, Sow. ; Oliva Branderi, Sow. ; O, Salisbu- riana, Sow.; Conus Dormitor, Sow. ; C. concinnus (2 var.), Sow. ; C. scabriusculus (2 var.), Sow. ; C. lineatus, Brander ; Nummulites l^vigata, Lam., cal. gros., Paris, Bordeaux, Traunstein ; Num variolaria, Sow.; Num. elegans, Sow. ; Nau- tilus imperialis Sow., cal. gros., Paris; N. centralis, Sow. ; N. ziczac, Sow. ; N. regalis, Sow.* Vegetable Remains. The Isle of Sheppy has long been known as affording a great variety of fruits and seeds ; and small portions and masses of wood are found in the London clay elsewhere, the argillo-calcareous concretions frequently enveloping pieces of it. Some fragments are pierced by a boring shell analogous to the Teredo navalis, which shows that the wood must have floated in the sea-j-. Bagshot Sands. These rest on the London clay, and con- sist, according to Mr. Warburton, of ochreous meagre sand, * Sowerby's Mineral Conchology ; Woodward's British Organic Remains ; Al. Brongniart, Tableau des Terrains qui eomposent 1'Ecorce du Globe. f Outlines of Geol. of Engl. and Wales. 264; Supracrctaceous Group. foliated green clay alternating with a green sand, and alterna- tions of white, sulphur-yellow, and pinkish foliated marls, con- taining abundant grains of green sand, and fossil shells of the genera Trochus ? Crassatella, Pecten *. Fresh-water Formations, Isle of Wight and Hampshire. We are indebted to Mr. Webster for the discovery of these beds, not long after the labours of MM. Cuvier and Brongniart on. the supracretaceous rocks round Paris so strongly excited the attention of geologists. The fresh-water strata of the Isle of Wight are divided into two deposits by a rock characterized by the presence of marine remains, and named the Upper Marine Formation, from being a supposed equivalent to the sands which intervene between the two fresh-water deposits of Paris. The lower fresh-water deposit of Binstead, near Ryde, consists of a limestone formed of fragments of fresh- water shells, white shell marl, siliceous limestone and sand ; at Headen the equivalent rock is composed of sandy, calcareous, and argillaceous marls. According to Mr. Pratt, one tooth of an Anoplotherium and two teeth of a Pal&otherium have been discovered in the lower and marly beds of the Binstead quarries ; and he further states, that these remains were " ac- companied, not only by several other fragments of bones of Pachydermata (chiefly in a rolled and injured state), but also by the jaw of a new species of Ruminant, apparently closely allied to the genus Moschus^" Prof. Sedgwick observes, that in the upper part of this de- posit there is a mixture of fresh-water and marine species, es- pecially in Colwell Bay, where a single specimen of rock con- tained the following genera : Ostrea, Venus, Cerithinm, Planor- bis, Lymnaa. The common fossils in the lower fresh-water deposit would appear to be: Paludina, Potamides, Melania, (more than one species), Cyclas (2 species), Unio, Planorbis, Lipnn&a (both tiie last more than one species), Mya, Mela- The Upper Marine Formation, first noticed by Mr. Webster, was called in question by Mr. G. B. Sowerby, who showed that all the shells detected in it were not marine ; and he hence inferred that there was no real separation between the fresh- water formations of the Isle of Wight . Subsequently to Mr. Sowerby's remarks, Prof. Sedgwick has presented us with an account of these strata, in which he remarks that " the lower calcareous beds appear to have been tranquilly deposited in fresh water. But if we ascend to the argillaceous marl which * Warburton, Geol. Trans., vol. i. 2nd series. f Pratt, Proceedings of the Geol. Soc. 1831. + Sedgwick, On the Geology of the Isle of Wight; Annals of Philos. 1822. G. B. Sowerby, Annals of Philos, 1821. Supracretaceous Group. 265 rests immediately upon them, we not only find a complete change in the physical circumstances of the deposit, but a new suite of organic remains ; some of which are of a marine origin, others of a doubtful character, and a few are identical with those in the lower beds*." With regard to the organic re- mains contained in this rock, Mr. Webster points out a thick oyster-bed in Colwell Bay; and Prof. Sedgwick gives the following list of shells: Murex (at least two species), Bucci- num, Ancilla subulata, Valuta (resembling V. spinosa\ Mosfel- laria rimosa (two last species rare), Murex effbssus, BRANDER, M. innexus, BRANDER, Fusus (fragments), Natica, Venus, Nucula, Corbula, Corbis? Mytilus, Cyclas, Potamides, Mela- nopsiS) Nerita (2 species, one approaching N. Jluviatilis), to- gether with other fresh-water shells. These beds would there- fore appear to have been deposited, as Prof. Sedgwick ob- serves, in an estuary. But to have produced this estuary, and the circumstances requisite for the presence of marine shells, some physical change, some alteration of the relative levels or of the geographical features of the sea and land, seems necessary, for the previous deposit does not contain marine remains. Upper Fresh-water Formation- This, according to Mr. Webster, principally consists of yellowish white marls, in which there are more indurated, and apparently more calca- reous portions. The organic remains are either fresh-water or terrestrial ; and therefore the circumstances, whatever they were, which permitted a mixture of marine shells in the beds beneath, no longer existed ; and a tranquil deposit in some lake was, probably, the mode in which these beds, about 100 feet thick, were formed. The fresh-water formation of Hordwell Cliff, Hampshire, was first described by Mr. Webster in 1821. The cliff is noticed as composed of alternations of clays and marls, some of a fine blueish green colour, in which there were also beds of hard calcareous marls, apparently derived from shells of the genera Lymncca and Planorbis. The whole is surmounted by a mass of transported gravel, which covers the various rocks of the vicinity. Mr. Webster observed that these beds seemed the equivalent of the lower fresh-water deposit of the Isle of Wight. Subse- quently to these observations of Mr. Webster, Mr. Lyell pub- lished a more detailed account of the Hordwell beds ; whence it would appear that the upper strata do not show a passage into a marine deposit, as was first supposed, but that all the fossil contents of the beds point to a fresh-water origin, equi- valent to the lower fresh-water rocks of the Isle of Wight. The following are the organic remains discovered at Hordwell, * Sedgwick, Annals of Philos. 1822. 266 Supracretaceous Group. according to Mr. Lyell : Tortoise scales, (a Tortoise found at Thorness Bay, Isle of Wight) ; Gyrogonites, or seed-vessels of Chara (C. medicaginula) ; seed-vessel named Carpolithes tha- lictroideS) AD. BRONG.; teeth of crocodile, and scales of fish? Helix lenta, BRANDER, abundant; Mclania conica ; Melanop- sis carinaia ; M. brcvis ; Planorbis lens ; P. rotundatus ; Lym- nccafiisiformis ; L. longiscata ; L. columellaris ; Potamides ; P.margaritaceus? Neritina ; Ancylus elegans ; Unio Solandri ; Myagregarea ; M.plana; M. subangulata, perhaps the young of M. plana; Cyclas (2 species). Mr. Lyell observes, that though the species are few, the individuals are numerous, a common characteristic of fresh-water deposits*. Both in the Isle of Wight and on the opposite coast of Hampshire, these fresh-water deposits rest upon a consider- able thickness of sand. As a similar sand occurs in the fresh- water rocks of Hordwell, Mr. Lyell considers that there is as much probability of its fresh-water, as of its marine origin. Be this as it may, there must have been a difference in the transporting power of water carrying the sands, from that which permitted the deposit of the marls, which seems to have been very quiet. The sands certainly do not require any considerable velocity of water; still there must have been a difference in the circumstances attending the deposit of the one mass and of the other, though those, which give rise to the mass of sand, partially returned during the formation of the marls. A very material difference, it will be observed, must have at- tended the deposit of the supracretaceous rocks in the Parisian and English districts (London and Isle of Wight), as far as respects their mineralogical nature. In the former we have deposits of carbonate of lime (calc. grossier), sulphate of lime, (gypseous deposits), and silica (millstones) ; formations only in part mechanical ; while in the latter we have little that may not be considered altogether mechanical, with the exception, perhaps, of the fresh-water marls and the calcareous concre- tions in the London clay, which latter may have been chemi- cal separations, after deposition, from the argillo-calcareous mass. There is, nevertheless, such an analogy between the organic character of the calcaire grossier of Paris and the London clay, that though not strictly identical, they may have been nearly contemporaneous ; so that however the mineral- ogical character of these deposits may vary, we may suppose them to have been formed at the same or nearly the same epoch, local circumstances and accidents having determined the character of each. * Lycli, Gcol. Trans. 2nd scries, vol. ii. Supracretaceous Group. 267 , Our limits prevent a proper notice of the labours of Pre- vost, Boue, Voltz, Parsch, Lill Von Lillienbach, Pusch *, Du Bois, and many other geologists, on the rocks of this age in various parts of Europe; but the following section seems so important that it requires a place here. Prof. Pusch, describing the rocks of Podolia and southern Russia, states, that near Krzeminiec, in Volhynia, (where mountains rise above a plain covered with chalk flints and sand,) upper supracretaceous sandstone, occupying a thick- ness of 396 feet above the river Ikwa and sixty feet beneath it, is composed of : 1. Twenty feet of sand, cemented by a little carbonate of lime, containing many small shells and ma- drepores, the latter approaching M. cervicornis. 2. Forty feet of calcareous sandstone, containing many shells of the genera Cardium, Venericardia, and Aica. 3. Sixty feet of a compact quartzose and porous sandstone, the cavities filled with sand ; contains many Verier icardia: ; lowest part most cal- careous. 4. Eighty feet of a marly limestone, containing many striated Modiolcc, Pectens, and other shells. 5. At sixty feet beneath the surface, a quartzose and slightly calcareous white sandstone, containing numerous Venericardicc^ Troclii^ and Paludma; or Phasianellcc. " According to M. Jarocki, while sinking a well in June 1829, the tusk and molar tooth of an elephant were found in the last-mentioned bed (No. 5), which are now preserved in the museum of Krzeminiec. Many other bones were also observed, but they were too firmly fixed in the rock to be extracted f." M. Pusch further remarks, that this rock is the same, both mineralogically and zoologi- * Amid a great variety of supracretaceous deposits in Russia and Poland, this author remarks some with an oolitic character, especially near Tiraspol, Latyczew, and Kaluez, on the Dniester, and in the Cecin hills at Czerno- witz. The pisolitic structure of some supracretaceous limestones is particu- larly remarkable in parts of Poland. The grains are either reuiform or rounded, and generally of the size of a pea or a hean, though they here and there become two or three inches in diameter. Good examples of this rock are seen at Rakow. M. Pusch states that repeated observations have con- vinced him that these concretions are derived from corals, especially Niilli- porcB. He observes that the large reniform concretions of Rakow are only the Nullipora byssoides, Lam., or the N. racemosa, Goldf. In some places, particularly at Skotniki, near Busko, a rock of this kind appears as if com- posed of bullets and cannon-balls. It should be stated that Prof. Pusch, from a careful comparison of the shells contained in the supracretaceous limestone of Poland with those figured by various authors, considers that the tertiary shells of Poland bear a much greater resemblance to those found at the foot of the Italian Alps and in the Sub-Apennine hills, than those discovered in England or the North of F.rance ; moreover, that the species which at first sight do appear identical with those of France and Italy, are found to be varieties of them when ex- amined with attention. t Pusch, Journal de Geologic, t. 2. 26 S Supracretaceous Group. cally, as the tertiary sandstone of Szydtow and Chmielnik, in Poland ; and that this fact is analogous to the occurrence of an elephant's molar tooth and tusk in the tertiary sandstone ofRzaka, Wieliczka, which contains Pecten polo?itcus 9 Saxi- cavcc, and many other marine shells. The reader will also observe that it corresponds with the occurrence of the remains of the great Pachydermata, previously noticed as found min- gled with marine exuviae in other parts of Europe. It will have been remarked, that throughout this detail of supracretaceous rocks, (perhaps too long for a work of this nature,) the observations have been confined to certain parts of Europe. Rocks of the same nature no doubt abound in other parts of the world ; indeed we are well assured that very extensive districts are composed of them, as for instance in India; but our knowledge of them is as yet so imperfect, that we cannot with safety compare them with known European deposits. Dr. Buckland, from the information which he ob- tained from Mr. Crawfurd, who collected an abundance of organic remains on the banks of the Irawadi, considered that supracretaceous rocks probably existed in the kingdom of Ava, containing shells of the genera Ancillaria, Murex, Ceriihium, Oliva, Astarte, Nticula, Erycina, Tellina, Teredo ; mixed with sharks' teeth and fish scales : these remains are contained in a coarse shelly and sandy limestone. A great abundance of mammiferous and other remains were discovered in the vici- nity of some petroleum wells, between Prome and Ava, ap- parently mixed with much silicified wood in a sandy and gra- velly deposit. The bones or teeth of vertebrated animals con- sist of those of the Mastodon lattdens, Clift; M. elephantoides^ Clift; Hippopotamus; Sus ; Rhinoceros ; Tapir / Ox ; Deer; Antelope; Trionyx ; Emys ; and Crocodiles (2 species)*. Mr. Scott met with beds, probably of the supracretaceous epoch, in the Caribari hills, left bank of the Brahmputra. The following section (order ascending) was observed: 1. Slate clay. 2. Ferruginous concretions and indurated sand. 3. Yellow or green sand. 4. Slate-clay. 5. Sand and small gravel. Fossil wood is found on the indurated clay; arid in a small isolated hill in the vicinity the following re- mains : Teeth and bones of sharks, fish palates and fin bones, teeth and bones of crocodiles, remains of quadrupeds, Os/raz-, Cerithia, Turritella, Balani, Patella, &c. f. These exuviae have subsequently been examined by Mr. Pentland, who found that the mammiferous remains were referrible to the genus Anthracotherium, Cuv., to a species allied to the genus Mos- * Buckland and Clift, Geol. Trans. 2nd series, vol. ii. f Colebrooke, Geol. Trans. 2nd scries, vol. i. Supracretaceous Group. chus, to a small species of the order Pachydermata^ and to a carnivorous animal of the genus Viverra. The Anthracothe- rium he proposes to name A. Silistrense*. These observations are sufficient to show that rocks, pro- bably supracretaceous, exist extensively in India. Accord- ing to Prof. Vanuxem and Dr. Morton, the supracretaceous or tertiary rocks are extensively distributed over parts of the United States, occurring in Nantucket, Long Island, Manhattan Island, the adjacent coasts of New York and New England; sparingly in New Jersey and Delaware, but ex- tensively in Maryland and to the southward. The deposit is stated to be composed of limestone, buhr-stone, sands, gra- vels, and clays ; and contains the remains of the genera Os- trea, Pecten, Area, Pectunculus, Turritella, Buccinum, Fenus, Mactra, Natica, Tellina, Nucula, Venericardia, Chama, Calyp- trcea, Fusus, Panopcea, Serpula, Dentalium, Cerithium, Car- dium, Crassatella, Oliva, Lucina, Corbula, Pyrula, Crepidu- la, Perna, &c. Of 150 species of these shells, found in a single locality in St. Mary's county, Maryland, Mr. Say has described and figured more than forty as newf. Accord- ing to Dr. Morton, the upper supracretaceous beds of Mary- land and the more southern states contain the following spe- cies of shells, still found in a recent state on the coasts of the United States : Natica duplicata, Say ; Fusus cmereus, Say ; Pyrula carica, Lam. ; P. canaliculata, Lam. ; Ostrea virginica, Linn. ; O.Jtabellula, Lam. ; Plicutula ramosa, Lam. ; Area arata, Say ; Lucina divaricata, Lam. ; Venus mercenaria, Linn. ; V. paphia ? Lam. ; Cytherea concentrica, Lam. ; Mac- tra grandis, Linn. ; Pholas costata. Linn. ; Balanus tintinna- bulum ? Lam. ; Turbo littoreus ? Linn. ; and a Buccinum J. That deposits of a similar age are not wanting in South Ame- rica seems also certain ; but as yet they have not been exa- mined in sufficient detail to enable us to institute any useful comparison with rocks of the same antiquity in Europe. Nei- ther can we, for the same reason, judge of the relative anti- quity of innumerable igneous formations scattered over various parts of the world. As the science of geology advances, great insight must be obtained into the superficial appearance of the world at this period, leading to the most important conclu- sions ; but we must anticipate very serious obstacles to this advancing knowledge, arising from hasty generalizations of local facts, and the too common endeavour to force conclu- * Pentland, Geol. Trans. 2nd series, vol. ii. f Vanuxen and Morton, Journal of the Academy of Natural Sciences of Philadelphia, vol. vi. J Morton, Ibid. 270 Supracretaceous Group. sions, more particularly as to the identity or parallelism of deposits. It is impossible to close this sketch of the supracretaceous rocks without noticing the important observations of Dr. Boue on those of Gallicia, wherein he establishes the fact, that the celebrated salt deposit of Wieliczka constitutes a portion of the supracretaceous series. Dr. Boue describes this deposit as 2560 yards long, 1066 yards broad, and 281 yards deep. The salt is termed green salt in the upper part of the mine, where it occurs in nodules with gypsum in marl. The salt sometimes contains lignite, bituminous wood, sand, and small broken shells. In the lower part the marl becomes more are- naceous, and there are even beds of sandstone in the salt. Beneath this is a gray sandstone, rather coarse, containing lignite, and impressions of plants, with veins and beds of salt. In the lower part of this stratum an indurated calcareous marl is observed, containing sulphur, salt, and gypsum. Beneath this is an aluminous and marno-argillaceous schist. From the fossils and various other circumstances, Dr. Boue concludes that this great salt deposit forms part of a muriatiferous and supracretaceous clay, subordinate to sandstone (molasse). Most frequently the marly clays are merely muriatiferous ; an abundance of salt, such as at Wieliczka, Bochnia, Parayd in Transylvania, and other places, being more rare *. Volcanic Action during the Supracretaceous Period. We have already seen that there was much difficulty in stating at what periods certain products of extinct volcanos had been thrown out. This difficulty is by no means lessened as we descend in the series ; for the seat of volcanic action seems to have continued nearly, or very nearly, in the same place for long periods ; and the mere circumstance of the interstratifi- cation of volcanic matter with aqueous rocks, whose relative age may to a certain extent be known, will not always give that of the igneous rocks so circumstanced, because we cannot be sure that they have not been injected among the aqueous deposits; and when this may have happened it would be diffi- cult to say. Thus Etna would appear to have been the seat of volcanic action through a long series of ages, commencing with the supracretaceous rocks, on which much of the igneous mass is now based. In Central France, amid the extinct volcanos which there constitute such a remarkable feature in the physical geogra- phy of the country, we certainly approach relative dates in some instances. Thus the volcanic mass of the Plomb du Bou6, Journal de Geologic, t. i. 1830. Supracretaceous Group. 271 Cantal appears to have burst through, to have upset, and to have fractured the fresh-water limestones of the Cantal, which, according to Messrs. Lyell and Murchison, may be equiva- lent to the fresh-water deposits of the Paris basin, and to those of Hampshire and the Isle of Wight. The following is a list of organic remains obtained by them in the fresh- water rocks of the Cantal : The rib of an animal resembling that of an Anc- plotherium or a Paltzotherium ; scales of a tortoise; fish teeth ; Potamides Lamarckii ; Limncea acuminata ; L. columellaris ; L.fusiformis ; L. longiscata ; L. injlata ; L. cornea ; L. Fa- bulum ? L. strigosa ? L. palustris antiqua ; Bulimus Terebra ; B. pygmeus ? B. conicus ; Planorbis rotundatus ; P. Cornu ; P. rotundus ; Ancylus elegans. Plants : Chara medicaginula, the seeds (gyrogonites), and stems; carbonized wood. It is remarked, that out of this short list there are eight or nine species identical with those found in the upper fresh-water rocks, and five or six with those in the lower fresh-water de- posits of the Paris basin *. Here we seem to obtain a rela- tive date for the upburst of the igneous products of the Plomb du Cantal ; one posterior to the deposit of the fresh-water rocks of Paris and the Isle of Wight. With regard to the relative date of the igneous rocks of Auvergne, it would appear from the labours of MM. Croizet and Jobert, that the Montagne de Perrier, N.W. from the town of Issoire (Puy de Dome), is divided into two stages or terraces, the first about twenty-five yards above the valley of the Allier, the second occupying a height of about 200 yards. The mountain may be considered as based on granite, above which there is a considerable thickness of fresh-water lime- stone, surmounted by numerous beds of rolled pebbles and sand, of which one in particular is remarkable for the abun- dant remains of mammalia found in it ; the whole crowned by a mass of volcanic matter. MM. Croizet and Jobert consider that in this locality and in the neighbouring country there are about thirty beds above the fresh-water limestone, which may be divided into four al- ternations of alluvial detritus and basaltic deposits. Among the beds there are four which contain organic remains : three belonging to the third of the ancient alluvions, that which suc- ceeded the second epoch of volcanic eruptions ; the third fos- siliferous deposit being referrible to the last epoch of ancient alluvion. The whole of these beds are not seen in the Mon- tagne de Perrier, but are determined from the general struc- ture of the country. * Lyell and Murchison, Sur Ics Depicts Lacustres Tertiaires du Cantal, &c. Ann. des Sci. Nat, 1829. 272 Supracretaceous Group. The principal ossiferous bed is about nine or ten feet thick, and can be traced a considerable distance at the foot of the Montagne de Perrier, and in the Vallee de la Couse on the opposite side. The fossil species, according to MM. Croizet and Jobert, are very numerous, consisting of : Elephant, one species; Mastodon, one or two ; Hippopotamus, one; Rhi- noceros, one ; Tapir, one ; Horse, one ; Boar, one ; Felis, four or five ; Hyrena, two ; Bear, three ; Canis, one ; Cas- tor, one; Otter, one; Hare, one; Water-Rat, one; Deer, fifteen ; and Ox, two. The animals were of all ages, and the various remains mixed pell-mell with each other. The bones are never rolled, though often broken, and sometimes gnawed. Mingled with these exuviae are the abundant fae- cal remains of the Carnivora, appearing to occupy the place where they have been dropped. Hence the authors con- clude that the remains have not been far removed from the places where the animals existed, and that the lignites found among these beds are the exuviae of the vegetation upon which many of them subsisted. MM. Croizet and Jobert notice the following remains in the fresh-water sands, clays, and limestone of the country, over which they consider that the first basaltic currents flowed : Anoplotherium ? two species ; Lophiodon, one ; Anthracothe- rium, one; Hippopotamus, one; a Ruminant; Canis, one; Marten, one; Lagomys, one; a Rat; Tortoise, one or two; Crocodile, one ; Serpent or Lizard, one ; Birds, three or four (among the latter remains are their eggs, perfectly preserved) ; Cypris faba , Helix ; Lymneea ; Planorbis ; Cyrena ; Gyro- gonites, and other vegetable exuviae. It should be observed that M. Bertrand- Roux* had some time previously observed the remains of a Palteotherium in a similar rock in the Puy en Velay, and that the fresh-water rocks at Volvic contain birds' bones f. M. Bertrand de Doue describes the occurrence of bones entombed in and beneath volcanic matter near St. Privat- d'Allier (Velay). After stating that the discovery was due to Dr. Hibbert, who communicated it to him, and that he pro- ceeded to the spot pointed out, accompanied by M. Deribier, he notices the following descending section : a, third and last flow of basaltic lava ; b, second flow, four yards thick ; c, gray- ish volcanic cinders, two to four decimetres thick ; d 9 agglu- tinated scoriag and tuff, one or more yards thick, in the upper part of which the bones were discovered ; e, oldest plateau of * Now M. Bertrand de Doue. f Croizet and Jobert, Recherches sur les Oss. Foss. du Depart, du Puy de Dome ; and Ann. des Sci. Nat. t. xv. 1828. Supracretaccous Group. 273 basaltic lava ; f, gneiss. The osseous remains were those of the Rhinoceros leptorhinus, Hyaena spelcea, and a large pro- portion of bones, referrible to at least four undetermined spe- cies of Cervi. The same author considers, from the fractured character and irregular distribution of the bones over a horizontal and limited space, that this place was the retreat of hyaenas, afford- ing them, from the nature of the country, the best shelter they could find. Into this it is considered they dragged their prey, as appears to have been done in the case of Kirkdale. It is observed that the lava-current which passed over the cinders containing these remains has very little altered the bones. M. Bertrand de Doue does not consider the detrital deposits of the country as produced by transport in a body of waters from a distance, but by a succession of local causes, the sub- stances being all derived from the vicinity. He supposes the distribution of the lateral valleys connected with the Allier, (among which is that where the bones were discovered,) the same now as when the neighbouring volcanos were in activity; and remarks on the " incertitude in establishing the chrono- logical relations between the epoch when the volcanos of the Velay became extinct, and that in which these animals dis- appeared from our climates*." M. Robert, describing the position in which numerous bones have been discovered at Cussac (Haute Loire), men- tions that marls, without fossils, rest on the granitic rocks o the country. At Solilhac these marls are surmounted by clayey marls about two or three feet thick, containing plates of mica, grains of quartz, volcanic ashes, basaltic gravel, and impressions of gramineous plants ; they also contain the entire skeletons of unknown Deer and Aurochs, with other bones. Above these are beds of volcanic sand two or three yards thick, with small basaltic and granitic pebbles, containing the remains of Ruminants and Pacliydermata^ the bones being more or less broken. On these rest alluvions of greater soli- dity, composed of the same volcanic sand, large granitic and basaltic blocks (of which the angles are not rounded), geodes of hydrate of iron, and bones, which appear to have been ex- posed to the air before they were enveloped. All these sub- stances are cemented by oxide of iron, and beds of ferrugi- nous sands either alternate with, or repose on, the alluvions. M. Robert extracted from these ferruginous beds at Cussac the remains of the Elephas primigenius ; the Rhinoceros lep- torhinus ; the Tapir Arvernensis ; the Horse, two species; .* Bertvand de Doue, Edin. Journal of Sci. vol. ii. new series, 1830. T -74- Supracretaceous Group. Deer, seven species (to two of which he assigns the names of Cervus Solilhacus, and C. Dama Polignacus)^ the Bos Urus and Bos Velaunus., and the Antelope. The same author re- fers the entombment of these remains to a more ancient date than the accumulation of bones at St. Prevatand Perrier, con- sidering it due to some particular cataclysm, which surprised the animals : thus explaining the occurrence of entire skeletons of young and old individuals found mingled at Solilhac ; a state of things differing from the accumulations at St. Prevat and Perrier, where the bones seem to have been dragged into their present position by carnivorous animals, whose bones are also mixed with those of their prey*. Dr. Hibbert considers that the lowest supracretaceous rocks of the Velay were deposited in fresh-water lakes, entombing the remains of the Palaeotherium and Anthracotlierium, of ter- restrial and fresh-water shells, and of the vegetation which then existed ; such deposit being of long continuance, as shown by its depth, which amounts to 4-50 feet. This deposit ceased, and the land became covered with forests and animals ; the forests being of a marshy growth. The common degradation of land taking place, parts of this vegetation were variously entombed, as were also the remains of animals which then existed ; such as various species of Cervi, some of large size, animals of the Bos kind, the Rhinoceros leptorhinus^ and the Hycena spelcea. Volcanic explosions now took place through various vents, ejecting trachyte and basalt, the latter predo- minating, piercing the fresh-water deposit in some places, and covering it with lavas in others. Notwithstanding these con- vulsions, vegetation still flourished in certain situations, and became entombed amid volcanic products, as is seen at Collet, Ronzal, and other places, where vegetable matter contained in black carboniferous clays, " accompanied with ferruginous sands, alternate with rolled masses of trachyte, phonolite, basalt, or volcanic cinders." During the progress of these eruptions, the water-courses became much deranged, lava- currents crossing these channels, damming up the passage, and forming lakes, in which various singular compounds and rock-mixtures were produced. It would appear from the large size and rounded angles of many of the fragments of basalt, that great currents of water had acted upon them in certain situations. After a time this great confusion seems to have ceased, and the large fragments became covered by a deposit of sand and clay, formed into regular strata, as may be observed near Cussac. During this state of things near Cussac, animals of the Bos kind, and gigantic stags, became * Robert,' Ferussae's Bulletin de Sci. Nat. et de Geologic, Oct. 1830. Supracretaceous Group. 275 entombed. After this, the district seems to have become the haunt of hyaenas, which, issuing from their dens in search of food, dragged their prey into their retreats*, in the manner of the Kirkdale hyaenas. In these various localities in central France, the evidence seems generally in favour of the great outburst of volcanos after the deposit of very extensive fresh-water rocks, the vol- canic action continuing more or less from that period up to a comparatively recent date. Quitting central France and proceeding either in the di- rection of Aix or Montpellier, we find remains of volcanos, which probably were more or less contemporaneous with those of Auvergne. Beaulieu near Aix has been known since the time of De Saussure. Spain, Italy, and Germany, present us with various igneous rocks, which appear referrible to the epoch in which the su- pracretaceous rocks were in the co'urse of formation. As yet, the volcanic rocks of Spain are little known; but those of Germany and Italy, and especially those of the latter, have long engaged the attention of geologists. The Euganean Hills, south of Padua, present a mass of trachytic and other volcanic products, which belong to the supracretaceous epoch ; as they rest in certain situations on scaglia, the equivalent of chalk. Dr. Daubeny mentions that the trachyte is associated with basalt at Monte Venda. The same author informs us, that at the hill of Belmonte in the Vicentine, a rivulet section exposes five basaltic dykes, which from their mode of occurrence might be mistaken for an in- terstratification of chalk and basalt. " Dykes of basalt are also frequently seen traversing this formation at Chiampo, Valdagno, and Magre, but without altering the adjacent rockf." An extensive formation of porphyritic augite rock covers the whole district, resting in some places on chalk, in others on older rocks, filling up the preexisting inequalities in each ; the upper part is amygdaloidal : this is surmounted by various alternations of calcareous beds, with others com- posed of fragments, basalts, volcanic sand, and scoriform lava; the aggregate or mixture of volcanic substances containing fossil remains, as well as the calcareous deposits, and being often as fully charged with them J. The long celebrated fossil fish from Monte Bolca are derived from the calcareous beds of this deposit. At Ronca there are six alternations of volcanic substances with the calcareous beds, the lowest volcanic pro- duct being a cellular basalt. * Hibbert, On the Fossil Remains of the Velay; Edin. Journ. of Sci. vol. iii. 1830. f Daubeny, Description of Volcanos. J Li-id. T 2 276 Supracretaceous Group. M. Al. Brongniart presents us with the following list of the shells and zoophytes in these beds of the Vicentine, the loca- lity of each being marked (R. for Ronca; C. G. Castel-Gom- berto ; V. S. Val-Sangonini ; M. M. Monteccio-Maggiore ;) : Nummulites nummiformis, Defr., R. ; Eulla Fortisii, Al. Brong., R. ; Helix damnata, Al. Brong., R. ; Turbo Scobina, Al. Brong., C. G. ; T. Asmodei, Al. Brong., R, ; Monodonta Cerberi, Al. Brong., V. S. ; Turritella incisa, Al. Brong., R. ; T. asperula, Al. Brong., R. ; T. Archimedis, Al. Brong., R. ; T. imbricataria, Lam., R. ; Trochus cumulans, Al. Brong., C. G. ; T. Lucasianus, Al. Brong., C. G. : Solarium umbro- sum, Al. Brong., R. ; Ampullaria Vulcani, Al. Brong., R. ; A.perusta, Defr., R.; A. obesa, Al. Brong., M. M. and C. G.; A. depressa, Lam., R.; A. spirata, Lam., V. S. ; A. cock- learia, Al. Brong., C. G. ; Melnnia costellata, Lam., (var. roncana, Al. Brong.,) R., and V. S. ; M. elongate Al. Brong., C. G. ; M. Stygii, Al. Brong., R. ; Nerita conoidea, Lam., R. ; N. Acherontis, Al. Brong., R. ; N. Caroms, C. G. ; Natica cepacea, Lam., Val de Chiampo ; N. epigiottina, Lam., R. ; Conus deperditus, Broc. (var. roncamts, Al. Brong.), R. : C. al- siosus, Al. Brong., R. ; Cyprtea Amygdalum, Broc., R. ; Cyp. iriflata, Lam., R. ; Terebellum obvolutum, AL Brong. ; Voluta subspinosa, Al. Brong., R. ; F. crenulata, Lam., V. S. ; V. af- Jinis, Broc., R. ; Marginella Phaseolus, Al. Brong., R. ; M. eburnea, Lam., R., and V. S. ; Nassa Caronis, Al. Brong., R. ; Cassis striata^ Sow., R.; C. Thesei, Al. Brong., R. ; C. JEnece, Al. Brong., R.; Murex angulosus, Broc., various parts of the Vicentine; M. tricarinatus^ Lam., Vicentine; Terebra Fulcani, Al. Brong., R. ; Cerithium sulcatum. Lam. (v&r.roncanum, Al. Brong.), R. ; C. multisulcatum, Al. Brong., R. ; C. undosum, Al. Brong., R. ; C. combustum, Defr., R. ; C. calcaratum, Al. Brong., R. ; C, bicalcaratum, Al. Brong., R. &c. ; C. Castellini, Al. Brong., R. ; C. Maraschini, Al. Brong., R. ; C. corrugalum, Al. Brong., R. ; C. saccatum, Defr., R.; C. ampullosum, Al. Brong., C. G. ; C. plicatum, Lam., R. ; C. lem?iiscatum, Al. Brong., R. ; C. Stropus, Al. Brong., C. G. ; Fusus intortm^ Lam. (var. ronca?ms 9 Al. Brong.), R. ; F. Note, Lam., R. ; F. subcarinatus, Lam. (var.) R. ; F. polygonus, Lam., R. ; F. polygoiiatus, Al. Brong., R. ; Pleurotoma clavicularis, Lam., M. M.; Pteroceras Radix, Al. Brong., C. G. ; Strombus Fortisii, Al. Brong., R. ; Rostel- laria corvina, Al. Brong., R. ; Ros. Pes-carbonis, Al. Brong., R.; Hipponyx Cornucopia, Defr., R.; Chamacalcarata,L,am., C. G. ; Spondyluscisalpinus, Al. Brong., C. G. ; Ostrea, R. ; Pecten lepidolaris? Lam., R. ; P.plebeius? Lam., R. ; Area Pandoras, Al. Brong., C. G.; Mytiluscotrugatus, Al. Brong., R. ; M. edulis? Liun., R. ; M. Antiquorum, Sow., R. ; Lu- Supracretaceous Group. 277 cina Scopulorum, Al. Brong., R. ; L. gibbosula, Lam., R. ; Cardita Arduini, Al. Brong., C. G.; Cardium asperulum, Lam., C. G. ; Corbis Aglaunz, Al. Brong., C. G. ; Cor. la- mellosa, Lam., R.; Venus? Proserpina, AL Brong., R. ; V.? Maura, Al. Brong., R. ; Venericardia imbricata, Lam., C. G.; Fen. Laura, Al. Brong., C. G.; Mactra? erebea, Al. Brong., R. ; M. ? Sirena> Al. Brong., R.; Cypricardia cycloptea, Al. Brong., R. ; Psammobia pudica, Al. Brong., V. S. ; Cassidu- lus testudinarius, Al. Brong., R. ; Nucleolites Ovulum? Lam., R. ; Astreafunesta, Al. Brong., R.; Turbinolia appendiculata, Al. Brong., R. ; T. sifiuosa, Al. Brong., Vicentine *. It has been concluded, and with great probability, that these rocks were produced by the alternate eruptions of vol- canos in the vicinity, and the deposit of calcareous matter in shallow seas. M. Brongniart mentions that parasitical shells and certain corals are seen adhering to fragments of igneous rocks, which shows that these rocks have had abundant time to cool and form the bottom of the sea previous to the deposits above them. And as in some places igneous products and calcareous deposits often alternate, we may infer that a long period elapsed during the formation of the whole. On the north and south of Rome there is abundant proof of extinct volcanic action. At Viterbo basaltic rocks rest on a compound of pumice and volcanic tuff, in which the bones of mammalia have been discovered ; reminding us of Auvergne. Rome itself is founded on rocks of volcanic origin, mixed with others which are aqueous, and mostly of contemporaneous formation. Proceeding hence to Sicily, we find it very diffi- cult to conceive when the volcanic action commenced which now finds a vent at Etna; as volcanic products are found mixed with supracretaceous rocks. Dr. Daubeny observes, that the supracretaceous blue marl which occupies a consider- able portion of Sicily, contains sulphur, various sulphuric salts, and muriate of soda ; all substances sublimed from mo- dern volcanos, and which may have been produced by exhala- tions from beneath. Among the variety of volcanic products in the vicinity of the Rhine and neighbouring parts of Germany, are many which seem clearly to belong to the supracretaceous epoch. Among these may be mentioned the Siebengebirge, the West- erwald, the Habichtswald near Cassel, and the Meisner near Eschwege. The Siebengebirge are composed of trachyte, basalt, and volcanic conglomerates, traversed by dykes. The Wester wald is composed of the like substances. Basaltic knolls are scattered over the country between the Westerwald * Brongniart, Terrains Calcareo-Trappeens du Vicentin, 1823. 278 Supracretaceous Group. and the Vogelsgebirge. The Kaiserstuhl and the igneous rocks on the north of the lake of Constance would appear to be examples of volcanic rocks which may have been ejected at the supracretaceous epoch. According to M. Beudant there are five principal volcanic groups in Hungary, referrible to the age with which we are now occupied: 1. That in the district of Schemnitz and Kremnitz. 2. That constituting Dregeley mountains, near Gran on the Danube. 3. That of the Matra, in the centre of Hungary. 4. The chain commencing at Tokai, and extend- ing north about twenty-five leagues. 5. That of Vihorlet, con- nected with the volcanic mountains of Marmorosch (borders of Transylvania). The whole composed of different varieties of trachytic rocks. According to Dr. Boue, volcanic rocks of undoubted su- pracretaceous origin occur in Transylvania. They constitute a range of hills separating Transylvania from Szecklerland, and extending from the hill of Kelemany, north of Remebyel, to the hill Budoshegy, on the north of Vascharhely. They are principally composed of varieties of trachyte, and trachytic conglomerate*. From the observations of Von Buch and Dr. Daubeny, it appears that Gleichenburg, not far from Gratz, Styria, is com- posed of trachyte, round which are mantle-shaped strata of volcanic products and supracretaceous beds, alternating with each other. If we turn from these igneous products on the continent of Europe to our own islands, we find that great igneous eruptions have taken place in the north-eastern parts of Ireland, after the deposit of the chalk, and consequently in the supracreta- ceous period. The basaltic ranges of the celebrated Giant's Causeway, Fairhead, &c. belong to this eruption, which in its upburst has torn and rent all which it encountered, entangling enormous masses of chalk, as may be seen at Kenbaan. We find the mass of this erupted igneous rock to be basaltic, sometimes columnar, at others not; the two varieties being so arranged on the coast between Dunseverie Castle and the Giant's Causeway, that they have the appearance of being in- terstratified. At Murloch Bay, Fairhead, and Cross Hill, the basalt rests on coal measures; at Knocklead and other places, on chalk t. As an intermixture with supracretaceous rocks has not yet been observed, the relative date of this erup- tion cannot be well determined. Both the basaltic mass and the rocks on which it rests have been traversed, at a period * Daubeny's Volcanos. f Buckland and Conybearc, Geol. Trans, vol. iii. ; and Sections and Views illustrative of Geological Phenomena, pi. 11). c ac a c a Siipracrelaceous Group. 279 posterior to the first overflow of the former, by dykes of igneous matter : one of these has pro- duced a singular change in the chalk, ?" . ' i . . 1*1 * ~- nnTn ~- -f -"uittt which it cuts, together with superin- cumbent basalt, in the Isle of Raghlin, as will be best explained by the annexed section. a a a, trap dykes cutting through chalk b b, which it has converted into granular lime- stone c c c c. It now only remains to consider those recent observations on the Alps, Pyrenees, and the vicinity of Maestricht, which seem to point to at least a zoological passage of this group into the next ; appearing to show, that from the progress of science, the clear line of separation once supposed to exist between the secondary and tertiary classes, as they are termed, cannot be drawn, but that the zoological character of the upper part of the one and the lower portion of the other would ap- proach each other, as indeed might be expected ; for we can- not conceive a natural destruction of life so general as to cause the complete annihilation of animals, particularly those which are marine, existing at any given time, so that a totally new creation should be necessary. Such a supposition would not appear to accord with what is observable in other rocks, as will be noticed in the sequel. It is not contended that there may not be great specific distinctions in the remains entombed in this and the next group in many parts of Europe, but merely that it does not necessarily follow, because Europe may pre- sent us with two classes of rocks, one of which may be named tertiary and the other secondary, from the general nature of their organic contents, that in many parts of the world the whole may not constitute a series in which lines of distinction cannot be drawn. Suppose some violent cause should pro- duce a great debacle which should rush over Europe, the land and fresh-water animals and plants would probably be destroyed ; and we will even consider, for the sake of the argument, that the marine inhabitants of our seas perished also, does it necessarily follow that the marine, fresh-water, and terrestrial inhabitants would also be annihilated in Aus- tralia ? Should we not rather consider that these would be entombed, if rocks were there forming, as well after and during the destruction of European life, as previous to it ? and that the rocks formed in those regions, about this supposed period, would by no means show any alteration in their zoological character ? That very great changes have taken place in the organic character of deposits in the same districts, and that somewhat suddenly, does not admit of a doubt; but it is 280 Snpracrclaccous Group. a subject on which we are, as yet, far from seeing our way clearly. There is always great difficulty in comprehending why the marine remains should be so suddenly changed in certain de- posits, which do not exhibit marks of being the results of violent commotion ; for although we can understand why ter- restrial and fresh-water animals should be destroyed by a change in the relative levels of sea and land, or an inroad of the sea, it is difficult to comprehend why, from these causes alone, the general character of the marine animals should be changed. The depth of a sea may, indeed, be so changed, from a movement in the bottom, that deep water may be ren- dered shallow, and shallow water deep; but it does not seem to follow that the inhabitants should necessarily all perish, the species being so completely destroyed as never to be found afterwards. On the contrary, we should be led to infer that, though such changes would by no means suit the habits of the respective inhabitants of such waters, they would merely induce them to seek situations better calculated for their various modes of live. From an examination of portions of the Austrian and Ba- varian Alps, in 1829, Professor Sedgwick and Mr. Murchison concluded that they had discovered a series of beds interme- diate between the chalk and commonly known supracretaceous rocks, affording as it were a passage of the, so called, tertiary class into the secondary ; yet, as they were above the true chalk, being considered as tertiary. The correctness of this deter- mination is questioned, more particularly by Dr. Boue, who contends that the disputed rocks belong to the cretaceous se- ries According to the former authors, the valley of Gosau, in the Salzburg Alps, presents a good example of the correct- ness of their views. This valley is described as about 2600 feet above the level of the sea, exhibiting these newer strata brought suddenly into contact with more ancient rocks on one side. The following is stated to be a section of them, in the descending order. "1. Red and green slaty micaceous sand- stone, several hundred feet thick (cap of the Horn). 2. Green micaceous gritty sandstone, extensively quarried as whetstone, succeeded by yellowish sandy marls (Ressenberg). 3. A vast shelly series consisting of blue marls alternating with strong beds of compact limestone and calcareous grit, the upper beds of which are marked by obscure traces of vegetables, and the middle and inferior strata by a prodigious quantity of well preserved organic remains*." The fossils found in the lowest * Proceedings of the Geol. Soc., Nov. 1829. Supracretaceous Group. 281 strata at Gosau bear the impress, according to these authors, of the cretaceous period ; while those of the overlying blue marls approach so nearly to many species of the lower supra- cretaceous or tertiary formations, that they refer the whole de- posit to an age intermediate between the chalk and those for- mations hitherto considered as tertiary *. Dr. Boue is by no means willing to admit this deposit of Gosau as a tertiary or supracretaceous rock, but as constitu- ting a part of the cretaceous series which extends along the Alps, as will be seen in the next section, from Austria into Savoy f. It may here be remarked that M. Brongniart long since (1823) considered that certain rocks constituting the upper part of the Diablerets ( Valais), were referrible to the supracre- taceous or tertiary series. From the section of this mountain, made by M. Elie de Beaumont, and produced by M. Brong- niart, it appears that the strata are singularly contorted, so that the newer beds have been twisted between the older strata in such a manner that the latter not only occur beneath the former, but also above them J. The beds considered supra- cretacecus are described as composed of calcareous sandstone, anthracite, and a black, compact, and carbonaceous limestone, containing Nummulites ; Ampullaria (two species); Melania costellata. Lam. ; Cerithium Diaboli, Al. Brong. (very abun- dant) ; Turbinella ? Hemicardium , Cardium ciliare, Broc. ; Caryophyllia ; Madrepora. The nummulites found so abundantly in the Alps by no means mark a distinct geological epoch, as they would appear to do in Northern France and in England ; for instead of being confined to the supracretaceous group, they pervade the cretaceous, and possibly also some older rocks. The observations of Dr. Fitton on the Maestricht beds would appear to throw light on these Alpine deposits, as far at least as their zoological character is concerned ; it being un- derstood that the celebrated deposit of the Mont St. Pierre contains a mixture, to a certain extent, of the, so called, se- condary and tertiary remains ; that " it is throughout superior to the white chalk, into which it passes gradually below, but the top bears marks of devastation, and there is no passage * The various labours of Prof. Sedgwick and Mr. Murchison on the Alps will be found in the second part of vol. iii. of the Geol. Transactions, 2nd series, where there are also figures of some of the fossils discovered by them at Gosau. For a list of the Gosau fossils, see Lists of organic remains at the end of this volume. t Boue, various memoirs, Edinburgh Phil. Journal, 1831; Journal de Geologic, 1830; and Proceedings of the Geol. Soc. of London, 1830. I Brongniart, Sur les Terrains Calcareo Trappeens du Vicentin, p. 47 ; and Sections and Views illustrative of Geological Phenomena, pi. 38, fig. 5. 282 Supracretaceous Group. from it to the sands above. The siliceous masses which it in- cludes are much more rare than those of the chalk, of greater bulk, and not composed of black flint, but of a stone approach- ing to chert, and in some cases to chalcedony ; and of about fifty species in the author's (Dr. Fitton's) collection, about forty are not found in Mr. MantelPs catalogue of the chalk fossils of Sussex*." According to M. Dufrenoy, a similar mixture of the organic remains, usually considered as characterizing the cretaceous and supracretaceous rocks respectively, is discovered in the upper part of the chalk series of the Pyrenees. This author observes, that out of numerous species obtained from this de- posit, many are such as are commonly referred to the supra- cretaceous epoch f. From these data, it would appear that at Maestricht, in the Pyrenees, and in the Alps, there do exist deposits con- taining organic remains common to the supposed great classes of secondary and tertiary rocks ; therefore it seems established that no line can, zoologically, be drawn between them. How far other characters may distinguish them, remains to be seen ; and probably minute researches will eventually afftfrd the ne- cessary information. * Fitton, Proceedings of the Geol. Soc. 1830. f Dufrenoy, Annales des Mines, 1831 and 1832. Cretaceous Group. 283 SECTION V. CRETACEOUS GROUP. SYN. Chalk, (Craie, Fr.,Kreide, Germ.,Scaglia, It.,). Chalk Marl, (Crai Tufau, Fr., Kreidemergel, Germ.) Upper Green Sand, (Glau- conie, Craycuse, Fr., Chlorifische Kreide, Planer/calk, Germ.). Gault. Lower Green Sand, (Glauconie Sableuse, Al. Brong., Grunsand, Germ. Part of the German Quadersandstein.) THE upper portion of the cretaceous group partakes of a com- mon character throughout a considerable portion of Western Europe, generally presenting itself under the well-known form of chalk. The upper part of the chalk throughout a large portion of England is characterized by the presence of nume- rous flints, more or less arranged in parallel lines : seams of this substance not only occur in a line with the flints, but also traverse the beds diagonally. The white chalk, when freed from the flints or siliceous grains mixed with it, is found to be a nearly pure carbonate of lime. According to M. Berthier, the chalk of Meudon, when the sand disseminated in it was separated by washing, contained in 100 parts, carbonate of lime 98, magnesia and a little iron 1, alumine 1 . In the lower parts of the English chalk deposit, the flints disappear, be- coming gradually more rare in the passage from the upper to the lower parts. From this circumstance, the white chalk has not unfrequently been divided into upper, or chalk with flints, and lower, or chalk without flints. This supposed character- istic is not available to any great distances ; for at Havre the lower chalk contains an abundance of flint and chert nodules, where it passes into the upper green sand. There is, however, along the line of coast irom Cap la Heve to the eastward, a considerable accumulation of chalk, in which flints are rare, apparently interposed between the Havre beds and the chalk with numerous flints. In the cliffs of Lyme Regis (Dorset), and Beer (Devon), we observe how little dependence can be placed on minute divisions of rocks, even within the distance of a few miles ; for considerable differences in the development of the cretaceous series will be observed between the two places, as I had formerly occasion to remark *. There are, * Geol. Trans. 2nd series, vol. ii. An example of a notable change in a much shorter distance is observable at Cap la Heve (Havre). Towards the town, and even under the light-houses, a marl or clay may be observed se- parating the arenaceous beds into two masses, which marl gradually disap- 284 Cretaceous Group. however, a few beds which are remarkably persistent through- out the district, extending to Weymouth ; they are character- ized by the presence of small and irregularly rounded grains of quartz, probably of mechanical origin, occasionally dissemi- nated through the mass in great abundance. These beds are also remarkable for a great variety of organic remains. Notwith- standing the very general presence of these beds, they sometimes become almost suddenly replaced by others, wherein the grains of quartz are not seen. Thus at Beer, the Beer stone, worked during centuries for architectural purposes, seems the equiva- lent of them, though composed of a white rock, principally carbonate of lime, with some argillaceous and siliceous matter. Probably the Beer stone may be the equivalent of the Malm rock of Hants and Surrey described by Mr. Murchison, and the Merstham firestone noticed by Mr. Webster, and consi- dered as the upper green sand. It may be here observed that the lower part of the chalk, or its passage into the green sand beneath, is extensively used as a building stone in Normandy, and that some of the inferior chalk beds of that country are considerably indurated, even approaching a whitish compact limestone, as may be well seen on the high road, bordering the Seine, between Havre and Rouen. M. Passy remarks, that many beds of the Norman chalk, divided from each other by flints, are so compact as even to approach a crystalline fracture (Elbeuf, Gouy, Duclair, &c.). The lower portion of the cretaceous group has, in England more particularly, received various names, though the mass is very commonly known as green sand. These subdivisions, for the accurate determination of which, and their separation from the Weald - en rocks, we are indebted to Dr. Fitton *, should be borne in mind, more particularly in the study of English geology ; as by tracing them as far as possible, we may obtain an insight into the causes which have produced them. These divisions are, Upper Green Sand, Gault, and Lower Green Sand ; and can be best studied in the south-eastern parts of Englandf. The upper green sand generally appears to graduate into the cretaceous mass above, and is charged with a large quan- pears, so that at a short distance eastward from the light-houses the bed (about 18 feet thick) becomes a slightly argillaceous green sand. M. Passy gives a tabular view of these changes in his Description Geologique de la Seine Inferieure : Rouen, 1828, p. 235. * Fitton, On the Beds between the Chalk and Purbeck Limestone: Annals of Philosophy, 1824; a memoir in which the general relations of all these beds were first pointed out. f The student should consult Dr. Fitton's Memoir (above cited) ; Mr. Murchison's Memoir on North-western Sussex, Geol. Trans. 2nd series, vol. ii. ; Mr. Mantell's Geology of Sussex ; Mr. Martin on West Sussex ; and Dr. Fitton's Geological Sketch of the Vicinity of Hastings, 1833. Cretaceous Group. 285 tity of green grains, which, according to the analysis of M. Berthier, made on those of the equivalent. deposit at Havre, contain: Silica 0'50, protoxide of iron 0'21, alumine 0'07, potash 0*10, water Oil. The same author found that the green or reddish nodules disseminated through the same rock, also at Havre, contained: Phospate of lime 0'57, carbonate of lime 0'07, carbonate of magnesia 0-02, silicate of iron and alumine 0-25, water and bituminous matter 0'07. The reader will at once observe the different composition of the nodules and grains. Respecting the former, M. Al. Brongniart ob- serves, that the phosphate of lime sometimes so abounds as nearly to constitute the whole substance*. The gault (or gait) is an argillaceous deposit of a blueish gray colour, frequently composed of clay in the upper, and marls in the lower part, containing disseminated specks of mica ; it effervesces strongly with acids. The lower green sand is formed of sands and sandstones of various degrees of induration, but principally of ferruginous and green colours, the former usually constituting the upper part, and the latter being most prevalent in the lower portions, which are not unfrequently argillo-arenaceous, particularly at bottom. Without entering further into the smaller divisions of the cretaceous group, it may be remarked that the whole, taken as a mass, may in England, and over a considerable portion of France and Northern Germany, be considered as cretaceous in its upper part, and arenaceous and argillaceous in its lower part. The divisions established in south-eastern England have been observed by Mr. Lonsdale in Wiltshire; and M., Dumont considers that the inferior portion of the cretaceous group, which occurs between the Meuse and the Roer, and is rather thick near Aix la Chapelle, may be well divided into Upper Green Sand, Gault, and Lower Green Sandf. In northern England the arenaceous deposit is scarcely observa- ble, the white chalk resting on red chalk, the latter based on an argillaceous rock, named Speeton clay by Mr. Phillips. In south-western England the chalk rests on a great are- naceous deposit somewhat variable in its composition, some- times containing thick regular seams of chert, at others being nearly without them ; the lower portion being very generally an argillo-arenaceous deposit, characterized by the presence of a great abundance of green particles, and a great variety of organic remains. The central part is formed of yellowish- brown and loosely aggregated sand, in which organic remains * Cuvier and Bvongniart, Desc. Geol. des Env. de Paris, 1822, p. 13. f Omalius d'Halloy, Elements de Geologic. 286 Cretaceous Group. are rare; the superior, of a mixture of brownish-yellow and green sands, with and without chert seams, the organic re- mains being frequently fractured. In Normandy the sands beneath the chalk assume a great variety of characters. Advancing into the interior of France, amid the sands which emerge from under the chalk, and ex- tend from the coasts of Normandy by Mortagne to the banks of the Loire at Tours, and thence by the vicinities of Auxerre and Troyes to the northward, we soon become sensible of the utility of abandoning the smaller divisions, so valuable in England, and of adopting two great divisions, Chalk and Green Sand. This group is extensively distributed over Europe. The chalk and mulatto, or green sand of northern Ireland, is the most western portion known in the British Islands. It occurs on the Spanish side of the Pyrenees, and would appear, from the observations of Colonel Silvertop, to extend much further westward into the kingdom of Spain. It is probably found in the provinces of Sevilla and Murcia. Of the geological structure of Portugal so little is yet known, that we are not aware of the existence of cretaceous rocks in that country. According to M Nilsson, the chalk of Sweden (the conti- nuation of that in Denmark,) is generally incumbent on gneiss, more rarely on rocks of the grauwacke group, and has only been observed resting on beds of the oolitic group, at one place near Limhamn, in Scania. In one locality, near Ham- mer and Kaseberga, it has a large capping of sand with bitu- minous wood, which M. Nilsson refers to the cretaceous group, as the vegetable remains are associated with cretaceous fossils. The chalk deposit of Sweden is occasionally of considerable thickness, and abounds in organic remains. The northern portion of the deposit is white or grayish white, more or less abundantly mixed with siliceous substances. The southern portion is stated to present the various modifications from green sand to white chalk *. From the observations of Professor Pusch it appears that the cretaceous group occurs extensively in Podolia and south- ern Russia, being a continuation of that of Lemberg and Po- land. It occupies the country in the shape of marly chalk be- tween the Bog and the Dniester round Janow, Lubin, Miko- lajew,Uniow, and Rohetyn. Concealed beneath the supra- cretaceous rocks it is prolonged from Halicz to Zalezczyki on the Dniester. On the west of this river it occupies the envi- rons of Tlumacz, Otynia, and other places to the foot of the * Nilsson, Petrificata Suecana Formationis Cretaceae descripta, et iconibus Ulustrata: 1827. Cretaceous Group. 287 Carpathians. On the north of the Dniester it exists beneath the supracretaceous rocks between that river and Brzezan ; it extends to Brody and into the plains of Volhynia. " In many places, and especially around Krzeminiec, it is covered by more recent deposits, but its presence is indicated by an abun- dance of flints and chalk fossils scattered through the sands." The chalk forms considerable eminences round Grodno in Lithuania. According to M. Eichwald, the chalk of the lat- ter country abounds in belemnites, which are wanting in Volhynia, where they are replaced by Echinites^ Terebratulce^ Ostrece, Placwuz, Inoceramus (Catillus), &c. The flints of the two countries contain Reteporce, Escharce, Ananchyte$ 9 EncriniteS) &c. *. According to M. Eichwald, chalk without flints, with shells of the genera Plagiostoma, Pecten, Ostrea, &c., rests on ar- gillaceous slate at Ladowa, on the Dniester. At about seven wersts from thence, near Bronnitza, it alternately rests on a coarse sandstone, grauwacke, and argillaceous slate -)-. Furthtr south, and in the plains of Moldavia, Podolia, and Bessara- bia, it only appears in detached portions, as between Ja- roszow and Mohilew on the Dniester, from Raszkow to Jaor- lik on the Pruth, near Kolomea, Sniatyn, Sadagora, Seret, lloswan, Illina, and Jassy. "Trie chalk is found on the south side of the granitic steppe, in the Crimea, and on the borders of the Sea of Azof, between the Berda and the Don ; it also occurs on the west of the Don, across the south-east and middle of Russia. In the country of the Don Cossacks, in the governments of Worenech, Koursk, and Toula, it here and there appears in hills, and on the banks of the rivers be- neath the vegetable soil, and probably constitutes the base of that great and fertile plain. The marly clay of eastern Gal- licia and of Podolia is connected, as in Poland, with gypsum, at Mikulnice, Seret of Podolia, to the east of Trembowla, but more particularly at Zbrycz near Czarnokozienice. The graphic chalk is there more abundant than in the centre of Poland, and more abounds in flints J." It further appears from the interesting details of M. Pusch, that there is a deposit of lignite upon the upper part of the chalk, reminding us of the lignite sand noticed by M. Nilsson in Sweden, which would thus appear to be similarly situated at various distant points. It seems to be wanting in central Poland, but is found in many situations in eastern Gallicia, and abundantly along the Carpathians, in Pocutia and Bu- kowine, from Otynia towards Maydan, Lanczyn, Kniazdwor, and mounting the Pruth, from Miszyn to Seret, and near * Journ. de Geol., t. ii. p. 62. f Ibid. p. 61 . J Pusch, Journ. de Geol. t. ii. 288 Cretaceous Group. Czorthow and Ulaszkowce, and on the Dniester near Cho- chim and Mohilew. This lignite deposit is described as a blueish or greenish gray calcareous sandstone, alternating with sand and clay, more or less calcareous, and with lami- nated marl : it sometimes contains amber, but more frequently pieces of bituminous wood, thin beds of lignite, and trunks of fossil trees. It contains many shells, among which are Pec- tunculus pulvinatus, P. insubricus, Pecten (smooth species), and more rarely Nummulites discorbinus, Dentalium eburnium, and small Cerithia. This sandstone is considered distinguish- able from the well-known lignite deposits of western and northern Poland by its fossil shells ; but it may perhaps ad- mit of a question, how far local circumstances may not have caused a great difference in this respect. Prof. Pusch describes the cretaceous rocks as extensively deposited in Poland, and as divisible into marly chalk and white chalk : the marly chalk is a soft calcareous marl, either white or light gray, becoming sandy in some districts (Mie- chow, Kazimirz) ; while other beds are coloured green by si- licate of iron (Czarkow, Szczerbakow) ; it alternates with more compact white limestone. A shaft sunk through this de- posit at Szczerbakow showed that it was 697 English feet thick at that place. M. Pusch considers that certain gypseous de- posits of Poland are connected with the marly chalk. The white chalk is described as identical with that of England, containing a much larger proportion of flints than the marly chalk *. Rocks of the cretaceous group occur in the great plain of Miinster, skirting the northern edge of the Westphalian slate mountains to the south-western border of theTeutoburgerwald. Though the rocks of this district undoubtedly form a conti- nuation of the English chalk series, they differ materially from it in mineralogies! structure. There is not throughout the whole tract a trace of white chalk with flints. On the south- ern borders of the district, whitish, yellow, blueish, and green marls predominate, containing beds of indurated marl full of green grains, which become sandy. These beds sometimes rest immediately on rocks of much greater antiquity, such as the coal measures of Frohnhausen near Essen, and sometimes occur, twenty feet thick, in the yellowish marl, as may be ob- served from Aplerbeck, near Werl, to the vicinity of Pacler- born. At the salt-works of Koenigsborn, where the marl deposit is 470 feet thick, two of these beds, in contact with each other, furnish excellent building stone. In the central portion of the district the light-coloured marls * Pusch, Journal de Geologic, t. ii. p. 253. Cretaceous Group. 289 become more calcareous, and pass into limestone. The fol- lowing section is observable in the chain of the Teutoburger- wald. At the top, a light gray calcareous marl, which readily crumbles in the air. To this succeeds a compact white lime- stone, often of a splintery fracture, which becomes green in the lower beds. Next follows, always in the descending order, a dark gray and friable calcareous marl, with globular de- tached masses of compact limestone (Horn, and on the Laus- berge at Bielefeld). Then sandy or gravelly yellowish gray clayey marls, striped black, and containing detached pieces of chert and chalcedony. The lowest portions mostly consist of a light-coloured, and sometimes ferruginous, thick sandstone, rarely associated with beds of conglomerate. The cretaceous rocks which occur on the borders of the hill country of northern Germany, join the northern side of the Hartz, and extend towards the north-west into the plain be- tween Brunswick and Hildesheim. Cretaceous marl some- times comes into immediate contact with the grauwacke rocks of the Hartz, and extends, under the sand, into the plain ot Peina. On the other hand, the sandstone is much developed in the country about Halberstadt and Blankenburg. It con- tains a thick bed of sandy marl, full of green grains, which passes into a gray white earthy limestone with flints, resem- bling that which occurs above the sandstone. Coal occurs in this sandstone near Quedlinburg. The great mass of the cre- taceous rock, named Quadersandstein, of Saxony and Bohe- mia, is separated from the border of the north German basin; but Prof. Weiss * has shown that the crystalline rocks which surround it on the N.E. side, from Weinbohla to Hohnstein, have been forced up since it was deposited, and thus caused the separation. The sandstone forming the well-known rocks of Schandau and Adersbach is very uniform, large-grained, and white, containing but little cementing matter. There are many patches of quadersandstein in Silesia, the most considerable of which rests on the northern edge of the Riesengebirge from Goldberg to Lowenberg, and thence from the Queiss to the Neisse. It extends N. W. towards the low country, and is bounded on the E. by older rocks. The low- est beds often become conglomerates, as in the mountains of Goldberg and Prausnitz, and on the ridge between Lowen- berg and Neuland, where millstones are quarried (Wehrau on the Queiss, Wartha). Similar beds of conglomerate are found in the Moiser quarry near Lowenberg, where numerous organic remains are seen in it. The sandstone is clayey at * Weiss, Ueber einige geogn. Punkte bei Meissen und Hohnstein, Kar- sten's Archiv fur Bergbau, &c. B. xvi. p. 3. u 290 Cretaceous Group. Tollendorf (Buntzlau), and contains beds, a foot and a halt" thick, of red and white potters' clay, out of which much pot- tery is manufactured. In the neighbourhood of Wehrau the sandstone passes into compact quartz rock, containing beds of clay and the well-known clay ironstone. Several coal beds, from six to twenty-four inches thick, are found in it near We- nig Rackwitz, Otlendorf, and Newen. The cretaceous group occurs extensively on the southern side of the great level to the eastward of the Oder. It fills the basin, forty-five miles broad, between the oolitic high range from Wielun to Cracow, and the Sandomirer grau- wacke range. It extends down the Vistula as far as Pulavy, and thence further eastward through the southern part of the province of Lublin to Lemberg and the Dniester. It is here connected with the great chalk plain of Volhynia and the plateau of Podolia. In the upper division of the cretaceous series on the Nida (west of the Vistula) in the province of Cracow, and in the basin of Lemberg and Podolia, at Mikulnice and Zbrycz as far as the river Podhorec, there is a gypsum deposit, 100 feet thick, consisting of large yellow and gray crystals. The isolated patches of gypsum in Upper Silesia probably belong to this bed *. The sulphur bed at Czarkow on the Nida occurs be- tween this gypsum and the chalk marl. Sulphurous springs, and occasionally weak saline springs, accompany this line from Busko to the Vistula, from Liibien near Lemberg to Jassy f. The fossils of these strata in general resemble those of the English and French chalk, though there are some small local differences J. The cretaceous rocks of France have been already noticed ; but it may be remarked that they rest on the coal measures of Mons and Valenciennes, and that the rocks of the Isle d' Aix and the embouchure of the Charente, are considered referrible to this group. They are well known as contained in some of the valleys of the Jura, and as ranging along a considerable portion of the French side of the Pyrenees. They occur on both sides of the Alps, and range down a large portion of the Apennines. This group occurs extensively in the maritime Alps, contain- ing among its fossils an abundance of Nummulites., remains once considered as wholly supracretaceous. Its usual appearance in that district is that of a marno-arenaceous limestone, the arena- ceous matter sometimes predominating, and forming a sand- stone. Beds of light-coloured limestone charged with green * German Transl. of Manual. f Pusch, Ueber die geogn. Konstitution der Karpathen und der Nord- Karpath en-Lander. Karsten's Arch, fur Min. &c. Bd. i. p. 29. t German Transl. of Manual. Cretaceous Group. 291 grains, and full of Belemnites, Ammonites, Nautili, and Pec- tines, constitute its lower part, and even appear intimately connected with the upper part of a light-coloured limestone deposit, among which crystalline dolomite abounds. The lat- t&r rocks are very difficult to classify, and may either belong to the lower part of the cretaceous, or upper part of the oolitic, group. Be the age of these beds what it may, they seem, ac~ cording to M. Elie de Beaumont, intimately connected with a large proportion of the Alpine nummulitic rocks, the light- coloured limestones of Provence, of Mont Ventoux, of the departments of the Drome, Isere, &c. ; the nummulitic rocks being connected with the cretaceous series of Brianconnet (Basses Alpes), of Villard le Lans (Isere), of the mountains of the Grande Chartreuse, of the Mont du Chat, of the high longitudinal valleys of the Jura, of the Perte du Rhone, of Thonne, and of la Montagne des Fis. Having premised thus much respecting the geographical distribution of the cretaceous group, we will take a slight sketch of the variations in its mineralogical character. Throughout the British Islands, a large part of France, the northern parts of Germany, in Poland, Sweden, and in va- rious parts of Russia, there would appear to have been certain causes in operation, at a given period, which produced nearly, or very nearly, the same effects. The variation in the lower portion of the deposit seems merely to consist in the absence or presence of a greater or less abundance of clays or sands, substances which we may consider as produced by the de- struction of previously existing land, and as deposited from waters which held such detritus in mechanical suspension. The unequal deposit of the two kinds of matter in different si- tuations would be in accordance with such a supposition. But when we turn to the higher part of the group, into which the lower portion graduates, the theory of mere transport appears opposed to the phenomena observed, which seem rather to have been produced by deposition from a chemical solution of carbonate of lime and silica, covering a considerable area*. For the reader will have observed, that white chalk, very fre- quently containing fiints, extends from Russia, by Poland, Sweden, Denmark, Northern Germany, and the British Is- lands, into France. The great European sheet of chalk and green sand, produced at the cretaceous epoch, has since been so covered up, shattered, upheaved and destroyed by various * If we regard present appearances, we find that silica is held in solution by thermal waters, which also, as in the case of those of St. Michael in the Azores, may contain carbonate of lime. No springs or set of springs that we can imagine are likely to have prodviced this great deposit of chalk, so uniform over a large surface. u 2 292 Cretaceous Group. causes, that we have mere remnants presented to our exami- nation. Still, however, we have enough to show that it over- lapped a great variety of pre-existing rocks, from the gneiss of Sweden to the Wealden deposits of south-eastern England inclusive. Thus far no very material difference in the arrangement and mineralogical character of the mass has been observed, of course disregarding small local variations : but arrived at the Alps we meet with rocks, which certainly, from their minera- logical characters alone, would never have been referred to the cretaceous group : yet, unless we disregard the evidence of organic remains, they have been formed at the same epoch. Instead of the soft and white chalk, and the abundance of loosely aggregated sands, which constitute so large a propor- tion of the group in England and northern France, we have compact limestones and sandstones vying in hardness with the oldest rocks, so as, in the earlier days of geology, to have been considered only referrible to them. Such is the hard black limestone (containing an abundance of Scaphites., Ha- mites 9 furrilites, and other fossils,) which crowns the sum- mits of the Fis, the Sales, and other mountains of Savoy, that range up to the Buet. The rocks referrible to this group, on the southern side of the Alps, and facing the great Lombardo- Venetian plains, are not so far removed from the mineralogical character of the chalk of western Europe, being often composed of white, green- ish, and reddish beds, occasionally very argillaceous. In some parts of the Apennine range, in which a large mass of rocks would seem referrible to this epoch, the character is quite cre- taceous. How far the Alpine rocks of this age have been altered since their deposit, in consequence of the disturbances they have ex- perienced, or how far their present condition can be attributed to original formation, which must always have been influenced by local causes, yet remains a problem to be solved : but it may be remarked, that we can scarcely imagine them to have been exposed to the various circumstances attending great dis- turbances, without having suffered from such circumstances. According to M. Dufrenoy, the cretaceous series of south- ern France not only contains a curious mixture of organic re- mains, but also presents mineralogical characters different from those of the contemporaneous deposit of the northern part of the same country. That portion which reposes on the central elevations of France, is composed, in its lowest parts, of marls and sandstones, more or less charged with oxide of iron, and containing lignite in some situations. M. Dufrenoy refers these beds, such as they are seen at Rochefort, Angouleme, Cretaceous Group. 293 Sarlat, Pont St. Esprit, and other places, to the inferior are- naceous rocks of the cretaceous series. At Angouleme, and some other localities, these deposits are surmounted by regu~ lar beds of a nearly saccharine limestone, a fact which shows that a slow chemical deposit here took place ; so that if we con- sider the white chalk of northern Europe as chemically formed, it would appear that there was a slower deposit in some loca- lities than in others. The same author also states, respect- ing that portion which either constitutes a part of the Pyre- nees, or is continuous with it, that although the limestones which rest on the arenaceous deposits (containing lignites and vegetable impressions) are commonly compact, there are some which are crystalline. It should however be observed, that there are evidences of mechanical action in the upper portion of the Pyrenean chalk, for it is stated that thick beds of cal- careous conglomerates alternate with the limestones in the upper part of the series *. The same author states that the cretaceous rocks of the Spanish Pyrenees closely resemble those on the French side of the same mountains. The celebrated salt mine of Cardona is contained in the upper part of the series, and the rock salt of Mon Real is also included in it. Saline springs occur near Orthez, between Jaca and Pampeluna, and other places, ac- companied by gypsum, trap rocks, and dolomite, always, it is stated, in lines of fractured country. Coal is discovered in this series at Pereilles near Bellesta, Ernani near Irun, at Saint-Lon in the Landes, &c., and sulphur and bitumen at Saint-Boes near Orthez f. M. Partsch describes a series of calcareous and arenaceous rocks containing nummulites in Dalmatia and the neighbour- ing provinces, which appears to belong to the cretaceous group. These rocks form high mountains, particularly in Croatia. From the direction of the mountain chains, M. Elie de Beaumont infers that these rocks may extend into Livadia and the Morea. Facts can alone determine how far this infe- rence is correct ; but in the mean time it may be remarked, that rocks of the Dalmatian character seem to prevail exten- sively in parts of Greece, and even along the coast of Kara- * mania. From the various memoirs of MM. Keferstein and Boue, Prof. Sedgwick, Mr. Murchison, and M. Lill von Lillienbach, it seems clear that the cretaceous group exists extensively in the Alps ot Austria and Bavaria, and in the Carpathians. There may be certain differences of opinion as to where the series commences, or where it ends, but the main fact of the * Dufrenoy, Annales dcs Mines, 1831. f /foU 1832, 29t Cretaceous Group. presence of the group itself would appear to be undisputed : it would also appear that the deposit was in a great measure arenaceous. After remarking on the stability of the cretaceous rocks of the Carpathians since their deposit, contrasted with their dis- location in the main chain of the Alps, (a fact subsequently fully confirmed within a certain distance from Vienna by Mr. Murchison,) M. Elie de Beaumont proceeds to observe that " nearly in the prolongation of the Carpathians, to the environs of Dresden, the right and northern side of the Elbe valley is bordered by a continuation of granite and sie- nite mountains, which extend from Hinterherms, on the fron- tier of Bohemia, to Weinbohla, about a league and a half east from Meissen, rising suddenly above the plain of quader- sandstein and planerkalk (cretaceous rocks). When the con- tact of the granitic and cretaceous rocks is examined, it is ob- served that the former cut, and even horizontally cover, the latter in many places ; clearly proving that the granitic and sienitic rocks were elevated to the surface since the deposit of the green sand and chalk : and it is not the less remarkable, that the little chain formed of them runs in the direction of the valley of the Elbe, and exactly parallel to that which reigns in the Pyreneo-Apennine system *." The most remarkable point is at the quarry of Weinbohla, where, according to M. Weiss, the chalk there worked con- tains the Plagiostoma spinosum, Podopsis, Spatangus, &c. This rock is in horizontal beds ; but near the sienite they gra- dually dip until they plunge beneath it, so that the sienite con- formably covers the chalk. A marly and clay bed, partly bi- tuminous, covers the chalk, occurring between it and the gra- nitic rock. M. Klipstein remarking on these appearances, observes, that mounting the valley of Polenz, from the foot of the Hochstein, the green sand beds on the right, which are generally horizontal, begin gradually to dip, the angle in- creasing with their approach to the granite ; near the latter, dipping at 46 or 48 beneath it; and he states that of this fact there can be no doubt. " Coming from Brand, the height of the green sand diminishes in such a manner in the descent of the valley, that a few feet of it are alone visible. In a val- ley extending into the mountains towards the Rothenwald, the chalk with its marls and clays appears between the green sand and granite ; and there are places where galleries have been driven through the granite and chalk into the green sand." * M. Elie de Beaumont cites these curious appearances of the superposi- tion of granitic rock, as obtained from the descriptions of Prof. Weiss, in- serted in Karsten's Archiv fur Mineralogie, &c. t. xvi., and new series, t.i. Cretaceous Group. 295 From these works it would appear that " the chalk with its clays and marls gradually diminishes, so that the granite at first resting on chalk, comes into contact with green sand. The superposition of the granite is quite evident at some di- stance from this point, when suddenly there is a change, and the granite cuts the arenaceous beds without at all deranging or altering them : it is even stated, that beneath it commences taking a position under the green sand *." Prof. Naumann remarks, that the fact of the increased dip of the cretaceous rocks as they approach the granite, so that they finally are covered by it, is also seen near Oberau ; and that near Zscheila and Niederfehre, the cretaceous rocks rest horizontally on the granite. The same author remarks that the connection of the two rocks is sufficiently evident at both these localities, for the limestone and granite are, as it were, entangled in each other, and irregular portions and veins of hard limestone with green grains and cretaceous fossils are here and there imbed- ded in the granite. The gorge of Niederwarta, on the left bank of the Elbe, is pointed out as a very interesting point. " The chalk is horizontal in the village, but at about the third of a league beyond it, the beds rise and dip at about 25 or 30 ; a hundred paces further on, the dip is from 70 to 80, and the rocks, fractured near the granite, rise in steep moun- tains above the chalk country." At Lichtenhain and Otten- dorf the limits of the sandstone and granite, are exposed, and at twenty paces from the granite the sandstone is seen to be horizontal ; but on approaching the granite, the beds, or frag- ments of beds, rise, and some dip at an angle of 60 f. For the following section, representing the contact of the sienite or granite with the chalk at Weinbohla, I am indebted to Mr. Killaly. Fig. 42. Fig. 44. Fig. 43. is a front view of the quarry from the west. , gravel ; b, granite or sienite ; c, gray clay containing nodules of iron * Journal de Geologic, t. ii. p. 182. t Naumann, Poggendorfs Annalen ; and Journal de G6ologie, t. iii. win. 296 Cretaceous Group. pyrites, varying in size from that of a nut to an egg ; d, the cre- taceous rock. Fig. 42. is a section of the quarry at m n (fig. 4-3) : , gravel ; b, granitic rock ; c, clay ; d, chalk. Fig. 44., a sec- tion atjfg, where the granitic rock rests immediately on the chalk, the clay bed being absent. The granitic rock on the northern side of the quarry, resting on the clay and chalk in that direction, contains numerous cubes of iron pyrites*. Before we terminate this sketch, we should notice certain beds found in the Cotentin (Normandy) and in Seeland ; which, if they do not show a passage of the chalk into the supracre- taceous rocks, exhibit an interesting juxtaposition of strata, containing chalk fossils, and those with organic remains com- monly referred to a more recent date. The baculite limestones, as they are termed, of the Cotentin had often been visited, and more or less noticed ; but their real position in the series was not pointed out until they were described by M. Desnoyers in 1825. The bacuiite limestone is white or yellow, and for the most part compact, varying, however, in its mineralogical character, being sometimes cretaceous and even arenaceous. It contains the organic remains of the cretaceous group, seve- ral being also found at Maestricht; such as Bacculites verte- bralis, Thecidea radians, T. recurvirostra, Terebratula, four or five particular species not named, &c. These beds are sur- mounted by others (the whole collectively being of no consi- derable thickness), composed principally of calcareous matter, not presenting any marked difference in appearance, though they do not precisely resemble those beneath. They contain organic remains, such as are found in the calcaire grossier ; and M. Desnoyers considers that a well defined zoological line can be drawn between the two deposits ; observing, how- ever, that at the contact of the lower portion of the one with the upper part of the other, when the rocks were without much coherence, there was sometimes an apparent mixture of the fossils. " But at the same time," observes M. Desnoyers, " it has appeared to me, that independent of this confusion, which may be accidental, the species of the compact chalk, Trochus and Bacculites, preserving their habitual mode of pe- trifaction, might have belonged to a previously formed bedf." The geological student can observe the baculite limestone at Freville, Cauquigny, Bonneville, Orglandes, Hauteville, and other places in the Cotentin. M. Desnoyers remarks on the absence of Turrilites, GrypJuza Columba, G. striata, Ostrea carinata, O. pectinata, Pecten spi- dela * Killaly, MSS. f Desnoyers, Sur la Craie et les Terrains Tertiaires du Cotentin ; Me"m. la Soc. d'Hist. Nat. de Paris, t. ii. Cretaceous Group. 297 nosus, Chenendopora, Hallirhoa, Ventriculites, Spongns, &c. amid a mass of fossils found in the chalk and green sand. MM. Passy and Graves have noticed some beds at Saint Germain-de-Laversines (Normandy), which are referred to the same age as the baculite limestone. The rocks of Saint- Germain-de-Laversines form two distinct beds, the highest of which consists of yellow limestone, containing numerous casts of shells and polypifers, and is about six or seven yards thick. The lowest bed, which rests immediately on white chalk, is very hard, and contains the same organic remains as the upper bed. These remains are stated to differ, with the exception of some cretaceous echinites, both from those usually found in the chalk and in the calcaire grossier*. The base of the cliff at Stevensklint (Seelarid) is formed of chalk with beds of nodular flints. Upon the chalk, which is re- presented as having an undulated surface, rests a thin bed (about six inches thick) of a bituminous clay, containing a Zoophyte, Sharks 1 teeth, a Pecten, impressions of a bivalve, and traces of vegetable remains. Incumbent on this is a hard yellowish white limestone, containing the remains of the genera Patella, 1 species ; Cyprcca, C 2 ; Fusus, 1 ; Cerithium, 2 ; Ampullaria, 1 ; Trochus, 1 ; Dentalium, 1 ; Area, 1 ; Mytilus, 1 ; Serpula, 1 ; Spatangus, 1 ; Favosites, 1 ; and Turbinolia, 1 ; with Fishes' teeth, and undeterminable univalves, bivalves, and corals. This limestone contains green grains, seldom exceeds three feet, and is sometimes only a few inches in thickness, but is nowhere entirely wanting. It is covered by another limestone, from thirty to forty feet thick, almost entirely composed of fragments of corals, and forming the upper part of the cliff. This is di- vided by chert into many beds, the chert being bent and curved. It is remarkable that the organic remains of this superior de- posit are such as are considered characteristic of the chalk, consisting of Ananchytes ovata, Ostrea vesicularis, Belemnites mucronatus, &c. Dr. Forchhammer observes that the remains of Ananchytes ovata are occasionally so abundant that the lime- stone consists almost entirely of them t. In this case it might be inferred that there was an alternation of the fossils, com- monly considered as cretaceous, with others of a supracreta- ceous character. From an inspection of the List of organic remains found in the cretaceous group J, it would appear that the remains of mammalia have not yet been detected in the cretaceous group ; while reptiles, one of them of considerable size, the Mososaurus Hoffmanni) have been observed in Yorkshire, Sussex, Maes- * Passy, Description Geologique de la Seine Inferieure : Rouen, 1832. f Forchhammer, Edin. Journ. of Science, vol. ix. 1828. I See List of organic remains at the end of the volume. 298 Cretaceous Group. tricht and Meudon. Fish have been observed in France, and in various parts of England. Sharks' teeth and the tritores of some fish are far from uncommon. Crustacea have been no- ticed in Denmark, Yorkshire, Sussex, the Isle of Wight, Dor- setshire, and Maestricht. Among the polypifers the most abundant would appear to be different species of the genera Spongia and Alcyonium of some authors ; genera, many spe- cies of which have been classed by Goldfuss under the heads of Achilleum, Manon, Scyphia, and Tragos, so that there is much difficulty in presenting a list which should give the dif- ferent species under any one arrangement. Marion pulvina- rium, and M. Peziza, Goldf., are found at Maestricht, and at Essen in Westphalia ; Spongia ramosa, Mant., is discovered in the chalk of Yorkshire, Sussex, and Noirmoutier ; Alcyonium globosum, Defr., at Amiens, Beauvais, Meudon, Tours, Gien, and in the baculite limestone of Normandy ; Hallirhoa costata, Lam., in the green sand of Normandy, and the upper green sand of Wiltshire ; Ceriopora stellata, Goldf., Maestricht and Westphalia ; Lunulites cretacea, Defr., at Maestricht, Tours, and in the baculite limestone of Normandy ; Orbilulites lenti- culata, Lam., in Sussex, and at the Perte du Rhone. Accord- ing to Goldfuss, numerous polypifers are discovered at Maes- tricht ; consisting of Achilleum, 2 species ; Manon, 4 ; Tragos, 1 ; Gorgonia, 1, Nullipora, 1 ; Millepora, 2 ; Eschara, 9; Cel- lepora, 6 ; Retepora, 5 ; Ceriopora^ 1 3 ; Fungia, 1 ; Diplocte- nium 9 2 ; Meandrina, 1 ; Astrea^ 1 3 ; to which should be added, according to M. Desnoyers, Lunulites^ 1. Among the Radi- aria, the Apiocrinites ellipticus, Miller, is found in the chalk of Yorkshire, Sussex, Westphalia, Maestricht, Normandy and Touraine ; the Cidaris variolaris, Al. Brong., in Sussex, and Normandy, at the Perte du Rhone, in Westphalia, and Sax- ony ; the C. granulosus, Goldf., at Maestricht, Aix-la-Chapelle, and Westphalia; the C. saxatilis, in Sussex and Normandy; the Galerites albogalerus, Lam. (Fig. 46.), in Yorkshire, Sus- sex, Dorset, Normandy, Quedlinburg, Aix-la-Chapelle, and Poland; the G. vulgaris, Lam., in Sussex and France, at Quedlinburg, and Aix-la-Chapelle ; the Ananchytes ovata, in Yorkshire, Sussex, Normandy, at Meudon, in Westphalia, Poland and Sweden ; the A. hemisphccrica, in Yorkshire and Normandy; the Spatangus Cor-anguinum, Lam. (Fig. 45.) in Yorkshire, Sussex, Dorsetshire, various parts of France, the Savoy Alps, various parts of Germany, Poland, and Sweden; the Sp. ornatus, at Aix-la-Chapelle, Normandy, and Bayonne; Sp. Bufo, Al. Brong., Sussex, Normandy, Maestricht, and Aix-la-Chapelle; theSp. Cor-testudinarium, at Maestricht and Quedlinburg. Among the shells, the most widely distributed would appear to be Lutraria Gurgitis, found at the Perte du Cretaceous Group. Rhone, and in Sweden; Mya mandibula, Sussex, Isle of Wight, Normandy, and in the South of France ; Trigonia alaeformis^ Sussex, Isle of Wight, West of England, South of France, Aix-la-Chapelle ; Inoceramus (or Catillus) Cuvieri (Fig. 47 and 48.)? discovered in the chalk of Yorkshire, Sussex, Normandy, Fig. 46. Fig. 47. Fig. 45. Fig. 49. Fig. 50. Fig. 51. Fig. 52. Meudon, the South of France, and Sweden ; Inoceramus (or Catillus) Brongniarti, in the chalk of England, Poland, and Sweden ; Ino. concentricus, in Sussex and in Wiltshire, West- phalia, at the Perte du Rhone, and in the Savoy Alps; Ino. sulcatus, in Sussex, at the Perte du Rhone, in the Savoy Alps, and in Sweden; Plagiostoma spinosum (Fig. 49. ), in the chalk of Sussex, Dorsetshire, Normandy, Meudon, the South of France, Saxony, Poland, and Sweden; Gervillia solenoides, Maestricht, Sussex, Wilts, Dorset, and Normandy; Pecten quinquecostatus (Fig. 50.), in Sussex, the West of England, Normandy, at Meudon, the Perte du Rhone, Sweden, &c.; P. quadricostatus (Fig. 51.), in Sussex, the West of England, Normandy, at Maestricht, and in the Alps of Dauphine ; P. asper, Wilts, Normandy, Germany, and Poland; Podopsis truncata (Fig. 52.), in Normandy, Dorset, Touraine, and Swe- den ; Pod. striata, in Yorkshire, Westphalia, and Normandy; Ostrea vesicularis* (Fig. 53.), in Sussex, Normandy and other places in France, at Maestricht, and in Sweden ; O. carinata, in Germany, Sussex, Normandy, and the South of France ; O. serrata, Sweden, Maestricht, and in the South of France; Gryph&a auricularis, at Perigueux, South of France, in the Alps of Dauphine, and Poland ; G. Columba (Fig. 54.), North- amptonshire, Normandy, South of France, Maritime Alps, Germany, and Poland ; G. sinuata, Yorkshire, Isle of Wight, Normandy, Dauphine, South of France, and the Pyrenees; Terebratula plicatilis, Moen, in Sussex, at Meudon, South of France, and the Alps of Savoy and Dauphine ; T. subplicata, * Grypkeea globosa, Sowcrby. 300 Cretaceous Group. in Yorkshire, Sussex, Maestricht, Normandy, and at Tours and Beauvais; T.Defi-ancii, in Yorkshire, Sussex, at Meudon, Maestricht, and in Sweden ; T. alata, Normandy, South of France, at Meudon, and in Sweden ; T. octoplicata, in Nor- mandy, South of France, Quedlinburg, and Sweden ; T. pec- tita^ in Wiltshire, Normandy, and Sweden ; T. semjglobosa, Sweden, Moen, Yorkshire, Bochum; Belcmnites mucronatm (Fig. 56.), in Yorkshire, Sussex, Normandy and other parts of France, Sweden, and Poland ; Ammonites varians, in Sussex, Wiltshire, Germany, Normandy, and the Savoy Alps; Am. Rhotomagcnsts, in Sussex, Wiltshire, and Normandy ; Am. Mantelli) Sussex, Saumur, Bochum, and Hanover; Am.Selli- guinus, Normandy, Savoy, Westphalia, and Poland ; Am. in- jlatus, Wilts, Normandy, and the Perte du Rhone ; Am. Hip- pocastanum, Dorsetshire, and Normandy; Bacculites Faujasii (Fig. 55.), Sussex, Norfolk, Maestricht, Bochum, Aix-la- Chapelle, and Sweden ; Bac. obliquatus, Sweden, Sussex, and Normandy; Hamites rotundus (Fig. 58. ), Yorkshire, Sussex, Normandy, the Perte du Rhone, and Aix-la-Chapelle. Fig. 53. Fig. 54. Fig. 55. Fig. 56. \A Fig. 57. Fig. 58. Fig. 59.' It will be observed that this list is far from large, when we consider the number of species enumerated in the catalogue, and that, perhaps, some of those considered identical may be different species. No doubt when we reduce our view to smaller distances and more minute divisions of the cretaceous group, other species than those above enumerated will be found oc- curring under similar circumstances in different situations; but even then, certain species do not seem to be so constant to particular beds as has been supposed, though some certainly are found over considerable distances in similar parts of the group. The following summary of the organic remains stated by various authors to have been discovered in the cretaceous * Fig. 57. Scaphites oUiquus, Sow. (So. strialus, Mant.) ; Fig. 59. Tur- rilites hibcrculatus. Figured to show the forms of these genera, common in the cretaceous group. Cretaceous Group. 301 group, though not pretending to perfect accuracy, may yet be useful, as presenting a general view of the subject, and as being an approximation towards the truth. PlantfB. Confervites, 2 species; Fucoides, 9; Zosterites, 4; Cycadites, I ; Thuites, 1 ; and various vegetable remains not yet determined. Zoophyta. Achilleum, 3; Manon, 7; Scyphia, 12; Spon- gia, 12; Spongus, 2; Tragos, 5; Alcyonium, 2; Choanites, 3; Ventriculites, 3; Siphonia, 4; Halirrhoa, 1 ; Serea, 1; Gor- gonia, 1; Nullipora, 1 ; Millepora, 5; Eschara, 10; Cellepora, 7; Coscinopora, 2; Retepora, 5; Flustra, 3; Cceloptychium, 3; Ceriopora, 21 ; Lunulites, 1 ; Orbitolites, 1 ; Lithodendron, 2; Caryophyllia, 2; Anthophyllum, 1 ; Turbinolia, 2; Fungia, 3; Chenendopora, 1; Hippalimus, 1; Diploctenium, 2; Mean- drina, 1; Astrea, 15; Pagrus, 1. Radiaria. Apiocrinites, 1 ; Pentacrinites, 1 ; Marsupites, 1 ; Glenotremites, 1 ; Asterias, 1 ; Cidaris, 9 ; Echinus, 5 ; Gale- rites, 9 ; Clypeus, 1 ; Clypeaster, 3 ; Echinoneus, 4 ; Nucleo- lites, 12; Ananchytes, 8; Spatangus, 29. Annulata. Serpula, 30. Cirripeda. Pollicipes, 2. Conchi/era. Magas, 1 ; Thecidea, 3 ; Terebratula, 54 ; Cra- nia, 8; Orbicula, 1; Hippurites, 8; Sphaerulites, 15; Ostrea, 22 ; Exogyra, 4 ; Gryphaea, 8 ; Sphrera, 1 ; Podopsis, 5 ; Spon- dylus? 1 ; Plicatula, 4 ; Pecten, 28 ; Lima, 3; Plagiostoma, 15 ; Avicula, 2; Inoceramus, 19; Pachymya, 1; Meleagrina, 1; Gervillia, 3; Pinna, 4; Mytilus, 5; Modiola, 2; Chama, 2; Trigonia, 11 ; Nucula, 12; Pectunculus, 3; Area, 6; Cucul- laea, 6 ; Cardita, 4 ; Cardium, 3 ; Venerecardia, 1 ; Astarte, 1 ; Thetis, 2 ; Venus, 9 ; Lucina, 1 ; Tellina, 3 ; Corbula, 6 ; Cras- satella, 2 ; Cytherea, 2 ; Lutraria, 2 ; Panopsea, 1 ; Mya, 2 ; Pholas? 1; Teredo, 1; Fistulana, 1. Mollusca. Dentalium, 4; Patella, 1 ; Emarginula, 2; Pile- opsis, 1 ; Helix, 1 ; Auricula, 3; Paludina, 1 ; Ampullaria, 2; Nerita, 1 ; Natica, 2 ; Vermetus, 4 ; Delphinula, 1 ; Solarium, 1 ; Cirrus, 4; Pleurotomaria, 3; Trochus, 8; Turbo, 4; Tur- ritella, 1 ; Cerithium, 1 ; Pyrula, 2 ; Fusus, 1 ; Murex, 1 ; Pte- roceras, 1 ; Rostellaria, 5 ; Strombus, 1 ; Cassis, 1 ; Dolium, 1 ; Eburna, I ; Nummulites, 2 ; Lenticulites, 2 ; Lituolites, 2 ; Planularia, 2 ; Nodosaria, 2 ; Belemnites, 7 ; Nautilus, 7 ; Scaphites, 2 ; Ammonites, 50 ; Turrilites, 6 ; Baculites, 5 ; Hamites, 21. Crustacea. Astacus, 4; Pagurus, 1 ; Scyllarus, 1 ; Eryon, 1; Arcania, 1; Elyaea, 1; Coryster, 1; Orythia, 1. Pisces. Squalus, 3 ; Muraena, ] ; Zeus, I ; Esox, 1 ; Sal- mo? 1 ; Amia? 1. Reptilia. Mososaurus, 1; Crocodile, 1. 302 Cretaceous Group. Thus making, Plants, 5 genera, 17 species. Zoopliyta, 35 genera, 146 species. Radiaria, 14 genera, 85 species. Annulata, 1 genus, 30 species. Cirripeda, 1 genus, 2 species, Conchifera, 48 genera, 300 species. Mollusca, 40 genera, 167 species. Crustacea, 8 genera, 11 species. Pisces, 6 ge- nera, 8 species. Reptilia, 2 genera, 2 species. Total, 160 genera, 768 species. Fossil vegetables are by no means common in the mass of the true or white chalk, and those that are found are stated to be principally marine. The distribution of vegetable remains would appear to be very unequal in the lower parts of the group; for while vegetable matter has been so abundant in some places as to constitute coal beds, at others traces of ve- getables are exceedingly rare. Dicotyledonous wood, pierced by some boring shell, seeming to show that it had been drifted about, is not rare in the green sands of Dorsetshire. The reader will observe that the genera Ammonites, Sea- phites, Hamites, Turrilites, Baculites, and Belemnites, are now first introduced into the lists of organic remains ; these genera not having as yet been noticed in the supracretaceous rocks. It was once considered that the genera Scaphites, Hamites, Turrilites, and Baculites, were confined to the series under consideration ; but though their species may be more abundant in the cretaceous group, they are not confined to it; for, as will be seen in the sequel, Hamites and Scaphites are found in the oolitic group. Moreover, a Turrilite has been men- tioned, though with doubt, as occurring in the Coral Rag of the North of France. The presence therefore of these genera in distant places may not be alone sufficient to identify the rocks containing them with the cretaceous group ; yet if the species are in any abundance, our present knowledge would lead us to suspect that such deposits might be contempora- neous with the cretaceous series. If we reason from the ana- logy of the existing state of things, there is nothing to oppose the inference that the same genera may equally characterize contemporaneous deposits in North America and in Europe ; for according to Dr. Morton, several species are now common to the shores of Europe and the United States. Dr. Morton considers that rocks equivalent to the creta- ceous group do exist somewhat extensively in North America. He has named it the Ferruginous Sand Formation of the United States, and describes it as occupying " a great part of the triangular peninsula of New Jersey, formed by the Atlantic, and the Delaware and Raritan rivers, and extending across the state of Delaware from near Delaware city to the Chesa- peak : appearing again near Annapolis, in Maryland ; at Lynch's Creek, in South Carolina ; at Cockspur Island, in Cretaceous Group. 303 Georgia ; and several places in Alabama, Florida, &c." In New Jersey there is a very extensive development of marl. Taken as a mass, the deposit varies considerably in its mine- ralogical character ; most frequently presenting itself in minute friable grains, with a dull blueish or greenish colour, often with a grey tint. The predominant constituent parts of this marl, as it is termed, are described as silica and iron. There are subordinate beds of clay, of siliceous gravel, (the pebbles varying in size from coarse sand to one or two inches in di- ameter,) and calcareous marl. The marl is sometimes yel- lowish brown and filled with green specks of silicate of iron, and sometimes contains a considerable quantity of mica. The following is a list, according to Dr. Morton, of the organic remains found in this deposit, and described by Mr. Say, Dr. Dekay, and himself*. Ammonites Placenta, Dekay ; A. Delawarensis, Morton ; A. Vanuxemi, Morton ; A. Hippocrepis, Dekay ; Baculites ovatus, Say ; Scaphites Cuvieri, Morton ; Belemnites America- nus, Morton, abundant, (allied to B. mucronatus); B. ambiguus, Morton; Turritella; Scalaria annulata, Morton; Rostellaria; Natica; Bulla? Trochus; Cypr&a (cast) ; Terebratula Har- lani, Morton; T.fragilis, Morton; T. Sayi, Morton; Gry- pli&a convexa, Morton ; G. mutabilis, Morton, (some varieties of this species closely approach Ostrea vesicularis. Lam.); G. Fomer, Morton; Exogyra costata, Say; Ostrea falcata 9 Morton; O. Crista-Galli , Ostrea, two other species ; Anomia Ephippium ? Lam. ; Pecten quinquecostatus, Sow. ; Pecten, another species; Plagiosloma,- Cardium; Cucullcca vulgaris, Morton; Cucullaca^ another species; My a,- Trigonia? Tel- Una; Avicula; Pectunculus; Pinna, resembling P. tetragona, Sow.; Venus; Vermetus rotula, Morton ; Dentalium Serpula ; Spatangus Cor-anguinum? Park.; Sp. Stella, Morton ; Anan- chytes cinctus, Morton ; An. Jimbriatus, Morton ; An. ? cru- cifer^ Morton; Cidaris? Clypeaster. Crustaceous remains: Anthophyllum atlanticum, Morton. Eschar a; Fluslra; Mete- pom, resembling R. clathrata, Goldf. ; Caryophyllia; Alcyo- niwn; Alveolites. Teeth and vertebrae of the shark. Sauro- don Leanus, Say. Remains of the Crocodile (frequent) ; of the Geosaurus; of the Mososaurus (Sandy Hook and Wood- bury, New Jersey); of the Plesiosaurus ; of a Tortoise; and of some gigantic animal. Lignite pierced by the Teredo, abundant. It is almost impossible not to be struck, in the foregoing list, with the great zoological resemblance of this ferruginous * Say, American Journal of Science, vol. i. and ii. ; Dekay, Annals of the New York Lyceum ; and Morton, Journal of the Acad. of Nat. Sciences of Philadelphia, vol. vi. ; and American Jour, of Sci. vol. xvii. and xviii. 304 Wealden Rocks. sand deposit with the cretaceous rocks of Europe. The Pec- ten quinquecostatus is a well known and widely distributed cretaceous fossil. But it is not so much by individual parts as by the general character of the whole, that Dr. Morton's inference seems in a great measure established. How far the cretaceous group of the United States may be separated be- neath and above from other deposits more or less contempo- raneous with those in Europe, remains an interesting problem, which it is hoped that Dr. Morton and other American geo- logists will endeavour to solve. From some notices scattered through the memoirs of Dr. Morton and other authors, it would seem far from improbable that the cretaceous rocks may pass into the supracretaceous group. Assuming that the American ferruginous sand formation be- longs to the group under consideration, of which there seems great probability} it would appear that the great white carbo- nate of lime deposit, or chalk, did not extend there, but that a series of sands, marls, clays and gravels, constituted the whole group. How far the marls or clays may be altogether mechanical is perhaps uncertain ; but the gravel would seem to attest the former presence of water, moving with some ve- locity, for the pebbles even attain one or two inches in dia- meter. WEALDEN ROCKS. SYN. Weald Clay, (Argile Veldienne, Al. Brong.; Wealdthon, Germ.) Hastings Sands, (Iron Sand ; Sable Ferrugineitx ; Kurzawka of Poland.) Purbeck Beds, (Calcaire Lumachelle Purbeckien, Al. Brong.) These rocks, characterized in England by the presence of abundant terrestrial and fresh-water remains, occur beneath the lower green sand of the English series. The Weald clay, which constitutes the upper part of the rocks under considera- tion, does not present a clear line of separation from the ma- rine deposits above it ; the lower part of the one and upper portion of the other alternating, according to Mr. Murchison* and Mr. Martin f, in the western part of Sussex; an impor- tant fact, as it shows that the change of circumstances, which permitted the residence of marine animals over a surface pre- viously only covered by fresh-water animals, was not sudden but gradual J. Weald Clay. According to Dr. Fitton, (to whom we are * Murchison, Geol. Trans. 2nd series, vol. ii. t Martin, Geol. Mem. on Western Sussex, 1828. J For particular descriptions of the Wealden rocks of Sussex, and their organic contents, the reader should consult the various works of Mr. Man- tell : Illustrations of the Geology of Sussex ; Illustrations of Tilgate Fo- rest, &c. Wealden Rocks. 305 indebted for our knowledge of the nature of the Wealden rocks of England, which were previously confounded with the marine argillaceous and arenaceous beds beneath the chalk,) this clay is composed in the Isle of Wight, where there are fine sections of it, of slaty clay and limestone, with beds of iron-stone ; the laminae of the clay frequently coated with the remains of Cypris Fdba, Desm.* Mr. Martin defines the clay of the Weald of Sussex (whence the name,) as " a stiff clay, brown on the surface, and blue and slaty beneath, containing concretionai iron-stone f." It appears that the iron-stone was once worked, and slags from the ancient furnaces are found in different situations. The thickness of the clay is estimated at 150 or 200 feet in western Sussex, Beneath this there is an alternation of clays and sands, including the lime- stones full of the Patudina vivipara, and known as the Pet- worth marble. Hastings Sands. The following is the order, according to Dr. Fitton, of the beds of Hastings sands, in Sussex, &c. 1. Ferruginous and fawn-coloured sands, and sandstone, in- cluding small portions of lignite, with stiffgray loam. 2. Sand- stone. 3. Sandstone containing concretionai courses of calci- ferous grit. 4. Dark-coloured shale. 5. White sandstone of Hastings cliffs, 100 feet. 6. Clay, shale, and thin beds of sandstone, containing lignite and silicified wood (Endogenites erosa). 7. Sandstone, without concretions; dividing into rhomboidal masses ; numerous veins of argillaceous iron ore, and of clay, approaching to pipe-clay at the lower part. 8. Dark-coloured shale, with roundish masses of sandstone, and several layers of rich iron-stone, thin layers of lignite, and innumerable fragments of carbonized vegetables J. The same author observes that the equivalent beds in the Isle of Wight are composed of sands and sandstones, " frequently ferrugi- nous, with numerous alternations of reddish and variegated sandy clays, and concretions of calcareous grit." There are certain local variations, which will be found de- scribed in the works treating of particular districts. The Hastings beds, however, would appear, as a mass, to be prin- cipally arenaceous. According to Mr. Mantel], the lower part of the Hastings deposits (the Ashburnham beds,) are composed of argillaceous limestone alternating with schistose marls, which are probably connected with the following. * Fitton, Ann. of Phil. 1824. f Martin, Geol. Mem. Western Sussex. J Fitton, Geological Sketch of the Vicinity of Hastings, 1833. Dr. Fitton considers the Tilgate heds, so well known from the remarkable fossil animals discovered by Mr. Mantell. as belonging to the upper part of the Hastings Sands. Ibid. p. 44. Fitton, Ann. of Phil. 1824. X 306 Wealden Rocks. PurlecJc Beds. These are composed of various limestone strata, alternating with marls, many of the former being ex- tensively used for the pavement of London. Mr. Webster observes, that at Warbarrow Bay, Lulworth Cove, and other places on the coast of Dorsetshire, the tipper bed of the Pur- beck strata, supporting the Hastings Sands, contains a large proportion of green earth, the calcareous matter being ap- parently derived from the fragments of a bivalve shell. From the lists of organic remains found in the Wealden rocks, it will appear that this deposit of limestones, sands, sandstones, and clays, was formed in water which permitted the existence of shells analogous to those which now live in fresh water. With these are discovered the remains of estuary animals, and a few shells (Corbula, Tellina, Build), which may be considered as marine, the species of the analogous genera of the present day inhabiting sea coasts. The presence of the latter will not, however, invalidate the general evidence in favour of a lake, river, or estuary ; for not only may these shells have been introduced accidentally, but the animals in- habiting them may also have been gradually accustomed to live in fresh or estuary waters, as is the case in the pre- sent day with the species of some genera usually considered marine. It would appear that the dirt-bed, first noticed by Mr. Web- ster in 1;he Isle of Portland, and which has since been observed in the vicinity of Weymouth and elsewhere, commences the phenomena which attest dry land, succeeded by submersion of the same land beneath fresh or estuary waters, in which the whole of the Wealden rocks of south-eastern England were formed ; not suddenly, for there are no conglomerates to mark a possible state of violence ; but quietly, the shells being tran- quilly enveloped by the calcareous, argillaceous, or arenaceous matter which now entombs them. It will be seen that the oolitic group, immediately preceding this state of things, was, judging from the nature of the organic remains, formed be- neath a sea. Therefore we must suppose a rise of the land, or depression of the sea, to such an amount as to permit the sea-formed rocks to become dry land, upon which Cycade- oidece and dicotyledonous plants of a tropical nature flourished. This land was then depressed ; but so tranquilly, that the ve- getable soil, mixed with a few pebbles of the subjacent rock, was not washed away; neither were the trees considerably displaced, but they were left much as we have seen other trees in the submarine forests which surround Great Britain in va- rious places, and occur on the coasts of France. Like them, also, the trees of the dirt-bed are found, some prostrate, others inclined, and others nearly in the position in which they grew ; Wealden Rocks. 307 the upright portions being partly included in the limestone strata above. The only difference in the trees in the dirt-bed^ and those in the submarine forests, would appear to consist in the tropical nature of those in the dirt-bed^ and the near ap- proach, if not the identity, of the submarine forest vegetation with that now existing in Great Britain and France. There is, therefore, nothing singular in the gradual depression of the land, so quietly as not to cause the removal of the trees and other vegetable matter, as this has happened at various pe- riods. Instead of the depression having been effected, in the first instance, beneath the waters of the sea, circumstances have so existed that it took place beneath fresh water, which gradually acquired sufficient depth to permit^ deposit of various mineral substances several hundred feet thick. The circumstances at- tending this deposit have not been constant. At first calca- reous matter was thrown down, with somewhat regular inter- ruptions, which introduced a sufficient quantity of argillaceous matter to produce marl. Although fresh-water and terrestrial animals were now imbedded, there would also appear to have been at least one time when the water near Weymouth and in the Isle of Wight was capable of supporting the life of oysters and cockles, and therefore at least brackish. After this first period, sands were accumulated in great abundance, and in them were entombed a great variety of land and fresh-water Tortoises, Crocodiles, Plesiosauri, Megalosauri, Hylaeosauri, and huge Iguanodons *, those monstrous terrestrial reptiles. These must have sported in the waters, or roamed along the banks of this lake or estuary, into which trees and different vegetables were drifted. A clay deposit crowns this succession of rocks, still however not showing any other than a fresh- water origin. How far we may consider the change of the relative level of sea to have produced a constant depression of the land, is uncertain; but be this as it may, the sea was des- tined again to cover the land and resume its empire, for above the last-noticed clay reposes the whole mass of the cretaceous rocks of south-eastern England, of marine origin. This change, like that which preceded it, was not sudden ; there are no marks of violence between the Weald clay and the green sand ; on the contrary, there is a passage of one into the other, an alternation of the two at their junction. There is every probability that the sea did not make a furious inroad over the land, but that there was a quiet and gradual change of level, as in the case of the dirt-bed. I shall not trace the subsequent changes that have taken place over this spot on * For descriptions of the remains of this creature, consult M antell, Phil. Trans. 1825, and Illustrations ofTilgate Forest, 1827. x 2 308 Wealden Rocks. the earth's surface, further than to remark, that the sea again disappeared (Isle of Wight), and fresh-water or estuary depo- sjts succeeded *. . These conclusions can scarcely be termed hypothetical, for they appear such, however remarkable, as may be considered honest deductions from the phenomena observed. The extent of the area over which the dirt-bed, or a con- temporaneous ancient soil, maybe traced, is very remarkable, when we reflect upon the various circumstances which must have combined to preserve such a surface of ancient dry land. Dr. Fitton notices an earthy bed in precisely the same geolo- gical position in the cliffs of the Boulonnois, and also in Buck- inghamshire and in the Vale of Warclour. It further appears that silicified wood is found in the bituminous bed from Bou- logne to Cap Gris-nezf. To form such a deposit as that we have been noticing would be a work of time, and therefore we may infer that equivalent formations were taking place elsewhere, the great operations of nature proceeding in their usual course. The fresh-water character of the deposit can only be considered accidental or local ; precisely as formations at the present day, though con- temporaneous, may be either marine or lacustrine. There- fore, even supposing various perpendicular movements in the land to have taken place extensively over certain portions of Europe, it does not follow that they should have produced a constant rise of that land above the surface of the sea. On the contrary, we may consider that such movements very fre- quently caused a mere change in the relative depth beneath the surface-water, and that all deposits in the course of forma- tion, and so circumstanced, partook of the marine character of the surrounding aqueous medium. The observations of MM. Graves J and PassyJ leave little doubt that beds of the same relative age with the Wealden rocks occur in the departments of the Oise and Seine Infe- rieure. The country usually known as the Pays de Bray, which runs N.W. from between Auneuil and Beauvais to and beyond Neufchatel, is a denudation in the midst of the great chalk district of that part of France, extending down to the beds of Kimmeridge clay. There are various sandstones and clays above the Kimmeridge clay, and beneath the chalk and a mass of the green sand series, containing a considerable * For further observations on these curious facts, accompanied hy sections of Portland, &c., consult a memoir on the Weymouth district by Dr. Buck- land and myself, Geol. Trans. 2nd series, vol. iv. f Fitton, Geological Sketch of the Vicinity of Hastings, 1833, p. 76. J Graves, Precis statistique du Canton d' Auneuil (Oise). Passy, Descr. G6ol. de la Seine Inferieure, 1832. Wealden Rocks. 309 number of vegetable remains, among which is the ' Lonchopteris Mantelli, well known as found in the Hastings Sands of Sussex. There would also appear to be a bed or beds of a limestone abounding in the remains of Paludintf, reminding us of similar beds in the WeaJden rocks of southern England. We may gather from the observations of MM. Graves and Passy, and from those of various members of the Geological Society of France who assembled at Beauvais in September, 1831 *, that though these Wealden rocks of the Pays de Bray con- tain abundant terrestrial, and some fresh-water remains, there are also numerous marine remains, characteristic of the green sand series. Dr. Fitton, who also notices the occurrence of contemporaneous rocks in the Boulonnois, suggests that the Wealden rocks of Sussex, the Boulonnois, and of the Pays de Bray may have been formed in a single estuary, the area in that case not being greater than that now occupied by some deltas. He, however, at the same time remarks that this, though a plausible explanation, should only be considered as provisional f. It certainly by no means follows that because these deposits should contain fluviatile shells, and are of the same age, they should necessarily have been produced in the same estuary, even when the shells have been observed within distances which might admit of this explanation. The estu- aries of the Thames and Seine are now in all probability the depositaries of mineral substances and organic remains which do not widely differ, and consequently in some future state of the world, when these deposits shall have been heaved above the ocean level and partially covered with other rocks, they would exhibit similar geological characters. M. Thirria describes a considerable superficial deposit of clay with pisiform iron-ore in the department of the Haute Saone, part of which he considers referrible to the green sand, and may be equivalent to the Wealden rocks. Above rocks which seem equivalent to the Portland beds of England, there are strata of sand and clay, apparently the denuded remains of a deposit, once more extensive, which has suffered aqueous destruction, the water mixing up portions of the removed strata with the bones of Bears and Rhinoceroses; so that the mass upon reconsolidation much resembles the mineralogical composition of the original beds. The following is a section of beds, which M. Thirria considers as in place, the list of fossils being increased by those which he discovered, also in place, in the department of the Haute Saone: 1. Unctuous green clay; 2. Fine and slightly argillaceous yellow sand; 3. Nodules of yellow limestone contained in greenish clay ; * Bulletin de la Soc. Geologique de France, t. ii. t Fitton, Geological Sketch of Hastings. 310 Wealden Rocks. 4. Yellow and slightly argillaceous sand ; 5. Greenish-yellow and unctuous clay; 6. Greenish clay, with nodules of marly limestone and grains of iron ore ; 7. Pisiform iron-ore, con- tained in an ochreous clay, with Ammonites binus, A. plani- costata, Sow., A. coronatus^ Schlot., and other species ; Ha- mites (new species) ; Ncriruzci; Cirrus; Terebratula coarctata, Sow., and other species; and P entacrinites ; 8. White marl, with nodules of greenish clay and concretions of marly lime- stone. The whole forming a thickness of about forty feet, and resting on beds considered equivalent to those of Portland *. The extraordinary mixture of fossils contained in the pisi- form iron ore is commented on by M. Thirria, who further remarks that the reniform pieces of ore sometimes contain the empty casts of Jura limestone fossils. In support of the opinion that some of these pisiform and reniform iron-ore beds are of contemporaneous formation with either the Wealden rocks or green sand and chalk of England, we may cite the observations of Professor Walchner on simi- lar beds near Candern in the Brisgau. He remarks, " that the reniform and pisiform iron-ore deposits in the vicinity of Candern belong to two formations of very different ages; one of which rests on a compact Jura limestone, apparently cor- responding with either the coral rag or Portland stone of the English. It is composed of a mass of sandy clay, containing reniform iron-ore in the lower, and pisiform iron in the upper part ; and at the same time spheroids of flint (silex) and jasper. The reniform ores, and the flints which accompany them, con- tain organic remains ; the former of Astreas and Ammonites, the latter of Pectines and spines ofddaris. The whole is cover- ed with the solid beds of conglomerate, more ancient than the molasse, or by the molasse itself. This iron-ore formation may be considered as one of the last of the Jura limestone (oolitic group), and it, without doubt, closely approaches the chalk ; perhaps it may be like the green sand, intermediate between the Jura limestone and the chalk f ." In further support of this conclusion, Professor Walchner quotes the remarks of MM. Merian and Escher, on parts of the Jura, both of whom describe a clay with pisiform or reni- form iron-ore, intermediate between the upper beds of the Jura limestone and the molasse (one of the supracretaceous rocks of Switzerland) ; but being sometimes wanting, so that the molasse rests directly on the Jura limestone. M. Merian states that, near Aarau, the ferriferous bed sometimes con- * Thirria, Notice sur le Terrain Jurassique du Departement de la Haute Saone ; Mem. de la Soc. d'Hist. Nat. de Strasbourg, torn. i. 1830. f Walchner, Sur les Minerais de Fer pisiforme et re"niforme de Candern tn Brisgau; Mem. de la Soc. d'Hist. Nat. de Strasbourg, torn. i. Wealden Rocks. 311 tains large angular fragments of the limestone on which it rests, as also nodules of flint and jasper ; angular fragments of the former containing organic remains, which are the same as those detected in the iron-ore itself. The same author ob- serves, that " the pisiform ore of Aarau is immediately covered by a sandstone and bituminous schist, passing into lignite, which sometimes clearly exhibits a woody texture." The schist, and its accompanying clays, contain an abundance of fossils, among which Planorbes and other fresh-water shells could be distinguished. M. Brongniart notices among the cretaceous rocks of the Isle d'Aix and the embouchure of the Charente, a marl, which he refers to the Wealden clay, containing nodules of amber, pieces of lignite and silicified wood, in which holes, formed by some perforating animal, are replaced by agates *. The latter fact agrees with the presence of pieces of silicified wood, occasionally of large size, found on the green sand of Lyme Regis, where the holes, formed by some perforating animal, are filled with chalcedony or agate; both examples appear- ing to show that the wood had drifted, and remained some time in the sea. According to Professor Pusch there is a ferriferous deposit in Poland, situated between the Jura limestone and the cre- taceous rocks, which may be considered as the equivalent to the Weald clay and iron sand (Hastings Sands) of England. The following is Prof. Pusch's account of these beds, which is too valuable to be abridged : " It fills the valleys (in Poland) of Czarna Przemsa as far as Siewirz, that of Mastonica, that of the Wartha from its origin at Kromolow towards Czensto- chau, and of the Liziwarta; extending across Higher Silesia to the Oder, and running up this river to the country of Ribnyk. It is composed of horizontal beds, often alternating and of little continuity, of a slightly calcareous and schistose clay, either blue or variegated, named kurzayoka; of a sili- ceous, quartzose, and compact conglomerate ; of a brown ferriferous sandstone ; of beds of loose sand, and of thin beds of white or variegated marly limestone. In the country of Kromolow, Poremba, and Siewirzce, this formation contains horizontal beds from six inches to fourteen feet in thickness, of a coarse coaly substance (moorkohl), often accompanied with bituminous wood and much pyrites. This combustible is little worked, as the deposit occurs in marshy valleys, but the want of wood may render it useful in the country between Pelica and Czenstochau. From Siewirz, the carbonaceous beds lose themselves on the north. Faint traces of them are * Tab. des Terrains, p. 218. 312 Wealden Rocks. found round Czenstochati, Krzepice, and Klobucho; while the unctuous and blue schistose clays are largely developed in these countries, with, as on the top of the carbonaceous de- posits, numerous beds of iron-ore, consisting of ranges of spheroidal nodules of compact argillaceous iron-ore, contain- ing numerous Ammonites, (especially Ammonites bifurcatus^) and bivalves, of the genera Cardittm, Venus^ Trigonia> San- guinolaria, &c., fossils which partly correspond with those of the Jura limestone. This ferriferous deposit abounds near Panki, near Krzepice, between this point and Wielun, and on the north of Upper Silesia. It furnishes iron for the foun- dries of Poremba, Miaczow, Panki, Zarki, and various places in Silesia, producing 50 per cent, of iron. A brown ferrugi- nous sandstone, agglutinated by hydrate of iron, covers the blue schistose clays, especially round Kozieglow, Panki, and Prauska *." The reader will at once perceive the great resemblance of this ferriferous deposit to that above noticed in the Jura; such resemblance being heightened by the occurrence of or- ganic remains, of which Ammonites constitute a portion, in the iron-stone nodules of both situations. There would appear to be little difficulty in considering this deposit, with M. Pusch, as the equivalent of the Wealden rocks of England, showing that where local circumstances did not interfere, and the de- posit continued to be effected beneath the sea, its zoological character marked a certain connexion with the oolitic group ; the species of animals existing during the formation of at least a portion of the latter rocks not being suddenly cutoff: thus exhibiting a zoological passage of the oolitic into the cretaceous groups, when local circumstances did not interfere, as they have done on the south-east of England. It is remarkable that, notwithstanding the different character of the organic remains, apparently entombed in beds of the same age, which would seem to point out deposits in different waters, iron-ore should be so common in the Wealden rocks of England, the Jura, and Poland. When the upper beds of the oolitic series formed dry land, and sustained vegetation in southern England, it seems rea- sonable to conclude that many parts of the land now consti- tuting Europe were similarly circumstanced ; and therefore contemporaneous deposits of various characters may have been produced in different situations; some, by the nature of their organic remains, marking the presence of large lakes, or the embouchures of considerable rivers : in fact, a state of things, during which there was a mixture of dry land, fresh waters, and sea in this part of the globe. Some cause, with which as * Pusch, Journal de Geologic, t. ii. Wealden Rocks. 313 yet we are imperfectly acquainted, subsequently produced a great change in the relative levels of sea and land, and the cretaceous rocks (chalk and green sand) became deposited over a very considerable area, one apparently extending over a much larger superficies than that in which the last-formed rocks of the oolitic series were deposited. 314. Oolitic Group. SECTION VI. OOLITIC GROUP. SYN. - Oolite formation, Engl. authors ; Calcaire de Jura, Calcaire Juras- sique, Fr. authors ; Oolitkenbildung, Jura/calk, Germ, authors. THIS group is, in the southern parts of England, composed of various alternations of clays, sandstones, marls, and lime- stones ; many of the latter being oolitic, whence the name oolitic series. At a very early period in the history of English geology, Mr. William Smith affixed names to various portions of this series, many of which are still employed by the geolo- gists of Europe. Several of the divisions and subdivisions are, undoubtedly, very arbitrary, and perhaps separate those things theoretically which nature has united; but their convenience seems proved by their very general adoption. In consequence of three great clay or marl deposits appearing to divide the series in the south of England into three natural groups, Mr. Conybeare has separated it into three systems, as follows, (the Purbeck beds only, for reasons before assigned, being omit- ted) : 1. Upper system, containing, in the descending order, a. Portland oolite ; b. calcareous sand and concretions ; c. an argillo-calcareous deposit, named Kimmeridge clay. 2. Mid- dle system, a. coral rag, and its accompanying oolites ; b. cal- careous sand and grit; c. Oxford clay. 3. a. Calcareous strata, (sometimes divided by clays or marls,) named corn- brash, forest marble, great or Bath oolite, and inferior oolite ; b. calcareo-siliceous sands, usually termed sands of the inferior oolite ; c. an argillo-calcareous deposit named lias. These three principal divisions, marked by argillaceous de- posits, have been traced to various distances, though their subdivisions have not been so readily identified. The extent to which a few fossil shells of each division can be observed, is also deserving of attention. Mr. Phillips distinguishes this group in Yorkshire into, a. Kimmeridge clay ; b. upper calcareous grit ; c. coralline oolite ; d. lower calcareous grit ; e. Oxford clay ; f. Kelloway rock (a name given to stony portions of the Oxford clay, near Kelloway Bridge in Wiltshire) ; g. cornbrash limestone ; h. upper sandstone, shale, and coal ; i. impure limestone (Bath oolite); A;, lower sandstone, shale, and coal; /. ferruginous beds (inferior oolite) ; m. upper lias shale; n. marlstone series; Oolitic Group. 315 and o. lower lias shale. It will be observed that these divisions do not very materially differ from those of the southern parts of England, except in the presence of certain shales and sand- stones containing coal, above and beneath a bed considered equivalent to the Bath oolite. These carbonaceous beds are stated to have a collective thickness of 700 feet, the supposed representative of the Bath oolite being abstracted. We are indebted to Mr. Lonsdale for a detailed and highly valuable account of the oolitic district of Bath, a district which, independently of other considerations, must always be inter- esting to British geologists from having been the scene of Mr. William Smith's early labours, and as having long con- stituted the type to which geologists directed their attention, when describing the oolitic series of Western Europe. M r Lonsdale divides the group into : a, Kimmeridge Clay (thick- ness unknown) ; b, Coral Rag, subdivided into Upper Calca- reous Grit, Coral Rag, and Lower Calcareous Grit (in all 190 to 130 feet) ; e, Oxford Clay, based on the calcareous sandstone named Kelloway Rock (thickness unknown); d, Cornbrash (a thin bed); e, Forest Marble (100 feet);^ Bradford Clay, which the author remarks should be united with the Forest Marble (40 to 60 feet); g. Great Oolite (40 to 125 feet); h, Fuller's Earth (about 14-0 feet); *, Inferior Oolite (130 feet); A-, Marlstone; /, Lias (280 to 290 feet)*. By a rigorous examination of the beds, from the marlstone to the cornbrash inclusive, in Gloucestershire, the same author was enabled to point out several important modifications of the oolitic rocks of the Bath district, even within that distance. It is remarked that in the South of Gloucestershire the inferior oolite "consists of nearly equal divisions of soft oolite and slightly calcareous sand ; but in the northern part of the county, the latter, for the greater part, is replaced by a yellow sandy limestone. The freestone beds, which are not to be lithologi- cally distinguished from those of the Great Oolite, gradually increase in number and thickness, from the neighbourhood of Bath to the Cotteswolds, east of Cheltenham, where they con- stitute the whole of the escarpment. This vertical importance is retained through the north of the country examined ; but to the eastward of the valley, ranging from Stow-on-the- Wold to Barrington, near Burford, a change takes place, both in the structure and thickness of the formation. The freestone beds are there replaced by strata of nodular coarse oolite, contain- ing numerous specimens of Clypeus sinuatus ; the sandy por- tion consists of only a thin bed, and the thickness of the whole formation is diminished from 150 to 50 feetf." Other impor- * Lonsdale, Geol. Trans., 2nd series, vol. iii. f Lonsdale, Proceedings of the Geol. Soc., Dec. 1832. 316 Oolitic Group. tant changes are remarked in the Fuller's Earth, Great Oolite, and Cornbrash, tending to show the variable nature of the oolitic subdivisions, even in short distances. In his memoir on the Bath district, Mr. Lonsdale also points out the thinning off of the Bath oolite in the vicinity of Norton, by. which the Bradford Clay and Fuller's Earth are brought into contact *. To this it may be added that the Bath Oolite is no more seen to the southward in England ; a thick bed of clay, probably the continuation of the two clays above noticed, separating the Forest Marble from the Inferior Oolite in Dorsetshire. The same author, in his memoir on the Gloucestershire oolites, establishes, by a close comparison of the lower part of the Great Oolite, as it exists at Burford, with the Stonesfield slate, so celebrated for its organic contents, that the latter, in- stead of being subordinate to the Forest Marble, is referrible to the lower part of the Great Oolite. Now this is a very important correction of an error, inasmuch as all conclusions at which we might previously have arrived, in our endeavours to trace the circumstances under which a particular part of the deposit might have been formed, would have been vitiated, by supposing the Stonesfield slate to occupy one portion of the series, when it really occurs in another. The oolitic series of Normandy presents a close analogy in its general, and even in some of its minor divisions, with those of southern England. Commencing with the vicinity of Havre, and extending our observations to the Cotentin, we find the following series : a. Kimmeridge clay, in which certain sandstones named Glos sandstones are subordinate ; b. lime- stone and oolitic beds, referrible, from their geological and zoological characters, to the coral rag ; c. a ferruginous and calcareous sandstone; d. Oxford clay; e. a series of beds, in- cluding the well-known Caen stone, and representing the forest marble and great oolite \f. inferior oolite ;g. liasj-. M. Boblaye divides the oolitic series of the north of France as follows J : a. beds referrible to the coral rag, (the highest of the oolitic series in the district); b. a sandy and ferruginous oolite; c. a series of beds representing the cornbrash, forest marble, and great oolite; d. ferruginous limestone, micaceous marls, and sandy lime- stones, equivalent to the inferior oolite and its sands ; e. lias. * Lonsdale, Geol. Trans. 2nd series, vol. iii. p. 254, where there is also a section representing the manner in which the Bath oolite fines off. f De la Beche, Geol. Trans, vol. i. 1822 ; De Caumont, Essai sur la Topo- graphic Geog. du Calvados, 1828 ; Herault, Tableau des Terrains du Calva- dos, Caen, 1832. \ Boblaye, Sur la Form, Jurassique dans le Nord de la' France; Ann. des Sci. Nat. 1829. Oolitic Group. 317 In Burgundy, M. Elie de Beaumont, who has remarked on the constancy of the geological facts observable in the oolitic belt of the great geological basin which contains London and Paris, has found beds which he considers referrible to those of Portland, beneath which is a marly limestone with the Gryplioea Virgula, a remarkable shell of the Kimmeridge clay, particu- larly in France. These beds are succeeded by compact earthy or oolitic limestones, beneath which is gray marly limestone, supposed equivalent to the Oxford Clay. This is followed, in the descending order, by a series of oolite and other beds, be- neath which there is a limestone remarkable for containing an abundance of Entrochi, and considered equivalent to the infe- rior oolite, under which are rocks corresponding with the lias*. M. Thirria describing the oolitic series of the department of the Haute Saone, where it constitutes the north-western limits of the Jura, notices the following beds (the lias being excluded from the list according to the views of some of the continental geologists) : a. inferior oolite, composed of various limestones, oolitic, sublamellar, lamellar, and compact, reddish, gray, and yellow ; some of the beds being studded with Entrochi, or joints of Crinoidea. One bed is remarkable for oolitic hydrate of iron, so abundant as to be worked for profitable purposes at Calmontiers, Oppenans, Jussey, and other places; b. a yellow marl, considered equivalent to the Fuller's earth of the English (two yards thick); c. great oolite, composed of oolitic beds, containing among other shells Ostrea acuminata and Avicula echinata ; d. limestones with much red oxide of iron, schistose, suboolitic, or compact, considered equivalent to forest marble ; e. marly limestone, gray or yellowish, full of oolitic grains, supposed equivalent to the cornbrash of England \f. schistose blackish gray marls with marly limestone, resting on gray schistose marls containing oolite grains of hydroxide of iron, worked for profitable purposes in the districts of Orrain and Saguenay. The whole of this subdivision,^ is based on dark gray and schistose argillaceous limestone, and contains many fossils, particularly in the ferruginous oolite, among which is Gryphaa dilatata, a very characteristic shell of the Oxford clay, to which, and to the Kelloway rock, the whole is referred; g. a series of clay and limestone beds, the latter mostly oolitic; the upper part containing Corals, and the lower portion num- bers of Nerirwete, the whole considered equivalent to the coral rag; h. gray marls and marly limestone, based on compact gray limestone, the latter containing abundant remains of * Elie de Beaumont, Note sur I'uniformite qui regne dans la constitution de la Ceinture Jurassique qui comprend Londres et Paris; Ann. des Sci. Nat. 1829. 318 Oolitic Group. Astarte, while the other parts present the Gryplicca Virgula ; these marls are consequently referred to the Kimmeridge clay ; i. various limestone beds, principally of a gray colour, some- times whitish and yellowish, at others of a deeper tint, consi- dered equivalent to the Portland stone*. M. Thurman divides the oolitic series of the central part of the Jura, named the Porrentry (the ancient Eveche de Basle, and the present Bernese Jura) into groups similar to those which have been formed in Normandy and southern England. a. fine oolites and various compact limestones (considered equivalent to Portland stone), 65 feet ; b. yellowish marls and marly limestones with Grypliaa Virgula (Kimmeridge clay), 50 feet; c. compact limestone, with Astarte minima (Phil.) 5 100 feet ; d. compact or cretaceous white limestone with Nerencete, 65 feet ; e. oolitic and pisolitic limestone, 65 feet ; f. compact gray polypiferous limestone, 18 feet (c, d, e, andyj are regarded as equivalent to coral rag); g. marly and sandy limestones, with the concretions named cliailles (calcareous grit), 75 feet; h. blue marls, smoke gray compact limestones, and ferruginous oolite (Oxford clay), 50 feet; i. oolitic lumachella limestone, 20 feet; k. reddish sandy limestones and marls, 30 feet; /. fine- grained oolite (considered equivalent to the great oolite), 18 feet ; m. marls and suboolitic limestone, with Ostrea acuminata (Fuller's earth), 13 feet ; n. subcompact oolite, 120 feet ; o. fer- ruginous oolite, 20 feet (n, and 0, considered as equivalent to the inferior oolite) ; p. reddish green and micaceous sandstones and marls (marly sandstone), 18 feet. The whole based on lias. M. Thurman observes that the thickness of the rocks here enumerated is often more considerable, and points out the Mont Terrible as exhibiting an excellent section of nearly the whole series f. M. Dufrenoy, in his remarks on the rocks of this age which occur in the south-western parts of France, divides the oolitic group into three distinct systems ; admitting, however, at the same time, that these divisions are not well pronounced, the beds which apparently correspond with the Oxford and Kim- meridge clays being replaced by marly limestone. He further observes, that "the numerous subdivisions noticed by the English geologists are but very imperfectly seen in the secon- dary basin under consideration; some, nevertheless, being sufficiently constant." The lower portion rests on lias, and is composed of micaceous marls, with Gryphtea Cymbium, Be- hmnites, and other shells, which, as he observes, may be re- ferred to the sands of the inferior oolite. There are beds of * Thirria, Notice sur le Terrain Jurassique du D6partement de la Haute- Saone; Mem. de la Soc.d'Hist. Nat. de Strasbourg, 1830. f Thurman, Essai sur les Soulevemens Jurassiques dn Porrentry, 1832. Oolitic Group. 319 limestone with oolitic iron, and oolites, considered equivalent to the Bath oolites, the latter only well developed at Mauriac, Aveyron. This lower division is represented as of consider- able thickness. Above this there is a system of marly lime- stone beds, in some places associated with considerable masses of polypifers and thick beds of irregular and earthy oolite (Marthon, forest of la Braconne, and other places). M. Du- frenoy infers, from the great abundance of the corals, the pre- sence of the oolite and many fossils, that these beds are equi- valent to the coral rag and Oxford oolite. Upon this system, rests another, composed of marls and marly limestone, abound- ing in the Gryphaa Virgula^ supporting an oolite (from the environs of Angouleme to the ocean), in which this gryphite is also found. These rocks are referred to the Kimmeridge clay and Portland oolite respectively, and are stated to be surmounted by rocks of the cretaceous group *. It would thus appear, that throughout a considerable por- tion of France and England, and in the Jura, the causes which have produced the deposit of the oolitic group have not varied materially. Before, however, we attempt any remarks on this apparent uniformity of mineralogical structure over a consi- derable area, it will be necessary to present a sketch of this deposit in Scotland, Germany, and Sweden. Our knowledge of the oolitic group of Scotland is more par- ticularly due to Mr. Murchison. The coal deposit of Brora, in Sutherlandshire, has been shown to correspond with the carbonaceous series of Yorkshire, described by Mr. Phillips as occurring between the inferior oolite and cornbrash, and including in its central part a rock considered equivalent to the Bath or great oolite. In the vicinity of Brora there would appear to be various sandstones and shales, containing coal and vegetable impressions. The freestone of Braambury and Hare hills is described as covered by a rubbly limestone, "an ag- gregate of shells, leaves, stems of plants, lignite, &c." Mr. Mur- chison considers the organic remains of this bed, and the casts in the freestone, as referrible to such as occur in the lower pnrt of the coral rag. At Dunrobin Castle calcareous sand- stones are succeeded by beds of "pebbly calciferous grit," covered by shale and limestone with fossils. Other varieties of this oolitic deposit occur on this coast, which consists, in the descending order, of rubbly limestone, white sandstone and shale, shelly limestone, sandstone, shale, and limestone, with plants and coal, considered the same with the Yorkshire car- bonaceous deposit. This oolitic deposit is not confined to the main land of Scot- * Dufrenoy, Annales des Mines, torn. v. 1829. 320 . Oolitic Group. land, but is found in the Hebrides. According to Mr. Mur- chison, it occurs at Beal near Portree, Sky, the higher part presenting a calcareous agglomerate of fossils, resembling many portions of the English cornbrash and forest marble : it is identical with the shelly limestone of Sutherland, above no- ticed. At Holm the sandstone rises to a considerable height from beneath the limestone. Impressions of plants are found in the sandstone on the north-east of Holm. Near Tobermory in Mull, sandstone, considered as equivalent to that of the in- ferior oolite, rests on lias, containing the Gryph&a incurva. It also appears that rocks of the oolitic series, including lias, occur in other parts of Mull, the opposite coast of Ross-shire, and in the islands of Rasay and Pabbla, often cut and covered by trap rocks*. M. von Decken observes, that the oolitic group of northern Germany, which occurs extensively from Bramsche on the Haase to Minden on the Weser, and thence to the country near Hildesheim and Eimbeck, as also northwards of the Hartz between Wolfenbiittel and Helmstadt, approaches in its cha- racters to the same series of Yorkshire and some parts of Scot- land. Marls and sandstones predominate, and the oolitic lime- stones are confined to subordinate beds. Beds of coal accom- pany the sandstone, and are worked for economical purposes, more particularly at Obernkirchen (Buckeburg). The connexion of the upper part of this group with the su- perincumbent cretaceous series is not clearly seen, and fixed points are wanting to compare it with the English divisions. The inferior division, the lias, is on the contrary well developed. The black bituminous marls contain layers of bituminous limestone, and the marl itself is in some places (Essen, Osna- burg ; Ostercappeln ) used for slate pencils. A thick sandstone bed, of a dark brown colour, and traversed by stripes of brown iron ore, rests upon the lias in the countries on the Weser. It contains beds of dark gray slate clay, in which are nodules of oolitic brown ironstone, and may be considered as the lowest member of the inferior oolite. Above this there is an oolitic limestone (in the Weser chain, eastwards of Hildersheim, Ith, and Lauensteinberge), which is at first sandy, and then contains veins of chalcedony and chert. There are also beds of dark slate clay, marl, and yellow brown sandstone, which sometimes resemble grauwacke and the quartz rock associated with it. The latter predominates from Liibbecke to Bramsche. M. von Decken remarks, that it is by no means decided to which part of the English series this mass of rock, 700 feet thick, should be referred, and that probably the various opinions on * Murchison, Geol. Trans. 2nd series, vol. ii. Oolitic Group. 321 this head can only be settled by an accurate examination of the fossils contained in it. If, however, it be assumed that the sandstone and coal of Obernkirchen and Bb'hlhorst, cor- respond with the oolitic coal of Yorkshire, this mass ot rock would answer to that part of the English series comprised between the inferior oolite and the upper part of the great oolite. Above this limestone formation there rests, at Bb'hlhorst and the Biickeberg, a mass of black marl 4-00 feet thick; then sand- stone, 200 feet thick, containing beds of coal, surmounted by black slaty marl, covered by sand and loose gravel. On the Deister and Osterwald, the schistose marl above noticed is wanting, and the sandstone rests immediately on the oolitic limestone. The series of beds between these last rocks of the oolitic group and the lias seem nowhere interrupted in northern Germany. Pursuing the comparison with the oolite coal of Yorkshire, the schistose marl above noticed (the highest mem- ber of the oolitic group of the Weser chain,) should be con- sidered equivalent to the Oxford Clay*. The oolitic series of southern Germany forms the immediate continuation of the Swiss Jura towards the N.E., cut through by the Rhine at Schaffhausen. These rocks extend to Sieg- maringen on the Danube, from whence they follow the left side of the valley of that river. They constitute the plateau known by the name of the Swabian Alps. The oolitic group extends northwards from Ratisbon to the Maine, to Banz, Lichtenfels, Staffelstein, and in detached portions to Coburg. On the east of this range the older strata rise on the slopes of the Bohmerwald and Fichtelgebirge. Some difficulty has attended the comparison of these rocks with the English and French divisions of the oolitic group. The lias is so completely de- veloped, and so similar to that in England, that its identity has long been placed beyond all doubt. Several divisions may be observed^in it, characterized by their organic remains. The lower portion contains numerous beds of limestone, and among other organic remains the Gryphcea incurva, Sow. Above this reposes aluminous marl, and dark smoky marl, remarkable for the prodigious number of Belemnites discovered in it, as also for containing the Gryphcea Cymbium, Lam. These are surmounted by black shales, with Posidonice, Fishes, and Sau- * M. Hoffman (Uebersicht der orog. und geogn. Verhaltniss evom N.W. Dentsehland,) compares this marl to the Weald Clay, the sandstones with coal of the Biickeberg to the Hastings Sands, and terminates the oolitic group with the marl beneath. The Kahlenberg near Eichte (on the N.W. flank or' the Hartz,) affords, however, sufficient proof of the age of these beds. Many fossils correspond with those of the coral rag, while some are referrible to the Kimmeridge clay and Portland stone. Von Dechen, German Transl. of Manual. 322 . Oolitic Group. rians. The sandstone with clay iron-stone at Aalen and Was- seralfingen, has been sometimes considered as forming part of the lias; it has, however, been shown by Count Minister to represent the inferior oolite of England *. The mass of the compact light-coloured Jura limestone succeeds, but from the want of marked differences in the compactness of the strata, sharp escarpments do not occur. Calcareous sandstone, with a bed of blue clay, rests on the ferruginous oolite of Wasse- ralfingen. This may comprise the rocks of the English series up to the Kelloway rock, represented by a repetition of ferru- ginous oolite in the succeeding clay, equivalent to the Oxford Clay, and containing the Grypheea dilatata. Upon this rests white marl and white compact limestones, the representative of the coral rag of England, more developed, but containing numerous characteristic fossils. A large portion of country is formed of these beds, and there is no clay stratum above them, which might be considered analogous to the Kimmeridge clay. The lithographic slates of Bavaria take their place above them, and contain such an extraordinary mixture of organic remains that they may be considered as local, not constituting an ex- tended bed which can be identified in distant places. These lithographic slates of Pappenheim, Solenhofen, and Monheim near Eichstadt, form the upper part of the oolitic group of Southern Germany; therefore their position in the series is doubtful. As, however, many fossils of the white limestone beneath occur in them also, they may be considered as not far removed from itf. In the whole range from the Danube to Coburg there are thick and extensive masses of dolomite, which take the place of a part of the white limestone (coral rag) . These masses are for the most part non-fossiliferous, and it is only in a few places that organic remains can be detected. They are immediately covered by the lithographic slates J. Von Buch was the first to point out that the coral rag con- stituted the elevated plateau between the Maine and Switzer- land, and that it was found in the mountains of Streitberg, at Donzdorf in Swabia, at Rathshausen near Bahlingen, and at Mont Randen near Schaffhausen. He observes, that at the latter place there are several beds of polypifers, in which Cne- midium lamellosum, Cn. striatum, and Cn. rimulosum, are the most characteristic fossils. Beneath these are beds full of * Munster, Uber den Oolithischen Thoneisen stein in Siid-Deutschland. f It is understood that M. von Buch is preparing a detailed account of the oolitic series of Southern Germany, which will no doubt afford us a mass of valuable information and points of comparison with the English portion of the same series. 1 German Transl. of Manual. Oolitic Group. Ammonites, such as A. placatilis, A. triplicatus, large and very abundant, A. perarmatus, A. biplex, A.jlexuosus, A. bifurcatus, and A. canaliculatus. These coral-rag beds rest on clays and marls, containing the gryphaa dilataia and Ammonites sub- lavis*. The List of Organic Remains will show that polypifers are abundant on this rock at Streitberg, Muggendorf, &c. M. Merian has afforded us very valuable details respecting the structure of the Jura near Bale, and of its continuation into Germany in the same vicinity ; whence it appears that the inferior oolite (Eisen Rogenstein) and the lias (Gryphiten Kalk) constitute clearly marked rocks of the series. The beds which rest on the Eisen Rogenstein are divided into older and newer Jura limestone (Alterer Rogenstein and Jungerer Jurakalk\ the former being considered in a great measure equivalent to the Great or Bath oolite, and separated from the latter by beds of clay f . So far, if we except the dolomite in Germany, we have found no great change in the oolitic group, taken as a mass : there is nothing which shows that in the particular parts of Europe above noticed any forces were called violently into ac- tion during its deposit. On the contrary, a greater or less degree of repose seems characteristic of it, as also the presence of a large proportion of calcareous matter. The lowest por- tion, or the lias, preserves certain general characters over a considerable area ; and why some geologists have separated it from the oolitic series is not easily understood ; for if an appa- rent passage into the rocks beneath in some situations be the reason, such a reason would hold equally good for not sepa- rating it from those above, into which it also passes: if its zoo- logical character be brought forward, there can belittle doubt that throughout Western Europe this would place it in the group under consideration. The lias of Western Europe may be considered, taken in the mass, as an argillaceous and calcareous deposit, in which sometimes one substance predominates, sometimes the other; sometimes presenting a great abundance of marls or clays, at others of limestones : the latter are however generally most common in the lower portions of the rock. In the Vosges di- strict the lower part of the lias is formed of a sandstone, de- scribed by M. Elie de Beaumont as yellow and quartzose, con- taining mica, a few flattened argillaceous nodules, and small * Von Buch, Recueil de Planches de Purifications Remarquables, Berlin, f Merian, Geognostischer Durchschnitt durch das Jura-Gebirge von Basel bis Kestenholz bey Aarwangen ; Denkschriften der allgemeinen Schwei- zerischen Gesellschaftfur die gesammten Naturwissenschaften. Zurich, 1829. Y 2 324 Oolitic Group. white or black quartz pebbles*. The presence more particu- larly of the pebbles seems to point to a transport by water. This sandstone extends into the neighbouring parts of Ger- manVj and is one of those to which the name of Qiiader sand- stein has been applied. Beneath the oolitic group which comes into contact with the granitic rocks of central France, M. de Bonnard has described an arenaceous rock, which he has named Arkose, and which may represent the arenaceous beds consti- tuting the lowest part of the same rocks in the district of the Vosges. M. Dufrenoy describes an arenaceous deposit cor- responding in geological position and external characters with the arkose of M. de Bonnard in the south-western part of France. He also states, that from Chatre, where the coal- measures terminate, to beyond Brives, the separation of the oolitic series and the granitic rocks is marked by the presence of this sandstone, composed of quartz grains and felspathic portions, cemented by matter generally marly, but sometimes siliceous; the silica in the latter case becoming sometimes so abundant as to obliterate its character of a sandstone, so that it passes into a jasper. This sandstone seems to pass into the lias limestone, presenting an arenaceous limestone between the two. M. Dufrenoy considers it as the inferior sand of the lias. The same author describes the lias of the south-west of France ; and states that it contains masses of gypsum. Although sulphate of lime, in the shape of crystals of selenite, is by no means un- common in the lias marls of other countries, its presence, in that form, does not appear to mark a chemical deposit so much as in the gypsum above noticed. Taken as a whole, the lias seems very persistent in its characters throughout a consider- able part of France, England, and Germany, pointing to a somewhat common origin. In the lias of Lyme Regis, Dorset, there would appear evidences of slow deposit in some parts, while in others the animals entombed seem to have been sud- denly killed and preserved, so that the animal substances hail not time to decay. The ink-bags of fossil Sepia?, noticed by Prof. Buckland, afford perhaps the best evidence we can ad- duce of this fact; for had the animal substances which con- tained ink been exposed but for a short time to decomposition or the attacks of other animals, the ink must have flowed out of the bags. Now the actual forms of this fossil ink are pre- cisely those of the ink-bags found in the Sepicc and other ani- mals possessing organs of a similar description at the present day ; and therefore they appear to have been preserved entire and suddenly in a soft deposit. * Elie de Beaumont, Mem. pour servir a line Description Geologique de la France, torn. i. Oolitic Groitp. 325 In the lias of southern England and many parts of France and Germany, the calcareous matter has been more abundant in the lower parts ; and limestone beds have been the conse- quence, interstratified with marl, the latter sometimes schistose. Above the lias we have an arenaceous deposit, into which the marls graduate; and these sandy beds would seem to have been formed over a considerable area, embracing a large por- tion of France and England, and parts of Scotland and Ger- many. These are surmounted by limestones, one of which, characterized by the presence of oolitic iron-ore, though not precisely continuous, is remarkable for its occurrence in a si- milar part of the series, whether it be in the southern parts of England, in the north of France, in the Jura, or in some parts of Germany. Above these beds, termed the Inferior oolite, there is a series which varies much in its mineralogical charac- ter, presenting modifications of clays, marls, and limestones; the latter, which are often oolitic, affording beautiful materials for architectural purposes, as is seen in the towns of Bath, Caen, Nancy, and other places. This variety is commonly known by the name of the Bath or Great oolite, while other portions have received the names of Fuller's earth, Bradford clay, Forest marble, and Cornbrash. There can be little doubt that in tracing these supposed minor divisions over many parts of Europe, too much attention has been given to them as they exist in southern England and in Normandy, and that conclusions respecting their complete identity elsewhere have been somewhat forced. This is not the case with the next di- vision, one like the lias composed of argillaceous and calca- reous matter, known as the Oxford clay, which, with certain modifications, seems to extend through England, and over a considerable portion of France, including the Jura, into Ger- many. The next superior rock, termed Coral rag, (from con- taining in certain situations a great abundance of polypifers,) separating an argillaceous deposit termed Kimmeridge clay from the Oxford clay, seems also to have a wide range, and presents a mixture principally calcareous, and often oolitic, the grains being not unfrequently so large that the rock is named Pisolite. The Kimmeridge clay is also an argillaceous and calcareous mixture, which has a considerable range, par- ticularly over England and France. Its covering, or the beds termed Portland beds, seems very irregularly dispersed, the causes that produced the beds not being so constant as those which formed the clay beneath : it will however have been seen that rocks considered equivalent occur in the south-west of France, and in the Jura. As yet we have seen the oolitic group composed of nearly similar mineral substances, and abounding in organic remains. 326 Oolitic Group. In Poland, however, there would appear, according to Prof. Pusch, to be a change in the general mineral structure, pre- paring us for other greater changes, which will be noticed in the sequel. M. Pusch describes the lower member of the group under consideration in that country as more or less white and marly. On this rests dolomite, generally of a dazzling whiteness, affording the forms so remarkable in the rocks of this nature, and composing the picturesque country between Oldkusz and Cracow, and near Kromolow, Niego- womie, and other places, rising to the height of 1200 or 1400 feet above the sea. The upper part of the dolomitic limestone from Oldkusz towards Zarki, and especially near Wladowice, contains pisiform iron-ore ; it there becomes mixed with a coarse sandstone, and constitutes a problematical agglomerate and red sandstones. The upper portion of the group is formed of gray and oolitic limestones and calcareous agglomerates, and is represented as passing into the beds considered equiva- lent to the Wealden rocks. The rocks of the oolitic group are seen to rest unconformably on the coal-measures and muschelkalk of Poland ; and it is necessary to use some cau- tion not to confound them with the latter rock, when they are in contact, as at Oldkusz and Nowagora. Taken on the large scale, the Polish rocks of this age are stated to have a general direction N.N.W. and S.S.E. From Wielun they plunge beneath the great plain of Poland, here and there appearing in islands above it, and are considered to be its support, being met with in sinking through it. The organic remains con- tained in this deposit are stated to be such as to establish its identity with the oolitic series of other parts of Europe*. We have now to consider a series of equivalent deposits, with little or no mineralogical resemblance to those noticed above, occurring in the Alps, the Carpathians, and in Italy. Numerous memoirs have been written by different geologists, and some have even considered that certain minor divisions might be established ; but it must be confessed, though the evidence is greatly in favour of a considerable development of the oolitic group, with altered mineralogical characters, in the situations above noticed, that the termination of the group either above or beneath is far from possessing that clear and certain character which could be desired. The mineralogical character being so different, recourse has generally been had to organic remains; there are, however, such singular mixtures of these, in the Alps more especially, that the determination of particular deposits is far from certain. Instead of tender, soft marls, clays, sands and light- coloured limestones, we have * Pusch, Journal de Geologic^ t. ii. Oolitic Group. 327 dark-coloured marbles, masses of crystalline dolomite, gypsum, and schists approaching talcose and micaceous slates. The Alps are also particularly difficult of examination, as from the convulsions by which they have been upraised or otherwise visited, whole mountain masses are thrown over, and the rocks really deposited the latest occur beneath the older strata; and this not in limited spaces, but over considerable distances. These dark- coloured rocks were during the prevalence of the Wernerian theory referred, as was natural, to the transition class ; and we are indebted to Dr. Bucklandfor first pointing out that they were of more recent origin : since that time, other geologists have shown the probable relative antiquity of different portions ; and among these, M. Elie de Beaumont holds a distinguished place, particularly as respects Savoy, Dauphine, Provence, and the Maritime Alps. In a note on the geological position of the fossil plants and Belemnites found at Petit Cceur near Moutiers in the Tarentaise, published in 1828*, this author observes that the system of beds described by M. Brochant in his memoir on the Tarentaise, and which in many places contains considerable masses of granular lime- stone and micaceous quartz rock, as well as large masses of gypsum, belongs to the oolitic group. He is of this opinion, as he considers that the most ancient secondary rocks of that country, in which no fossil shells have been found that have not been discovered in the lower part of the oolitic series, can be traced to the environs of Digne and Sisteron (Basses Alpes), where they afford a great abundance of those remains supposed to be characteristic of the lias. In a notice on the geological position of the fossil plants and graphite found at the Col du Chardonnet (Hautes Alpes), M. Elie de Beaumont observes, that as the traveller quits the Bourg d'Oisans (Piedmont) and approaches the continuous range of masses, termed primitive, that extend from the Monte Rosa towards the mountains on the west of Coni, he will per- ceive that the secondary rocks gradually lose their original character, though certain distinguishing marks may still be seen, thus resembling a half-burnt piece of wood, in which the ligneous fibres may be traced far beyond the part that re- mains woodf. He has also remarked on the original differ- ences that may have existed between these secondary rocks of the interior of the Alps, and those in the same series of other countries; and thence concludes, that very little importance should be attached to the difference of mineralogical structure observed in the beds above mentioned, and in the lower part of the oolitic group, occurring undisturbed in other parts of * Annalcs des Sciences Naturelles, t. xiv. p. 113. t Ibid. 1828, t. xv, p. 3513. 328 Oolitic Group. Europe, and of which these Alpine rocks appear to him the enlarged prolongation. The vegetables found by M. Elie de Beaumont in the situations above noticed, were examined by M. Ad. Brongniart, and many were found by him to be ge- nerally the same with those discovered in the coal-measures. The following is a list of those which he obtained from the Alps, apparently all similarly situated as to geological position : Catamites Suckowii, Ad. Brong., at Pey-Ricard, near Briancon (also in the coal-measures of Newcastle and other places) ; C. Cistii, Ad. Brong., the same locality (also at Wilkesbarre in Pennsylvania); Lepidodendron, C 2 sp., Pey-Ricard and Pey- Chngnard, near Lamure ; Sigillaria, the above localities, and La Motte near Lamure; Stigmaria., Pey-Chagnard ; Neuro- pteris gigantea^ Ad. Brong., Servoz, Savoy (also in the coal- measures of Bohemia) ; N. tennifolia, Ad. Brong., Petit-Cceur, and Col de Balme (also in coal-measures of Liege and New- castle); N.Jlexuosci) Stern., La Roche Macot, Tarentaise (also coal-measures of Liege and Bath); N. Soretii, Ad. Brong., same locality ; N. rotundifolia, Ad. Brong., La Roche Macot, and Col de Balme (also in the coal-mines of Plessis, Calvados) ; Odontopteris Brardii, Ad. Brong., Petit-Cceur (also coal-mines of Terrasson, Dordogne) ; Od. obtusa, Ad. Brong., Col de TEcuelle, near Chamonix; Petit-Coeur (also at Terrasson); Pecopterispolymorpha*, Petit-Coeur (also in the coal-measures of St. Etienne, Alais, Litry, Wilkesbarre); Pe. pteroides, Ad. Brong., Pey-Chagnard (also in coal-measures at Liege, Man- nebach, St. Etienne, and Wilkesbarre) ; Pe. arboresccns. Ad. Brong., Val Bonnais, near Lamure; Petit-Coeur (also at Mannebach and Aubin, Aveyron); Pe . platyrachis 9 Ad. Brong., Val Bonnais (also at St. Etienne); Pe. Beaumontii, Ad. Brong., Petit-Coeur ; this new species is described as resembling the Pe. nervosa, Pe. bifurcata. Stern., and Pe. muricata^ Schlot., found in the coal-measures, and Pe. tennis^ found in the oolitic series of Whitby and Bornholm; Pe. Plukenetii? Petit-Coeur; Col de 1'Ecuelle (also at Alais) ; Pe. obtusa, Ad. Brong., Petit- Coeur (also in coal-measures near Bath) ; Asterophyllites cqui- setiformiS) Tarentaise (also at Alais and Mannebach) ; Annula- ria brevifolia., Col de Balme (also at Alais and Geislautern)f. These vegetable remains are so far associated with Belem- nites, that the latter occur both above and beneath them ; so that there can be no doubt as to the Bclemnites having existed previous to and after the vegetable deposit ; and therefore these localities would involve the question of the preference that should be given to the Belemnites or to the vegetables, if * This species is common in the coal-measures of France according to M. Ad. Broiigniart. f Ad. Brungniart, Ann. dcs Sci. Nat. vol. xiv. pp. 129, 130. Oolitic Group. 329 M. Elie de Beaumont did not appear certain that the same series of beds was continued to Digne and Sisteron, and there contained characteristic lias remains. M. Necker ck Saussure has described a series of beds that composes the upper part of the Buet (Savoy), and which con- stitutes the lowest calcareous deposit of that portion of the Alps, resting, like those above noticed at Petit- Cceur and the Col de Chardonet, on older and non-fossiliferous rocks. The following is a section, in the ascending order : 1. Mica slate, which may form part of the protogine rocks of this district. 2. A sandstone, formed of numerous grains of quartz, mixed with a few crystalline grains of felspar, and sometimes with a little talc or chlorite. 3. Red and green argillo-ferruginou!? schist. This rock is sometimes wanting in the section ; but on the east of the Vallee de Vallorsine it alternates with the well-known Vallorsine conglomerate, which is but a similar schist, filled with rounded pebbles of gneiss, mica slate, pro- togine, &c., among which we neither observe true granite nor limestone; an important fact, as is observed by M. Necker, for it appears to show that the Vallorsine granite, which cuts through the gneiss, did not exist before the formation of the conglomerate. 4. A black schist, with impressions of ferns, the vegetable remains being converted into thin talc*. 5. Black or dark bluish-gray limestone, filled with grains of quartz. 6. A black argillaceous schist, containing nodules of Lydian stone. Ammonites are found in this rock, as also in an ar- gillo-talcose schist which alternates with it. 7. A gray cal- careous and arenaceous schist, containing Belemnites^. The last bed constitutes the summit of the Buet, 10,099 English feet above the sea. It has been observed by M. Elie de Beaumont, that the cal- careous portions of these regions of the Alps are separated from the older and non-fossiliferous rocks by a sandstone more or less coarse, which passes into a conglomerate, seen not only at the Vallee de Vallorsine above noticed, but also at Trient, Ugine, Allevard, Ferriere, and Petit-Cceur. The same cir- cumstance is observable to the east of the Bourg d'Oisans and * When crossing and wandering over the Col de Balme in 1819, I picked up specimens of sandstone with impressions of plants upon them ; these plants I then considered, from their general character, to be such as are usually found in the coal-measures (Geol. Trans. 2nd series, vol. i. p. 162) ; an opinion which has since been confirmed by M. Ad. Brongniart, though it now appears that they may belong to a more modern deposit. t Necker, Mem. sur la Vallee de Vallorsine, Mem. de la Soc. de Phys. et d'Hist. Nat. de Geneve, 1828. For a section of the Buet, see the same Memoir ; and Sections and Views illustrative of Geological Phsenomena, pi. 27. fig. 5. 336 Oolitic Group. Huez, and in other places *. This evidence of the action of water possessing sufficient velocity to transport coarse sands and pebbles should be borne in mind, as, however such sands and pebbles may have been since altered in appearance, it shows that the deposits were not produced quietly; though subsequently, from a change of circumstances, and the esta- blishment of' comparative tranquillity, limestones were formed. These appearances are not confined to the Savoy and French Alps, but are seen on the shores of the Lake of Como and of the Gulf of La Spezia. The calcareous beds, of which such fine sections are afforded in the Lakes of Como and Lecco, are separated from the gneiss and mica-slate of the higher .Alps, by a conglomerate composed of rounded pieces of quartz, red porphyry, and other rocks, associated with sandstone beds. Monte del Nova. d I d a d c g m d, d, d, dolomite. 1 9 limestone. , gypsum included in do- lomite, c, red conglomerate and sandstone separating the calcareous and dolomitic rocks from the gneiss arid mica-slate. g 9 gneiss. ?, mica slatef. The limestone series incumbent on the conglomerate is in some situations strangely mixed with dolomite more or less crystalline, as will be noticed in the sequel. Taken as a mass, the limestones occupy a thickness of many thousand feet, and are more or less gray. They are siliceous, and contain seams of chert in the upper part (near Como), become slaty, with apparently little siliceous matter in their central parts, and are finally compact and more thickly bedded in their lowest situ- ations. Ammonites greatly resembling A. Bucklandi and A. heterophyllus are discovered in it, as are also Turritella, and other shells. Anthracite is here and there found. I have little doubt that the oolitic group is represented by at least a part of this calcareous mass; but how much, and what other equivalents there may be, my present information will not permit me to hazard an opinion. The general circumstances are however so similar, that it does not seem unreasonable to conclude that the causes, whatever they were, which produced * Elie de Beaumont, Ann. des Sci. Nat. t. xv. p. 354. f For a map, other sections, and a description of this district, see Sections and Views illustrative of Geological Phenomena, pi. 31, 32. Oolitic Group. 331 the Vallorsine conglomerates and the sandstone associated with them in that part of the Alps, were contemporaneous with those which formed the conglomerates and associated sandstones of the lakes of Como and Lugano. To present a detail of the various observations on those Alpine rocks which are considered as referrible to the oolitic group, would far exceed our limits ; the student will consult with advantage the various labours of Studer, Boue, Sedgwick, Murchison, Lill von Lillienbach, Lusser, and others. There may be occasionally some difference of opinion among authors, as to where the series may commence, or where it may end ; but the main fact, the existence of the group itself seems esta- blished beyond all doubt. When we consider the disturbed nature of the country to be examined, and the difficulty of attaining certain situations perfectly necessary to a right un- derstanding of the subject, except under very favourable cir- cumstances, we should be more surprised that so much has been accomplished in so short a time, than at finding discor- dant opinions on certain minor points. Mr. Murchison observes that, accompanied by M. Lill von Lillienbach, he found in the dark-coloured limestone and shale, at the gorge of the Mertelbach, below Crispel (Austrian Alps), Ammonites 2 species (one approaching A. Con&beari), Pec- ten 3 species, small Gryphcca, Mya, Penia 2 species, Ostrea, Corallines, &c. This group is referred to the lias. An overlying red encrinite limestone contains several species of Ammonites, and some Belemnites. According to Professor iSedgwick and Mr. Murchison, most of the salt-mines of the Austrian Alps are contained in the oolitic group (Halstadt, Aussee, &c.). The upper part of the oolitic series of this part of the Alps contains semi-crystalline, brecciated, compact, and dolomitic limestones*. I cannot conclude this sketch of the oolitic group, without adverting to certain limestones of La Spezia which may be referrible to it. On the west side of the celebrated Gulf of La Spezia, there is a range of mountains extending along the coast nearly to Levanto, their breadth augmenting as they advance N. W. The sections of these mountains expose the following rocks, easily observed up any of the cross valleys. The an- nexed wood-cut exhibits a section over Coregna. Coregna. Fig. 61. * Proceedings of the Geological Society, 1831. Phil. Mag. and Annals, vol. ix. 1831. 332 Oolitic Group. S. Gulf of La Spezia. M. Mediterranean, a. Limestone series : Upper beds compact and gray, varying in intensity of tint ; more or less traversed by calcareous spar ; here and there interstratified with schistose beds, and even argillaceous slate. The beds most commonly thick. The limestone with light-brown veins, so long known by the name of Porto Venere marble, forms part of these beds. b. Dolomite : varying in appearance; not unfrequently crystalline; when most so nearly white ; in some places beds may be distinguished, in others stratification cannot be traced, c. Numerous thin beds of dark-gray limestone, d. The same kind of beds alternating with light-brown schist, containing an abundance of small no- dules of iron pyrites, Belemnites, Orthoceratites, and Ammo- nites, enumerated beneath. The limestones which alternate with the schist become occasionally light-coloured as they ap- proach the next rock, from which however they are separated by a repetition of the dark-coloured limestone and brown schist, e. Brown shale which does not effervesce with acids. J. Variegated beds : greenish-blue and argillo-calcareous rocks; more or less schistose, the calcareous matter being often in very small quantity, g. Brown sandstone; princi- pally siliceous, though some of it does contain calcareous mat- ter. It is sometimes micaceous, and occurs either in thick, thin, or schistose beds. It has sometimes been called grau- wacke, and it is one of the macignos of the Italians. The organic remains from Coregna were first discovered by M. Guidoni, of Massa ; a few indications only of the pre- sence of such bodies in the limestone under consideration having been noticed by M. Cordier some years previously. The strata being perpendicular, the weather acts on the edges of the shale beds, in which the remains are found, and they are thus brought to light. At my request Mr. Sowerby ex- amined the remains that I brought from thence, and he con- siders that out of fifteen different species of Ammonites^ one seemed the same with the A. erugatus, Phil., discovered in the lias of Yorkshire, while two resembled A. Listeri * and A. bi- Jbrmis, shells discovered in the coal-measures of the same part of England. The remainder he considers undescribed. From the great scarcity of organic remains of these limestones in Italy, I have inserted Mr. Sowerby's descriptions of the va- rious species, together with figures, considering that they may be of service in the examination of other parts of Italy, as well as Greece, and various countries eastward. * This shell is also discovered, according to M. Hceninghaus, in the coal- measures at Werden. Oolitic Group. 333 Fig. 62. Fig. 63. Fig. 64. Fig. 65. Fig. 66. Fig. 67- Fig. 68. Fig. 62. Ammonites cylindricus. Inner whorls perfectly concealed; sides slightly concave about their centres, flat towards the margin ; surface smooth ; aperture oblong, deeply indented by the preceding whorl ; the front square, which distinguishes it from A. heterophyllus, Sow. Fig. 63. A. Stella. A small portion of the inner whorls exposed ; the sides rather convex, largely umbilicated ; of the inner whorls plain ; of the outer, two thirds covered by large convex rays ; aperture elongated, its front elliptical, its inner angles truncated. Fig. 64. A. Phillipsii. Inner whorls almost wholly exposed ; whorls slowly increasing, about four, their sides flat, irregularly and obscurely un- dulated ; aperture four-sided, rather longer than wide, the sides ^nearly straight. The cast is contracted at distant intervals by the periodical thicken- ing of the edge of the aperture. Named in honour of Mr. Phillips *. Figs. 65 and 67. A. biformis. Inner whorls partly visible ; whorls three or four, rapidly increasing, crossed by many prominent sharp ribs; each rib suddenly becomes obscure, and spreads into two as itjpasses over the broad convex front ; aperture transversely oblong, twice as wide as long, slightly arched. Upon the inner whorls, which have the front plain, the ribs are contracted into round tubercles. The extremities of the longer ribs almost form spines. This species is .found in the coal-measure near Leeds. Fig. 66. A. Listen. See Min. Conch, tab. 501. Also discovered in the coal-measures of Yorkshire. Fig. 68. A. Coregnensis. Inner whorls much exposed ; whorls three or four, crossed by many straight, prominent, sharp ribs, which bend forward, and suddenly terminate upon the nearly plain front ; aperture transversely obovate. This shell is intermediate between A. biformis and A. planicostata, Sow. : it is, however, nearer the former, as it has tubercles upon the inner whorls, where A. planicostata is quite smooth. Fig. 69. Fig. 70. Fig. 71. Fig. 72. Fig. 73. Fig. 74. Fig. 75. * Author of Illustrations of the Geology of Yorkshire. 334- Oolitic Group. Fig. 69. A. Guidoni. Inner whorls much exposed; whorls few, their sides flat and crossed by distant flattened ribs ; each rib split, the posterior branch most prominent, and raised into a low tubercle before it passes over the narrow convex margin. Named in honour of Sig. Guidoni, the disco- verer of these remains at Coregna. Fig. 70. A. articulatus. Inner whorls nearly exposed ; whorls few, each divided by eight or ten furrows into as many imbricating joints ; the anterior edge of each joint elevated, and crossed by the edges of the septa. Fig. 71. A. discretus. Inner whorls partly exposed in a large umbilicus ; globose ; whorls three or four, crossed by many prominent ribs, which split as they cross over the convex front; keel sharp, entire; aperture transversely oval, slightly arched. Fig. 72. A. ventricosus. Inner whorls slightly exposed ; whorls about three ; half the fourth whorl much inflated ; sides ornamented with arched ribs, that are often flattened and united in pairs as they pass over the front, which in the last whorl has a furrow along it ; aperture circular, large. Fig. 73. A. comptus. Inner whorls almost wholly exposed, rapidly in- creasing in size ; sides flat ; whorls crossed by very numerous, sharp, straight radii, which terminate in obscure spines near the narrow concave front; aperture oblong, narrowest towards the front. Fig. 74. A. catenatus. Inner whorls much exposed ; whorls rapidly in- creasing, crossed by strong curved ribs, which enlarge as they approach the margin ; front ornamented with a chain of hollow squares ; apertures rather square, notched by the preceding whorl ; the hollow squares around the margin united by two of their angles to the extremities of corresponding radii. Fig. 75. A. trapezoidalis. Inner whorls exposed; whorls three or four, rapidly increasing in size, crossed by many prominent nearly equal ribs reaching to the narrow front ; aperture trapezoidal, indented by the pre- ceding whorl ; the acute angle truncated by the front. The above figures are all of the natural size of the Ammo- nites. The remains of Orthoceratites, which abundantly ac- company the Ammonites, resemble the O. Steinhaueri, found in the coal-measures of Yorkshire ; they also approach the O. ? elongatus of the Dorsetshire lias. The remains of Belem- nites consist only of their alveoles, and are somewhat com- mon. As far therefore as the evidence of the Ammonites and Or- thoceratites extends, we may refer the limestone of La Spezia either to the lias or the coal-measures. There will be observed a curious correspondence in the organic character of the rocks of the Savoy and French Alps above noticed, and considered as lias by M. Elie de Beaumont, with that of the limestones of La Spezia. In the former, coal-measure plants are found with Belemnites; in the latter, coal-measure Ammonites also occur with Belemnites. The organic character of the oolitic group in the Alps is far from being well ascertained, and the undescribed organic remains found in the same series of the South of France are exceedingly numerous, so that it may be possible to discover some of the La Spezia Ammonites in both situations; and the organic remains of the south-east of France, the Alps, and La Spezia, may hereafter mutually assist in de- Oolitic Group. 335 termining the relative ages of the rocks in which they are discovered *. The dolomite found among the limestones of La Spezia rises so perpendicularly, that it might be considered as a dyke elevating the strata; while at the same time it has the appear- ance of an included bed, or series of beds. It preserves a very constant position, and extends in a line across the mountains of La Castellana, Coregna, Santa Croce, Parodi, and Ber- gamo, towards Pignone. M. Laugier, at the request of M. Cordier, very obligingly made for me an analysis of some crystalline dolomite of La Castellana. One hundred parts were found to contain, carbonate of lime, 55*36 ; carbonate of magnesia, 41*30; peroxide of iron and alumine, 2 ; silica, 0-50; loss, 0-84.. These limestones occur on the other or eastern side of the Gulf of La Spezia, and dolomitic rocks are also found among them. The mode on which they repose on the older rocks is particularly instructive, and is well seen at Capo Corvo, of which the annexed wood-cut is a section, laid bare by the sea. Fig. 76. ed e f g h i k I m G. Gulf of La Spezia. M. Embouchure of the Magra. a. Gray compact limestones mixed with schist, ft. Thick beds of gray compact limestone, c. Schist with mica. d. Thick beds of hard conglomerate, containing pieces of quartz, vary- ing from the size of a pea to that of a walnut, and even larger, agglutinated by a siliceous cement. Two or three beds of coarse sands are associated with this. e. The same, mixed with chlorite schist, often in the same bed. The quartzose beds contain veins of specular iron-ore, f. Brown micaceous and schistose beds, with a small proportion of limestone, g. A mixture of brown and white crystalline limestone, h. Com- pact chloritic rock. i. White saccharine limestone. Jc. Brown micaceous beds. I. White saccharine limestone, rendered schistose by mica. m. Brown semi-crystalline limestone, mixed with white, n. Micaceous schist, curving round to the eastward. The crystalline limestones and micaceous schist of this sec- tion would seem to form part of the system of rocks, which in * It should be observed, that M. Passini states he has discovered red am- monitiferous limestones in the midst of sandstones in Tuscany, which he considers may be referred to the same age as the limestones of La Spezia. Journal de Geologie, t. h". p. 98. 336 Oolitic Group. the neighbouring mountains of Massa Carrara, now again known by the name of Alpi Apuani, furnishes the long cele- brated Carrara marbles. The gray limestones appear the same as those on the western side of the Gulf of Spezia; but instead, like them, of resting upon a mass of sandstone, they repose upon a conglomerate, seen, between the mouth of the Magra and Ameglia, to become far more developed than at the Capo Corvo section, where it is in some manner squeezed between the crystalline limestones and the compact gray lime- stones. Amid this greater development, which appears to mark an unconformable superposition, a conglomerate will be observed (particularly on the shore of the Magra), closely re- sembling that commonly known as the Vallorsine conglome- rate, and noticed above. I cannot avoid connecting this conglomerate, and that of the Lake of Como, with the conglomerates and sandstones of the Vallorsine and other parts of the Western Alps, and re- ferring them to the same epoch of formation; one in which water, with a certain velocity, ground down portions of pre- existing rocks, and which was succeeded by a state of things when a great abundance of carbonate of lime was deposited. This deposit appears to have been extensive, not only in the Alps, but in Italy ; and in both situations, where it occurs close to the rocks of an older date, such as protogine, gneiss, micaceous slates, associated saccharine marble, and talcose rocks of that age, it seems to be separated from them by strata which mark a mechanical origin. As we may suppose great inequalities to have existed during this deposit, and others im- mediately preceding it, we may perhaps in this way account for the almost close contact of the gray compact limestones with the saccharine limestone and other associated rocks at Capo Corvo, while on the western side of the gulf they rest on arenaceous rocks of considerable thickness, which again repose on gray siliceo-calcareous schists and sandstones, that extend over a considerable part of Liguria. How far these beds, which separate the limestones of the Alps, Liguria, and Tuscany, may be equivalent to the sandstone found beneath the lias in Southern Germany and various parts of France, may perhaps be now difficult to determine, but there is a cer- tain general resemblance which seems to point to that con- clusion. Supposing that these Italian and Alpine limestones do re- present the oolitic series of Western Europe, (and it seems very possible that they may do so,) it remains to account for the very great abundance of organic remains in the one, and their very great scarcity in the other. It has often struck geo- logists, that some deposits may have taken place in shallow Oolitic Group. 33 7 seas, and others in deep water. This mode of viewing the subject has, if I mistake not, induced M. Elie de Beaumont to consider that the oolitic series of the Western Alps was deposited in a deep sea, at the same time that the same series was in the course of formation in shallow seas in other places. This observation may be extended into Italy and Greece, where the absence or very great scarcity of organic remains at this epoch seems to afford it support. That great inequa- lities existed at all periods on the earth's surface it seems lair to infer, as well beneath the sea as on land. It would be un- philosophical to conclude that marine animals were ever more capable of supporting very considerable differences of pressure than at the present day. Kovv we know that certain kinds of marine animals, particularly some Mollusca and Conchifera, are only found on coasts where they can find support beneaih a moderate pressure of water; while others, such as the Nau- tilidte, are so provided with floating apparatus, that they are discovered in parts of the ocean where there may be consider- able depth. We have only to consider that in those parts of Western Europe where organic remains are abundant, shallow seas existed, while the same ocean was deep, with some ex- ceptions, over that part of the globe's surface where we find Italy and Greece, and an explanation would seem to be afforded, not only of the abundance of shells in one place, and their scarcity in another, but also of the kind of shells found; for, as yet, camerated shells, such as Belemnites, Orthoceratites, and Ammonites, have been principally discovered in the oolitic rocks of central Italy ; in other words, animals capable of swimming in deep seas*. Organic remains are not only scarce in the limestones in Italy, but also in the sandstones or macignos, which occur in great thickness above and beneath them. The organic remains as yet noticed in these sandstones are Fucoides^ marine plants which may easily be drifted to considerable distances, as the Sargasso Weed now is. The differences of depth may also in some measure account for the different mineralogical structure of the rocks composing the oolitic group in different situations. Still, however, the ques- tion whence all this great mass of carbonate of lime was de- rived, remains unanswered. To attempt to account for it by means of springs neither more numerous nor abundant than those we now see, seems quite unphilosophical ; and to con- * M. Guidoni states in a memoir published in the Nuovo Giornale de letterati de Pisa, 1830 ; and the Journal de Geologic, 1831, that he has dis- covered in the limestone of La Spezia, not only a variety of Ammonites re- ferrible to the oolitic group, hut also many other univalves and bivalves ; among the rest, the Gryph&a arcuala, Lam. (G. incurva, Sow.), which would appear to show a state of things at that place more resembling the oolite of Western Europe. 338 Oolitic Group. sider it entirely due to animals which have separated lime from the water, leaving their shells produced through millions of ages to be gradually converted into limestone, appears also a cause inadequate to the effect required, though it cannot be denied that the mass of many limestones is nearly made up of organic remains. With every allowance for the limestone de- posits of the oolitic series formed by springs and organic bodies, there remains a mass of calcareous matter to be accounted for, distributed generally over a large surface, which requires a very general production, or rather deposit, of carbonate of lime, contemporaneously, or nearly so, over a great area. It appears from the lists of fossils discovered in the rocks of the oolitic group*, that our knowledge of the vegetable re- mains is too limited to enable us to form any general conclu- sions respecting them. Mammalia have been found in one locality only, Stonesfield ; where there are the remains of more than one species of Didelphis. Pterodactyles have been dis- covered at Solenhofen, where there would appear to be many species ; and at Lyme Regis, where there is another species found also at Banz, in Bavaria, The remains of this strange genus probably also occur at Stonesfield. The Macrospondyli, nearly allied to Crocodiles, are found in Northern France and Germany. The Teleosaurus is discovered near Caen, Nor- mandy. The Megalosaurus is found in Oxfordshire, in Nor- mandy, and near Besancon. The Geosaurus has as yet been noticed only in the lias of Wurtemberg, and in the Solenhofen beds. Two species of Lacerta are discovered in the Solen- hofen beds, which also contain the remains of the genera jElodon, Rhacheosaurus, and Pleurosaurus. Ichthyosauri and Plesiosauri would appear to have been somewhat widely distributed, and to have existed during the formation of the whole oolitic series. Neither Pterodactyles, Crocodiles, nor any of the above-noticed reptiles have as yet been detected in the oolitic deposits of Southern France, of the Alps, or of Italy. Tortoises have been noticed in England and Germany. Fish would appear to be by no means rare ; those of Germany, however, have only been examined with attention. Insects have been detected in the oolite of Stonesfield and at Solen- hofen. Polypifers occur in considerable abundance in parti- cular places, more especially in the beds which have been named Coral Rag, and in the upper part of the great oolite, which has thus obtained, in Normandy, the name of Calcaire a Polypiers. Of Radiaria, the genera detected in the oolitic series are numerous, consisting of Cidaris, Echinus, Galerites, Clypeaster, Nucleolites, Ananchytes, Spatangus, Clypeus, En- * See lists at the end of the volume. Oolitic Group. 339 crmites, Eugeniacrinites, Apiocrinites, Pentacrinites* Solano- crinites, Rhodocrinites, Comatula, Ophiura, and Asterias. Respecting the shells, the following summary will show some of those that have been discovered in the same division of the oolite series*, in more than one moderately distant locality; and the places where they have been observed will be found by reference to the list of oolite fossils. Fig. 77. Fig. 78. Fig. 79. Kimmeridge Clay. Ostrea deltoidea (Fig. 78.), a very cha- racteristic shell in England; Gryphcea virgula (Fig. 77.), a characteristic shell of this part of the oolitic series in France; Pinna granulata; Trigonia clavellata ; T. costata; My a de- pressa; Pholadomya acuticostata ; Pteroceras Ponti. Coral Rag. Ostrea gregarea ; PectenLens; P . incequicos- tatus; P. vimineus; P. vagans; Lima rudis; Plagiostoma rusticum; P. laviusculum ; P. rigidum; Modiola bipartita; Gervillia aviculoides ; Trigonia costata ; T. clavellata; Turbo muricatus; Trochus Tiara ; Melania Heddingtoncnsis ; M. stri- ata ; Ammonites plicatilis; A. vertebralis ; A. Sutherland^. Oxford Clay. Terebratula ornithocephala ; Ostrea pal- metto; 0. Marshii ; O. gregarea; Gryph&a dilatata ( Fig. 79. ), a very characteristic shell in England and France ; Pectenji^ brosus; P. Lens; Gervillia aviculoides ; Trigonia clavellata; T. costata; Ammonites armatus; A.Kcenigi; A. Calloviensis ; A. Duncani; A. sublccvis; A. plicatilis ; Patella latissima. Compound Great Oolite, including Fuller's Earth, Great Oolite, Bradford Clay, Forest Marble, and Cornbrash. - Te- rebratula subrotunda ; " T. intermedia; T. digona ; T. obsoleta; * The student will have noticed, that in the list of oolite fossils, the same shell is stated to have been discovered in places distant from each other, but in various beds. Such shells are not here enumerated; and it may be questionable how far some of those stated to be found in remote situations in equivalent strata may really be so ; for conclusions respecting the smaller divisions of the oolite frequently appear much forced. z 2 340 Oolitic Group. T. reticulata : T. globata; T. coarctata ; T. media; Ostrea Marshii; O. costata; O. acuminata; Pecten fibrosus; Plagi- ostoma cardiiforme ; Avicula echinata ; Av. costata; Lima gib- bosa ; Modiola imbricata ,- Perna quadrata ; Trigonia clavel- lata; T. costata; Nucula variabilis; Isocardia concentrica; Patella rugosa. Inferior Oolite with its Sands. Tcrebratula spharoidalis ,- T. ornithocephala / T. obsoleta; T. media; T. concinna; T. bullata; T. emarginata ; T. punctata; T. resupinata; T. ovoides; Gryplicea Cymbium; Pecten Lens; Avicula incequi- valvis; Lima proboscidea ; L.gibbosa; Plagiostoma giganteum ; P.punctatum; Modiola plicata ; Trigonia clavellata ; T. stri- ata; T. costata; Isocardia concentrica; Cardita similis; C. lunulata ; Astarte excavala ; Mya V scripta ; Myoconcha crassa; Melania Heddingtonensis ; M. lineata; Turbo orna- tus; Trochas arenosus; T.fasciatus; T.promineus; T. punc- tatus; T.elongatus; T. abbreviatus ; T. Tiara; T. angulattis ; T. duplicatus; Pleurotomaria ornata; Ammonites l&viusculus ; A. discus ; A. contractus ; A. Blagdeni; A. Brocchii ; A.acu- tus; A. Stokesii; A. Murchisona ,- A, Braikenridgii ; A. ele- gans; A. annulatus; A. Parkinsoni ; Nautilus lineatus; N. obesus; Belemnites compressus. Fig. 80. Fig. 81. Fig. 82. Fig. 83. Fig. 84. Fig. 85. Lias. Spirifer Walcotii (Fig. 85. ), a very characteristic shell ; Terebratula ornithocephala ; T. acuta ; T. tetraedra ; T. punctata ; T. triplicata ; T. bidens ; T. serrata ; Grypli&a incurva (Fig. 81.), a very characteristic shell; G. obliquata ; G. gigantea; G. Maccullochii ; Pticatula spinosa ; Pecten tfquivalvis ; P. barbatus ; Plagiostoma giganteum (Fig. 82.); P. punctatum ; P. Hermanni ; Lima antiqua ; Avicula in- fcquivalvis (Fig. 84-.) ; A. cygnipes ; Inoceramus dubius ; Mo- diola Scalprum ; M.Hillana; Unio crassissimus ; Amphidesma rotundatum ; Pholadomya ambigua ; Trochus Anglicus ; T. im- bricatus ; Belemnites sulcatus ; B. elongatus ; B. apicicurvatus ; B. pistilliformis ; Ammonites Walcotii (Fig. 80.), character- istic ; A.Jimbriatus ; A. Henleii ; A. communis ; A. planicos- talus ; A.falcifer ; A. hetcrophyllus ; A. brcvispina ; A. Jame- soni ; A. Turneri ; A.stellaris; A. Bucklandi (Fig. 83.), cha- Oolitic Group. 341 racteristic; A. oltusus ; A. Slokesii (A. Amaltheus) ; A. sig- mifer ; A. Conybeari ; A. concavus ; A. Humphresianus ; A. Birchii ; A. Bechii ; Nautilus lineatus. Although this list may assist the student, so far as to show the shells stated to be found in the same rock in various situ- ations, he must be cautious in referring any particular beds, wherein he may detect any of the above remains, to the rock under the 'head of which such remains are here noticed ; but rather look at the general character of all the shells he may find in such beds, and thence infer their probable similarity, yet with much reserve, when the type and the rock considered equivalent to it are far distant from each other. The following summary will convey an idea of the genera, with their respective number of species, stated by various au- thors to have been discovered in the beds of the group under consideration. Plants. Fucoides, 3 species ; Equisetum, 1 ; Pachypteris, 2 ; Pecopteris, 6 ; Sphaenopteris, 5 ; Taeniopteris, 2 ; Cyclo- pteris, 2; Glossopteris, 1; Neuropteris, 2; Lycopodites, 1; Pterophyllum, 4; Zamia, 11; Zamites, 4- ; Thuytes, 4 ; Tax- ites, 1 ; Bucklundia, 1 ; Mamillaria, 1. Zoophyta. Achilleum, 6 ; Manon, 3; Scyphia, 41 ; Tragos, 9; Spongia, 2 ; Alcyonium, 1 ; Cnemidium, 9 ; Limnorea, 1 ; Siphonia, 1 ; Myrmecium, 1 ; Gorgonia, 1 ; Millepora, 6 ; Madrepora, 1 ; Cellepora, 2; Retepora? 1 ; Flustra, 1 ; Ce- riopora, 9; Agaricia, 3; Lithodendron, 3 ; Caryophyllia, 7; Anthophyllum, 3; Fungia, 2 ; Cyclolites, 1; Turbinolia, 2; Turbinolopsis, 1 ; Cyathophyllum, 6; Meandrina, 5; Astrea, 25 ; Thamnasteria, 1 ; Aulopora, 3 ; Entalopora, 1 ; Favo- sites, 1 ; Spiropora, 4 ; Eunomia, 1 ; Crysaora, 2 ; Theonoa, 1 ; Idmonea, 1 ; Alecto, J ; Berenicea, 1 ; Terebellaria, 2 ; Cellaria, 1 ; Sarcinula, 1 ; Intricaria, 1. llcidiaria. Cidaris, 18; Echinus, 6; Galerites, 3; Cly- peaster, 1 ; Nucleolites, 6 ; Ananchytes, ] ; Spatangus, 4 ; Clypeus, 6 ; Encrinites, 2 ; Eugeniacrinites, 6 ; Apiocrinites, 8; Pentacrinites, 14; Solanocrinites, 3; Rhodocrinites, 1; Comatula, 4; Ophiura, 3; Asterias, 8. Annulata. Lumbricaria, 6; Serpula, 53. Conchifcra. Spirifer or Delthyris, 3; Terebratula, 59; Orbicula, 3; Lingula, 1 ; Ostrea, 28 ; Exogyra, 3; Gryphaa, 15; Plicatula, 4 ; Pecten, 28 ; Monotis, 4; Plagiostoma, 18; Posidonia, 1; Lima, 5; Avicula, 12; Inoceramus, 1; Ger- viliia, 7; Perna, 3; Crenatula, 1; Trigonellites (Phil.\ 2; Pinna, 7 ; Mytilus, 6 ; Modiola, 22 ; Lithodomus, 1 ; Chama, 3; Unio, 6; Trigonia, 15; Nucula, 18; Pectunculus, 2; Area,?; Cuculloea, 14; Hippodium, 1; Isocardia, 11; Car- dita, 3 ; Cardium, 1 1 ; Myoconcha, 1 ; Astarte, 9 ; Cras- 342 Oolitic Group. sina, 7 ; Venus, J ; Cytherea, 4 ; Pullastra, 2 ; Donax, 2 ; Corbis, 3 ; Tellina, 2 ; Psammobia, 1 ; Lucina, 4 ; Sangui- nolaria, 2 ; Corbula, 4 ; Mactra, ] ; Amphidesma, 5 ; Lu- traria, 1 ; Gastrochaena, 1 ; Mya, 8 ; Pholadomya, 20 ; Pa- nopasa, 1 ; Pholas, 2. Mollusca. Dentalium, 2 ; Patella, 8 ; Emarginula, 1 ; Pi- leolus, 1 ; Bulla, 1 ; Helicina, 4 ; Auricula, 1 ; Melanea, 5 ; Paludina, 1 ; Ampullaria, 1 ; Nerita, 4 ; Natica, 5 ; Verme- tus, 2; Delphinula, ]; Solarium, 2; Cirrus, 5; Pleuroto- maria, 3; Trochus, 21; Rissoa, 4; Turbo, 8; Phasianella, 2 ; Turritella, 5 ; Nerinaea, 6 ; Cerithium, 3 ; Murex, 2 ; Rostellaria, 3 ; Pteroceras, 3 ; Actaeon, 5 ; Buccinum, 1 ; 'Terebra, 4; Belemnites, 65; Orthoceratites? 1; Nautilus, 10; Hamites, 1; Scaphites, 2; Ammonites, 173; Aptychus, 4 ; Onychoteuthis, 1 ; Sepia, 1. Crustacea. Pagurus, 1 ; Eryon, 4 ; Scyllarus, 1 ; Palae- mon, 3; Astacus, 6. Insecta. Libellula, 1 ; JEschna, 1 ; Agrion, 1 ; Myrme- leon? 1; Sirex? 1; Solpaga? 1. Pisces. Dapedium, 1 ; Clupea, 5 ; Esox, 2 ; Urasus, 1 ; Sauropsis, 1 ; Ptycholepis, 1 ; Semionotus, 1 ; Lepidotes, 3 ; Leptolepis, 3; Tetragonolepis, 4. Reptilia. Pterodactylus, 7; Macrospondylus, 1 ; Croco- dilus, 3 ; Teleosaurus, 1 ; Megalosaurus, 1 ; Geosaurus, 2 ; Lacerta, 2 ; Rhacheosaurus, 1 ; JElodon, 1 ; Pleurosaurus, 1 ; Plevsiosaurus, 6 ; Ichthyosaurus, 4 ; Tortoise, 1 . Mammalia. Didelphis, 2 *. Thus making; Plantce, 17 genera, 51 species. Zoophyta, 43 genera, 175 species, Radiaria, 17 genera, 94 species. Annulata, 2 genera, 59 species. Conchifera, 55 genera, 406 species. Mollusca, 39 genera, 372 species. Crustacea, 5 ge- nera, 15 species. Insecta, 6 genera, 6 species. Pisces, 10 ge- nera, 22 species. Reptilia, 13 genera, 31 species. Mamma- lia, 1 genus, 2 species. Total 208 genera, 1233 species. Although this summary cannot be considered as strictly ac- curate, the present state of our knowledge respecting organic remains rendering the catalogues of those contained in any series of beds imperfect, it may still be useful as an approxi- mation to the truth, and as affording a general view of the fossils stated, for the most part by those now considered as good authorities, to have been discovered in the group under consideration. It has been above remarked, that the surface on which the * When the genus has been noticed, and the species is represented as undetermined in the lists at the end of the volume, the genus has been con- sidered to have only one species in this summary. Two species have been assigned to Didelphis, as there seems little doubt of that fact. Oolitic Group. 343 oolitic group was deposited, was probably at very various depths beneath that of the sea; and that even during the de- posit itself, the sea varied in depth over the same point, in consequence of movements in the land. The nature of the organic remains also points to the proximity of dry land in some places, while it may have been comparatively remote in others. It does not seem unphilosophical to infer that the bays, creeks, estuaries, rivers, and dry land, were tenanted by animals, each fitted to the situations where it could feed, breed, and defend itself from the attacks of its enemies. That strange reptile the Ichthyosaurus * (one species of which, 7. platyodon, was of a large size, the jaws being strong, and occasionally eight feet in length,) may, from its form, have braved the waves of the sea, dashing through them as the Porpess now does; but the Plesiosaurus, at least the species with the long neck (P. dolichodeirusy fig. 87.)fj would be better suited to Fig. 86. * It is attempted in the annexed wood-cut (fig. 86.), to convey an idea of the probable form of /. communis, and of the head of /. tenuirostris. The former is represented on dry land, where probably it never reposed, for {lie purpose of exhibiting its form. f The animal is represented in the act of catching a Pterodactyk. It is 344- Oolitic Group. have fished in shallow creeks and bays, defended from heavy breakers. The Crocodiles were probably, as their congeners of the present day are, lovers of rivers and estuaries, and like them destructive and voracious. Of the various reptiles of this period, the Ichthyosaurus, particularly the /. platyodon, seems to have been best suited to rule in the waters, its power- ful and capacious jaws being an overmatch for those of the Crocodiles and Plesiosauri. Thanks to Professor Buckland, we are now acquainted with some of the food upon which these creatures lived : their fossil faeces, named Coprolites^ having afforded evidence, not only that they devoured fish, but each other ; the smaller becoming the prey of the larger, as is abun- dantly testified by the undigested remains of vertebrae and other bones contained in the coprolites*. Amid such voracity, it seems wonderful that so many escaped to be imbedded in rocks, and after the lapse of ages on ages to tell the tale of their exist- ence as former inhabitants of our planet. And strange inha- bitants they undoubtedly were : for, as Cuvier says, the Ich- thyosaurus has the snout of a dolphin, the teeth of a crocodile, the head and sternum of a lizard, the extremities of cetacea (being, however, four in number), and the vertebrae of fish ; while the Plesiosauras has, with the same cetaceous extremi- ties, the head of a lizard, and a neck resembling the body of a serpentf. It is almost needless to remark that these two genera have disappeared from the surface of our planet ; and, as the stu- dent may have collected from the various lists of organic re- mains, even previous to the deposit of the supracretaceous rocks, at least as far as regards Europe. The vegetable remains have been accumulated in particular places, such as Yorkshire, Brora, and parts of Germany, at about the same period. Circumstances therefore must have existed at such situations not common to the whole area. These deposits do not seem the result of violence ; for the ve- getables are well preserved, as if, like the hortus siccus of the botanist, for the purpose of examination. By their aid we learn that the vegetation which then clothed some parts of this por- tion of our planet, no longer resembles that which we now see, but one widely different. Perhaps we may, in anticipation, look forward to times when the geologist may speculate on figured as swimming high above the water for the purpose of showing its general form. It more probably swam beneath the surface, in the manner of crocodiles, which would enable it the better to support its great length of neck. * For an interesting account of Coprolites and their contents, see Buck- land's Memoir, Geol. Trans. 2nd series, vol. iii. f Cuvier, Oss. Fossiles, t. v. This notice of the Plesiosaurus applies more particularly to P. dolichodeirus. Oolitic Group. 345 the proximity of certain lands near places where the abundant remains of vegetables and certain animals would seem to point to such conclusions, even though, from the various movements in the land, no part of such ancient continents or islands may now appear on the surface. We of the present day, however desirous we may be to elucidate this subject, seem to possess too few data to proceed in the inquiry. One thing, however, the student should bear in mind : he must not consider that all older rocks in the vicinity of others of more recent origin, though now rising in mountains high above them, necessarily formed the dry land previously to the deposit of the newer rocks: for amid the various surface-changes that have been effected, such older rocks have frequently been upheaved after the for- mation of the more recent, as is shown by the mode of strati- fication near the junction of the two, the one being tilted up with the other. We may, perhaps, more closely approach the truth, when we find, as in Normandy, the oolitic group resting quietly on, and surrounding, disturbed older strata; so that in that country (and the same observation applies to other situ- ations) we may conceive the sea in which the oolitic rocks were formed, to have bathed the slaty and granitic districts of Nor- mandy and Brittany. Fig. 88. Scale = i of nature. Those strange flying creatures, the Pterodactyles, must have sported on dry land, probably subsisting on insects, such, among others, as that figured above, which was obtained by Mr. Murchison from the quarries at Solenhofen, where the remains of Pterodactyles are also discovered. That the Pterodactyles should be scarce fossils, is what we might expect, for the circumstances favourable to their preser- 34:6 Oolitic Group. ration must have been extremely rare. Even supposing that they dashed out to sea in pursuit of their insect prey, there must have been a combination of fortunate accidents to have prevented the Pterodactyles and their intended prey from being devoured by the fish and other inhabitants of the sea, among the exuviae of which their remains are now detected. It is curious, and seems to establish a connexion between the insects and the Pterodactyles, that in the spot where the remains of the latter are most abundant (Solenhofen), the greatest quantity of fossil insects yet noticed in the oolitic group has been detected. At Stonesfield also, where the re- mains of insects are stated to have been discovered, the exuvise of Pterodactyles, according to Prof. Buckland, are also ob- served. Not so, however, with the Pterodactyle of Lyme Regis, whose remains are mixed with those of Ichthyosauri and other marine animals, and where insects have not yet been detected. But when we consider the abundant exuviae of Plesi- osauri, perhaps we may not err greatly, in considering dry land not very far distant from the spot where we now find their bones entombed. Be the case as it may, a Pterodactyle in a sea, amid Ichthyosauri and other voracious creatures, must have had but a slight chance of escape; and geologists should be grateful that any combination of circumstances should have so far prevailed, as to permit the preservation of even a single individual, to show us the strange terrestrial creatures that then existed. In the lias of Lyme Regis, the Ichthyosauri, Plesiosauri, and many other animals, seem to have suffered a somewhat sudden death ; for in general the bones are not scattered about, and in a detached state, as would happen if the dead animal had descended to the bottom of the sea, to be decomposed, or devoured piecemeal, as indeed might also happen if the crea- ture floated for a time on the surface, one animal devouring one part, and another carrying off a different portion ; on the contrary, the bones of the skeleton, though frequently compressed, as must arise from the enormous weight to which they have so long been subjected, are tolerably connected, frequently in perfect, or nearly perfect order, as if prepared by the anatomist. The skin, moreover, may sometimes be traced, and the compressed contents of the intestines may at times be also observed, all tending to show that the animals were sud- denly destroyed, and as suddenly preserved. Not only has this apparently happened to these reptiles which, breathing air, might under favourable circumstances be drowned simultane- ously in great numbers, but also to the mollusca, to which constant, or nearly constant, immersion in water is absolutely necessary. Among the multitude of Ammonites discovered in Oolitic Group. 347 the lias, I have often observed individuals, of which the large terminating chamber of the last whorl, where the body of the animal seems to'have been placed, was hollow for half its di- stance upwards towards the aperture or mouth, as if the ani- mal, when overwhelmed, had retreated as far as possible into this part of the shell, so that the muddy matter was prevented from completely filling it. This idea is rendered more pro- bable from the condition of the calcareous matter filling the remaining part of the great cavity, which is exceedingly bitumi- nous, as would happen from the decomposition of the animal within the remainder of the chamber. The student should not, from what has been above remarked respecting the lias at a particular point, Lyme Regis, consider that such observations are applicable to the same rock gene- rally ; or even that the lias of Lyme Regis has suddenly been produced in its whole thickness at once : on the contrary, the lias varies materially at different points, as we should expect it to do, from different local causes; and the lias of Lyme Regis bears evidence of successive deposition, in part during a state of comparative tranquillity, and partly in consequence of a se- ries of small catastrophes, suddenly destroying the animals then existing in particular spots. One observation is, however, necessary, and it will be often applicable to other parts of the oolitic rocks in various situations, that during the formation of the lias in this part of England, there has been a certain change in the animal life of the same place. Thus the animals and shells in the upper part of this rock differ in the mass from those in the lower portion. Very frequently, also, par- ticular strata afford certain organic remains, while all others are exceedingly rare. Notwithstanding the temptation to treat of the probable circumstances that have accompanied the deposit of a particu- lar rock, even within the distance of a few miles, we must abstain, as it would lead us into detail not compatible with this work. It may, however, be remarked, that the destruction of the animals, whose remains are known to us by the name of Belemnites, was exceedingly great at this place. When the upper part of the lias was deposited, multitudes seem to have perished simultaneously, as is attested by a bed composed of little else, beneath Golden Cap, a cliff' between Lyme Regis and Bridport Harbour. Not only are millions entombed in this bed, but in the upper part of the lias generally. The pro- duction of such a bed would seem by no means difficult; for we have only to consider the occurrence of some circumstance destructive to molluscous creatures in the fluid containing, or otherwise carrying, the belemnites, such as might happen to those swarms of mollusca which sometimes surround the navi- 348 Oolitic Group. gator in warm latitudes, and the floating animal mass, if not immediately, would eventually descend to the bottom ; at least the part that escaped the predaceous animals, which indeed might be driven away by the circumstance, whatever it was, that destroyed the molluscous animals. Suppose a multi- tude of the common cuttle-fish to be suddenly killed by the irruption of, or their entrance into, water charged with car- bonic acid; their internal bones, as they are commonly. termed, would be distributed over a common surface after the decom- position of the animals, which were not likely to fall a prey to other creatures ; for those which were not destroyed with the cuttle-fish, would avoid the water so charged with carbonic acid. The vegetables of the lias of this place occur in two differ- ent states : the one showing that they have been scarcely in- jured before they were imbedded ; the other seeming to point to the fracture of wood into junks, the small branches trun- cated, as if they had been broken either during, or previous to, their drift. These latter most frequently occur in argillo-cal- careous nodules, often of large size : but the nodules are not concentric concretions ; on the contrary, both these nodules, and those that frequently envelope the Ammonites and Nautili in the argillaceous beds, are fissile, the line of the laminae being parallel to that of the general stratification ; so that though the nodules, particularly those containing Ammonites and Nautili, are spheroidal, their fracture is lamellar ; and a suc- cessful blow in the line of the laminge, through the centre, dis- closes a fossil, not unfrequently a fish. It being a very interesting inquiry to ascertain the chemical condition of organic remains entombed in various rocks, Dr. Turner was kind enough to analyse certain fossils from the lias of Lyme Regis. He found that a vertebra, a rib, and a tooth of an Ichthyosaurus examined by him, had all a highly cry- stalline texture, owing to the deposit of carbonate of lime, of which they chiefly consist. The colour is nearly, and in parts quite black, in consequence of bituminous matter, which in general amounts to not more than |, and he has not found it exceed j per cent. The phosphate of lime in the vertebra amounted to about 29 per cent., while in the rib and tooth it was about 50 per cent. In fact, as might have been expected, the phosphate of lime remains in greater or less quantity in different specimens, probably depending on the situation where it was preserved, and on the compactness of the original bone. Dr. Turner also ascertained that the scales of Dapedium politum, cleared as much as possible from adhering limestone, consisted of the same ingredients as the ichthyosaurian bones; but the phosphate of lime amounted only to 19 per cent. Oolitic Group. 349 Of course care was taken to select such specimens as were not impregnated with sulphuretof iron, as sometimes happens; and those examined were found to be remarkably free from iron, manganese, alumina, and silica. When we view the oolitic group as a whole, we cannot but remark a certain general uniformity of its structure over a considerable portion of Western Europe ; showing that at the time of its production some similar general causes were in ac- tion over a particular portion of the European area. While, however, this general uniformity is sufficiently obvious over such area, it is equally obvious that the various attempts which have been made to detect certain minor divisions of the oolitic group in the Alps, in Italy, and other places, have been by no means successful. There can be little doubt that a very large portion of the oolitic series has been mechanically produced. Granting this, we can scarcely expect that perfect uniformity to exist which was once considered probable. In point of fact, minor changes in the nature of the beds are constantly taking place; and from a multiplication of these minor changes, very considerable differences in the subdivisions of the group are produced. The annexed proportional sections (Fig. 89. and 90.) will exhibit the different development of the oolitic series in the northern and southern portions of England ; the superincum- bent cretaceous rocks being also represented, to render the general changes still more apparent. Fig. 89. Wilts Sf Somerset. Fig. 90. Yorkshire. Cretaceous Group. Oolitic Group. ... t I Cretaceous j r Group. lH'iil'tr ''-> - liltl k. K I j " I Oolitic Group. 350 Oolitic Group. As the same proportional scale has been adopted for both sections, the eye will readily seize the different depths of the cretaceous and oolitic groups in each. The same letters also have been used for the minor divisions, so that the rocks in one section can easily be compared with those in the other. Creta- ceous Group: a, chalk; b, upper green sand; c, gault; d, lower green sand. (b. and d. appear to be absent in the Yorkshire section, being represented bye.) OoliticGroup: e, Kimmeridge clay ;j coral rag, and its calcareous grits; g, Oxford clay, and the Kelloway rock in its lower part; 7z, cornbrash, and forest marble ; i, Bradford clay ; &, great oolite ; I, Fuller's earth ; m, inferior oolite ; n, marl stone ; 0, lias ; (t, gravel, &c.) It will be observed that so far as regards the divisions e,f, and g, the two sections do not much vary ; but that a very considerable differ- ence exists between the beds h, i, k, I, is developed in South- ern and in Northern England. Not only is their mineralogi- cal character different, but their organic contents are for the most part distinct. The clays and limestones of the South are full of marine remains, while the shales and sandstones of the North abound with terrestrial plants, which have been so thickly accumulated in certain beds as to form coal strata. The marl- stone (n) and the lias (o) are also far more developed in York- shire than in Somersetshire; and the former, which in the South may be considered as the passage of the inferior oolite sands into the lias, has in the North become of more importance, and is separated from inferior oolite by what is termed the upper lias shale. A thin bed of clay or marl is indeed interposed be- tween the inferior oolite sands and the marlstone in the South, which may be the upper lias shale of the North. The terrestrial character of the organic remains contained in a certain portion of the oolitic group is not, as has been seen, confined to Northern England and Scotland, but extends into Germany. Now if it be fair to infer, as it seems to be, that when we find accumulated vegetable matter sufficiently abundant to constitute coal strata, accompanied by beds full of delicately preserved plants, dry land could not have been far distant at the time such vegetable matter was deposited ; we may conclude, that when the beds of Northern England, Scotland, and a portion of Germany corresponding with the compound great oolite of Southern England, were deposited, dry land existed in that part of our planet now known as Northern Europe. Mr. Murchison has observed vertical stems of Equisetum columnar e, apparently in the position in which they grew, reminding us of the dirt bed of Portland and other places, in the lower carboniferous shale and sandstone of the Yorkshire oolite, not only on the coast, but also at a distance of forty miles, on the north-western escarpment of the Yorkshire moorlands. This author very properly concludes, from finding Oolitic Group. 351 the same plants in the same bed, and in the same position, over so large an area, that it is almost demonstrable that the vertical position of the Equiseta could not have been the effect of chance, but must have been the result of some general cause acting over the area. This cause, he further observes, has very pro- bably been a great submergence of the area, permitting the plants to retain their original positions, and their gradual en- velopment by mud, silt, and sand*. Now this fact shows, as in the case of the Portland dirt bed for another period, that the conversion of dry into subaqueous land has been ex- ceedingly gradual in that part of England during the deposit of the oolites. How far this fact may be found more general remains to be seen ; but it is, as far as it goes, extremely im- portant. If we continue our view of the course of the beds, contem- poraneous with those above noticed, from Northern to South- ern Europe, we observe that the character of the organic con- tents becomes marine, and that many of the beds, particularly the clays, can be traced over extensive areas. The multitude and character of their general organic contents may lead us to suppose that over a considerable area, one extending from Southern England to the Jura, and embracing a large portion of France, and a part of Southern Germany, the sea was by no means deep in which the beds deposited. At Stonesfield in- deed, the slates of which place have recently been shown by Mr. Lonsdale to be a lower portion of the great oolite, we have evidence, in the remains of the Didelphis and other or- ganic exuviae, tending to show that land may not have been far distant. Now if we connect this with the vertical stems of the Yorkshire oolite, they would, both taken together, lead us to infer that the dry land was somewhat extensive at this period over what now constitutes a considerable portion of England. It has been above remarked that the Alpine and Italian de- posits, equivalent to the oolitic group generally, may have been formed in a deep sea, while other portions of the same group may have been deposited in shallow waters. If we now regard that particular part of the series corresponding to the compound great oolite, it appears very probable, that dry land was then abundant over what now constitutes Northern Europe, that over Southern Europe there was deep water, while an exten- sive area between the two was shallow water, with here and there, perhaps, dry land. It would not accord with the plan of this volume further to investigate the probable condition of the European area during the deposit of various portions of the oolitic series; but we may * Murchison, Proceedings of the Geol. Soc., and MSS. 352 Oolitic Group. remark that the clay beds, which can be traced with nearly si- milar characters over considerable areas, while the changes in the calcareous and arenaceous masses are much more compli- cated, is precisely what we should expect from the transport of comminuted matter mechanically suspended in water. For the more comminuted the matter, the further will moving water transport it; and consequently the uniformity of the deposit of such a mixture will be greater than from less comminuted matter, supposing always the transporting force to be equal, or nearly equal. While on this subject it should be remarked that the sands of the inferior oolite cover a very considerable area, extending from Northern far into Southern Europe. Of all the divisions of the oolitic series the lias is most uniform over the whole space occupied by it, for its variations are com- paratively inconsiderable in Great Britain, France, and Ger- many. It is not until we enter the oolitic districts of the Alps, and other portions of Southern Europe, that we find even the mineralogical structure very materially altered. Red Sandstone Group. 353 SECTION VII. RED SANDSTONE GROUP. SYN. Red or Variegated Marls (Marnes Irisees, Fr. ; Keuper, Bunte Mer- gel, Ger.). Muschelkalk (Calcaire Conchylien, Al. Brong.). Red or Va- riegated Sandstone (New Red Sandstone, Eng. Auth. ; Ores Bigarre, Fr. ; Bunter Sands tern, Ger.). Zechstein (Magnesian Limestone, Eng. Auth.; Calcaire alpin, Fr. ; Alpenkalkstein, Ger.). Rothliegendes (New Red Conglomerate, Lower New Red Sandstone, Exeter Red Conglomerate, Eng. Auth.; Todtliegendes, Rothe Todtliegende, Ger.; Ores Rouge, Fr. ; Psepliite Rougedtre, Al. Brong.). THIS group, which is often one of very considerable thickness, succeeds, in the descending order, that previously noticed. Perhaps very fine lines of distinction should not be drawn be- tween the two ; for when the lower part of the one and the up- per part of the other have been consideraly developed, they seem in some measure to pass into each other. This led M. Charbaut, who first observed the circumstance in the vicinity of Lons le Saulnier, to class the lias with the variegated marls which constitute the upper portion of the group under consi- deration. The rocks composing the red sandstone group oc- cur in the following descending order: 1. Variegated Marls; 2. Muschelkalk; 3. Red or Variegated Sandstone; 4% Zech- stein; and 5. Rothliegendes. Variegated Marls. In the district of the Vosges and in the neighbouring countries, these commence beneath the sandstone named lias sandstone, into which they gradually pass; the up- per part of the variegated marls, which are green, presenting thin beds of black schistose clay, and of quartzose sandstone, nearly without cement, which latter gradually becomes the lias sandstone, a rock that passes into the lias, and contains the same organic remains*. M. Elie de Beaumont observes, that in many countries the variegated marls can scarcely be sepa- rated from the lias sandstone, even artificially, as is done in the Vosges ; for they appear to become one deposit, as in the en- virons of St. Leger-sur-Dheune, and Autun, and in the arkose of Burgundy. The variegated marls of the Vosges generally are, as their name implies, marked by different colours, among which the principal are wine-red and greenish or blueish gray ; they break into fragments, which have no trace of a schistose structure. In the central portion of these marls there are beds * Elie de Beaumont, Mem. pour sevvir a une Desc. Geol. de la France, t. i. 2 A 354? Red Sandstone Group. of black schistose clay, blueish gray sandstone, and grayish or yellowish magnesian limestone. The sandstone and clay con- tain vegetable impressions, and even coal. Masses of rock- salt occur in the lower part of the marls at Vic, Dieuze, and other parts of that district ; and masses of gypsum are found in the upper and lower portions, but principally in the latter*. According to M. Charbaut, limestone beds, almost entirely composed of shells, are found in the upper part of this deposit. M. von Dechen remarks, that the superior white sandstones are, in the neighbourhood of Stuttgard and Tubingen, covered by variegated clayey marls, by which they are separated from the lias. They are coarse grained; and the fragments of quartz, limestone, &c. are sometimes a foot in diameter, partly rounded, partly angular. Dark gray rolled pieces of limestone sometimes predominate in the lower beds, and are cemented by calcareous matter, containing grains of quartz and felspar. The cement of the other beds is quartzose. The sandstone, especially the upper portion, often contains so much carbona- ceous matter that it becomes dark gray and even black. Fi- brous anthracite and pitchcoal occur in it near Spiegelberg. At Erlaheim, near Balingen, similar coal, with large masses of iron pyrites, are found close to the lias boundary. These carbonaceous deposits are separated from the sandstones be- neath by variegated marls, inclosing thin sandstone beds and traces of coalf . The variegated marls, not differing considerably in their mi- neralogical characters, occur in various parts of the north of France and Germany, and according to M. Dufrenoy they crown the red sandstone rocks of the South of France. How far the variegated marls may be traced in England remains questionable ; but it would appear far from improbable, that the upper part of the red sandstone deposit of this country would answer sufficiently well in its mineralogical structure to the rocks above noticed in the Vosges. There is with us no apparent passage of the lias into the red sandstone series ; on the contrary, we sometimes have, as at the Old Passage near Bristol, a kind of conglomerate of pieces of limestone, bones, teeth, and other remains of saurians and fish, with their fossil faeces or coproiites, which would seem to mark a period when comminuted deposits ceased, and currents of water sufficient to transport pebbles were in action, accumulating bones and other substances, as at the bottom of some seas. Where seen on the southern coast of England, between Lyme Regis and Sidmcuth, the upper part of the red sandstone series is so like * Elie de Beaumont, Memoir above cited. T Von Dechen, German Transl. of Manual. Red Sandstojie Group, 355 the variegated marls of the Vosges and parts of Germany, that I have little hesitation in considering them contemporaneous deposits. In this part of England these marls contain vege- table remains, and, though rarely, scales of fish and bones of pterodactyJes (?). As the lower lias sandstone passes into the variegated marls, and even seems in some measure equivalent to them, a deposit of sands having possibly taken place in one situation, while marls were produced in another, we should not, when consi- dering the general subject, force our conclusions too far, nor carry those divisions which may be locally useful beyond the countries where they may be advantageously employed. Prof. Pusch, in his very interesting account of the Polish rocks, states that between the oolitic series of Poland and the muschelkalk there is an extensive and important deposit of sandstone, usu- ally termed white sa?idstone, from its colour. The deposit is divisible into two portions; the upper being formed of the white sandstone, while the lower part is composed of alterna- tions of fine white marly sandstone, schistose sandstone, shale, and other schistose and dark-coloured rocks, the whole inclo- sing beds of coal from three to twenty-five inches thick. The white sandstone of the upper part alternates with thick beds of gray blue marls, partly red, and more rarely variegated. Beds of limestone are also found in it; but the most valuable product is iron ore, which furnishes the largest amount of iron of any rock in Poland, twenty-seven furnaces affording annually 560,000 quintals of metal. Fossils are rare in this deposit, with the exception of vegetable remains. M. Pusch refers this rock to the lias sandstone, the same as it occurs in Suabia, in Scania, and in the Isle of Bornholm, in all which places it is rich both in iron and coal*. M. von Dechen considers it very doubtful whether we should refer this white sandstone to the lias sandstones and variegated marls, or to the inferior oolite, and rather inclines to the latter opinion, as it occurs between what is considered a middle mem- ber of the oolitic series and the muschelkalk. He moreover remarks, that the true geological position of the sandstone of Hor and the north of Lund in Scania, of which it is thought to be the continuation, is by no means settled. Muschelkalk. A limestone varying in texture, but being most frequently gray and compact. It is occasionally dolo- mitic, and passes into mads above and beneath. When very compact, with numerous remains of the Encrinites moniliformis, Miller (a very characteristic fossil of at least a considerable * Pusch, Esquisse Gognostique du Milieu de la Pologne : Journal cle Geologic, t. ii. 1 2 A 2 356 Red Sandstone Group. portion of the deposit), it has much the appearance ^of some varieties of the carboniferous limestone of England. The mus- chelkalk is sometimes, though rarely, oolitic (between Stiih- lingen and Bonndorf), and contains beds of chert ( Wurtem- berg, and some places in Germany). Gypsum and marl are not unfrequently mixed with it. Copper ore is found in a bi- tuminous marl-slate in the neighbourhood of Horgen, on the eastern border of the Swarzwald, which has led to the erro- neous impression that these beds belonged to the zechstein. Disseminated copper ore is also found in the lower portion of the muschelkalk in other parts of Wurtemberg*. It is some- times, as at Epinal (Vosges), sufficiently hard to be employed as marble. In some situations organic remains would appear to be very abundant, while in others they are somewhat rare. According to M. Alberti, salt is contained in the muschelkalk of Wurtembergf ; and M. von Dechen states that salt is also found in it at Buffleben, between the Thliringerwald and the Hartz. This rock would appear to be unknown in England and in the North of France ; but on the east and south of the latter country, and in parts of Germany, it is found interposed, in its place, between the variegated marls and the red or varie- gated sandstone. According to Prof. Pusch it occurs in Po- land, and is described as being gray and yellow. Red or Variegated Sandstone. This rock is, as its name im- plies, of different tints, these being red, white, blue, and green; the former, however, greatly predominating. It is principally siliceous and argillaceous, occasionally containing mica, masses of gypsum, and rock-salt. In the Vosges, the upper part of the variegated sandstone often presents, according to M. Elie de Beaumont, thin beds of marly limestone and dolomite, which gradually become more abundant; so that, finally, they constitute the lower part of the muschelkalk J. An oolitic and calcareo-magnesian rock is found in this deposit in some parts of Germany, and conglomerates are also included in it. A very extensive deposit, varying but little in its character, occurs in the Vosges, and has thence obtained the name of the Gres de Vosges. A difference of opinion seems to exist be- tween M. Elie de Beaumont and M. Voltz respecting the exact member of the red sandstone series to which this rock should be referred ; the former considering it the equivalent of * Von Dechen, German Transl. of Manual. f Alberti, Die Gebirge des Konigreichs Wurtemberg, 1826. M. Bronn notices salt in the Muschelkalk of Hasmerheim, and other places in the vi- cinity of Heidelberg. Gaea Heidelbergensis, 1830. J Elie de Beaumont, Terrains Secondaires du Systeme des Vosges. The grains forming this oolitic rock are radiated from the centre to the circumference. Red Sandstone Group. 357 Fig. 91. the rothe todte liegende, which occurs beneath the zechstein ; the latter, that it is the lower portion of the red or variegated sandstone, which rests on the zechstein : as the zechstein is wanting in the district, there is perhaps but little essential difference in these opinions. The Gres de Vosges is essentially composed of amorphous grains of quartz, commonly covered by a thin coating of red peroxide of iron; among which are discovered others which appear fragments of felspar crystals. It is often marked by cross and diagonal laminae so common in arenaceous rocks, the result, probably, of deposit by cross currents of water. The rock contains quartz pebbles, sometimes so abundantly as to present a conglomerate with an arenaceous cement. From the mineral character of these pebbles, M. Elie de Beaumont con- siders that they are derived from the destruction of the older rocks, and are merely larger portions which have better resisted trituration than the smaller grains composing the body of the sandstone. The variegated or red sandstone of some countries affords a good building-stone, and when nearly free from colour, as at Epinal, (Vosges,) one of handsome appearance. In situations where it becomes schistose from mica, it is often employed, like some vari- eties of the old red sandstone of the English, for flag-stones, and even tiles for houses. According to Prof. Sedgwick, the red sandstone occurring above the magnesian limestone, in the North of England, repre- sents the Bunter sandstein of Germany, the variegated marls surmounting it being the equivalent of the keuper of the same country. This sandstone is represented as of a com- plex character, from the variable mixtures of sand, sandstone, and marl. In its range from Nottinghamshire into Yorkshire it is generally coarse, often nearly incoherent, and here and there passes into a fine con- glomerate. The superincumbent marls are red and gypseous*. At Wasselonne, Marmoutier, and Sulz- les-Bains, more particularly at the latter place, numerous vegetable remains have been discovered in the red sandstone. These have been described by M. Adolphe Brorigniart. The annexed figure (Fig. 91.) represents a spe- Sedgwick, Geol. Trans. 2nd series, vol. iii. 358 Red Sandstone Group. cimen of Voltzia brevifolia, from Sulz-les- Bains, remarkable as exhibiting the fructification of the plant*. Zechstein. This name has, fortunately, been applied by Humboldt to distinguish a limestone series of a very variable character, to which different names were given, the term zech- stein having been previously applied to only one of the varie- ties. The various beds were known to the German miners by the names of Asche (friable marl), Stinkstein (fetid lime- stone), Rauchwacke, Zechstein, and Kupferschiefer (copper- slate). According to Daubuisson, the mean thickness of the copper-slate in these countries is about one foot. The zech- stein is represented as sometimes from twenty to thirty yards thick ; the rauchwacke, when pure and compact, one yard thick, when cellular sometimes attaining fifteen to sixteen yards ; the Stinkstein, from one to thirty yards thick ; and the asche, very variable. Notwithstanding these minor divisions, to which an extraordinary value has been attached, it does not appear that they can always be observed in the countries where they have been established ; for Daubuisson observes, that the upper portions pass into each other, and even some- times into the zechstein. The zechstein of Germany is composed at its lowest part of a thin bed of marl-slate, known as the copper-slate (Kupfer- schiefer), from containing finely- disseminated ores of yellow copper, purple copper, and vitreous copper, worked for cen- turies. Upon this rests the Zechstein, properly so called, which is a compact dark-coloured limestone. At Thalitter and Stadtbergen, fine seams of marl-slate, containing disse- minated malachite, alternate from ten to thirty times with the beds of zechstein, and must be considered as forming part of it. On the Upper Saale, at CamsdorffJ the marly slate lies entirely in the zechstein. The upper series of beds is more irregular, and appears to have been more influenced by local causes. Masses of dolomite and gypsum sometimes attain such thickness that the other rocks disappear. The dolomite is known in Mansfeld by the name of Rauchwacke. Rock- salt is found in the gypsum. Beds of bituminous limestone (Stinkstein) commonly rest on the gypsum and the accompa- nying marls and clays. A pulverulent loose-grained mass, con- sisting of limestone fragments, frequently occurs in the upper division, and is named Asche. In this division there are also masses of carbonate of iron. They are irregularly combined with limestone (named Eisenkalkstein). At Camsdorffa marly slate accompanies the iron-stone. On the south border of the * Taken from a figure by M. Ad. Brongniart, in the Ann. des Sci. Nat. t. xv. pi. 16. Red Sandstone Group. 359 Hartz, where the largest masses of gypsum occur, limestones are found among them, but are so interrupted that no regular order of stratification can be observed. Beds of an oolitic cha- racter are discovered in the upper division of the zechstein series of Germany *. According to Professor Sedgwick, the magnesian limestone of the North of England, which is the equivalent of this de- posit in Germany, is divisible into, 1. Marl-slate and com- pact limestone, or compact and shelly limestone, and varie- gated marls. 2. Yellow magnesian limestone. 3. Red marl and gypsum. 4. Thin-bedded limestone. The same author considers No. 1. as equivalent to the kupferschiefer and zech- stein, and Nos. 2. 3. and 4?. to the rauchwacke, asche, stink- stein, &c. of Thuringia f. Rothliegendes. This name is given to a series of red con- glomerates and sandstones which occurs between the zechstein or magnesian limestone and the rocks of the next group. The term was originally applied to those beds of Thuringia and other adjacent countries upon which the copper-slate reposes, with the intervention only of portions which are white. M. von Dechen remarks that the rothliegendes appears to be more developed on the east side of the Hartz than at any other point at which it has been observed. Under the Weisliegen- des (the upper white portion of the rothliegendes), which disappears with the marl-slate, there is a red slate clay, and a fine-grained argillaceous sandstone. Beneath this is a por- phyry conglomerate. The porphyry pebbles are remarkable, and do not always resemble the nearest quartziferous porphyry of the Saale, often becoming of greater volume as the distance from the latter increases. They vary in size from that of a walnut to that of the fist. This conglomerate is widely ex- tended throughout the country, and it is worthy of remark that porphyry pebbles are not discovered in the remaining portion of this rothliegendes. Red, greenish, yellowish, and white sandstone, with an uniform grain, are sometimes asso- ciated with it; the quartz grains retaining their crystalline structure, and being cemented by argillaceous matter. On account of their roughness the sandstones are well adapted for millstones, as at Siebigkerode. These beds, with com- mon red sandstones, schistose sandstones, and slate clays, constitute the upper division of the rothliegendes. In the middle division, several thin seams of red and dark blue gray limestone alternate with argillaceous beds, sand- * Von Dechen, German Transl. of Manual. t Sedgwick, on the Geological Relations and Internal Structure of the Magnesian Limestone, &c. : "Trans. Geol. Soc. 2nd Series, vol. iii. 360 Red Sandstone Group. stone, and breccia. They occur regularly on both sides of a ridge at Rothenburg on the Saale. The lower division is characterized by a conglomerate, in which large pebbles of splintery gray quartz occur. The ce- ment is a red friable clay. Rocks, which by their partial de- struction may have furnished these quartz pebbles, are not known in Northern Germany. They occur only in the higher part of the lower division of the rothliegendes. No regularity is observable in this lower division, the inferior portions of which consist of schistose sandstone, schistose clay, and brec- cias. In the immediate vicinity of the Hartz grauwacke, peb- bles of this rock occur in it. The general colour of the roth- liegendes is cherry and violet red. The connexion of this rock with the porphyry at Wettin and Loebejun, at Ihlefeld, on the southern flank of the Hartz, and in the Thuringerwald, render the relations of the whole deposit difficult to determine. While in the Hartz, the pebbles and rock fragments corre- spond but little with the nearest rocks, in the Thuringerwald they invariably do so *. Such are the characters of the rocks, known as Rothlie- gendes, in the countries whence the name has been derived. Taken as a whole, they are for the most part conglomerates, formed from the partial destruction of those rocks on which they rest, the fragments being sometimes angular, as well as rounded, and of considerable size. The researches of Professor Sedgwick have shown that an arenaceous deposit, of a somewhat variable character, and known as the Pontefract rock of Smith, is in all probability the equivalent of the rothliegendes of Germany. It may be traced between the coal measures and zechstein (magnesian limestone), with a few interruptions, from the mouth of the Tyne to the confines of Derbyshire. Though of a very vari- able structure and thickness, it possesses a certain uniformity of character when viewed on the large scale. Conglomerates are rare, and are sometimes seen at the junction of this rothlie- gendes with the magnesian limestone above it. A coarse sili- ceous sandstone, usually of a red or purple tint, but sometimes gray or yellowish brown, seems most common. It contains pebbles of quartz more than an inch in diameter, generally ranged in lines parallel to the stratification, though sometimes irregularly disseminated. It is often a mere sand, with little or no cohesion ; and this character is attributed to an abun- dance of earthy felspar. Though the want of cohesion would appear common, it occasionally becomes sufficiently hard to Von Dechcn, German Transl. of Manual. Red Sandstone Group. 361 afford building-stone (Wetherby, Knaresborough, and Hart Hill, Yorkshire). The sandstones are associated with varie- gated micaceous sandy shale, and variegated marls. This rothliegendes sometimes becomes a brown or gray micaceous sandstone, not to be distinguished from some of the sand- stones of the coal measures*. Mr. Hutton, though he fully admits the variable thickness and structure of this rock, considers that it may, in the county of Durham, be satisfactorily divided into two parts; the up- per consisting of an incoherent sand, generally of a buff co- lour, while the lower portion is more consolidated, even fur- nishing building materials. Though the colour of the latter varies materially, red or purple is by far the most prevalent tint. The division line between these two portions is gene- rally well defined, though they sometimes pass insensibly into each other f. Prof. Sedgwick remarks not only that this rothliegendes, or lower (new) red sandstone, rests unconformably on the coal measures beneath, an opinion confirmed by the obser- vations of Mr. Hutton, but also that its upper surface, that on which the magnesian limestone rests, is uneven. In some places (Branham Moor, North Deighton, Knaresborough,) considerable degradation of the upper surface of the lower red sandstone has taken place prior to the deposit of the lime- stone ; showing, as Prof. Sedgwick remarks, that the conti- nuity of the two deposits was partially interrupted by disturb- ing forces J. It seemed necessary to premise the above notices of the more remarkable mineralogical structures of the various rocks of this group, known as Variegated or Red Marl, Muschel- kalk, Red or Variegated Sandstone, Zechstein, and Rothlie- gendes, in order that the student might be acquainted with the whole when fully developed. Taken as a mass, the group may be considered as a deposit of conglomerate, sandstone, and marl, in which limestones occasionally appear in certain terms of the series ; sometimes one calcareous deposit being absent, as the muschelkalk is in England ; sometimes the zechstein, as in the East and South of France ; and some- times both being wanting, as in Devonshire. The conglo- merates, or rothliegendes, commonly occupy the lowest posi- tion, though conglomerates are occasionally noticed higher * Sedgwick, On the Geological Relations and Internal Structure of the Magnesian Limestone, &c. : Trans. Geol. Soc. 2nd series, vol. iii. t Hutton, On the New Red Sandstone of the County of Durham Trans. Nat. Hist. Soc. Newcastle, vol. i. I Sedgwick, Geol. Trans. 2nd series, vol. iii. p. 74. 362 Red Sandstone Group. in the series ; the sandstones form the central part, and the marls occur in the highest place. When we look for the causes which have produced this mass, we may, perhaps, in some measure approach them, by observing the state of the rocks on which it rests. These are found in the greater number of instances highly inclined, con- torted, or fractured ; evidences of disturbance which the in- ferior and older rocks have suffered previous to the deposit of the red sandstone group upon them. These appearances are not confined to one particular district, but are, with a few ex- ceptions, more or less general in Western Europe. From an examination of the lower beds, no doubt can exist that the fragments of rock contained in them have, for the greater part, been broken off from the older rocks of the more imme- diate neighbourhood. It therefore does not appear unphilo- sophical to conclude, that, as far at least as regards these lower conglomerate beds, we have approached to something like cause and effect, the cause being the disruption of the strata, the effect being the dispersion of fragments, consequent on this violence, over greater or less spaces by means of wa- ter, probably thrown into agitation by the disturbing forces. That these forces have, in some places at least, not been small, is attested by the large size of the fragments driven off, and the rounded condition of some of them, as may be well seen in the vicinity of Bristol, where the rolled masses of car- boniferous limestone are sometimes considerable. Of the evi- dence of the great force employed, I know of no better or more easily observed example, than that at the cliff' named Petit Tor, in Babbacombe Bay, Devon, whence so large a portion of Devonshire marble is obtained. Of this the follow- ing is a section : Fig. 92. P. Petit Tor Cliff, a. Fractured limestone, the rents filled, when sufficiently open, with the finer matter of the conglo- merate above ; when small, with carbonate of lime. b. A brec- cia composed of large blocks (some many tons in weight) of the same marble limestone as that on which it rests, mixed with others which are smaller. The cementing matter is sometimes a red sandstone, at others a reddish clay. The marble (known as Babbacombe marble) is wholly derived Red Sandstone Group. 363 from these blocks, which are detached from their situations, and either partially worked on the spot or removed elsewhere. Upon this rest beds of fine conglomerate, sandstone, and red marl, at c, which are surmounted by a considerable thickness of red conglomerate d, extending many miles eastward, and com- posed of angular pieces of limestone, numerous pieces of slate, such as is of common occurrence in the surrounding coun- try, as also of pebbles of flinty slate, grauwacke, &c. Among these are rounded pieces of various red quartziferous porphy- ries, f. A fault or dislocation of the strata, bringing down the conglomerates on the left hand against the fractured lime- stones on the right. Such faults or dislocations are common in the district. The annexed figure (93.) represents one of Fig. 93. the fissures in the fractured limestone at Petit Tor, filled with the matter of the superincum- bent conglomerate. &, b. Limestone, a. Fissure filled with the smaller matter of the red conglo- merate above. It will, I think, be scarcely doubted that the angular blocks of the conglomerate b (Fig. 92.) have been detached by violence from the lime- stone a, and that during the commotion they were thrown upwards, in such a manner that other and smaller detrital substances were insinuated between them ; the watery mass being highly charged with sand, mud, and other substances held in mechanical suspension. It may be right, while on the subject of these Devonshire conglomerates, to adduce evi- dence of the unequal action of currents of water, in this vici- nity, at the same period. There is perhaps no situation where better examples of this can be observed than on the line of cliffs between Babbacombe and Exmouth. The alternations of conglomerates and sandstones at the upper part of the con- glomerate series are very frequent, more particularly in the vicinity of Dawlish ; showing that the water had sometimes the power of carrying forward rounded fragments of the size of the head and even larger, while at others it merely accom- plished a transport of sand. Not only do the alternations ex- hibit this difference in the velocity of water, but the structure of the beds themselves shows that the directions of the cur- rents have continually varied, as will be seen by the annexed wood-cuts. Fig. 94. Fig. 95. 364? Red Sandstone Group. Fig. 94-. in the cliff west of Dawlish. Fig. 95. on the east of the same place, a. Conglomerate, b, b, b, b. Sandstones deposited by changing currents, c. Wavy sandstone. The velocity of the currents must have varied considerably in the immediate neighbourhood of these sections; for amidst sand- stones and moderately sized conglomerates on the west side of Little Haldon Hill, there are blocks of quartziferous por- phyry, generally rounded, of a ton or more in weight. Be- ing scattered on the side of the hill, they might be mistaken for superficial erratic blocks, did we not find them in their proper situations on the sea cliffs, imbedded in the mass of rock. The transport of these must have required water mo- ving with considerable velocity, so great, possibly, as to grind down, by attrition against each other, the rock fragments of inferior hardness, while the pieces of quartziferous porphyry being exceedingly hard, and of very difficult fracture, have better resisted attrition. The presence of these porphyries in the red conglomerate of South Devon is remarkable, inasmuch as, though rolled, masses of the same kind are not observed unconnected with the red conglomerate of the same country. The absence of such rocks on the exposed surface is certainly no proof that they may not be near ; for when we consider the area covered by the red sandstone series in that district, there is ample space for the abundant occurrence of such rocks beneath the sandstone ; and there are also many unexplored situations, where they may yet be detected among the rocks now unco- vered by the sandstone series. The student must be careful not too hastily to generalize on such facts as have been above noticed in Devonshire, for the appearances may be more or less local. When however we extend our observations, we find that conglomerates are very characteristic of deposits of the same age in other parts of Britain, France, and Germany, and they most frequently, though not always, rest on disturbed strata. As we can scarcely conceive such a general and si- multaneous movement in the inferior strata, immediately pre- ceding the first deposits of the red sandstone series, that every point on which it reposes was convulsed and threw off frag- ments of rocks at the same moment, we should rather look to certain foci of disturbance for the dispersion of fragments or the sudden elevation of lines of strata, sometimes, perhaps, producing lines of mountains, in accordance with the views of M. Elie de Beaumont. The accumulation of the larger frag- ments, and the relative amount of conglomerate, would, under this hypothesis, be greatest nearest to the disturbing cause ; and amid such turmoil we might anticipate the occurrence of igneous rocks thrown up at the same period. If we return Red Sandstone Group. S65 for the moment to that part of Devonshire with which we com- menced these remarks, we shall observe facts which seem to afford support to this view ; for where the conglomerates are abundant, there is no want of trappean rocks in the vicinity, such as various greenstones and porphyries, which have cut and broken through the slates, limestones, and other older rocks, in various directions : and I had recently an opportu- nity of observing, that red quartziferous porphyry, precisely resembling some of that which occurs so abundantly in rolled fragments in the red conglomerate of the district, is found in mass among the lower portion of the latter, and even (at Ides- ton, near Exeter,) surmounts a portion of it. But notwith- standing the abundance of the greenstones and dark-coloured porphyries, fragments of them have not yet been discovered among the conglomerates, though rolled portions of the red porphyries are so abundant : and it should be observed, that good sections are by no means rare, particularly on the coasts. We have every reason to consider that the eruption of trap rocks did accompany, if partly not produce, the disruption of strata, whence the fragments in the conglomerate were de- rived : for we have seen that red quartziferous porphyry, in mass, surmounts a portion of the red conglomerate; and the occurrence of trappean rocks (principally of a red or brown tint, and containing much siliceous matter,) so blended with the conglomerates that lines of separation cannot be drawn between them, is by no means rare in the district (Western Town, Ideston, and other places in the vicinity of Exeter). Now if igneous rocks were ejected a conclusion which the facts appear to justify at the time of the production of the conglomerate, there would seem no reason why, under fa- vourable circumstances, the two should not be in some mea- sure blended with each other. Another circumstance also lends probability to this view, and that is the occurrence of peb- bles cemented in certain inferior beds, (well observed on the coast and in-land between Babbacombe Bay and Teignmouth, at the Corbons, Turbay, in the vicinity of Exeter, and other situations,) by a kind of semi-trappean paste, containing cry- stals of that variety of felspar named Murchisonite by Mr. Levi. Such a cement might possibly have resulted from the upburst of igneous rocks, accompanied by various gases be- neath a mass of water, when some of the erupted matter may have so combined as to form a cement, in which crystals of Murchisonite became developed : without some such hypo- thesis this cement seems of very difficult explanation. We must now turn from this scene of disturbance, which may be one of the extreme cases, (though many analogous facts might be adduced,) to that state of things where no vio- 366 Red Sandstone Group. lent disrupting cause is to be surmised, but where, on the contrary, the causes which produced the arenaceous rocks that constitute the upper portion of the next, and inferior group, have not been interrupted by any sudden violenee, one series of rocks passing into the other so that the exact lines of demarcation are imaginary. Such a state of things is perfectly consistent with local and violent disturbances ; for the consequences of a violent disruption of the inferior rocks would extend no further than to distances proportioned to the agitating cause ; and the effects would gradually become less, until finally the deposits at remote places would not be interrupted, though the disturbing causes may have produced such a general state of things in the fluid mass, and in the re- lative positions of land and water, that future deposits would have an altered character ; one more common over a large area. This supposed passage of certain lower parts of the red sandstone group into the upper part of the coal measures, seems also supported by facts; for such is stated to be the case in certain parts of the continent of Europe ; so that some geologists, and among them Humboldt, Daubuisson, and others, consider the two rocks as one. Between such extremes there would be every variety of de- posit, produced either by difference in the intensity of the dis- turbing forces, or by local circumstances. Thus, sands and little or no conglomerate might be found resting unconform- ably upon older rocks, even in the vicinity of greatly disturbed situations, as may be occasionally observed in the district first noticed. After the causes, whatever they were, which produced the conglomerates and sandstones known by the name of Roth- liegendes, had in some measure been modified, a considerable deposit of carbonate of lime, often charged with carbonate of magnesia, took place over certain parts of Europe. This is the Zcchstein, which, though somewhat extensively developed in certain parts of Germany and England, seems little known in France. The causes which produced this limestone have therefore not been so general as those which have furnished the limestones formerly noticed under the head of the Oolitic Group, which are distributed over a far larger area. A de- posit of bituminous or marly slate appears to have been con- temporaneous at distant places, in parts of Germany and in the North of England, containing the remains of a marked genus of fishes, Pal&othrissum. There is nothing in itself re- markable that the same fish should be discovered in rocks formed within the same geological epoch, at such distances as Mansfeld and Durham ; for if these districts were now beneath Red Sandstone Group. 367 a common sea, no naturalist would be surprised that cod-fish, turbots, and many other fish, should be caught at the two places, being aware that cod-fish are found on the shores of North America and Europe, and that salmon ascend the ri- vers of both continents. The geologist, therefore, should expect to find the remains of similar fish entombed in con- temporaneous deposits within certain reasonable limits of la- titude and longitude. As yet, these fish seem only to have been observed in the copper-slate, or its equivalent marly slate, and they have ap- parently perished by some common cause ; what that cause was, is by no means clear ; but certainly waters which held the component parts of the copper-slate of Thuringia either in chemical solution or mechanical suspension, would be far from favourable to their existence ; and if the fish should by any chance be enveloped by, or enter into, such a medium, they are little likely to escape from it alive. When we con- sider the numerous marine animals always ready to prey upon fish either dead or alive, and the small chance that any part of them will remain undevoured, their occurrence in a fossil state would seem to show that the fossil individuals have been so circumstanced, that the creatures which preyed on them were either destroyed with them, avoided those situations which had been fatal to the fish, or were otherwise unable to get at them. By reference to the lists of organic remains, it will be ob- served that marine vegetables occur with the fish in the cop- per-slate. Now certainly these could no more exist in a me- dium impregnated with copper, than the fish ; and therefore one would suppose they existed prior to the presence of such medium : but as we cannot be certain that these grew near the spot where now entombed, (for marine plants may, like the Sargasso Weed in the Atlantic, be drifted considerable distances,) they do not afford direct proof that the copper- slate was of sudden formation. The remains of the Monitor and some plants seem to indicate a certain proximity of land. The remainder of the zechstein deposit is of a very mixed character ; part being such as we may consider mechanical, while much seems a deposit from a solution of carbonate of lime, carbonate of magnesia, and sulphate of lime. The very frequent occurrence of the two latter in rocks that have appa- rently originated from some common causes, is very remark- able, and has not yet received any satisfactory explanation. In Somersetshire and the neighbouring districts, as will be found detailed in the valuable memoir of Prof. Buckland and Mr. Conybeare, the lower part of the red sandstone group is 368 Red Sandstone Group. very frequently a conglomerate, composed of the broken frag- ments of inferior and older rocks, united by a cement contain- ing much magnesia, whence the term Magnesian or Dolo- mitic conglomerate. This rock sometimes graduates into a limestone of a more homogeneous character, apparently con- taining also much magnesia. This conglomerate seems the result of violent action on the carboniferous rocks of the di- strict, detaching various portions of them ; in fact, producing effects similar to those noticed under the head, of Rothliegen- des. How far it may be the exact equivalent of the latter de- posit, that is, how far the epoch of disturbance may be precisely contemporaneous, may admit of doubt ; for the disturbance which caused the deposit of the rothliegendes in Germany may have preceded that in Somersetshire, so that the latter may have been brought more within the influence of a spread of calcareous and rnagnesian matter. Still, however, the pro- duction of the magnesian conglomerates of Somersetshire and the rothliegendes of Thuringia would not appear to be widely separated from each other as to time; they both constitute the lower part of the red sandstone group in their respective situ- ations, and both contain fragments of rocks commonly derived from their more immediate vicinities. The organic character of the zechstein approaches, as far as researches have yet gone, that of the next, or carboniferous group ; Producta?) which abound in the carboniferous lime- stone, being not only discovered in the zechstein, (and the student will observe that these shells are now introduced for the first time to his attention,) but also Spirifers, shells which likewise abound in the carboniferous limestone. This resemblance in organic character will at all times ren- der the determination of the two rocks difficult, when their geological position cannot be ascertained with certainty, as it may be in Germany and England ; and this difficulty may in some cases be considered as insurmountable, should the de- posit of the two groups have been continuous, without a vio- lent break, the limestones of the carboniferous group being dispersed through the coal measures (the upper part of the next group,) in such a manner that they should approach the upper terms of the series on the one hand, while the zech- stein should descend towards the lowest parts of the red sand- stone group on the other. We might under these circumslances have a series of arenaceous and limestone rocks representing the carboniferous group and the lower part of the red sandstone group, with one common, or nearly common, organic character. The zechstein is surmounted by a mass of rocks for the most part arenaceous, though occasionally argillaceous, gyp- Red Sandstone Group. 369 seous and saliferous. The predominant colour is red, though it is not un frequently variegated, whence the names Bunler Sandstein, Gres bigarre. Where the zechstein is wanting, this sandstone graduates into the inferior conglomerates; and when the muschelkalk is absent, as is commonly the case in England, into the red or variegated marls: therefore where both calca- reous deposits are wanting, as in Devonshire, the whole group is composed of conglomerates in the lower part, sandstones in the central, and marls in the upper part ; an arrangement which suggests the possibility of the whole, in that district, being the result of some violent commotion, which, as the dis- turbing causes ceased, deposited the various matters held by water in mechanical suspension, somewhat in the order of their specific gravities; always considering the deposit in the mass: for not only are there alternations of conglomerates and sand- stones, sandstones and marls, where these pass into each other on the large scale, but frequent mixtures of them also occur on the small scale, a circumstance easily accounted for by the various directions and velocities of the currents produced. Viewed in the mass, circumstances appear to have been un- favourable in those parts of Europe which have been best ex- amined, if not to the existence of animal and vegeteible life, at least to their envelopment and preservation ; for, with the ex- ception of Alsace and Lorraine, few or no organic remains have been detected in it. The vegetables have been enume- rated in the lists of organic remains, from the descriptions of M. Adolphe Brongniart, as also the remains of shells noticed by M. Voltz and others. It will be observed that these shells are not analogous to those found in the zechstein, but to those discovered in the muschelkalk, a rock well developed in the same district; and it is further important to observe, that the remains discovered by M. Elie de Beaumont in the sandstone under consideration in the district of the Vosges, were not far beneath the muschelkalk. I obtained numerous fragments of vegetables from- the sandstone near Epinal, Vosges, and the quarrymen informed me that they very commonly discovered them. We next arrive, in the ascending order, to the Muschelkaljc, a limestone the general characters and known extent of whieh have been above noticed. We here have evidence, that pro- bably at the same epoch a deposit of calcareous matter, mixed sometimes with carbonate of magnesia, sulphate of lime and salt, took place, if not continuously, at least at various places, from Poland to the South of* France inclusive, and that the marine animal life distributed over this surface was nearly of the same kind. But it is a remarkable circumstance, that this 2B 370 Red Sandstotie Grottp. life was riot of the same kind as that which existed at the time when the zechstein was formed ; the organic character of the two rocks is distinct, and therefore those who found their di- visions of strata solely on this character, do well in drawing a line between the zechstein and muschelkalk. If, however, we look at these rocks on the large scale, and see the minera- logical passages which exist between the muschelkalk and the rocks above and beneath it, and observe that the latter is far from being a constant rock in the series, and that when it is absent, those beds between which it is interposed graduate into one another, there seems, theoretically, a difficulty in se- parating them, and practically, a very great inconvenience in doing so. In whatever manner this may be considered, the fact appears certain, that circumstances had arisen, changing the character of marine life over certain portions of Europe; that certain animals abounding previously, and apparently for a great length of time, (for, as will be seen in the sequel, they are enveloped in various thick and older deposits,) have dis- appeared never to reappear, at least as far as we can judge from our knowledge of organic remains. While on the subject of the muschelkalk, it may be useful to show the manner in which this rock gradually disappears in Germany, so that the Red or Variegated Marls (Kenper) above, and the lied or Variegated Sandstone (Bunter Sand- stein) beneath, approximate and finally constitute one mass, such as is seen in England. In some places thick beds of a red or variegated marl occur in the muschelkalk. These marls gradually become more developed at the expense of the muschelkalk, so that the latter appears merely as a few sub- ordinate beds in a mass of red marl, which may be considered as common both to the keuper above or the bunter sandstein beneath. This fact is observable at the termination of the north-western range of hills of Germany. It is very clearly seen on the southern declivity of the coal range of Ibben- buhren, in the district of Oster Ledde. The bunter sandstein here chiefly consists of red marls, with a few beds of sandstone. Above this follows limestone, about fifty feet thick. It is of a light gray colour, thinly laminated and friable, intermixed with some thicker beds of dark-blue, gray, and yellowish brown. Encrinites and other characteristic fossils are not wanting in it. Above this rests red marl, exactly resembling that beneath, in which there are two or three limestone beds, which can only be considered as belonging to the muschelkalk. The upper part of this marl is not to be distinguished from the keuper. The limestone beds finally vanish between the red marls by passing into rocks, which resemble the calcareous marls of the Red Sandstone Group. 371 keuper*. Thus a rock, which has formed a very marked feature in the red sandstone group, disappears, even in a coun- try where it is largely developed, and where its geological im- portance is considerable. Among the organic remains of i?ig. 95. the muschelkalk, two of the most characteristic of which are consi- dered to be the Ammonites nodosus (Fig. 96.) and Encrinites monilifor- mis(E.liliiformis Schl.), are reptiles of various forms. That extraordi- nary genus the Plesiosaurus, and perhaps also his common fossil com- panion the Ichthyosaurus, then ex- isted near what now constitutes the eastern part of France, and the adjoining portion of Germany. How far these singular Saurians now first appeared in any numbers on this part of the globe, it would be premature to say ; for it must always be re- collected, that the preservation of such remains would seem in some measure to depend on local circumstances, possibly also, in some cases, on the proximity of land, and the chance that if drifted about at sea when dead, they escaped the other preda- ceous animals of the deep, all, great and small, ready to devour them. The Red or Variegated Marls, which surmount the mus- chelkalk, possess a common mineralogical character over very considerable surfaces, such as would lead us to suppose some cause or causes exerting an influence of a similar kind over a large area. At least some of the deposit would appear che- mical, more particularly the masses of gypsum and rock-salt which exist in certain situations. How far the mass of marls may be partly chemical or wholly mechanical, may in the pre- sent state of science admit of a doubt; but the sandstones with which they are in some countries connected, and which even seem to replace them in others, (as has been above noticed,) are of mechanical origin, inclosing beds of coal, the result pro- bably of accumulated vegetable matter. Some of the vege- table remains are still sufficiently preserved to be determined, as in the red or variegated marls of the Vosges. If we now abstract our attention from these divisions, and regard the group as a mass, it would seem to constitute the base of a great system of rocks, which when not deranged by local accidents has filled numerous hollows and inequalities of land over considerable parts of Europe. Such a hollow is well * Hoffman, liber die grognostischen Verhaltnisse der Gegend von Ibben- biihren und Osnabvuck, Karsten's Archiv fur Bergb. B. 12. 2 B 2 372 Red Sandstone Group. seen in our own island, where the central counties are occupied by the red sandstone series, apparently filling up a previously existing depression in that situation ; but it is here without that great capping of the oolitic group, which for the most part rests so conformably upon it ; so that taken as a whole, and abstraction being made of minor derangements, they would both seem to fill up great depressions in Europe; sometimes, as is the case in Normandy, the oolitic rocks overlapping and coming in contact with strata older than the red sandstone group, upon which latter they nevertheless rest so conformably that the one seems a tranquil deposit on the other. We must of course consider that numerous local disturbances would pro- duce a marked difference in the deposits, even amounting to a perfectly unconformable position ; yet the conformable nature of the two groups taken in the mass is somewhat striking. During their deposit, great and remarkable changes were ef- fected in animal and, perhaps, vegetable life ; and it seems some- what necessary to admit, that considerable differences in the relative levels of sea and land were produced at various times, causing changes in the character of the inhabitants of the sea, from variations of pressure and other circumstances, while no small difference might be effected from the filling up and rise of the bottom. The following summary of the fossils stated to be found in the Red Sandstone Group of Europe, will exhibit our present knowledge of its organic contents. In the Variegated Marls. Plantte, Equisetum, 3 species; Pecopteris, 1 ,- Taeniopteris, 1; Filicites, 2; Marantoidea, 1; Pterophyllum, 3. Radiaria, Ophiura, 1. Conchifera, Plagt- ostoma, 1; Cardium, 1; Trigonia, 3; Mya, 2 ; Avicula, 3 ; Posodonia, 2 ; Modiola, 1 ; Venericardia, 1 ; Lingula, 1 ; Sax- icava, 1. Mollusca, Buccinum, 1. Pisces, genera not deter- mined. Reptilia, Phytosaurus, 2 ; Mastodonsaurus, 1 ; Ich- thyosaurus, 1 ; Plesiosaurus, 1. Thus making, Plants, 6 genera, 1 1 species. Radiaria, 1 genus, 1 species. Conchifera, 10 genera, 16 species. Mollusca, 1 genus, 1 species. Pisces, number of genera and species not determined. Reptilia, 4 genera, 5 species. Total, 22 genera, 34- species. In the Muschelkalk. Plantce, Neuropteris, 1 ; Mantellia, 1. Zoophyte, Astrea, 1. Radiaria, Cidaris, 1; Ophiura, 2; As- terias, 1 ; Encrinites, 1 ; Pentacrinites, 1. Annulata, Serpula, 2. Conchifera, Terebratula, 4; Delthyris, or Spirifer, 1; Lin- gula, 1 ; Ostrea, 9 ; Gryphsea, 1 ; Pecten, 4 ; Plagiostoma, 5 ; Avicula, 4; Mytilus, 1; Trigonia, 6; Area, 1; Cardium, 2; Mya, 5; Venus, 1; Mactra? 1; Cucullaea, 1. Mollusca, Calyptrsea, 1 ; Capulus, 1 ; Dentalium, 2; Trochus, 1 ; Turn- Red Sandstone Group. 373 .; Buccihum, 2; Strombus, 1; Natica, 2; Turbo, 2; Nautilus, 2; Ammonites, 2. Crustacea, Palinurus, 1. Pisces, genera not determined. Reptilia, Plesiosaurus, 1 ; Ichthyo- saurus, 1 ; Crocodilus, 1 ; Great Saurian, genus not named ; Chelonia, 1. Thus making: Plantoe, 2 genera, 2 species. Zoophyte, 1 genus, 1 species. Radiaria, 5 genera, 6 species. Annulata, 1 genus, 2 species. Conchifera, 16 genera, 47 species. Mol- lusca, 11 genera, 21 species. Crustacea, 1 genus, 1 species. Pisces, number of genera and species not known. Reptilia, 5 genera, 5 species. Total, 42 genera, 85 species. In the Red or Variegated Sandstone. Plants, Equisetum, 1 species; Calamites, 2; Anomopteris, 1; Neuropteris, 2; Sphenopteris, 2; Filicites, 1; Voltzia, 5; Con valia rites, 2; Paleoxyris, 1 ; Echinostachys, 1 ; Ethophyllum, 1. Conchifcra, Plagiostoma, 2; Avicula, 2; Mytilus, 1; Trigonia, 1; Mya, 2. Mollusca, Natica, \ ; Turritella, 2; Buccinum, 1. Thus making: Plants, 11 genera, 19 species. Conchifera, 5 genera, 8 species. Mollusca, 3 genera, 4 species. Total, 19 genera, 31 species. In the Zechstein. Plants, Fucoides, 6 species ; Pecopteris, 2; Lycopodites, 1; Asterophyllites, 1. Zoophyta, Gorgonia, 3 ; Calamopora, 1 ; Retepora, 2. Radiaria, Encrinus, 1 ; Cyathocrinites, 1. Conchifera, Delthyris, or Spirifer, 4; Te- rebratula, 9; Prod u eta, or Leptsena, 7; Orbicula, 1 ; Axinus, 1 ; Ostrea, 1 ; Pecten, 1 ; Plagiostoma ? 1 ; Avicula, 1 ; Myti- lus, 3 ; Modiola, 1 ; Area, 1 ; Cucullaea, 1 ; Astarte ? 1 ; Venus ? 1 . Mollusca, Turbo ? 1 ; Pleurotomaria ? 1 ; Melania ? 1 ; Ammonites, 1. Pisces, Palaeothrissum, 8; Stromateus, 2; Clupea, 1. Reptile, Monitor, 1. Thus making: Plantce, 4 genera, 10 species. Zoophyta, 3 genera, 6 species. Radiaria, 2 genera, 2 species. Conchifera, 15 genera, 34 species. Mollusca, 4 genera, 4 species. Pisces, 3 genera, 1 1 species. Reptile, 1 genus, 1 species. Total, 32 genera, 68 species. It would appear, more particularly from the descriptions of Humboldt, that very extensive tracts of red sandstones and conglomerates exist in Mexico and South America ; how far these may be of contemporaneous production with the red sandstone series of Europe, the state of science does not per- mit us very satisfactorily to determine. The porphyries and slates of New Spain are surmounted by red conglomerates and sandstones, forming the plains of Celaya, Salamanca, and Burras, and supporting a limestone which mineralogically re- sembles that of the Jura. The conglomerates contain fragments of pre-existing rocks, cemented by an argillo-ferruginous, and yellowish brown or brick red, paste. In Venezuela, the vast 374 Red Sandstone Group. plains are in a great measure covered by red sandstones and conglomerates, with limestones and gypsum ; the former being deposited in a concave manner between the coast mountains of the Caracas and the mountains of Parima, resting on slates, termed transition, on the north, while on the south they repose upon granite. This arenaceous deposit is covered, at Tisnao, by compact whitish gray limestone. An immense extent of red sandstone is described as " not only covering, nearly with- out interruption, the southern plains of New Grenada, between Mompox, Mahates, and the mountains of Tolu and Maria, but also the basin of the Rio de la Magdalena, between Teneriffe and Melgar, and that of the Rio Cauca, between Carthago and Cali." The conglomerates of this country are composed of angular fragments of lydian stone, clay-slate, gneiss and quartz, cemented by argillaceous and ferruginous matter. These con- glomerates alternate with schistose and quartzose sandstones *. According to Humboldt, the Cordilleras of Quito presented him with the greatest extent of red sandstone which he had ob- served, covering the whole plateau of Tarqui and Cuenca for twenty-five leagues. The sandstone is generally very argilla- ceous, with small grains of slightly rounded quartz ; but it is sometimes schistose, and alternates with a conglomerate con- taining fragments of porphyry from three to nine inches in di- ameter. The same author considers that the red sandstone of Cuenca also occurs in High Peru, and remarks on the resem- blance of these rocks of New Grenada, Peru and Quito, to the red sandstone or rothliegendes of Germany f. A series of red sandstones, intermixed with conglomerates, occurs extensively in Jamaica, particularly in the Port Royal and St. Andrew's mountains, stretching thence north-west towards the north side of the island. The sandstone is gene- rally siliceous and compact, intermixed with marly red sand- stone and marl, and, though rarely, with gypsum (Hope Val- ley). The conglomerate is formed of pebbles (from an inch to four inches in diameter) of granite, large-grained greenstone, sienite, quartz, hornstone, &c. Beds of a gray colour are in- termixed with these rocks ; and subordinate to them are strata of compact gray limestone and of shale, and schistose sandstone intermixed with coal. The higher portion of the mass is formed of a conglomerate in a great measure composed of pieces of trap rocks, principally porphyry, the cementing matter being most frequently reddish brown and argillaceous, varying in in- duration, sometimes so obscure that the pebbles seem joined by a trappean cement. Mixed more particularly with this supe- rior portion there is a great variety of trappean rocks, such as * Humboldt, Gisement des Roches dans les deux Hemispheres, f Ibid. Red Sandstone Group. 375 sienite, greenstone, porphyries, &c. appearing as if an upburst of igneous matter had accompanied the production of the con- glomerates. The red conglomerates and sandstones pass be- neath into a rock, which at first differs from them only in co- lour, and finally presents the mineralogical character of grau- wacke. The aggregate thickness of the whole is considerable, amounting to several thousand feet. These rocks appear to me the equivalent of those named red sandstone in the neigh- bouring continent of America*. The mere mineralogical resemblance of this deposit, in Ame- rica and Jamaica, with the sandstones and conglomerates of the red sandstone group of Europe, is in itself of no great va- lue, and therefore we can only at present conclude that consi- derable forces have been exerted in both parts of the world (whether contemporaneous or not remains to be determined), which have dispersed fragments of pre-existing rocks, scatter- ing them, most probably by the medium of agitated water, in various directions, the transporting powers being unequal, so that sandstones and marls alternate with conglomerates. These sandstones and conglomerates would appear, from the descrip- tions of geologists and intelligent travellers, to extend from Mexico far into the heart of North America ; so that if differ- ent deposits have not been confounded under one head, as might easily happen in England if it were an uncultivated country and rapidly examined, (the old red sandstone of En- glish geologists being confounded with their new red sand- stone,) these sandstones and conglomerates of America would appear not the result of a limited disturbance, but of one com- mon to a considerable surface. * For a more detailed account of these and other Jamaica rocks, with sec- tions, consult my Remarks on the Geology of Jamaica, Geol. Trans, 2nd series, vol. ii. 376 Carboniferous Group. SECTION VIII. CARBONIFEROUS GROUP. SYN. Coal-measures, Engl. Auth. (Terrain Houttler, Fr. Auth. SteinJcohlen- gebirge, Germ. Auth.). Carboniferous limestone, Conyb. (Mountain lime- stone, Engl. Auth. Calcaire carbonifere, Calcaire anthraxifere, Calcaire de Transition, Fr. Auth. Bergkalk, Kohlenkalk, Uebergangskalk, and Neuere Uebergangskalk, Germ. Auth.). Old red sandstone, Engl. Auth. (Ores rouge intermediate, Fr. Auth. Jiingeres Grauwackengebirge, Germ. Auth. Alter r other sandstein, Von Dechen.) Coal Measures. THESE are composed of various beds of sandstone, shale, and coal, irregularly interstratified, and in some countries inter- mixed with conglomerates; the whole showing a mechanical origin. The coal-measures abound in vegetable remains, and the coal itself is now, by very general consent, referred to a vegetable origin, being considered the accumulation of an im- mense mass of plants. The sandstone, shale, and conglomerate beds vary much, as might be expected, in different situations. Some being, even in the same districts, more continuous than others ; a sandstone bed, for example, becoming gradually thinner and finally disap- pearing, so that the beds above and beneath, should their conti- nuity continue, come into immediate contact with each other. The like happens with the shale, conglomerate, and even coal- beds. Undoubtedly some beds are persistent over a great area, but nothing can be more various than the areas occupied by any given beds. It thus becomes highly important to trace given beds with accuracy, noting whether they terminate by gradually fining off, or by an almost insensible change in their mineralogical character ; such, for instance, as a sandstone bed acquiring argillaceous matter by degrees, and thus passing through a sandy into an argillaceous shale. It is stated by Mr. Buddie, that as the strata of the New- castle coal-field rise or crop upwards, the sandstone beds in- crease in number and thickness; whereas the argillaceous shales increase in the opposite direction *. The same author re- marks, that the quality of the coal in the same district greatly depends on the kind of bed which immediately covers the coal stratum, and that the coal is always deteriorated when covered by sandstone, becoming in that case more or less mixed with * Buddie, Trans. Nat Hist. Soc. Newcastle, vol. i. p. 238. Coal Measures. 377 iron pyrites ; whereas it is of comparatively good quality be- neath argillaceous shale*. This difference in the comparative value of the coal itself beneath sandstone or shale, would seem readily accounted for by the difference in the character of the two rocks, the first being generally pervious to gases and liquids, while the shale would by no means so readily permit the passage of either. Hence it would follow, not only that the vegetable matter would be better preserved beneath the shale from the comparative difficult escape of gaseous matter upwards, but that it would also be preserved from injury by the difficulty with which water, not only pure, but charged with foreign matter, would have in percolating downwards. In tracing coal-beds, it has been observed, that the same bed frequently varies much in thickness, sometimes being much thinner than at others. Moreover shale and sandstone often get interstratified, as it were, with the coal-beds, thus parting them into minor beds. The area and depth occupied by these interstratified portions of sandstone and shale, though of little importance in some beds, become not unfrequently in other coal strata so considerable as seriously to injure the value of the coal. It is by no means uncommon to describe the coal deposits as basins; but it may be doubted how far this term is generally correct; for admitting that many accumulations of these beds have been deposited within depressions of the surface, it would by no means seem to follow, reasoning at least from the mode in which vegetable accumulations are now formed, that all coal- measures have been thus produced. Suppose plants to be car- ried down rivers, such as now happens at the mouth or delta of the Mississippi, we should ill characterize the deposit, by the term basin-shaped, which would seem to imply a hollow or de- pression, bounded by a circumference of nearly equal eleva- tion. The following analyses by Dr. Thomson exhibit the va- rious proportions of elementary substances existing in some of the coals of England and Scotland : Carbon. Hydrogen. Oxygen Nitrogen. Newcastle Caking Coal . . . 75-28 ... 4-18 ... 4.58 . . 15-96 Glasgow Splint Coal . . , . 75-00 . . . 6-25 . . . 12-50 . . 6-25 Glasgow Cherry Coal .... 74-45 . . . 12-40 . . . 2-93 . . 10-22 Cannel Coal 64-72 . . . 21-56 . . . 0-00 . . 13-72 Carburetted hydrogen, ovjlre damp of the collieries, is well known as an abundant product of most coal-mines. Mr. Buddie observes that the discharge of this gas in the ordinary work- ings of a colliery much depends on the pressure of the atmo- * Buddie, Trans. Nat. Hist. Soc. Newcastle, vol. i. p. 217. 378 Coal Measures. sphere; being greatest when that pressure is least, the gas being then enabled more freely to escape from the pores and fissures of the coal. The same author remarks, that a sand- stone roof (the bed immediately above the coal stratum) from being commonly split into innumerable fissures, and, conse- quently, receiving the gas when evolved from the coal beneath, becomes a great natural gasometer, ready to pour out the car- buretted hydrogen when circumstances are favourable*. To this it may be added, that those sandstones which are suffi- ciently porous to permit the percolation of water downwards, would likewise allow of the escape of gas upwards, particularly if it were in a compressed state: hence a sandstone roof would probably, in many cases, be saturated with carburetted hydro- gen, of greater or less density, according to circumstances. Mr. Hutton considers that this gas exists in a highly con- densed, and even liquid, state in the pores of the coal. This opinion is rendered probable by the fact that small explosions from coal, technically called eructations, are not uncommon when the coal is struck with the pick ; these explosions being apparently due to the sudden expansion of condensed gas. Mr. Buddie relates, that in a part of Jarrow colliery, near Newcastle, the eructations were as loud as the report of a mus- ket, large fragments of coal being thrown off at the same timef. Coal varies considerably in the quantity of bitumen it con- tains, and is more or less valuable for economical purposes ac- cording to the admixture of this substance. The quantity of coal raised in the British Isles is very considerable, and it may be said that to this substance and the iron-ore found in the same deposit, England owes a great part of her commercial pros- perity; for to the abundance and cheapness of both these sub- stances in various districts, we are indebted for a large pro- portion of our manufactures, the same series of beds not only furnishing fuel for working the steam-engines, but also iron for their construction. In the present condition of the coal-measures, opportunities of observing design seem to be afforded us, even when these rocks are so disposed that at first sight such design does not appear very obvious. * Buddie, on the Explosion in Jarrow Colliery : Trans. Nat. Hist. Soc. New- castle, vol. i. f Buddie, Ibid. The explosion which took place in Jarrow colliery, Au- gust, 1830, was caused by a volume of cai*burettcd hydrogen being highly compressed near a small fault. When the workmen approached this volume of compressed gas, they necessarily weakened the resistance of the coal in the direction of the workings. Finally, the resistance being unequal to the pressure of the gas, the coal was driven out into the galleries, the carburetted hydrogen mixed with the common air, and exploded at the contact with flame, killing many of the miners. Coal Measures. 379 The accumulation of vegetable matter at a remote epoch in the history of the world, for the consumption of creatures which should afterwards exist on its surface, must strike the least in- quiring ; but when the upturned, twisted, and shattered strata, so common in the districts composed of the coal-measures, are before us, design is not so apparent, more particularly when the miner complains of the dislocations (faults) which interrupt his progress*. We might therefore regard this apparent con- fusion as a bar to the ingenuity and industry of man in extract- ing the combustible so valuable to him. When, however, we look more closely into this subject, we find that the shattered and contorted condition of the rocks, though it may embarrass mining operations for a time, is in reality highly advantageous. The fractures, termed faults, frequently so cross each other, that the surface, if it could be examined without its covering of vegetation and detritus, would present much the same ap- pearance on the great scale, as the frozen surface of a lake broken to pieces and reunited by subsequent frost. Masses of fractured strata are thus often bounded by faults which pre- vent the passage of subterraneous waters from one mass into the other; and the miners, in collieries situated in one parti- cular mass, have only to contend with the waters in it ; whereas if the strata were always horizontal, unbroken, and continuous, the abundance of water that would flow into the workings would render them so difficult and expensive, that the extraction of the coal must be abandoned. It should be observed, that though the two sides of a fault often come into close contact, there is very frequently an in- terposed clayey substance impervious to water; and it rarely happens that water on the one side passes to water on the other, so as to form a continuous and abundant percolation in one direction. On the contrary, the water is commonly thrown out along the line of the fissure, particularly on mountain sides, in the shape of springs, often good guides to the geologist, not only in tracing faults among the coal-measures, but in other rocks. The appearance of springs along lines of fault is what we should expect; for not only do they act as main drains to the strata which they traverse, but as Artesian wells, producing the same effects as these artificial perforations. For supposing faults to be abundant, in countries where Artesian wells are now so valuable, such countries would possess an abundant supply of water, upon the same principle that these wells now act. Knowing, therefore, that faults are abundant on the sur- * For sections of faults in coal-measures, see Sections and Views illustra- tive of Geological Phenomena, pi. 5. 6. 7. ; Geol. Trans. 2nd series, vol. i. pi. 32. ; Transactions of the Nat. Hist. Soc. of Newcastle, vol. i. ; La Ri- chesse Minerale, by M. Heron de ViHefosse, &c. 380 Coal Measures. face of our planet, we may infer that no inconsiderable portion of water is conveyed by them to that surface. It will be obvious that in a district where the coal-measures are greatly contorted, the relative position of strata alone would prevent the abundant percolation of water from one situation to another. The following section (Fig. 97), at Jarrow colliery, near Newcastle, will afford an idea of the manner in which the coal- measures are sometimes fractured *. Fig. 97. The mode in which they sometimes are contorted without fracture, is shown in Fig. 98., a section of the coal-measures at Little Haven, St. Bride's Bay, Pembrokeshire. Fig. 98. Little Haven. The organic remains discovered in the coal-measures are principally terrestrial plants ; with these are a few fresh- water shells, and certain marine exuvia?, which for the most part would rather appear to occur in beds alternating with the coal- beds and their accompanying shales and sandstones, than mingled with the terrestrial remains. The following is a sum- mary of the fossils stated to have been found in the coal-mea- sures. Plants. Stigmaria, 9 species. Pinites, 3. Peuce, 1. Spe- nophyllum, 10. Annularia, 7. Asterophyllites, 12. Beche- ra, 1. Flabellaria? 1. Noeggerathia, 2. Cannophyllites, 1. * A small portion of Mr. Buddie's beautiful sections of the Newcastle coal-field, Trans. Nat. Hist. Soc. Newcastle, vol. i. pi. 21. 22. 23. Carboniferous Limestone. 381 Sternbergia, 4. Poacites, 2. Trigonocarpum, 4. Musocar- pum, 3. Equisetum, 2. Calamites, 13. Sphenopteris, 32. Cyclopteris, 9. Neuropteris, 17. Odontopteris, 6. Peco- pteris, 62. Lonchopteris, 1. Schizopteris, 1. Caulopteris, I. Lycopodites, 8. Selaginites, 2. Lepidodendron, 43. Ulo- dendron, 1. Lepidophyllum, 2. Lepidostrobus, 2. Car- diocarpon, 5. Sigillaria, 37. Volkmannia, 4. Cyperites, 1. Polyporites, 1. Conchifera. Pecten, 2. Mytilus, 1. Lutricola, 3. Unio, 3. Nucula, 2. Mya? 3. Mollusca. Turritella, 2. Bellerophon,2. Orthoceratites, 5. Ammonites, 5. Pisces. Palaeothrissum, 2. Acanthaessus, 1. . Thus making: Plants, 35 genera, 310 species. Conchi- fera, 6 genera, 14 species. Mollusca, 4 genera, 14 species. Pisces, 2 genera, 3 species. Total, 47 genera, 341 species. Carboniferous Limestone. This rock, in the South of England, Wales, the North of France, and Belgium, seems to possess a somewhat general character, being a compact limestone, frequently traversed by veins of calcareous spar, and at times being in a great measure composed of organic remains, while at others not a trace of these remains can be observed. The colours are mostly gray, varying in intensity of shade ; other tints are however observed in it, and in some situations it affords good marble. It is oc- casionally of an oolitic structure, as near Bristol ; and some- times contains parts of encrinital columns in such abundance, that the rock is in great measure made up of them, whence the name Encrinal Limestone. It has also been known by the name of Metalliferous Limestone, in consequence of the quan- tity of lead-ore obtained from it, more particularly in the cen- tral and northern parts of England. Though these charac- ters may suffice for a considerable portion of the carboniferous limestone, it will be seen by the sequel, that even within the distance of two or three hundred miles, the series of limestone beds known by this name becomes mixed with shale, sand- stone, and even with coal. The following is a summary of the organic remains enume- rated as having been discovered in the carboniferous limestone, the plants only being omitted ; as when such exuviae occur, they are generally found to correspond with the plants of the coal-measures. Zoophyta. Gorgonia, 4 species. Cellepora, 1. Ketepora, 1. Caryophyllia, 3. Cyathophyllum, 4. Astrea, 1. Tubipora, 1. 382 Old Red Sandstone. Syringopora, 3. Calamopora, 2. Aulopora, 1. Favosites, 3. Lithostrotion, 3.* Radiaria. Pentremites, 3. Poteriocrinites, 2. Platycri- nites, 7. Actinocrinites, 5. Melocrinites, 1. Rhodocrinites, 1. Cyathocrinites, 2. Annulata. Serpula, 2. Conchifera. Spirifer, or Delthyris, 29. Terebratula, 21. Atripa, 1. Producta, or Leptaena, 29. Crania, 1. Inoce- ramus, 1. Pecten, 2. Megalodon, 1. Nucula, 1. Area, 1. Isocardia, 1. Cardium, 4-. Lucina, 2. Sanguinolaria, 1. Solen, 2. Mollusca. Pileopsis, 1. Melania, 1. Ampullaria, 2. Ne- rita, 3. Delphinula, 1. Euomphalus, 12. Trochus, 7. Tur- bo, 6. Rotella, 3. Helix?!. Turritella, 10. Buccinum, I. Phasianella, 3. Bellerophon, 12. Conularia, 2. Orthocera- tites, 13. Nautilus, 9. Ammonites, 2. Crustacea. Asaphus, 1. Other Trilobites not determined. Pisces. Ichthyodorulites and fish palates. Thus'making: Zoophyta, 12 genera, 27 species. Radiaria, 7 genera, 21 species. Annulata, \ genus, 2 species. Conchi- fera, 15 genera, 97 species. Mollusca, 18 genera, 89 species. Crustacea, 1 genus, I species. Pisces, 1 genus, 1 species. Total, 55 genera, 238 species. Old Red Sandstone. This rock is of very variable thickness, sometimes consist" ing of a few conglomerate beds, while at others it swells out to the depth of several thousand feet. As might be expected, this variation in thickness is accompanied by differences in mi- neralogical structure, conglomerates being abundant in some situations, while in others they are exceedingly rare. The sandstone possesses different degrees of induration, and is not unfrequently schistose and micaceous, affording flag-stones and coarse materials for roofing. The prevalent colour is red, generally dull, which, as commonly occurs in the red marls and sandstones of all ages, is occasionally intermixed with different tints of greenish blue (Pembrokeshire, &c.). The conglomerates of course vary in their contents, but pieces of quartz are very common, so much so in the southern parts of England and Wales, that the greater portion of such beds is wholly composed of them. The sandstones also are prin- * This catalogue is exceedingly meagre ; for, unfortunately, British natu- ralists have published next to nothing on the zoophytic contents of the car- boniferous limestone, though this rock is extensively developed in the British Isles, and contains a great abundance of Polypifers. Carboniferous Group. 383 cipally siliceous, so that if the mass be considered as wholly of mechanical origin, it must have resulted from a considera- ble destruction of pre-existing siliceous rocks. Few organic remains have been discovered in this rock, and those that have been observed would appear, for the most part, to be the same as in the grauwacke beneath, or the carbonife- rous limestone above. According to Dr. Fleming, Orthocera- tites cordiformis, O. giganteus, Nautilus bilobatus, and N.pen- tagomiS) are found in a limestone associated with the old red sandstone of Dumfriesshire* ; and M. Dumont notices Producta concinna in the old red sandstone of the Liege district. The student being now acquainted with the general zoolo- gical and botanical characters, and with the more marked mi- neralogical composition of the three rocks comprised within this group, we will proceed to a more general notice of the same rocks taken in the mass. It has been above remarked, that the rothliegendes is con- sidered, in certain parts of Europe, to pass into the coal-mea- sures, so that the two rocks constitute the upper and lower portion of the same mass. Some geologists have gone further, and considered the coal-measures as subordinate to the roth- liegendes, the carboniferous portion bearing the same re- lation to the general mass, which certain lignites, such for instance as those noticed by M. Elie de Beaumont in Dau- phine and Provence, bear to the transported matter in which they are found included. It would appear from the observations of Von Veltheim on the district of Wettin and Loebejun, and on the eastern Hartz, that the coal-measures are there subordinate to the rothlie- gendes, and that the whole mass may be divided into three portions: 1. The lowest, composed of red sandstone, with slaty clays, schistose sandstone, and conglomerate, 500 feet thick (considered equivalent to ihe English old red sand- stone). 2. The middle, carboniferous rocks and limestone, 250 feet thick (supposed equivalent to the carboniferous lime- stone, coal-measures, and millstone grit of the English series). 8. The upper, formed of red sandstone, with slate-clays, con- glomerates, and porphyry breccia, 2600 feet; thick (referred to the red sandstone group). M. von Dechen observes that it is particularly difficult to afford a better explanation of these facts, as the coal depo- sit of Wettin and Loebejun, wedged in between two kinds of porphyry, is disturbed and irregular. The same author re- marks, however, the important fact, that there can be no doubt of the coal deposit being covered by the upper beds of * Fleming, History of British Animals, 384? Carboniferous Group. the real rothliegendes, and resting on red sandstone. The identity of the latter sandstone with the lower strata of the rothliegendes on the borders of the Hartz seems not so evi- dent, though it would appear probable from the careful re- searches of M. von Veltheim *. It is certainly exceedingly de- sirable that the identity of these latter rocks should be more clearly proved ; for as the sandstones of the coal-measures are often red, it might be considered that the lowest rocks at Wettin and Loebejun were merely such red sandstones, and that the coal-measures were covered by the upper members of the rothliegendes, because the lower beds were wanting. A good example of coal subordinate to the red sandstone is observed in the coal series of Waldenburg and Neurode, in Lower Silesia, especially in that part of the deposit which ex- tends into Bohemia, between Schatzlar and Nachod. In the neighbourhood of Altwasser, red sandstone and conglomerate rest in a narrow band upon grauwacke, and, in an adit, a bed of coal, with slate-clay, is seen apparently in a continuation of the red conglomerates. North from Schatzlar, above Liebau, to the environs of Waldenburg, the coal-measures rest imme- diately upon grauwacke, without any intermediate beds. Be- neath the Bohemian coal, at Rohnow on the Metau, there is a thick mass of fine-grained red sandstone. Its beds form a saddle from Nachod to Schatzlar, and on the south-west side of it, towards the basin of Bohemia, no coal-mine is known, there being merely a continuous uninterrupted mass of red sandstone. A bed of bituminous shale, with the remains of fish, indistinct impressions of plants, and disseminated copper- ore, occurs beneath the coal near Eipel, not far from Saug- witz ; above this occurs limestone, about 20 feet thick, and beneath the whole, white sandstone. North-west of Rohnow, red sandstone appears between the coal strata, and soon at- tains such thickness, that with a dip of about 20, it occupies a breadth of 2j miles on the surface. It in this manner di- vides the coal-measures, for a length of 17 miles, into two clearly defined parallel lines. This sandstone, though it per- fectly resembles that beneath, must necessarily be considered as an integral portion of the coal-measures. The two lines of coal above noticed do not unite (as far as is yet known) with the coal-field of Schatzlar, but cease in the middle of the red sandstone district at Goldenelse and Teich- wasser ; and the upper line is also covered by red sandstone, and that uniformly for the distance above mentioned. The same red sandstone covers the coal deposit on the Silesian side. At Lang- waltersdorf, the red and gray coal sandstones alternate with each * Von Dcchen, German Transl. of Manual. Carboniferous Group. 385 other, but in general the limits of both are well defined. Conglomerates are rare in the upper sandstone, which is usu- ally fine-grained ; pebbles of quartz and flinty slate are found only in its lower portion. Red slate clays are not frequent. Single, but long continued, beds of limestone and dolomite are found in it (Griisow, and Conradswalde); they are usually red- dish gray, seldom dark gray (Ottendorf, and Scheidwinkel). It contains beautiful impressions of Neuropteris coiiferta. Im- pressions offish are found in it at Ruppersdorf, which, like those of Saugwitz, are not yet determined. As the red sand- stone is covered by quadersandstein (green sand), many of its relations are uncertain, but very probably it is parallel with the rothliegendes of Mansfeld. If at Waldenburg a separa- tion of the upper red sandstone from the coal deposit is pos- sible, because an unconformable stratification sometimes takes place, the latter cannot be separated from the lower red sand- stone, as the beds of this sandstone are quite conformable with those of the coal strata *. By consulting Mr. Weaver's observations respecting the views of the German geologists previous to those of MM. von Veltheim and Hoffmann, it will be seen he was led to the con- clusion, that the old red sandstone of the English geologists might be equivalent to the lowest part of the German rothlie- gendesf. Perhaps by following up the views of Mr. Weaver, and con- sidering that the coal is not necessarily constant to a particular part of the series, and that the mutual relations of different portions of the whole mass vary materially, we may approach towards a solution of this apparent difficulty. In the first place, we obtain no help from organic remains ; for it will have been observed that the general zoological character of the marine exuviae is the same in the zechstein (above the roth- liegendes), in the carboniferous limestone, and, as will be seen in the sequel, in the grauwacke series. The general character of the vegetable remains was probably also similar, as we know it was in the descending order. Assuming, therefore, (and it does not appear unphilosophical to do so,) that organic remains will not aid us in the investigation, we can only ap- peal to mineral ogicai structure and relative geological posi- tion. Our first inquiry therefore should be, Are these con- stantly the same, allowances being only made for smaller * Von Dechen, German Transl. of Manual. f It should be stated, while on this subject, that, according to M. He- rault, th'e coal-measures of Littry, Normandy, which rest unconformably on grauwacke, pass into the red sandstone series above them. Tableau des Terrains de Calvados : Caen, 1832. 2c 386 Carboniferous Group. variations ? We can only reply to this question by a statement of facts. In the southern part of our island the three divisions of old red sandstone, carboniferous limestone, and coal-measures, are well marked, and there is clearly no passage of the latter into the red sandstone (commonly termed new red sandstone) above it ; on the contrary, the coal-measures and the inferior rocks have been upset prior to the deposition of the magnesian con- glomerates and limestones with their associated red sandstones and conglomerates ; and it seems more than probable that the lower portions of the latter series of rocks resulted from the disturbance produced by the fracture, contortion, and elevation of the coal-measures and older rocks. With respect also to the carboniferous group itself, the masses of the old red sand- stone, carboniferous limestone, and coal-measures, are well separated from each other, though there may be small alter- nations at their contact, as the student can observe at Clifton gorge near Bristol, and other places in that district *. As we advance northwards into the central part of England, we find that the lower part of the coal-measures and the upper part of the carboniferous limestone, which in the south only alternated at their contact in a comparatively moderate degree, have now assumed a new character as they approach each other, presenting a mass of shales, sandstones (most frequently coarse), and limestones, with occasional seams of coal, the whole being of very considerable thickness, and known as Millstone Grit. Prof. Sedgwick has shown, that still further north in England the great lines of distinction between the carboniferous lime- stone and the coal-measures are broken up, and that the one rock is lost in the other. As the student can have no better or more condensed view of the subject than in Prof. Sedgwick's own words, I shall offer no apology for inserting them here. " On the re-appearance of the carboniferous limestone at the base of the Yorkshire chain, we still find the same general analogies of structure : enormous masses of limestone form the lowest part, and the rich coal-fields the highest part of the whole series; and we also find the millstone -grit occupying an intermediate position. The millstone-grit, however, becomes a very complex deposit, with several subordinate beds of coal ; and is separated from the great inferior calcareous group * For the necessary details respecting the coal-measures and the carboni- ferous series generally, consult the labours of Mr. Conybeare in his Outlines of the Geology of England and Wales ; and for that of the southern part of our island in particular, the Observations on the South-western Coal Dis- trict of England, by Dr. Buckland and Mr. Conybeare, Geol. Trans., 2nd series, vol. i. Carboniferous Group. 387 (known in the north of England by the name of scar limestone], not merely by the great shale and shale-limestone, as in Derby- shire, but by a still more complex deposit, in some places not less than 1000 feet thick, in which five groups of limestone strata, extraordinary for their perfect continuity and unvarying thickness, alternate with great masses of sandstone and shale, containing innumerable impressions of coal plants, and three or four thin seams of good coal extensively worked for domes- tic use. " In the range of the carboniferous chain from Stainmoor, through the ridge of Cross Fell to the confines of Northumber- land, we have a repetition of the same general phenomena. On its eastern flanks, and superior to all its component groups, is the rich coal-field of Durham. Under the coal-field we have, in regular descending order, the millstone-grit, the alter- nations of limestones and coal-measures nearly identical with those of the Yorkshire chain, and at the base of all is the great scar limestone. The scar limestone begins however to be sub- divided by thick masses of sandstone and carbonaceous shale, of which we had hardly a trace in Yorkshire, and gradually passes into a complex deposit, not distinguishable from the next superior division of the series. Along with this gradual change is a greater development of the inferior coal-beds al- ternating with the limestone, some of which, on the north-east- ern skirts of Cumberland, are three or four feet in thickness, and are now worked for domestic use, with all the accompani- ments of rail-roads and steam-engines. " The alternating beds of sandstone and shale expand more and more as we advance towards the north, at the expense of all the calcareous groups, which gradually thin off and cease to produce any impress on the features of the country. And thus it is that the lowest portion of the whole carboniferous system, from Bewcastle Forest along the skirts of Cheviot Hills to the valley of the Tweed, has hardly a single feature in common with the inferior part of the Yorkshire chain ; but, on the contrary, has all the most ordinary external characters of a coal-formation. Corresponding to this change is also a gradual thickening of carbonaceous matter in some of the lower groups. Many coal-works have been opened upon this line, and near the right bank of the Tweed (almost on a paral- lel with the great scar limestone) is a coal-field with five or six good seams, some of which are worked, not merely for the use of the neighbouring districts, but also for the supply of the capital *." * Seclgwick, Address to the Geological Society, 1831. Phil. Mag. and Annals, vol. ix. pp. 286, 287. 2c 2 388 Carboniferous Group. We thus observe that a very material change has been ef- fected in the carboniferous rocks, the limestone beds having become mixed up with, and even disappearing among, the are- naceous and shaly coal-measures. Two rocks of the series are therefore as it were amalgamated, and no line of distinction can be drawn between them. Not only have the separate cha- racters of coal-measures and carboniferous limestone disap- peared, but the remaining rock, the old red sandstone, no longer presents the common arenaceous aspect which it pos- sesses in the South of England and in Wales. It is here a conglomerate, which, instead of offering the appearance of a passage into the grauwacke strata beneath, actually rests on the upturned edges of those strata, and is frequently absent, so that, as has been shown by Prof. Sedgwick and Mr. J. Phil- lips, the carboniferous limestones repose directly on the pre- viously disturbed and upset grauwacke rocks*. If we now proceed to Scotland, to that mass of conglomerate and arenaceous deposits intermixed with limestone and coal described by Dr. Fleming, Prof. Jameson, Dr. MacCulloch, Mr. Bald, Dr. Boue, Prof. Sedgwick, Mr. Murchison, and other geologists, there would appear to be some difficulty in establishing distinctions such as can readily be made in the southern parts of our island ; and this difficulty is increased by the presence of rocks, referrible, at least in part, to the (new) red sandstone group. In the northern English districts noticed by Prof. Sedgwick, the red sandstone rocks have clearly been deposited on the carboniferous limestone and coal-measures after the two latter rocks had suffered great disturbance and violent dislocations; but it may be question- able, at least in parts of Scotland, how far fine lines of distinc- tion can be drawn between the upper part of the coal-measures and the lower portion of the red sandstone group. Organic remains will be of little assistance, for reasons before stated, neither is the mineralogical character of much avail, for it will have been seen that this also changes; and there is nothing, that we are aware of, which should prevent the zechstein, if produced under general similar circumstances, from assuming the character of the carboniferous limestone, such more par- ticularly as the latter appears when divided and included in the coal-measures. The colour of the rocks is, if possible, of still less importance, for the coal-measures are not unfrequently red ; so that should the whole get mixed up together, more * See Phillips's Sections in the Geol. Trans., 2nd series, vol. iii. ; and Sedgwick, Proceedings of the Geol. Soc. 1831. When the sections and descriptions of the latter author shall have been made public, geologists will be in possession of a highly valuable and illustrative series of documents on this subject. Carboniferous Group. 389 particularly without discordant stratification, it would appeal- highly theoretical to distinguish the various portions by par- ticular names, each portion being considered the decided equi- valent of divisions that may be established elsewhere. It is by no means intended to infer that during any considerable de- posit, such as the one under consideration, there should not be equivalents in age: such there must always have been ; but the contemporaneous effects produced by different causes may have varied most materially; so that distinctions which mark particular events in one situation are not always useful when applied generally; for, perhaps without being aware that we are doing so, we theoretically consider circumstances to have been generally the same at a given time, whereas we should, in the first place, consider them as more local, resulting from the operation of more limited causes. I am fully aware that this view may be carried too far, and that the minor divisions of rocks should be established as much as possible; but we should also avoid extremes, arid not pass that point where the distinctions may admit of very great doubt, for by doing so we seem in a great measure to preclude ourselves from tracing the causes which have produced the great changes in the minera- logical and zoological characters of rocks on the surface of the earth generally. Dr. Boue considers the conglomerates, sandstones, lime- stones and coal of the great arenaceous deposit of Scotland, as subordinate portions of one great whole, which he believes equivalent to the red sandstone (gres rouge). To what extent this opinion may be correct would not yet appear to be well decided ; and it no doubt may startle English geologists to compare, in any manner, the old red sandstone with a system of rocks containing the rothliegendes: but supposing a series of conglomerate, sandstone, and other rocks of a certain com- mon character, to have been produced, so that during the de- posit the inferior should not in any manner have been disturbed, but the strata to have been laid regularly on one another, and that we obtain little or no aid from organic remains, it would seem difficult to regard the mass in any other light than as re- sulting from the operation of nearly similar and uninterrupted causes. I am far from stating that this actually is the case in Scotland, being merely desirous to show that an union of the rothliegendes and zechstein with the carboniferous group may not be impossible elsewhere. It may, however, be interest- ing to the student to learn, while on this subject, that the Ca- lamites Mougeotii, first discovered in the red sandstone group of the Vosges, is, according to Prof. Lindley and Mr. Hutton*, also found in the Edinburgh coal-measures. Undoubtedly we * Lindley and Hutton, Fossil Flora of Great Britain. 390 Carboniferous Group. cannot hence infer the exact contemporaneous deposit of the two rocks, any more than the identity of the zechstein and car- boniferous limestone, because certain fossils are stated to be common to both ; but the connexion, as far as respects organic remains, is important. In certain districts, such as Pembrokeshire, the old red sandstone passes into the grauwacke series beneath : in such situations, therefore, we may regard this rock as resulting from the continuance of causes similar to those which have produced the grauwacke; for the zoological character of the two deposits is the same, as is also their mineralogical structure; the dif- ference between them is in the colour, a circumstance of no importance, for red rocks are often present in the body of the grauwacke group itself. In the north of England, the old red sandstone, as has been above noticed, rests on upturned grauwacke : causes therefore have acted violently in one situation at a given period, which have scarcely, if at all, produced marked effects in another at no very considerable distance; leading us to infer that at still greater distances the differences observable in deposits of this period may be more remarkable. For the following sketch of the range of the old red sand- stone through Great Britain, I am indebted to my friend Pro- fessor Sedgwick. It will no doubt be perused with much in- terest, presenting as it does this distinguished author's views on a part of a system of rocks, which, as is well known, have for many years more particularly engaged his attention. He observes that the old red sandstone not only insensibly passes into grauwacke beneath it in the south-western parts of En- gland, but that it may also be said to graduate into the carbo- niferous limestone above it in the same districts by the inter- vention of alternating beds of sandstone. He also observes, as a general fact, that the coal-measures, the carboniferous limestone, and the old red sandstone, are obviously affected by a common system of flexures, produced by disturbing forces, posterior to them all. The truth of this observation is evident from the various sections of Professor Buckland, Mr. Cony- beare, and Mr. Weaver*. Professor Sedgwick points out, as the next remarkable fact connected with the history of these deposits, the entire, or nearly entire, absence of the old red sandstone along the whole base of the carboniferous limestones, which, commencing at Llanymynech, ranges along the eastern skirts of Denbighshire and Flintshire, and, doubling to the N.W., runs to the Great Orms Head; thence taking its course to the Menai Straits, * Gcol. Trans., 2nd series. Carboniferous Group. 391 and forming a small part of the interior of Anglesea. He con- siders that, in this long range (partly examined by Mr. Mur- chison and partly by himself), there is not perhaps a single trace of old red sandstone, unless we designate by that name some beds of reddish shale and sandstone, which here and there form the base of the carboniferous group, and are, lie thinks, only varieties of the lowest limestone shale. This conclusion is confirmed by the observations of Professor Henslow, pub- lished about ten years since *. Professor Sedgwick believes that some part of the Isle of Anglesea, coloured by the latter as old red sandstone, is in fact a red sandstone of the grau- wacke group. Prof. Henslow himself states the impossibility of separating the old red sandstone in all cases from the grau- wacke ; and it should be recollected that at the time his paper was written, extensive deposits of red sandstone in the grau- wacke group were not generally known. Along the whole line above mentioned, where the old red sandstone is wanting, the carboniferous group rests unconformably on the grauwacke group ; a fact which seems to show that the old rocks of North Wales underwent a great movement, anterior to the period of the old red sandstone ; and that, by this movement, the bottom of the neighbouring seas was raised out of those causes which produced the old red sandstones. In no part of Denbighshire do we see the base of the car- boniferous group; but on the confines of Yorkshire, Lanca- shire, and Westmoreland, we find, in several places, the great escarpments of the carboniferous limestone resting unconfor- mably on the edges of the grauwacke, or separated from it by masses (sometimes of great thickness) of coarse red conglome- rate. The same statement applies to the carboniferous zone wrapped round the Cumbrian mountains, and also to the chain of Cross Fell. The old red conglomerates sometimes, though rarely, alternate with green and red marls, and with red, green- ish red, or white sandstone ; and we occasionally find in them, not only pebbles derived from the older calcareous rocks, but calcareous concretions like the Herefordshire Cornstojie. On the confines of Scotland, the red conglomerates appear (though Professor Sedgwick considers rarely) at the base of the carboniferous series. He has seen them occupying this position on the flanks of the Cheviot Hills, in two or three places pointed out by Mr. Culley, of Coupland Castle. He remarks the importance of this fact, as it proves that the very old carboniferous system of the Tweed (the sandstone beds of which are generally of a red colour) is newer than the old red conglomerates last mentioned ; and therefore probably newer than the old red conglomerates of Cumberland, Westmoreland, * Transactions of the Cambridge Philosophical Society. 892 Carboniferous Group. and Yorkshire. He does not therefore believe that the carbo- niferous red sandstone of the Tweed is the representative of the old red sandstone of Herefordshire; but that it is superior to the old red sandstone, and is of about the age of the great scar limestone of Yorkshire and Cross Fell. The carboniferous red sandstone, appearing here and there in the great Caledonian trough (between the chains of the Grampians and the grauwacke chain which stretches from St. Abbs Head to the Mull of Galloway), is probably in no in- stance older than the carboniferous red sandstone of the Tweed. Professor Sedgwick doubts whether in any part of this great trough there be a true representative of the (new) red and va- riegated sandstone of central England. In the Isle of Arran we have an old red sandstone and conglomerate, a carboniferous series of considerable thickness (but obscurely developed), and an upper red sandstone and conglomerate. Guided by analogy, it was concluded by Prof. Sedgwick and Mr. Murchison, in 1827, that the upper red sandstone and conglomerate were equivalents of the (new) red sandstone of England. Knowing by subsequent experience the great development of red sandstone in the carboniferous system of Scotland, Prof. Sedgwick now doubts the truth of this conclusion, and thinks that the upper red sandstone and conglomerate of the Isle of Arran may perhaps be only an unusual development of a portion of the carboniferous series. To settle this point it would be necessary to connect the Ar- ran sections with those on the coast of Ayrshire, and, by a northern traverse, to connect the Ayrshire red sandstone with the red conglomerate system flanking the Grampians. This task has not yet been attempted by any one well acquainted with the English types, and at the same time with the Scotch carboniferous deposits. In regard to the vast masses of red sandstone and conglo- merate on the shores of the Highlands, in the old oceanic val- ley of the Caledonian canal, and on the south flank of the Grampians, they were referred, by Prof. Sedgwick and Mr. Murchison, for the most part to the old red sandstone of En- fland; and this classification has been subsequently confirmed y the observations of Dr. Fleming and Mr. Lyell, who have traced a part of the system under the great coal-field of Fife- shire, and thus left no doubt respecting its relative situation. In Caithness this system is divided into (a) old red sand- stone and conglomerate ; (b) calcareo-bituminous schist, with numerous impressions offish* (Dyptcri^ fyc.) ; and (c) upper red * For figures of these fish, and a detailed account of the beds in which they occur, see the paper by Prof. Sedgwick and Mr. Murchison, Geol. Trans. 2nd eeries, vol. iii. Carboniferous Group. 393 sandstone. The first and second of these divisions cannot be separated from each other, and must therefore be included in the old red sandstone, a conclusion amply confirmed by re- cent observations. The upper division (c) forms the highest part of an imperfect section, and it is perhaps impossible to determine its exact place, but was hypothetically referred by Prof. Sedgwick and Mr. Murchison to the lowest division of the (new) red sandstone series. With his present views, Professor Sedgwick would wish this hypothetical adjustment of the up- per groups (c) of Caithness to be changed, and to see them classed with the old red sandstone; as he considers them iden- tical in age with the upper part of the series which descends from the southern Highlands, and passes under the carboni- ferous system of Fifeshire*. The carboniferous group occupies the surface of a large portion of Ireland, the limestones being exceedingly abundant. Mr. Weaver describes sandstones and conglomerates as fre- quently, though not constantly, interposed between the older deposits and the carboniferous limestone, and refers them to the old red sandstone. The Gaultees mountains are mentioned as wholly composed of them. They occur along the flanks of the clay-slate districts, and isolated caps of the sandstone often rest on these older deposits. The red sandstone emerges from beneath the interior of the great limestone plain at Moat, Ballymahon, and Slievegoldry Hill. The same author notices that the strata of this sandstone deposit are most inclined as they approach, and are in contact with, the older rocks ; but that as they accumulate and recede from the latter, they be- come more and more horizontal. The carboniferous limestone may be considered as the pre- valent rock in Ireland ; for, as Mr. Weaver observes, all its counties, with the exception of Derry, Antrim, and Wicklow, are more or less composed of it. This limestone is described as coming in contact with, and sweeping round, various moun- tain chains, " filling up every interval and hollow between them." It supports the coal-measures, properly so called ; and thus the analogy between the carboniferous series of central and southern England, and that of the corresponding portions of Ireland, is complete ; the arenaceous and conglomerate de- posits of the old red sandstone in the latter country being sur- mounted by a sheet of limestone, varying in thickness, some- times attaining a depth of 700 or 800 feet, but generally averaging 200 or 300 feet; and being in its turn covered by coal-measures f . * Sedgwick, MSS. t Weaver, On the Geological Relations of the East of Ireland, Geol. Trans. 394? Carboniferous Group. It would appear from the observations of Archdeacon Ver- schoyle on the north-west portions of Mayo and Sligo, that the carboniferous limestone of that part of Ireland has been largely developed; for Benbulden, 1700 feet, Knocknodie, 1025 feet, and Knocknashie, 980 feet, are stated to be wholly composed of it*. The carboniferous rocks of the north of France and Bel- gium have a direction from east-north-east to west-south-west from the vicinity of Aix-la-Chapelle to and beyond Valenci- ennes, and rise from beneath cretaceous or newer rocks. The carboniferous limestone and coal-measures of the Boulognais can be considered only as a continuous portion of the same deposit. According to M. de Villeneuve the coal-measures and lime- stone alternate at their contact with each other between Liege and Chaude Fontaine. The limestones are metalliferous, bluish and compact, and contain subordinate conglomerates of blue limestone. The alternating sandstones are sometimes reddish, and at others greenish brown ; they are sometimes compact, at others fissile with mica, and the lines of cleavage are in some beds not the same with those of stratification. The upper part of the limestone and sandstone contains aluminous shale, worked for profitable purposes (Huy and other places) f. The same author informs us that the coal-measures, which are composed of the usual mixture of sandstones, shales, and coal- beds, present at the Montague de St. Gilles no less than sixty- one beds of the latter, varying from six feet to a few inches in thickness ; and M. Dumont states that the coal-measures of Liege contain eighty-three beds of coal. The strata of the di- strict are greatly disturbed, as is well seen at Mons, and are traversed by faults, as may be observed at St. Gilles. Coal is worked far down in the lower beds, and even amid the lime- stones at Mons, which circumstance, however, M. de Ville- neuve attributes to the contortions of the strata. The line of demarcation between the coal-measures and the carboniferous limestone in the north of France, Belgium, and the country extending beyond Aix-la-Chapelle to Eschweiler, is generally well defined ; the occurrence of beds of coal be- tween the beds of limestone is rare, and partly apparent, be- ing produced by contortions of the beds. That portion of the strata which is known in England by the name of Millstone vol. v. Also consult Griffith's Account of the Connaught and Leinster Coal Districts; and Sections and Views illustrative of Geological Phaenomena, pi. 29. * Verschoyle, Proceedings of Geol. Soc., Nov. 1832. f De Villeueuve, Ann. des Sci. Nat. t. xvi. 1829. Carboniferous Group. 395 Grit, and in Westphalia as Ranker sandstein, is by no means thick, and a bed of aluminous shale occurs in it between Huy and Liege, as above noticed. This aluminous shale also oc- curs in Westphalia, at Lintdorf, and between Werden and Velbert, as far as Schwelm. Between Namur and Huy, beds of coal rest immediately on carboniferous limestone. On the south of Werden the rnillstone-grit is not much developed, but becomes more so to the north of Elberfeld. Very thick beds of conglomerate here occur in it. The lower division, immediately above the carboniferous limestone, is composed of a series of strata, consisting of slaty clays, shales, lime- stones, and sandstones, and is, in essential points, absolutely identical with the limestone shale of England. These beds become still more developed further east, and attain a thick- ness, in the neighbourhood of Arnsberg, Merscede, and W r arstein, which has not yet been observed at any other point. No old red sandstone has hitherto been observed in Westphalia. The carboniferous limestone here rests imme- diately on grauwacke, the continuation of that of the slate di- stricts of the Netherlands, the Rhine, and Westphalia. In Belgium, especially on the Meuse, the carboniferous limestone occurs extensively. From the undulating character of the stratification, it is difficult to determine the various beds of which the mass is composed. The intervening rocks have been usually termed grauwacke and clay-slate, but probably the red conglomerates may be considered as equivalent to the old red sandstone. These are found on the Meuse at Lustin and Profondeville ; on the Hoyoux at Masleye, south of Huy ; and on the Vesdre at Pepinster; on the Vichtbach; &c.* In all cases where the coal-measures come into contact with grauwacke, without intervening beds, we should be careful to recollect that the chances of an overlap are as great with the coal-measures as with any other rock. We have in Pem- brokeshire an instructive example of an overlap of this kind. The coal-measures of Pembrokeshire are well known as form- ing the western continuation of the South Wales coal-field. On the eastern part of Pembrokeshire they are still retained within bands of carboniferous limestone, which separate them from the old red sandstone to the north and south. On the west- ern side of the same county circumstances are different ; for instead of resting on carboniferous limestone, they repose on grauwacke, having passed over carboniferous limestone and old red sandstone in an oblique direction. The direct conti- nuation of the coal-measures of the eastern part of the county * Yon Dechcn, German Transl. of Manual. 396 Carboniferous Group. is seen, at Broad Haven and Little Haven, in St. Bride's Bay, to rest on grauwacke on the north. Trap rocks have disturbed its quiet relations on the south, but it is seen to rest against at least a small portion of carboniferous limestone in that direc- tion. To the north of Broad Haven, a patch of coal-mea- sures is observed to rest wholly on grauwacke, evidently the continuation of the coal strata on the south, though this con- tinuity is not evident on the surface from the indentation of St Bride's Bay. This overlap of the .coal-measures has not arisen from a greater development of coal-measures than of carboniferous limestone or old red sandstone, for both these rocks occur in great abundance to the south in the same di- strict, but to a passage over these rocks in a direction different from their general range *. To return to the carboniferous rocks of the Netherlands : It would appear that they are continued into Germany, to the deposits between Essen, Werden, Bochum, Hattingen, Wet- ter and Dortmund, which repose on the north-west corner of the great exposure of grauwacke rocks in that part of Eu- ropef. To the north of these deposits, on the northern side of the great gulf of cretaceous and supracretaceous rocks which enters easterly into Germany, and on which stands Miinster, there is, according to M. Hoffmann, an outcrop of carbonife- rous strata at Ibbenbuhren, between Osnabriick and Rheine J. Coal-measures occur at Seefeld, in Saxony. At Wettin, north of Halle, is another deposit ; and at Saarbruck, and the * For a description of this countiy, with maps and sections, see my Me- moir on Southern Pembrokeshire, Geol. Trans. 2nd series, vol. ii.; and Sec- tions and Views illustrative of Geological Phenomena, pi. 12. f M. von Dechen remarks that the carboniferous limestone of north- western Germany, Belgium, and the north of France, is so connected with the grauwacke group, that it has hitherto been impossible to distinguish them. The limestone of the Meuse from Namur to Vise, and from Ratin- gen to Arnsberg and Warstein, is decidedly carboniferous limestone. The limestone beds of the Dillenburg country, on the Luhn, from Stromberg near Bingen, may perhaps, without much objection, be considered as grau- wacke limestone. But the position of the limestone from Pfaffrath and Bensberg on the right bank of the Rhine, and which may be traced east- wards to Gummersbach and Mittelacher, is wholly doubtful. The Eiffel limestone from Schonecken near Priim to the Erft below Munstereifel, on the left bank of the Rhine, seems proved by the observations of M. Schulze merely to fill cavities in the grauwacke, and on the large scale never to be covered by it. This Eifel limestone may therefore, as far as present obser- vations extend, be referred either to the carboniferous or grauwacke groups. German Transl. of Manual. J For the localities of the coal in the north-west of Germany, consult M. Hoffmann's map of that country ; and for descriptions, " Ucbersicht der oro- graphischen und geognostischen Verhaltnisse voin Nordwestlichen Deutseh- larid," by the same author. Carboniferous Group. 397 neighbouring country, the coal-measures are abundant, and rest, when trappean rocks are not interposed, upon part of the grauwacke mass previously mentioned *. The coal-measures of Saarbruck are rendered particularly interesting by the development of the upper part of the series, such upper part being apparently a passage from the coal- measures into a rock equivalent to the rothliegendes, reminding us of the connexion of this rock with the red sandstone group previously noticed. The coal-beds of the upper portion of the coal-measures, and which occur at considerable intervals from each other, are intermingled with beds of red sandstone, in many places not to be distinguished from the rothliegendes of Mansfeld. This red sandstone alternates with thin beds of limestone and dolomite, and with shale. It contains layers of nodules with impressions offish (Lebach, and Borschweiler), and of the same ferns which are discovered in the lower coal- beds. This sandstone, which is of great thickness, would, as M. von Dechen observes, be certainly considered as rothlie- gendes, if it were not so decidedly connected with the coal- measures beneath it. It must not, however, be confounded with the Bunter sandstein, which rests unconformably on the coal f . M. Pusch describes the coal-measures in Poland as extend- ing from Hultschin to Krzeszowice, the more ancient beds passing into the grauwacke on which they rest : but the same author remarks, that in the rocky valleys of Czerna Szklary, and near Debnik, not far from Krzeszowice, a black marble, employed in the arts, supports the coal-measures. M. Pusch considers this marble as equivalent to the carboniferous lime- stone of the English geologists, and observes that the calca- reous conglomerates which accompany the coal, sandstones and shales in the gorges of Miekina and Filipowice are referrible to the same marble beds. The same author states that the coal-measures contain the plants so commonly observed else- where in similar deposits, and that he has identified thirty- six species with those noticed in the works of MM. Stern berg and Ad. Brongniart J. M. Kovalevski describes a very rich carboniferous deposit * The student will find instructive plans and sections of the coal-mines at Werden, Essen, Eschweiler, Valenciennes, Mons, Fuchsgrube (Silesia), and Saarbruck, in the Atlas to la Richesse Minerale, by M. Heron de Ville- fosse, pi. 24, 25, 26, 27, and 28. He should also consult the geological map and sections of the countries bordering the Rhine, by MM. Oeynhau- sen, la Roche, and von Dechen, for the coal-measures of Saarbruck and the adjacent country. Parts of these sections are inserted in Sections and Views illustrative of Geological Phsenomena, pi. 18. fig. 1. and 2. f Von Dechen, German Transl. of Manual. J Pusch, Journal de Geologic, t. ii. 398 Carboniferous Group. as existing in the mountains, which extend for a distance of 150 wersts, on the right bank of the Donetz in Southern Rus- sia. The exact age of this deposit, which is of considerable importance, is perhaps somewhat doubtful ; but it would ap- pear referrible either to that of the group under consideration, or to the upper part of the grauwacke. The rocks in which the coal principally occurs are described as arenaceous, and commonly red, with a mixture of argillaceous and schistose beds. The whole is described as passing into the grauwacke on which it rests. The coal, occurring in beds from a few inches to seven feet in thickness, is bituminous among the sandstones and shales, but becomes anthracitic where the rocks pass into grauwacke, the latter rock containing thin beds of limestone, and being traversed by quartz veins. Va- rious fossil plants are discovered in this carboniferous deposit, which is principally worked in the districts of Bakmout and Shavianoserbskoi, of the Government of Ekaterineslavsk, and partly in the country of the Mious, in the military district of the Don Cossacks *. The coal deposits of central France repose on granite, gneiss, mica-slate, &c., without the intervention of any lime- stones, sandstones, or slates, which can be distinctly referred to the carboniferous limestone, old red sandstone, or grau- wacke : such are the coal-fields of St. Etienne, Rive de Gier, Brassac, Fuis, &c. At St. Georges-Chatellaison the coal- measures also rest on gneiss and mica-slate f. The carboniferous deposits of the United States are, ac- cording to Prof. Eaton, of different ages ; one being contained in the argillaceous slates (argillite) of Worcester (Mass.) and Newport; another being considered equivalent to the coal- measures of Europe ; and a third being of a more recent epoch, though older than certain lignites. The deposit referred to the same epoch as the carboniferous series of Europe, occurs at Carbondale, Lehigh, Lackawaxen, Wilkesbarre, and other places f . Mr. Cist describes the coal of Wilkesbarre as alternating with various sandstones and shales, the latter containing a great abundance of fossil plants >, many of which it will have been seen by the lists of organic remains are identical with some discovered in the coal-measures of Europe, and all are of the * Kovalevski, Gornoi Journal, 1829; and in Boue's Memoires Geol. et Paleontologiques, vol. i. Other carboniferous deposits exist in the same country, but appear to be more modern. f Mr. Grammer remarks that the coal deposit of Virginia rests on granite. American Journal of Science, vol. i. J Eaton, Amer. Journ. of Science, vol. xix. Cist, Amer. Journ, of Science, vol. iv., where a map of the deposit will be seen. Carboniferous Group. 399 same general character with those obtained from the carboni- ferous and grauwacke series. The sandstone beds vary from five to one hundred feet in depth, and the coal is sometimes from thirty to forty feet thick, its general thickness being from twelve to fifteen feet. Prof. SilJiman states that the beds at Mauch Chunk (Pennsylvania) consist of conglomerates, sand- stones, and argillaceous slate. The pebbles in the conglome- rate are described as pieces of quartz rounded by attrition, and the cementing matter of the conglomerates and sandstones as siliceous *. According to Prof. Eaton, the limestone which supports the strata containing the Pennsylvanian coal extends along the foot of the Catskill Mountains, and is continued from the southern part of Pennsylvania to Sackett's Harbour on Lake Ontario f. Mr. Hitchcock informs us that coal is associated with trap- pean rocks, fetid, siliceous, and bituminous limestones, red and gray sandstones, and conglomerates, in Connecticut. It is described as bituminous; whereas the Wilkesbarre coal is frequently termed Anthracite by the American geologists J. It occurs at Durham, Chatham, Berlin, Enfield, and other places in Connecticut, and is described as passing into the so- called old red sandstone of the country, which is composed of a series of sandstones and conglomerates, generally of a dark red colour. An excellent section of this coal deposit, de- scribed in great detail by Mr. Hitchcock, is exposed where the Connecticut river cuts through it between Gill and Mont- ague. Fossil fish are obtained from associated bituminous shale at Westfield (Connecticut) and Sunderland (Massachusetts); and one species is considered referrible to the genus Palaothris- sum of Blainville , a genus which the student has seen no- ticed both under the head of zechstein, and of the coal-mea- sures. The relative antiquity of the coal deposit of India is not very clearly ascertained. According to the observations of Dr. Voysey, Mr. Calder||, and Mr. Koyle, it rests upon gneiss and other rocks of that character. The latter author states that the coal-field of Damuda, discovered by Mr. Jones in 1815, is composed of various beds of bituminous coal, (of * Silliman, Amer. Journ. of Science, vol. xix. f Eaton, ib. vol. xix. t This distinction would not, in itself, appear to be of any great im- portance ; for the continuous coal deposit of South Wales is anthracitic in Pembrokeshire, and bituminous in its eastern prolongation through Mon- mouthshire. Hitchcock, American Journal of Science, vol. vi. || Calder, Trans. Phys. Class. Asiatic Soc. Bengal, vol. i. 1829. 400 Carboniferous Group. good quality,) shales, sandstones, and conglomerates ; the lat- ter, mixed with red sandstone, constituting the upper part of the series. The shales contain the abundant remains of Ver- tebraria Indica, Royle (commonly termed by Indian geolo- gists the Ranijung reed); of Sphenophyllum ? speciosum^ Royle; of Glossopteris angustifolia. Ad. Brong.; of Glossopteris Brown- iana.> Ad. Brong. ; of Pustularia Calderiana, Royle ; Peco- pteris Lindleyana, Royle; and of various other plants*. Mr. Jones observes that the shales and coal-beds of Damuda crop out in many places, and that the strata are generally undu- lating. The coal is principally worked at Ranijung colliery, where eight coal-beds, varying from four inches to nine feet in thickness, are associated with shales and sandstones f. Mr. Everest observed the effects of pseudo-volcanic action in this coal district, which appear evidently to have been caused by the combustion of the coal, the various shales and sand- stones being more or less acted upon by the heat evolved. The same author remarks that some of the sandstones of the Damuda coal-field contain calcareous matter, and that the sandstones are raised in large slabs for economical purposes on the banks of the Adji J. The coal of India is by no means confined to the Damuda district. Mr. Jones considers that the coal of Sylhet and of Cachar constitutes an eastern prolongation of the same depo- sit; and Mr. Royle infers that it is continued a considerable distance to the westward, having been observed resting on gneiss and other rocks of that class, as in the district first noticed, at Goomeah, Palamow, Jubbulpoor, and Hosanha- bad . Capt. Franklin states that five beds of coal were found in a part of the Palamow coal district, which, judging from the distances of the localities mentioned by this author, seems somewhat extensive || . Coal is therefore by no means rare in India : on the con- trary it extends, probably at intervals, along an east and west line of several hundred miles. There is indeed no direct evi- dence, such as that of a perfect correspondence in organic re- mains, to show that the various deposits observed are contem- poraneous, but there is sufficient evidence to make such an inference highly probable. Although the fossil plants disco- vered in the shales are not specifically the same with any yet noticed in the carboniferous rocks of Europe or America, they * Royle, Illustrations of the Botany, &c. of the Himalayan Mountains, London, 1833, where further detail and figures of these plants are given, f Jones, Trans. Phys. Class. Asiatic Soc. Bengal : Calcutta, 1 829. J Everest, Gleanings in Science, vol. iii. : Calcutta, 1831. Royle, Illustrations of the Botany, &c. of the Himalayan Mountains. || Franklin, Gleanings in Science, (Calcutta,) vol. ii. p. 217. Carboniferous Group. 401 are still of the same general character as far as respects the tem- perature in which they probably flourished. The presence of Glossopteris Browniana in the coal district of Damuda is re- markable as it points to a connexion, with regard to date, be- tween that deposit and the carboniferous series of Eastern Aus- tralia, where it was first observed by Dr. Robert Brown. If we, for the moment, abstract the limestone beds, there would appear little doubt that the carboniferous series is of mechanical formation, a deposit from water varying in its transporting powers. Thus, at one time the velocities were sufficient to force forward gravels, while at others they only accomplished the transport of silt or mud. If proportional sections be made of coal deposits, it will be observed that the coal-beds occur at very unequal intervals, showing that the causes which produced them have acted irregularly. From the careful examination of the Forest of Dean by Mr. Mushet, we have a detailed list of the various beds of the coal-measures, carboniferous limestone and old red sandstone, the whole con- stituting a collective thickness of about 8700 feet; the coal- measures being 3060 feet in depth, and the limestone 705. The mass reposes on the grauwacke (transition) limestone of Long Hope and Huntley *. The sandstones of the old red sandstone in Gloucestershire, Somersetshire, and the neighbouring parts of England, afford us no great evidence of a quick deposit, more particularly as the conglomerates are not common ; the latter are, however, sufficient to show that the velocities of the transporting waters were not constant, but liable to variation. A great change in the depositing and transporting powers subsequently took place ; and instead of the siliceous and arenaceous sediment, carbonate of lime, often enveloping a variety of marine animal remains, was produced ; and this not for a short time, but apparently during a long period ; for the carboniferous limestone of this district bears marks of slow formation, many beds being composed of a mass of fossils, the remains of myriads of animals, which have apparently lived and died where we now find them entombed. It must however be admitted that the origin of many beds, which do not present a trace of animal exuviae, remains ob- scure, and we have no direct evidence that they may not have been produced more suddenly by deposits from water, either holding carbonate of lime in chemical solution or in mechani- cal suspension. After a thickness of seven or eight hundred feet of calcareous rock had been formed, another great change in the matter deposited was effected ; not however so suddenly but that the arenaceous sediment which afterwards became so * Mushet, Geol. Trans., 2nd series, vol. i. p. 288. 2 D 402 Carboniferous Group. abundant, and the calcareous matter, were alternately pro- duced for a comparatively limited period. An immense mass of sandstone, shales, and coal was then accumulated in beds one above another, which, though irregular with regard to the relative periods of deposit, are frequently persistent over considerable areas. By general consent the coal is considered as resulting from the distribution of a body of vegetable remains over areas of greater or less extent, upon a previously deposited surface of sand, argillaceous silt, or mud, but principally the latter, now compressed into shale. After the distribution of the vegeta- bles, other sands, silt, or mud, were accumulated upon them ; and this kind of operation was continued irregularly for a con- siderable time, during which there was an abundant growth of similar vegetables at no very distant place, to be suddenly, at least in part, destroyed and distributed over considerable areas on the more common detritus. Great length of time would be requisite for this accumula- tion, because the phsenomena observed would lead us to con- sider the transporting power, though variable, to have been generally moderate : moreover a very considerable growth of vegetables requiring time, would be necessary at distinct in- tervals ; for coal-beds only now six or ten feet thick, must, before pressure was exerted upon them, have occupied a much greater depth. It is a remarkable circumstance connected with the coal-measures of the South of England, that marine remains have not been detected in them, which, though it does not prove the deposit of coal to have been effected in fresh water, does appear to show that there was something which prevented the presence of marine animals, a circum- stance the more remarkable, as we have seen that such ani- mals swarmed during the formation of the carboniferous lime- stone. These remarks are not only applicable to the small di- strict above noticed, but to a large extent of country, one stretching from Belgium, through the North of France, and the southern parts of England and Wales, into Ireland ; and for the most part concealed beneath masses of newer rocks. As we advance northward, however, the marked distinctions, as before noticed, disappear; so that the causes, whatever they were, which produced such a separation of arenaceous and calcareous rocks on the south, became modified, and the limestones were more intimately blended, in alternating beds, with the sandstones and shales, affording a greater mixture of marine with terrestrial remains. It has long been known that the Yorkshire coal-measures presented a bed containing the remains of Ammonites and Pectines, and that the fossils of Carboniferous Group. 403 the carboniferous limestone and coal-measures were detected in the millstone grit ; or, in other words, that there was an alternation of terrestrial with marine remains, showing that the causes which effected the deposit of calcareous matter and envelopment of marine remains sometimes predominated, while at others a transport of mud and sand entombed an abundance of vegetables. The occurrence of marine remains amid the coal-measures, it will be observed by reference to the foregoing lists, is not confined to Great Britain, but is also remarked in different parts of Germany ; so that the same modification of circumstances which has produced a mixture, or rather alter- nation, of marine and terrestrial remains in Great Britain, has extended into the continent of Europe. Mr. Buddie observes, respecting the Newcastle coal-field, that the sandstones increase in number and thickness as they rise or crop upwards, whereas the argillaceous shales become thicker in the opposite direction*. This is precisely the ap- pearance we should expect from a deposit from water moving with moderate velocity, and which should hold detritus of dif- ferent degrees of fineness in mechanical suspension. The sands would be first deposited, while the silt and mud would be carried greater distances. It would hence follow, that the sands would be more abundantly deposited in proportion as they approached the situations whence the detritus- bearing water proceeded, and consequently the resulting beds would become thinner as they receded from the same situations. The very reverse would happen with the mud deposits. So that sup- posing the transporting current of water to have been nearly uniform and not rapid, we should have the effects produced stated to be now observed in the Newcastle coal-field. There is another class of appearances connected with these rocks which demands our attention. From a considerable mixture of porphyry in certain situations with the coal-mea- sures, it has sometimes been considered that this rock was an essential and component part of the group under consideration. From all analogy it may be concluded that porphyries are of igneous origin ; and for the same reason it is inferred that the coal-measures and their accompanying beds were produced by aqueous deposition. We therefore should be led a priori to consider, that two substances of such different origin did not necessarily constitute parts of a common whole, but that their admixture was accidental. And we may consider this at once proved, by the abundant occurrence of coal-measures without porphyry, such as is so commonly the case in England. In the sections which M. Hoffmann has presented us of the * Buddie, Trans. Nat. Hist. Soc. Newcastle, vol. i. p. 238. L 2 D 2 40 4- Carboniferous Group. coal-measures of Wettin, and other places in North-western Germany, it is easy to conceive, although porphyry occurs both above and beneath the coal strata, that the latter are not necessarily of contemporaneous formation ; on the contrary, the fractured and contorted state of the beds shows that great violence has been exercised upon them, precisely such as would be expected if igneous rocks had burst in amidst them, when among other accidents we should expect to find large masses of coal-measures caught up and included in the porphyry, as we find masses of chalk caught up and enveloped by basalt in the North of Ireland. . As we shall return to the subject of the igneous rocks found among the carboniferous group in an- other place, the above notice has been introduced merely to show that the supposed connexion of porphyry and coal strata has not been overlooked. From the similarity of general circumstances attendant on the coal strata, we have reason to conclude, although the series may contain more limestone at one place than at another, that in Poland, Western Germany, Northern France, Belgium, and the British Isles, there were some common causes in ope- ration at the same epoch, producing the envelopment of a great abundance of terrestrial vegetables, of a nature that could not, from the want of the necessary heat, now flourish in the same latitudes. Proceeding to the central part of France, we find several smaller coal deposits, which more particularly from their or- ganic character are referred to the carboniferous epoch of which we are treating. How far they may have been once more extensive and continuous, and how far they may have suffered from movements in the land, dislocations, and denu- dation, we are not certain ; but we are certain that they were directly deposited on granite, mica-slate, gneiss, and other rocks of that character. The causes therefore which produced the calcareous beds, sometimes very abundantly, in the countries above noticed, have not extended to them. The observed phagnomena are however sufficient to show that a vegetation similar to that of the more northern carboniferous rocks is there entombed, though we are not quite assured to what precise period their formation can be referred; for, as will be seen in the sequel, vegetables of the same general character are de- tected in the grauwacke series, and it is also possible that they might be discovered in the rothliegendes under the zechstein. The precise period of any particular deposit of similar vege- tables may thus be sought through a considerable lapse of time, and it becomes hazardous to fix, without very good evidence, on any relative portion of that time. The conglomerates usu- ally referred to the old red sandstone in Northern England, Carboniferous Group. 405 sometimes intervening between the upturned grauwacke rocks and the carboniferous limestone beds resting upon them, and noticed by Prof. Sedgwick and other geologists, may have been followed by a coal deposit where circumstances were fa- vourable ; and the result would have been a formation similar to the deposits of central France, except that the subjacent rocks would be, perhaps, more ancient in the latter case. It may however have also happened, that during the grauwacke deposit, to be noticed in the next section, circumstances fa- voured a production of rocks similar to those of St. Etienne and other places; as also that during a subsequent period, one equivalent to the lower part of the red sandstone group, a si- milar deposit might be effected ; for as rocks may be violently disturbed in one place and not in another, so may they also be quietly formed in one situation, while a few hundred miles di- stant, disruptions of strata and the trituration of organic re- mains may have taken place, so that no trace of organic life may be left. While on the subject of the coal series of France deposited in gneiss, mica-slate, &c., we may remark the very unequal manner in which the St. Etienne coal-field has been deposited. The rocks beneath (gneiss, mica-slate, and talc-slate) have a very irregular surface, in which the series of conglomerates, sandstones, shales, and coal are moulded ; and the result is that the whole is divisible into several minor basins or cavi- ties, as far at least as the coal is concerned. From such a state of things, we could scarcely expect that comparative even thickness of the coal-beds observable in the coal-measures of England. The coal-beds of St. Etienne suddenly swell out, and as suddenly become mere seams ; sometimes being from fifty to sixty-five feet thick, at others presenting mere traces of coal. The extraordinary thickness of this coal may surprise some readers ; it nevertheless rests on excellent authority *. The mean thickness of the beds worked is rarely beyond sixteen or twenty feet, and more rarely below three feet and a quarter. Let us now consider the mode in which the remains of ter- restrial vegetables, so abundantly preserved in the coal strata, occur. They are. for the most part laid flat, and the leaves and stems parallel to the line of stratification ; but there are other cases where they repose at various angles in the beds; and finally they are found vertical, with their roots downwards. The student will recollect that this is precisely the manner in which the vegetables of the submarine forests are found ; and if several submarine forests, such as those which are discovered around the shores of Great Britain, occurred above each other, * Chevalier, Annales des Mines, 3eme Series, 1832, t. ii. p. 444. 406 Carboniferous Group. with the intervention of sandy and clay beds, we should have a series of deposits not very unlike the coal strata, so far as regards the position of the vegetable remains. If we are to consider parts of the coal-measures as in any way resulting from a series of similar deposits, we are certainly called upon to admit a very remarkable series of changes in the relative surface levels of land and water; but there are also very great difficulties attending the supposition that the vegetables have been swept by strong currents of water into the positions where we now find them ; for not only have similar effects been pro- duced over considerable areas, but the vegetables have suffered very little injury, their delicate leaves being most beautifully preserved. Now, though we know that vegetables are abun- dantly borne down by river floods into the sea, they by no means remain uninjured; and if they be of a soft nature, such as the bulk of the coal plants are considered to have been, the damage done them by transport is considerable, as I have had occasion to remark on the coast of Jamaica, where arborescent ferns and other tropical productions are sometimes, though very rarely, carried by floods from the neighbouring moun- tains into the sea. In the few cases which passed under my observation, the fern-trees were so damaged in the river- courses as to be with difficulty recognisable*. We have now so many cases in France, Germany, and Great Britain, of the occurrence of some coal plants in a ver- tical position, with their roots downwards, that such cases can scarcely be considered as accidental, but, in some measure, as characteristic of the deposit in particular situations. Mr. Witham has brought forward some good examples of vertical stems in the carboniferous rocks of Durham and New- castle. Two stumps or stems of Sigillarite are described as standing erect, with their roots imbedded in bituminous shale, in the Derwent Mines, near Blanchford, Durham : the space round them was cleared out to obtain the lead ore ; and one plant is stated to have been about five feet high, and two feet in diameter. A more curious case was observed by the same author in the Newcastle district, where, in sandstone beneath the High Main coal, numbers of fossil vegetables, chiefly Sigillaria, are discovered erect, their roots imbedded * The height at which arborescent ferns are found, would seem much to depend on local causes. Thus, on the southern side of Jamaica they do not flourish much under an elevation of 2000 feet above the sea ; while on the northern side of the same island I have seen them at not more than 400 or 500 feet above the same level. The cause would seem to be the greater moisture of the northern side. It would therefore appear that a considerably moist climate would be necessary for the abundant production of this class of plants in the low situations, such as it has been imagined the lands were which produced the mass of the coal plants. Carboniferous Group. 407 in a small seam of coal under the sandstone, while they are all truncated on the line of the High Main coal-bed, to the forma- tion of which their higher ends have in all probability partly contributed*. A still more curious example of vertical stems of plants in coal-measures is noticed by Mr. Wood, as having been ob- served above the High Main coal, at Killingworth colliery in the same district There were many of them, and they rose ten feet through various strata of shale and sandstone. That figured beneath (a stem of Sigillaria pachyderma\ though not the largest, was upon the whole better exposed to view than the others. It was about two feet in diameter in the lower part, and the roots could be traced running into the shale, for about four feet from the stem. The roots of the various stems were interlaced with each other, and the interior of the plants was filled with a sandstone, resembling, not that of the lower beds through which the plant rose, but that of the upper beds. This fact proves that the interior of the plant was sufficiently hollow, when the upper beds were deposited, to permit the in- filtration of the sands downwards. About thirty of these stems were visible within an area of fifty square yards. We can scarcely refuse to admit with Mr. Wood that these stems of Sigillarice are exactly in the position in which they grew, the shale being the soil or mud in which they vegetated f. The following sketch (Fig. 99.) will illustrate better than words the manner in which one of the stems was preserved J. Fig. 99. * Witham, Observations on Fossil Vegetables, 1831, p. 7, where there is an illustrative section. f Wood, Trans. Nat. Hist. Soc. Newcastle, vol. i. p. 205, and pi. 19 ; and Lindley and Mutton's Fossil Flora of Great Britain. % Taken from Lindley arid Button's Fossil Flora of Great Britain, pi. 54> the beds bsing enumerated from Mr. Wood's figure. 4-08 Carboniferous Group. a, High Main coal; b, argillo-bituminous shale; c 9 blue shale; e, compact sandstone ; g, alternating shales and sand- stones ; ^, white sandstone ; 2, micaceous sandstone ; , shale. The beds dip gently to the westward. Such cases as these, and that long since noticed by M. Ad. Brongniart at St. Etienne*, where numerous stems are also included upright in coal sandstone, without however being trun- cated by a coal- or shale-bed, are sufficient to show the very great analogy which exists between them, certain submarine forests, the dirt-bed at Portland, and the vertical stems in the York- shire oolite s inasmuch as they all apparently point to a quiet submergence f. We may have some difficulty in considering the deposition of sand to have been effected so quietly amid the stumps of trees as not to have washed away the substances in which they were imbedded ; but we have only to recollect, that among the submarine forests round our shores, if once any of them were at such a depth beneath the surface of the sea as to be suffici- ently beyond the influence of the waves, they would become quietly covered by sand ; for the velocity of water sufficient to transport this sand, would scarcely disturb the trees. It seems impossible to come to any other conclusion, respect- ing the vertical stems, not only in the coal-measures, but also in the other rocks above noticed, except that they occur in the re- lative situations in which they grew, were submerged quietly, and as quietly entombed in sands, shales, or calcareous matter. The supposition that these stems have been transported in mass by water, with their roots downwards, is quite untenable when we consider the various facts adduced. The analogy between the trees in the dirt-bed of Portland, the vertical stems of Equi- seta observed by Mr. Murchison in the oolitic series of York- shire, and the vertical stems of the coal-measures, is quite per- fect; in all cases the plants are too numerous to admit of any other explanation than that of their having grown in the beds in which their roots are now found, and they all point, as be- fore stated, to quiet submergence and tranquil envelopment. That this explanation is only applicable to cases of large areas, occupied by vertical stems of plants, will be at once granted. It does not account for the mass of coal generally, * Annales des Mines, 1821. f It cannot be denied that, under particular circumstances, stems of trees preserving a vertical position may be forced onwards by river inundations. Thus, snags, or trees with their roots downwards, and only forced by the cur- rent from a vertical position, are common, so as to be very dangerous, in the Mississippi ; and trees were forced down the valley, during the debacle of the Valle'e de Bagnes, and left standing with their roots down wards, at Mar- tigny. These facts admit of easy explanation ; for if trees be suddenly de- tached from the soil, and their roots loaded with stones and other heavy matter, they would naturally float with the branches upwards. Carboniferous Group. 409 which has every appearance of having been drifted at unequal intervals. Drifts of vegetable matter now take place into lakes and estuaries ; but admitting that much coal may have accu- mulated in such situations, there are serious difficulties attend- ing this as a general explanation of coal accumulations, more especially where, among the marine remains contained in lime- stones, alternating with coal strata, we find corals. Now co- rals are generally very abundant in these alternating limestone beds, and it is well known that the creatures which construct the corals of the present day avoid fresh and brackish waters. It therefore follows, either that the coal in these alternations could not have been deposited in fresh or estuary waters, or that the habitats of corals were not then of the same kind as we now find them. The occurrence of limestone strata, continuous over consi- derable areas, and alternating with the shales and sandstones, would seem to require comparative tranquillity for their pro- duction, more particularly as the marine shells entombed in them have evidently not been subjected to violence, but appear to have been imbedded at no great distances from the places where they lived and died. The vegetable remains are often of considerable size. M. Brongniart observes that in the coal strata of Dortmund, Es- sen, and Bochum, stems are found in the planes of the strata, more than fifty or sixty feet long, and that they may be traced in some of the galleries for more than forty feet without ob- serving their natural extremities*. Vegetables of large size have also been detected in Great Britain. Mr. Witham men- tions one in Craigleith quarry as being forty-seven feet in length, from the highest part discovered to the root. The bark is de- scribed as converted into coalf. Lepidodendra have been discovered in the northern parts of England, from twenty to forty-five feet long, and four feet and a half in diameter:):. Respecting the general character of the vegetation of this period, such as we find it entombed in the carboniferous rocks of the northern hemisphere, M. Ad. Brongniart observes, that it is remarkable ; 1. for the considerable proportion of the vascular cryptogamic plants, such as the Equisetacetf, Filices, Marsileacea, and Lycopodiacece ; C 2. for the great development of the vegetables of this class, so that they have attained a mag- nitude far beyond those of the same class now existing ; thus proving that circumstances were particularly favourable to their production during the period under consideration. It * Brongniart, Tableau des Terrains qui composent 1'Ecorce du Globe, t Witham, Edinburgh Journal of Natural and Geographical Science. April, 1831. I Lindley and Hutton, Fossil Flora, vol. i. p. 17. 410 Carboniferous Group. has been ascertained within the last few years, that dicotyledo- nous plants, which were once considered to be exceedingly rare, if not altogether absent, in the coal-measures, exist in great abundance in the same rocks. Coniferas are sufficiently common. It would appear that authors are by no means agreed as to what families certain genera of fossil plants should be re- ferred. Thus M. Adolphe Brongniart refers the genus Stig- maria, very common in the coal-measures, to the family of Z/j/- 1 copodiacecE\ while Prof. Lindley and Mr. Hutton consider that, if an existing analogy must be found, it is with greater proba- bility of accuracy referrible to Euphorbiacece or Cactece, most probably to the former; a difference of considerable impor- tance, as upon it depends whether the genus in question be- longs to the class Cellulares, or to the class Vasculares. No doubt this difference of opinion arises from the obscurity of the subject, fossil botany being beset with very great difficulties ; difficulties far beyond those which attend the study of fossil zoology, though the latter are by no means either small or rare. The possibility of this variety of opinion among distinguished botanists, should teach geologists caution, and prevent them from indulging in those hasty generalizations, which, though often brilliant, too frequently impede the progress of the sci- ence they cultivate, not only by rendering authors satisfied with rapid conclusions, but also by throwing doubt, when such conclusions are found to be erroneous, upon others which are firmly based, and which may be considered as exact as those of any other science. As, in the opinion of botanists, islands in warm countries are favourable to the growth of Ferns and other plants of the same natural class, not only from the presence of the necessary heat, but from the moisture so congenial to them, it has been considered by MM. Sternberg, Boue, and Ad. Brongniart, that the vegetation of this period, such as we find it in the car- boniferous deposits of Europe and North America, was the growth of islands scattered in archipelagos. When we come narrowly to look into the structure of the coal-measures, the vast accumulations of shale and sandstones, sometimes amounting to the depth of 460 feet (Forest of Dean), do not precisely accord with mere oscillations of islands above and beneath the level of the sea, which might in some cases appear probable ; for these accumulations of detritus require considerable drift, and must have resulted from the destruction of pre-existing rocks, mostly siliceous, and therefore, if solid, requiring much time for their degradation, with the assistance of other forces than the mere battering of the surf on clusters of low islands, perhaps defended, like those in the Pacific, by coral reefs. Carboniferous Group. 411 The presence of larger masses of land, with mountains, rivers, and other physical features necessary for the production of a larger amount of detritus, would seem requisite, inde- pendent of volcanic eruptions, and other exertions of internal force, for the accumulations we observe. The oscillation of low islands is mentioned merely as the possible explanation of some of the observed phenomena, and the student must be careful to consider it only in that light. While on this sub- ject, however, it may be as well to notice a possible explana- tion of some of the minor alternations of limestones with marine remains, with shales and coal containing terrestrial remains such as are found in the millstone grit; because such hints, without attributing any particular value to them, very fre- quently lead to further inquiry. Suppose a tract of low land covered with a dense vegetation, such as is found in the tropics, to be, by a movement in the earth, an earthquake, for in- stance, submerged a few feet beneath the sea; marine animals would establish themselves on the submerged surface, which would become in the condition of the submarine forests pre- viously noticed ; and the consequence would probably be, that not only millions of testaceous creatures would leave their ex- uviae, but that the corals would also swarm, and might even- tually produce coral islands, upon which vegetation might again establish itself, to be again submerged. That coral is- lands are sometimes raised above the sea, is what we should expect; and evidence of it has been adduced by Captain Beechey, who describes Henderson's Island (in the Pacific) as apparently upheaved by one effort of nature to the height of eighty feet. It is composed of dead coral, bounded by per- pendicular cliffs, which are nearly encompassed by a reef of living coral, so that the cliffs are beyond the reach of the spray*. Now depression to this amount might as easily have taken place; in which case the vegetation of the island would have been submerged eighty feet, and the amount of destruction it would suffer would depend on the greater or less suddenness of the movement. Such movements cannot be considered great when regarded, as they always should be, with reference to the mass of the world ; for we have proof that far greater have occurred, and the differences which have been produced in the relative levels of land and water are, when viewed on the great scale, of very trifling importance. According to M. Ad. Brongniart, if we look at the mass of the coal plants, we must consider the vegetation of the carbo- * Beechey, Voyage to the Pacific Ocean and Behring's Straits, p. 194. Descriptions of other coral islands, with sections of their general structure, will be found, pp. 160 and 186 of the same work. 4-12 Carboniferous Group. niferous group to have been produced in climates at least as warm as those of the tropics ; and as we now find plants of the same class increase in size as we advance towards warm lati- tudes, and as the coal-measure plants exceed the general size of their existing congeners, he concludes, with much apparent probability, that the climates in which the coal plants existed were even warmer than those of our equinoctial regions. This view leads us to another consideration. There cer- tainly was a similar vegetation about the same period (for whe- ther the American coal-measures may be, like those in parts of Europe, somewhat older, does not alter the question,) over parts of Europe and North America : we may therefore infer a similar climate over a large portion of the northern hemi- sphere, such as we have not at present, for it was at least tro- pical, and very probably ultra-tropical. The question natu- rally arises, Is there any evidence to show that the same tem- perature existed at the same period in the southern hemisphere? for if there is, there must have been some common cause to produce such an equality of climate, at present unknown to us. Unfortunately, the actual state of our knowledge will not per- mit an answer to this question ; but by it we learn the impor- tance of ascertaining the botanical character of the various rocks in the southern hemisphere, more particularly those of the earliest formation, such as may be considered the equiva- lents of the carboniferous and grauwacke groups of the North. With respect to the testaceous remains, the limestones con- tain a great abundance, not only of species, but of individuals of the genera Spirifer and Producta. Of the form of these shells, and of the Cardium hibernicum and C. alceforme, (the latter by no means a rare fossil in the limestones of the next group,) the following figures will afford examples. Fig. 100. 101. 102. Fig. 105. Fig. 106. Fig. 10*. Fig. 103. Carboniferous Group. 4? 13 Fig. 100. Producta Martini; Fig. 101. Spirifer glaber ; Fig. 102. Spirifer attenuates; Fig. 103. Spirifer cuspidatus ; Fig. 104?. one of the two spiral appendages contained in Spirifer trigonalis*; Fig. 105. Cardium hibernicum ; and Fig. 106. Cardium alaeforme. Of the vertebrated creatures which may have existed at this period our knowledge is very limited ; but it may be observed that the Tritores (or palates of fish) still retain phosphate of lime; for Dr. Turner ascertained that a palate from the carbo- niferous limestone of Bristol, contained 24?'4) per cent, of phos- phate of lime, the remainder being carbonate of lime and bitumi- nous matter, the latter abundant. A palate from the chalk, ex- amined for the purpose of comparison, was found to contain 18-8 per cent, of phosphate of lime, the remainder being car- bonate of lime, with traces of bituminous matter. * Their position in the shell will be seen by reference to Sowerby's Mine- ral Conchology, pi. 265, fig. 1. Grautsoacke Group. SECTION IX. GRAUWACKE GROUP. SYN. Grauwacke (Traumate, Daubuisson). Grauwacke Slate (Grauwacke schistoide, Fr. ; Scfdste Traumatique, Daubuisson ; Grauwackenschiefer, Germ.). Grauwacke Limestone (Transition Limestone, Engl. Authors ; Calcaire de Transition, Calcaire Intermediate, Fr. Authors ; Uebergangs- kalkstein, Germ. Authors). IT has been observed that the old red sandstone of some coun- tries graduates into grauwacke ; whence it may be inferred that the causes, whatever they may have been, which produced the latter deposit, were not violently interrupted in such situations, but that they were gradually modified, if indeed it be neces- sary to consider the old red sandstone in any other light, taken generally, than the upper portion of the grauwacke series. That it is so, is the opinion of most continental geologists ; and where the one graduates into the other, such an opinion seems well founded. Variations in the classification of the old red sandstone would appear solely to arise from its mode of occurrence in the particular countries where geologists have been accustomed to observe it. When accidents have hap- pened to the grauwacke, throwing the strata on their edges, and a red sandstone or conglomerate deposit intervenes between the carboniferous limestone and the upturned beds, classifica- tions made in the countries where such phaenomena prevail, would naturally be framed so as to separate the (old) red sand- stone from the grauwacke : but when such accidents have not happened, and the carboniferous limestone, the intervening red sandstone, and the grauwacke are so circumstanced that the two former rest conformably on the latter, and they all gradu- ate into one another, it is altogether as natural that the old red sandstone should be pronounced the upper part of the grau- wacke series. Nor should we be surprised that the carboni- ferous limestone should also be included in the group ; for the general organic character of the whole is similar, and does not differ more, if so much, as the upper part of the oolitic group from the lower portion of the same deposit, or as the chalk from the green sand. Viewed on the large scale, the grauwacke series consists of a large stratified mass of arenaceous and slaty rocks intermingled with patches of limestone, which are often continuous for con- siderable distances. The arenaceous and slate beds, consi- Grauwacke Group. 415 clered generally, bear evident marks of mechanical origin, but that of the included limestones may be more questionable. The arenaceous rocks occur both in thick and schistose beds ; the latter state being frequently owing to the presence of mica disposed in the lines of the laminae. Their mineralogical cha- racter varies materially ; and while they sometimes, though rarely, pass into a conglomerate, they very frequently graduate into slates, which become of so fine a texture as to lose the arenaceous character altogether. Roofing-slate is far from rare among the grauwacke rocks ; and if we consider it of mecha- nical origin, like the mass of the strata among which it is in- cluded, we must suppose it to have originated from the depo- sition of a highly comminuted detritus. If the size of transported substances be considered as the necessary evidence of rapid currents of water, the grauwacke rocks, taken as a mass, have been slowly deposited ; for though evidences of cross currents are sufficiently abundant in the va- rious directions of the laminae, and in the mode in which are- naceous and slaty beds are associated with each other, the substances are generally fine-grained, rarely passing into con- glomerates. A rapid current would however not appear to require large pebbles as the necessary evidences of its existence at any given period, although when large pebbles are present we infer that small currents of water could not have transported them ; for the size of the substances transported by a current moving with considerable velocity will greatly depend on the surface over which it passes and the nature of the substances carried onwards. Perhaps from the absence of organic re- mains in a large proportion of the arenaceous part of this de- posit, and their abundance in some of the included limestones, we might infer that there had been something in the transport and deposit of the sands unfavourable to their preservation, such as trituration in waters moving with rapidity. There is, however, a general appearance in the mass of the grauwacke which would lead us rather to consider a great portion of it of slow deposition. It is by no means an uncommon circumstance for the laminae of the slates of this group to be so arranged as to form various angles with other lines, wKich may be considered as those of the beds, or of stratification. Of this structure the annexed section of grauwacke slates at Bovey Sand Bay on the east side of Plymouth Sound, affords us an instructive example. a a, curved beds of slate, the laminae of which meet the ap- parent lines of stratification at various angles, being even per- pendicular to them. The beds are cut off by the fault (/) from the slates c 9 the laminae of which are more confusedly disposed, having however a general horizontal arrangement. The whole 416 GrauwacJce Group. is covered by a detritus (b b) composed of fragments of the same kind of slate as that on which it reposes, and of the va- rious grauwacke rocks of the hill behind. Fig. 107. b a a f It is often interesting to remark, in countries where vertical or highly inclined strata render such observations clear, the various mineralogical changes which frequently take place in minor divisions of the grauwacke group. Thus, for instance, in some parts of Devonshire, where quartz rocks are associated with the grauwacke, there are often opportunities of observing various changes of this kind within the distances of ten or twenty miles. At first the series of beds may consist of fine- grained argillaceous slates. After a range of two or three miles the continuation of these beds will become more arenaceous. The arenaceous structure of the rocks will then gradually in- crease, until they assume the character of compact sandstones. Finally the arenaceous appearance is lost by a kind of union of the grains with each other, and the rocks consist of beds of quartz, in which traces of a mechanical origin can seldom or rarely be observed. If now we continue to follow up the same system of beds, we shall find that, after a course of some miles of quartz rock, the latter character gradually disappears, the strata at first becoming arenaceous, and afterwards passing into schistose rocks, in which the grain is more or less fine. It not unfrequently happens that the quartz rock suffers its first change by the acquisition of mica, which renders it to a certain extent schistose, though at first it only causes it to assume the cha- racter of avanturine. In such situations mineralogical mica- slate, that is, a slaty rock solely composed of mica and quartz, is by no means uncommon, and is merely the quartz rock ren- dered highly schistose by an abundance of mica. Of course the change is not always sufficiently great to produce decided quartz rocks. It frequently arrives only at that state on which the arenaceous rock becomes exceedingly hard, but in which its mechanical origin and sandstone character are quite apparent. Changes of minor importance are innumerable; yet it not un- frequently happens that these minor divisions retain their mineralogical characters unimpaired, or with only slight mo- difications, for considerable distances: thus showing that cer- tain minor causes have varied considerably during the produc- tion of the grauwacke, and that, though the mass of this group, viewed on the large scale, presents considerable uniformity of Grauwacke Group. 417 structure, there is abundant evidence of much variation and change of mineralogical character on the small scale. The mode in which the calcareous beds occur in the group under consideration is also exceedingly variable. They are at times little else than the finer grained strata, which may con- tain calcareous matter, such calcareous matter varying much in its proportion to the other substances of which the beds are composed. Hence we have every change from complete limestone to beds in which the calcareous matter is sparingly disseminated*. This calcareous character, though liable to be lost at unequal intervals, can often be traced for consider- able distances in the general direction of the strata. Whence it may be inferred, that, during the deposit of the grauwacke, calcareous matter was often disseminated over large areas, sometimes being merely mixed with the more common sub- stances of the deposit, while at others it was sufficiently abun- dant to constitute bands of limestone. Though this mode of occurrence is far from uncommon, the more sudden appearance and disappearance of carbonate of lime is by no means rare. Limestone beds will often become abundant, and constitute an important part of the series, without any variation in the mine- ralogical character of the grauwacke in their line of bearing, which would lead us to expect so sudden a development of calcareous matter. Examples of this fact can readily be ob- served in several parts of Southern Devon. The origin of the limestones is of far more difficult expla- nation than the sandstones and slates in which they are inclu- ded. We cannot well seek it in the destruction of pre-existing calcareous rocks ; for as far as our knowledge extends, such rocks are of comparative rarity among the older strata. In fact, the quantity of calcareous matter present in the grauwacke group greatly exceeds that discovered in the older rocks ; and the same remark applies to many of the newer deposits when considered with reference to the grauwacke series. If we take the mass of deposits up to the chalk inclusive, we shall find that, instead of a decrease of carbonate of lime, such as we should expect if that contained in each deposit originated solely from the destruction of pre-existing limestones, the calcareous mat- ter is more abundant in the upper than in the lower parts of the mass; and we may hence conclude that this explanation is insufficient. If, as has often been done with other limestones, we attri- bute the origin of the grauwacke limestones in a great measure * It may not perhaps be generally known that many of these beds, which are little else than argillaceous limestones, with the addition of a small pro- portion of silica, are highly useful as water-setting limestones, being scarcely, if at all, inferior to the lias limestones, so much valued on this account. 2 E 4J8 Grauwacke Group. to the exuviae of testaceous animals and polypifers, we must grant the animals carbonate of lime with which to construct their shells and solid habitations. This they may have ob- tained either in their food or from the medium in which they existed. The marine vegetables are not likely to have sup- plied them with a greater abundance of carbonate of lime at that time than at present. Those that were carnivorous might acquire much carbonate of lime by devouring other animals more or less possessed of this substance : but the difficulty is by no means lessened by this explanation; for the creatures de- voured must have procured the lime somewhere. It would appear that we should look to the medium in which testaceous animals and polypifers existed, for the greater proportion, if not all, of the carbonate of lime with which they constructed their shells and habitations. Now if we consider the mass of limestone rocks to have originated from the exuviae of marine animals, we are called upon to consider that carbonate of lime was once far more abundant in the sea than we now find it, and that it has been gradually deprived of it. This supposi- tion would lead us to expect, that as the sea was gradually de- prived of its carbonate of lime, limestone deposits would become less and less abundant ; and consequently, that calcareous rocks would be most common, when circumstances were most fa- vourable, that is to say, during the formation of the older rocks. This, however, is precisely the reverse of what has happened. Hence we may infer that the origin of the mass of limestone deposits must be sought otherwise than in the attrition or so- lution of older and stratified rocks, or from the exuviaB of ma- rine animals deriving their solid parts from a sea which has gradually been deprived of nearly all its carbonate of lime. Both these causes may have eventually produced important modifications on the surface of the earth ; but the great pro- portion of lime necessary for the formation of the calcareous masses covering a considerable part of it, would appear to have been otherwise obtained. It has been usual to consider the lime of calcareous deposits as derived from limestone rocks, through which waters charged with carbonic acid percolated, the carbonic acid dissolving a certain portion of the lime, which is thus held in solution by the water until it reaches the surface, where it is thrown down in the shape of limestone. This explanation may suffice for the small deposits we observe in calcareous countries, but is insufficient for the productions of limestones generally; for it assumes that the solution of a small quantity of lime obtained from older rocks is, as previously noticed, capable of produ- cing an immense deposit of the same substance. We know that carbonic acid is now discharged into the atmosphere from the Grauwacke Group. 4? 19 earth by means of volcanos, fissures, and springs, and we have no reason to doubt that this has been the case during a long succession of ages ; indeed we have every reason to believe that such discharge of carbonic acid formed* a part of the great economy of nature, for without this aid we should have much difficulty in explaining the abundance of carbon and carbonic acid now locked up in coal deposits and limestones, all of which have clearly been produced successively on the earth's surface. The reason why extensive tracts of carbonate of lime have been produced at one time more than at another is not quite so apparent; but it may be observed, as a mere conjecture, that as this substance is not very unfrequent in volcanic re- gions, great disruptions of strata may have produced circum- stances favourable for its deposition, and that without distur- bances, carbonate of lime may have been thrown upwards in water, through fissures, more abundantly at one time than at another, from causes unknown to us. It is worthy of attention that when the limestones occur, then also do the organic re- mains generally become more abundant, appearing as if the calcareous rocks and the organic remains were connected with each other. That the animals, by secreting carbonate of lime from the medium in which they lived, sometimes contributed considerably to the mass, we are certain, as their remains now constitute a large portion of it ; but that they were the means through which all the carbonate of lime was derived from the waters, may very justly be doubted, more particularly as in certain districts not a trace of animal exuviae can be detected in such limestones. If carbonate of lime were present in some situations and not in others, animals, such as the Crinoidea, Testacea, and Polyparia, would naturally flourish more in the former than in the latter, as they could there more readily ob- tain the lime necessary for them, and we should consequently expect to find their remains more common there than elsewhere. In limestones devoid cf organic remains we appear to have evidence of carbonate of lime being abundant in such situations unconnected with animal life, and we may consider it derived from the interior and dispersed through the waters over a given space, where it has been gradually deposited. When, however, the remains of shells and corals are present, and nearly con- stitute the mass of the rock, other causes may have produced the effects required, precisely as coral reefs and accumulations of shells now occur in one place and not in another, either in consequence of shelter, proximity to the surface of the sea, or other favourable circumstances. Be the general origin of the grauwacke limestones what it may, the causes which produced them were destined to cease during the deposit of the grauwacke itself, and a series of sand- 2 E 2 420 Grauwacke Group. stones and slates, similar for the most part to those beneath, were accumulated upon them. In some districts, such as the North of Devon, there has been a return of causes favourable to the deposit of limestone, and two bands parallel to each other have been produced. In other districts more limestones have been formed, while in some they are nearly absent; a state of things we should expect from variations produced by local circumstances on si- milar general causes in operation over a considerable area. The grauwacke sometimes assumes a red colour in the midst of beds of the usual gray and brown tints (South of Devon, Pembrokeshire, Normandy} the grauwacke district of the Mo- selle, &c.), and is then undistinguishable from the old red sandstone of English geologists*. This red colour seems, for the most part, due to little else than the highly oxidized state of the iron disseminated through the rock. In the red grau- wacke the iron is in the state of a peroxide, while in the gray or brown grauwacke it is combined with a less proportion of oxygen. This change of colour may sometimes be seen in the vicinity of trap-rocks, which have been protruded through the grauwacke. In many cases we may suppose the iron con- tained in the grauwacke to have acquired an additional quan- tity of oxygen, converting it into a peroxide, after the formation of the rock. We may indeed consider the red tint as sometimes caused by heat, from observing that common brown grauwacke, when kept for a long time at a heat insufficient to fuse it, will become red, the iron having acquired additional oxygen from the atmosphere. It will, however, be obvious that the rock when heated beneath water would not be precisely under the same conditions ; and we must be careful not to attribute the occasional red colour of the strata under consideration solely to this or a similar cause, for it will be clear that the dissemi- nated iron may easily have been in a state of peroxide from other causes before such red grauwacke was deposited. In- deed that this has been the case is often sufficiently evident ; and if the old red sandstone be included in this group, there can be little doubt that the iron, during the deposit of the up- per part of the series, was most commonly in that state. Beds and even accumulations of strata are sometimes min- gled with the common grauwacke and grauwacke slate, which * This circumstance renders the determination of those limestones of Southern Devonshire which are much broken by faults, greatly disturbed and contorted, or much concealed by superincumbent (new) red sandstones, exceedingly difficult. The difficulty is particularly felt in the vicinity of Tor Quay, where, however, the limestones of the southern side of Tor Bay would certainly appear to be included in the grauwacke series, as is shown by coast sections, and their prolongation to the Dart. Grauwacke Group. 4-21 at least show a variation in the mode of deposit. Thus, flinty slate, sometimes associated in this series (Devonshire, &c.), is exceedingly compact, and, as its name implies, is principally composed of silica, the rock having much the appearance of a deposit from water in which silica was chemically dis- solved *. Associated with the grauwacke, more particularly in its older portions, we often find beds which, in mineralogical composition, precisely resemble certain greenstones, corneans, &c., known from their mode of occurrence to be of igneous origin. These associated beds do not cut the other strata ; on the contrary they are clearly included in them, and have every appearance of being interstratified. After a course of, per- haps, a few miles, they are seen to terminate on either side, first becoming slaty; at least this frequently takes place. When this slaty condition arises, the rock is then frequently undistinguishable from hornblende slates. The most satisfac- tory mode of explaining these facts, seems that proposed by Prof. Sedgwick for the association of similar rocks with the slates of Cumberland. He considers that they were igneous rocks ejected during the time that such slates were depositing. This hypothesis is applicable to certain of these rocks asso- ciated with the grauwacke of Southern Devon. But care must be taken to distinguish them from other greenstones and por- phyries of the same country, which occur in dykes and masses, and which have clearly been ejected at a much more recent epoch. It is also very necessary to distinguish them from the altered rocks of the same part of England, as these last are more particularly deceptive. Another difficulty, and one by no means easy to surmount, attends the examination of such included beds, or apparent beds, of greenstone and porphyry. If a mass of schistose or stratified rock, such as grauwacke, be exposed to the action of a disruptive force, such mass would rend in the parts of least resistance. Now the lines of stratification would neces- sarily be those of least resistance, and hence the mass would be most likely to part in those lines, permitting the injection of igneous rocks, should these endeavour to escape through the fissures of the grauwacke. We should thus have tabular masses of greenstone, porphyries, and other rocks of that character presenting every appearance of included beds. In- structive examples of such deceptive tabular masses are well seen in Pembrokeshire and Devonshire, as they can fre- quently be traced to larger masses of similar rocks, which * The reader will recollect that under the head of Deposits from Springs, siliceous hcds were noticed as having been produced by deposition from thermal waters in Iceland and the Azores. 422 Grauvoacke Group. have evidently been protruded through the grauvvacke after its consolidation. Towards the lower part of the grauwacke group, its fossili- ferous character disappears, and the presence of crystalline rocks, apparently of contemporaneous origin, becomes more common. This change varies much in different countries, but in general the slaty rocks gradually prevail, presenting a very great thickness of argillaceous schists. We seem to have ar- rived, in the descending order, at a state of the world when there was a combination of those causes which have produced fossil iferous and non-fossiliferous rocks. That there should be a transition or passage, even effected by the alternate ope- ration of particular causes, from that condition of the world's surface when chemical action prevailed to that when mecha- nical action became more abundant, is what we should ex- pect, since it is in accordance with our knowledge of rock deposits generally; for we observe, however sudden certain changes may have been produced in particular situations, that viewed on the large scale, a general change of circumstances attending rock formations has been more or less gradual. Chlorite slates are not unfrequently associated with the low- est portions of the grauwacke series. I have often observed this to commence by small alternations of chloride argillaceous slate with common argillaceous slate, the former gradually becoming more charged with chlorite. There are often also very ambiguous rocks, to which it is exceedingly difficult to assign names, as they are constantly varying in their minera- logical composition. Their general character is, however, far more chemical than mechanical. The study of this part of our subject must always be at- tended with great difficulty ; for, independently of the mixture of chlorite, talcose, and other slates with the lowest fossili- ferous deposits, we have to contend with the presence of ig- neous rocks injected among these deposits in their line of stra- tification, producing, as before observed, the most deceptive appearances. No small difficulties are also caused by the al- teration of strata, arising from the protrusion of granite and other igneous rocks among them, such products often causing, when the masses are large, very remarkable changes, the va- rious grauwacke beds assuming the appearance of a great va- riety of older rocks. Mica-slates, gneiss, and hornblende rocks are often thus re-produced, but the changes so effected are limited in extent, and by careful examination can be easily traced. This reproduction of gneiss, mica-slate, and other rocks of the same kind, to account for which nothing can be more simple, as I shall have occasion to show elsewhere, has given rise, I am informed* to the hypothesis, that all the Gmuwacke Group. 423 crystalline and non-fossiliferous rocks are but fossiliferous rocks out of which the organic remains have been driven by heat. This may certainly be a convenient hypothesis as tar as regards a particular theory, but can scarcely be seriously en- tertained by those who have examined any of those vast tracts of the true inferior or non-fossiliferous rocks, so common in various parts of the world, more especially with respect to the mode in which the various mineral masses occur. MM. Brongniart and Omalius d'Halloy long since pointed out the apparent alternation of the granitic and schistose rocks of the Cotentin and Brittany, as also that the deposits thus associated with the granitic compounds were fossiliferous *. Thegrauwacke of these districts certainly appears associated, more particularly in its lower parts, with rocks, the mechani- cal origin of which is far from evident; but while studying them we must be on our guard against granite veins, and other intrusions of the same rock, which are also observable in that country. Independently, however, of decidedly intruded rocks, there are associated crystalline rocks which render it exceed ingly hazardous to affirm where the series, in which confusedly crystalline compounds prevail, may commence, or where the mechanical and fossiliferous deposits may terminate. The highly indurated sandstones also, which are clearly, like those of South Devon, associated with the fossiliferous rocks of Normandy, so pass into quartz rock, that, as M. Brong- niart has observed, they often present the appearance of ha- ving been produced by confused crystallization. The following is a summary of the various organic remains stated to have been detected in the grauwacke series ; it is necessarily one which does not pretend to more than an ap- proximation to the truth, for the catalogue on which it is founded will no doubt receive both important additions and corrections ; it may, however, be found useful as affording a general view of the subject. Plantce. Fucoides, 2 species. Calamites, 2. Sphenopte- ris, 1. Cyclopteris, 1. Pecopteris, 1. Sigillaria, 2. Le- pidodendron, 1. Stigmaria, 1. Asterophyllites, 1. Zoophyta. Manon, 2. Scyphia, 5. Tragos, 2. Gorgo- nia, 2. Stromatopora, 2. Madrepora, 1. Millepora? 1. Cellepora, 2. Retepora, 3. Flustra, 1. Ceriopora, 6. Glauconome, 1. Agaricia, 1. Lithodendron, 3. Caryo- phyllia, 1. Fungites, 2. Anthophyllum, 1. Turbinolia, 3. Cyathophyllum, 20. Strombodes, 1. Astrea, 2. Colum- naria, 1. Sarcinula, 3. Coscinopora, 1. Catenipora, 3. * Journal des Mines, t. xxxv. ; 181 L 424 Grawmacke Group. Syringopora, 4. Calamopora, 9. Aulopora, 5. Favosites, 6. Mastrema, 1. Amplexus, 1. Pleurodyctium, 1. Cy- clolites, 1. Radiaria. Apiocrinites ? 2. Pentremites, 1. Pentacri- nites, 1. Actinocrinites, 7. Cyathocrinites, 4. Platycri- nites, 4. Rhodocrinites, 5. Melocrinites, 2. Cupressocri- nites, 3. Eugeniacrinites, 1. Eucalyptocrinites, 1. Sphae- ronites, 4. Annulata. Serpula, 5. Conchifera. Thecidea? 1. Pentamerus, 2. Gypidia, (Pentamerus?) 3. Spirifer, or Delthyris, 41. Terebratula, 31. Strygocephalus, 3. Calceola, 1. Atrypa, 14. Pro- ducta, or Leptaena, 22. Orbicula, 1. Crania,]. Gryphaea, 1. Pecten, 5. Plagiostoma, 1. Inoceramus, 1. Avicula, 1. Pterinea, 9. Posodonia, 1. Area, 1. Nucula, 5. Tri- gonia, 2. Megalodon, 1. Modiola, 3. Mytilus, 1. Cras- satella, 1. Cardium, 9. Cardita, 4. Isocardia, 2. Vene- ricardium, 1. Luciria, 3. Cyprina, 1. Corbula, 1. Cy- there, 8. Sanguinolaria, 8. Pholadomya, 1. Mollusca. Patella, 5. Pileopsis, 3. Melania, 1. Na- tica, 1. Nerita, 2. Delphinula, 5. Cirrus, 1. Pleuroto- maria, 1. Euomphalus, 16. Trochus, 6. Rotella, 1. Turbo, 2. Turritella, 7. Pleurotoma, 1. Murex? 1. Buccinum, 5. Phasianella, 3. Bellerophon, 9. Conularia, 3. Ortho- ceratites, 30. Cyrtoceratites, 6. Spirula, 7. Lituites, 2. Nautilus, 9. Ammonites, 15. Aptychus, 2. Crustacea. Calymene, 17. Asaphus, 21. Ogygia, 4. Paradoxides, 9. Nileus, 2. Illaenus, 3. Ampyx, 1. Agnostus, 1. Isotelus, 2. Pisces. At least 1 genus, and 2 or 3 species. Thus making: Plantce, 9 genera, 12 species. %oophijta^ 33 genera, 98 species. Radiaria, 12 genera, 35 species. Annulata, 1 genus, 5 species. Conchifera, 35 genera, 191 species. Mollusca, 26 genera, 144 species. Crustacea, 9 genera, 60 species. Pisces, 1 genus, 2 species. Total, 126 genera, 547 species. From the above it would appear that the grauwacke series contains a mixture of genera inhabiting the seas and oceans of the present day, and of others which are not now known. It may be doubtful how far all the genera have been correctly determined ; for possibly some of them may have been rather hastily referred to those now existing, while others may have been considered extinct without sufficient evidence. But, ad- mitting these sources of error, some genera are certainly, as far as our actual knowledge extends, extinct, while others do not differ from those now existing. This catalogue also shows Grauwacke Group. 425 that, at the early epoch when the grauwacke was produced, there was not that poverty of organic structure which was once supposed. From the various forms of the fossils imbedded in the grau- wacke, we may infer that the animals, of which they consti- tuted the solid parts, occupied situations as different as those of the present day ; some preferring deep waters, while others were fitted for shallow seas, and not a few swam freely in the open ocean ; certain creatures frequenting one kind of bottom, while others sought another of a different description. The most abundant shells belong to the genera Orthoceratites, Producta, Spirifer, and Terebratula. The former often at- tain a large size, even reaching a yard or more in length ; so that if they really once constituted a part of swimming mol- lusca, analogous to the Nautilus of the present day, some of such creatures must have far exceeded the size of the animals of that kind now known to us. The three latter most abundant genera constitute a natural group, which the Swedish naturalists have arranged under the heads of Lep- tccna (Producta\ Orthis^ Cyrtia^ Delthyris (Spirifer), Gy- pidia, Atrypa, Rhynchora^ and Terebratula ; the characters being considered such as to justify the formation of the ge- nera. Supposing this arrangement to be well founded, it would appear from the lists of those who have proposed it, that Terebratula are rare in the older rocks, such as those under consideration, while they are abundant in the newer strata. Products are, as has been seen, common in this and the carboniferous groups, and existed during the deposit of the zechstein. Spirifers, which also abounded during the depo- sit of the grauwacke and carboniferous series, have been ob- served as high up as the lias, where three species of the genus Spirifer have been detected, one (Spirifer Walcotii) being a very common and characteristic shell. The Terebratulae, which, even admitting the Swedish divisions, are found in the preceding series, if not in the higher part of this, extend upwards to the present day, many species being now known. Taking, therefore, this natural group as it existed at this early period, in which we should probably include the car- boniferous limestone, and tracing it upwards through the va- rious rocks, we find that the Productae first disappeared, and then the Spirifers, while the Terebratula have been preserved through all the changes which have taken place on the surface of our planet. The family of the Trilobites was one, the individuals of which must have swarmed in particular places during the deposit of the grauwacke. In some parts of Wales the 4-26 Grauwacke Group. Fig. 108. Fig. 109. Asaphus Debuchii (Y\g. 108.) is so abun- dant that the laminae of the slates are charged with them, so that millions have probably lived and died not far distant from those places where we now discover their remains. This species has not been confined to Wales, though it is there very abundant, but has also been discovered in Norway and Germany. The Trilo- bite long known in museums as the Dud- ley Trilobite, because found so commonly at that place, is the Calymene Blumen- bachii of M. Al. Brongniart (Fig. 109.). This species existed over a considerable area, having not only been discovered in England, Germany, and Sweden, but also in North America. Although many parts of these creatures are found distributed in such a manner that we may conclude they were separated by decomposition after the death of the animal, the perfect preservation of others, and their frequent contracted at- titudes, such as we should expect creatures of this structure to assume when disturbed, would lead us to conjecture that they had been often suddenly destroyed, and as sud- denly enveloped in that matter which subse- quently became hard rock ; thus preventing the separation of the harder parts by decom- position. The forms of the Trilobite family vary more con- siderably than might be supposed from the Asaphus and Caly- mene represented above, as will be seen by the annexed figure of Agnostus pisiformis, Fig. 110. being the natural size of the animal, and Fig. 111. a magnified re- Fig. 110. presentation of it. The Trilobite family seem now to have entirely disappeared from among existing animals ; and we may perhaps venture to infer, from our present information respecting or- Fig. 111. ganic remains, that it became extinct before the Products? ; and we are nearly certain it ceased to exist long before the Spirifers, for neither in the muschelkalk nor in the lias has the smallest trace of them ever been detected. Unlike the Trilobites, the Crinoidea common in this early period are continued up to the present day, though many ge- nera observed in the grauwacke series and in the carboniferous group seem to have disappeared previous to the deposit of the oolitic series, when other genera were called into existence. Grauwacke Group. 427 The genus Pcntacrinites being, according to M. Goldfuss, found in the rocks under consideration, and being well known in the present seas, this genus has also survived the various changes that have taken place on the earth's surface. The discovery of the defensive fin bones, named Ichthyo- dorulites, in the grauwacke series is worthy of attention, as it shows that the class of animals to which they belong was among the earliest inhabitants of the globe, and that it con- tinued to exist over what now constitutes Europe, up to the cretaceous rocks inclusive, though differing in species, as far at least as we can judge from the various forms of the bones. The Ichthyodorulites are usually accompanied by palates; these latter have not yet been detected in the grauwacke. Among the corals will be found several genera now existing; and it deserves notice, that throughout the series of fossilile- rous rocks, wherever there is an accumulation of polypifers, such as would justify the supposition of coral banks or reefs, the genera Astrea and Caryophyllia are present, genera which, according to the more recent observations of natural- ists, in addition to Meandrina and one or two others, are the principal architects of coral reefs at the present day. Our knowledge of the kind of vegetation existing at, and entombed during, the epoch of the grauwacke group, is in- sufficient to warrant any general conclusions respecting it, further than it was probably much the same as that, the re- mains of which are abundantly preserved in the carboniferous series. Anthracite has been long known in the grauwacke of North Devon, and may have been derived from the re- mains of vegetables. Where there are vegetables entombed in rocks we may expect to find accumulations of them, and there seems no good reason why grauwacke should not con- tain its coal-beds as well as other great deposits. It will have been observed that all the fossiliferous groups of rocks have their accumulations of vegetable matter in some part or other of the areas respectively occupied by them. In 'Europe these accumulations have been more abundant in our carboniferous group than at any other time, more especially in the mass of sandstones and shales thence named the Coal-measures ; but it by no means follows that this should have been the case as re- gards the whole surface of the world. On the contrary, all analogy with other rock deposits would lead us to infer that the coal-measures would have their equivalent marine depo- sits, in which, if terrestrial plants occurred at all, they would be found merely scattered here and there, as they are in other rocks abounding in marine remains, and hence termed Ma- rine deposits. M. Elie de Beaumont observes that the grauwacke rocks 428 Grauwacke Group. of the Bocage (Calvados), and of the south-eastern angle of the Vosges, contain vegetable impressions differing but little from those discovered in the coal-measures, as also anthracite, sometimes worked for profitable purposes *. According to M. Voltz, certain anthracitic rocks of Baden are of this age ; and M. Virlet refers the coal of St. Georges Chatelaison to the grauwacke series f. Mr. Weaver considers that all the coal in the province of Munster, excepting that in the county of Clare, is of this age. He states that thin beds of anthracite, inclined at various angles from 70 to vertically, are included in the grauwacke at Knockasartnet, near Killarney, and on the north of Tralee. Mr. Weaver further remarks that this old coal is more developed in the county of Cork, particularly at Kanturk, and that large quantities of it are annually raised at Dronagh collieries. He also enumerates beds in the county of Limerick, on the left bank of the Shannon, north of Abbey- feale and at Longhill. The remains of plants, described as chiefly those of Equiseta, and Calamites* with some indica- tions of Fucoides, are stated to be common J. Assuming the foregoing observations to be correct, we ob- tain evidence that the accumulation of vegetables sufficient to produce beds of anthracitic coal commenced at the epoch of the grauwacke in Europe. Prof. Eaton states that anthra- cite is found in an equivalent deposit in America (Wor- cester, and Newport) . If the relative age of these latter rocks be also correctly determined, it proves the existence of dry land, at different distant points, with vegetation upon it, contemporaneously, or nearly so, with the first appearance of animal life. Although when we regard the mass of the grauwacke rocks we are struck with the minute proportion that organic remains bear to the whole, we must still perceive that the atmosphere was capable of supporting vegetation, and the seas of sus- taining zoophytes, crinoidea, annulata, conchifera, mollusca, Crustacea, and fish. What other creatures existed we are unable, from the absence of their remains, to judge: it may however be by no means unphilosophicul to conclude that ve- getation did not exist alone on dry land, but that, consistently with the general harmony of nature, it afforded food to ter- restrial creatures suited to the circumstances under which they were placed. * Elie de Beaumont, Researches on some of the Revolutions which have taken place on the Surface of the Globe ; Phil. Mag. and Annals, vol. x. p. 247. f Bulletin de la Soc. Geol. de France, t. iii. J Weaver, Proceedings of the Geological Society, June 4, 1830. Eaton, American Journal of Science, vol. xix. Grauwacke Group. 429 To describe the precise limits of the fossiliferous deposits, and draw fine lines of distinction between them and the non- fossiliferous rocks, is obviously impossible. We can only in- fer that the remains of organic life were generally entombed in deposits of mechanical origin, as well at this early period as subsequent to it. Respecting the abundance and different structures of the animals first called into existence, we shall never perhaps have any definite ideas ; for the preservation of any portion of their more solid parts must always have de- pended on a great variety of circumstances, not likely to have been most favourable during a state of things, in which a change was effected from the formation of such rocks as gneiss, mica-slate, and others of the same description, to the deposit of those evidently of mechanical origin. Whatever the kind of animal life may have been which first appeared on the surface of our planet, we may be certain that it was consistent with the wisdom and design which has always prevailed throughout nature, and that each creature was pe- culiarly adapted to that situation destined to be occupied by it. Bearing therefore in mind this general adaptation of ani- mals to the circumstances under which they are placed, we may be led so far to speculate at this early condition of life, as to inquire, what kind of creatures, judging from the gene- ral character of those known to us, might flourish at a period when there might have been a comparative difficulty in pro- curing carbonate of lime for their solid parts. It will be ob- vious that fleshy and gelatinous creatures, such as Medusae and other animals of the like kind, might have abounded, as far as regards a comparative scarcity of this substance. Hence it would be possible to have the seas swarming with these and similar animals, while testaceous creatures and others with solid parts were rare. These remarks are merely intended to show, that the scar- city of organic remains observed in the lowest part of the grau- wacke by no means proves a scarcity of animal life at the same period, though from it we may infer that testaceous and other animals with solid parts were not abundant. Mere fleshy crea- tures may have existed in myriads without a trace ol them having been transmitted to us. In proof of this, if any were requisite, we may inquire what portion of those myriads of fleshy animals, which now swarm in some seas, could be transmitted, as organic remains, to future ages *. * Dr. Turner has suggested to me, that under this supposition of an abundance of Medusae or of analogous creatures among the early inhabit- ants of our globe, we may perhaps account for the bituminous nature of some of the earlier limestones, more particularly of the carboniferous series, in which not a trace of solid organic remains can be observed ; for the de- 4-30 Graittoacke Group. It may be remarked, while on this subject, that though an extensive distribution of carbonate of lime is essential to a great variety of animals, it is surprising how little may supply the wants of some, even those with vertebrae, such as sharks and cartilaginous fish generally. To consider that there may have been some connexion between the animals with solid parts and a facility of procuring carbonate of lime on the sur- i'ace of the globe, appears perfectly consistent with the design manifested in the creation, because it assumes such design at all periods, and constant harmony between the forms of crea- tures and their mode of existence. If we imagine a mass of animals to be suddenly called into life, each properly pro- vided with its solid parts, the carbonate of lime contained in their bodies would no doubt be sufficient for a constant quan- tity of the same animal life during a succession of ages ; for, by devouring each other, this necessary substance would be transmitted from one creature to another. We are however certain that this has not been the case ; for the solid parts of animals which have been successively imbedded in various rocks, constitute a very large proportion of certain of those rocks, and if withdrawn from the fossiliferous deposits gene- rally, would very considerably diminish their thickness. There- fore if the exuviae of animals had not been entombed, and if the supply of carbonate of lime had not been greater than that which could have been derived from the mere destruc- tion of one animal by another, for the purpose of food, the surface of our planet would not have been what it now is ; and consequently, the fitness of things for the end proposed being constant in creation, the general condition of animal and ve- getable life would not have been such as we now find it. From the advance of Geology, many districts which were formerly considered as composed of grauwacke, are now re- ferred to less ancient deposits, and consequently the surface occupied by grauwacke is much less extensive than was for- merly supposed. Thus large portions of the Alps and Italy have been deprived of their supposed antiquity, which had been founded on the mineralogical structure of the deposits. The grauwacke group occurs in Norway, Sweden, and Russia. It forms a portion of southern Scotland, whence it ranges, with breaks, as far as regards the surface, formed by newer deposits or the sea, down western England and Wales, into Normandy and Brittany. It appears abundantly in Ire- land. A large mass of it is exposed in the district constituting composition of a mass of such creatures would produce much bituminous matter, which may have entered largely into the composition of limestones then forming. Grauwacke Group. 431 the Ardennes, the Eifel, and the Taunus. Another mass forms a large portion of the Hartz mountains, while smaller patches emerge in other parts of Germany, on the north of Magdeburg, and other places. In all these situations there is, notwithstanding small variations, a general and prevailing mineralogical character, which points to a common mode of formation over a considerable area. From all the accounts which have been presented to us by Dr. Bigsby and the Ame- rican geologists, we have every reason to consider that a de- posit closely agreeing in relative antiquity, and in its general mineralogical and zoological characters, exists extensively in North America : so that there is evidence to show that some general causes were in operation during the same epoch over a large portion of the northern hemisphere, and that the re- sult was the production of a thick and extensive deposit enve- loping animals of similar organic structure over a consider- able surface*. * It was considered useless to present a long detail of the exact areas oc- cupied by the grauwacke rocks, as the reader will comprehend more by a single glance at good geological maps of any given country, such as Green- ough's Map of England, Hoffmann's North-western Germany, Oeynhausen, La Roche, and Von Dechen's Countries near the Rhine, and De Beau- mont's and Dufrenoy's France, than by long and tedious descriptions. Fig. 112. Cyathophyllum turbinatum, Goldf. 432 Inferior Stratified Rocks. SECTION X. INFERIOR STRATIFIED OR NON-FOSSILIFE- ROUS ROCKS. SYN. Clay slate (Schiste Argilleux, Fr. ; Phyllade, Daubuisson ; Thon- schiefer, Germ.). Aluminous slate (Ampelite Alumineux, Brong. Schiste Alumineux, Fr. ; Alaunschiefer, Germ.). Whetstone slate (Schiste co~ ticule, Brong. ; Wetzschiefer, Germ.). Flinty slate (Schiste siliceux, Fr. ; Jaspe Schistoide, Brong. ; Kieselschiefer, Germ.). Chlorite slate (Schiste Chloriteux, Fr. ; Chloritschiefer, Germ.). Talcose slate (Schiste Talqueux, Fr. ; Talkschiefer, Germ.). Steachist. Hornblende slate (Amphibolite Schistoide, Fr. ; Hornblendschiefer, Germ.). Hornblende rock (Amphibolite, Daubuisson). Quartz rock (Quartzite, Brong. ; Quarzfels, Germ.). Serpentine (Ophiolite, Brong. ; Serpentin, Germ.). Diallage rock (Euphotide, Haiiy ; Schillerfels, Germ.). Whitestone, (Eurite, Daubuisson ; Weisstein, Germ.). Mica slate (Schiste Micace, Micaschiste, Fr. ; Glimmerschiefer, Germ.). Gneiss (Gneiss, Fr. ; Gneuss, Germ.). Protogine. WE have now arrived at that early condition of our planet, when, as far as our knowledge extends, neither animal nor vegetable life existed on its surface. The student, instead of wandering in imagination amid forests and over lands and seas, surrounded by strange vegetables and still stranger ani- mals, should now direct his attention to those laws which go- vern inorganic matter. This may not at first sight be so at- tractive as the contemplation of the varied forms of organic life and the probable conditions under which it may have ex- isted; but it will nevertheless be found equally, if not more delightful, as the inquirer obtains more certain results, from the investigation being conducted through the medium of the exact sciences. It must, on the outset, be confessed that little has yet been accomplished respecting the causes which may have produced gneiss, mica-slate, and other rocks of the same character. Names of the various compound and confusedly crystalline rocks we have in abundance, and if the investigation required no other aid we might sit down satisfied ; but unfortunately the abundance of these names has confused the subject, and the student has more frequently contented himself with ar- ranging and disarranging particular mineral compounds in a cabinet, than in investigating their general relations to each other, and the occurrence of the whole in the mass. It will readily be admitted, that the difficulty of the sub- ject is very considerable, requiring no small insight into the Inferior Stratified Rocks. 4.33 exact sciences; but the subject being difficult would seem a good reason why the more advanced cultivators of those sci- ences should attack it, offering as it does such an ample field for the exertion of their abilities. The inferior stratified rocks are of various compositions, sometimes so passing into each other, that it is almost impos- sible to affix definite names to the different mixtures. The strata rarely present a simple mineral substance constituting a large tract of country, without the admixture of other sub- stances, unless we consider clay slate as such. Before how- ever we proceed further, the student should become acquainted with the following rocks, which more particularly appear to deserve distinguishing names. Argillaceous or Clay Slate. This rock, as its name implies, is schistose, and contains a considerable portion of argillaceous matter. It varies mate- rially as to induration, fissility, and composition ; and is com- monly {indistinguishable, except in its geological relations, from the argillaceous slates of the grauwacke series. Its origin therefore becomes very ambiguous, and is not the less so from often containing cubical and other regularly formed crystals of iron pyrites, affording evidence that the rock was once in the condition to permit the free arrangement of sulphuret of iron into crystals, a fact observable in the argillaceous deposits of all ages, some decidedly of mechanical origin : therefore we have no direct evidence to show that the argillaceous slates of this epoch may not also have been mechanically produced ; for the fineness of grain will by no means assist us, the texture of the roofing slates obtained from the grauwacke series being altogether as fine as that of the argillaceous slates associated with the mica slate, or gneiss. Like, also, the argillaceous slates of the same series, the lines of cleavage are frequently not the same with those which appear to be lines of stratifica- tion, but meet them at various angles. Argillaceous schist passes into chlorite slate, talcose slate, and other rocks, by gradually acquiring particular minerals, which finally replace the matter of the argillaceous slate. Chlorite Slate. This is by no means an unfrequent associate of the preceding, into which it passes on the one hand, while it graduates into mica slate, &c. on the other. It is of course essentially com- posed of chlorite, which occurs alone or mixed with quartz, felspar, hornblende or mica, in various proportions. 2F 434 Inferior Stratified Rocks. Talcose Slate. This is also a rock into which argillaceous slate graduates, at first acquiring a few plates of talc, and afterwards becoming replaced by that mineral, generally associated with quartz, or quartz and felspar. There is not unfrequently a transition from this rock into mica slate. Quartz Rock. Quartz rock, as has been observed by Dr. Macculloch, when viewed on the large scale, so that the grauwacke series be included, sometimes appears of chemical, at others of me- chanical origin. We should, however, carefully separate the quartz rock which occurs in the grauwacke series, from that associated with the rocks under consideration. That it should possess an arenaceous character in the former case, would be in accordance with the structure of grauwacke generally, though I would be far from stating that some of the grauwacke quartz rocks may not have been chemically produced. The quartz rocks usually interstratified with gneiss, mica slate, &c. are commonly either granular, or resemble common quartz. As it is a subject of much interest to determine if they really present marks of mechanical origin when associated with the true non-fossiliferous rocks, and not those which may appear such from alteration, quartz rocks should be very carefully examined, for there is much reason to believe that some of the quartz rocks stated to occur among the inferior rocks, are really not so as- sociated. The quartz rocks, intermingled with mica slate and gneiss, are observed to pass into both those rocks, by acquiring mica in the one case and mica and felspar in the other. This rock often occupies extensive areas. It is well known in Scot- land and its isles; and according to MM. Humboldt and Eschwege, it is of an extent and thickness in the Cordilleras of the Andes and in Brazil, far exceeding what we are ac- quainted with in Europe. Some of these Brazilian rocks are auriferous, and M. Eschwege attributes the auriferous and platiniferous deposits of that country to their decomposition or destruction. Hornblende Rock and Slate. Under this head are included, following the suggestions of Dr. Macculloch, all those compounds, clearly contempora- neous with the rocks among which they occur, of which horn- blende constitutes an essential and prevailing ingredient. Much of this rock has been known by the names of primitive green- stone, and greenstone slate, being composed of hornblende Inferior Stratified Rocks. 435 and felspar. The hornblende sometimes so predominates as to exclude other minerals. As the names imply, these rocks occur both compact and fissile; in the latter case the felspar is frequently green. Some curious changes in the structure of continuous beds may occasionally be observed. I have seen thick beds of a compound consisting of nearly equal parts of hornblende and felspar, not differing in mineralogical character from the common unstratified greenstones, become gradually schistose by acquiring mica, so that the compound resembles certain varieties of gneiss. .After a time the hornblende would become scarce, and the rock would become a mixture of mica and felspar, with probably some quartz. Changes of this kind are innumerable, and serve to distinguish the hornblende rocks from the greenstones, with which, without careful examination, they may be confounded. In the southern part of Devon, hornblende rocks insensibly become converted, in the line of their direction, into chlorite slate. This is observable in the di- rection of the strata between the promontory named the Bolt Tail and the neighbourhood of Salcombe. From the infor- mation of Mr. Royle, it appears that large tracts of country are occupied by hornblende slate in India, particularly in the central range of mountains. It occurs also in the Himalah mountains, associated with gneiss and mica slate. In both situations it often contains disseminated grains of magnetic or titaniferous iron ore, which in the central range of mountains is found abundantly in the river courses, being washed out, by the rains, from the decomposed hornblende rock. In the Himalah mountains the natives pound up this variety of horn- blende rock, and obtain the iron ore by washing*. Professor Sedgwick informs me that the menaccanite (titaniferous iron ore), found abundantly in the bed of a stream near Tregonwell mill, Menaccan, Cornwall, is derived from the decomposition of a hornblende rock, composed of hornblende and felspar f. In these various cases the titaniferous iron ore appears to form a constituent part of the rock. Limestone. This rock occurs variously associated among the inferior stratified rocks. The saccharine variety is, however, by no means confined to them ; for, as has already been noticed, it is discovered among the fossiliferous deposits, as for instance, amid the belemnitic rocks of the Western Alps. The lime- stone is of various colours, but principally white and crystal- line, affording the well known statuary marbles of Greece and Italy. It is sometimes large-grained, as, for example, that * Royle, MS. f Sedgwick, MS. 2 F 2 4-36 Inferior Strati/led Rocks. included in mica slate on the lake of Como, which afforded the mass of materials for the construction of the celebrated Duomo at Milan. From a mixture of talc or mica, it sometimes be- comes schistose. Some of the crystalline dolomites are asso- ciated with these marbles and others of the rocks under con- sideration. The limestones not only vary in their crystalline character, but pass into compact substances, and become mixed with various minerals, such as hornblende, augite, quartz, &c. A remarkable compound, consisting of nearly compact lime- stone with small crystals of felspar, and thus forming a kind of porphyry with a calcareous base, occurs at the Col de Bon- homme, near Mont Blanc, constituting the calcipTiyre felspa- thique of M. Brongniart. Eurite. A rock principally, and in many cases entirely, composed of the substance named compact felspar. It does not appear to constitute any extensive tracts in nature, but to be generally subordinate to gneiss or mica slate. Mica Slate. This rock is essentially composed of mica and quartz, and forms extensive tracts of country, as well as thin beds included among other rocks. Mica slate sometimes contains garnets so abundantly, that they may almost be regarded a regular com- ponent part of the rock. It graduates on the one hand into gneiss, and on the other into talcose slate, chlorite slate, and other compounds. Gneiss. This rock is either schistose or divided into beds which vary in thickness. It is composed of quartz, felspar, mica, and hornblende, with the occasional mixture of other minerals. Sometimes one of. these minerals is absent, sometimes another: from this loss of either the quartz, felspar, mica, or hornblende, and from the occasional absence of even two of them, as well as the admixture of other substances, there results a very va- riable general compound. When it occurs confusedly crystal- lized in regular beds, the mica not being distributed in plates parallel to the strata, as is the case in the fissile and schistose gneiss, it is really, as far as mineralogical characters are con- cerned, nothing but that much disputed substance, stratified granite. And this is rendered even more apparent, when, as happens in the Alps, Scotland, and other situations, large crystals of felspar are disseminated through it, precisely as in the granite of Dartmoor, &c. When blocks have been detached Inferior Stratified Rocks. , 437 from this gneiss, as has happened with many of the erratic blocks of the Alps, they cannot be distinguished from those of true granite. Gneiss, with its variations, constitutes very consi- derable tracts of country. Protogine may conveniently be arranged with gneiss, the only difference between its decidedly stratified varieties and the gneiss being the substitution of talc and steatite for the mica. Protogine is the well known granitic rock of Mont Blanc, which certainly has the appearance of graduating into a more massive compound ; but in this it does not differ from gneiss, which also seems to pass into granite in a similar manner. Although the above are the most remarkable of the inferior stratified rocks, they are far from being the whole of them. The varieties and transitions of one to the other appear end- less, and, occurring in no determinate order, set classifications utterly at defiance. It was at one time considered that gneiss was the inferior rock, and was succeeded by mica slate ; but this is found to be by no means the case, the two being inti- mately blended with each other as well as with other com- pounds. It must however be confessed that the mass of the gneiss frequently appears to occupy an inferior position. All this apparent confusion, and this passage of one rock into another, though it embarrasses arrangements, may be precisely the circumstances which may lead to some know- ledge of the causes that have produced the lowest stratified rocks. These irregular passages, and the possibility of dis- covering any given rock at the top as well as at the bottom of the series, show that the causes, whatever they may have been, which produced this variety in the substances, were secondary, and that there was some general cause upon which the forma- tion of the whole depended. If we also consider what minerals have entered most largely into the composition of the whole mass, we find that quartz, felspar, mica, and hornblende, are those with which it most abounds, and which impress their characters upon its various portions; chlorite, talc, and carbonate of lime, are certainly not wanting; but if we, as it were, withdraw ourselves from the earth and look down upon such parts of its surface as are geologically known, we find that these latter mineral sub- stances constitute a very small portion of the whole. The inferior stratified rocks which form the largest part of the ex- posed surface of our planet are gneiss and mica slate, and when viewed on the great scale, the others are more or less subor- dinate to them. Supposing this view an approximation to the truth, we ar- rive at another and Important conclusion ; namely, that the 438 Inferior Stratified Hocks. minerals which compose the mass of these stratified rocks are precisely those which constitute the mass of the unstratified rocks, rocks which, from the phenomena attending them, are referred to an igneous origin. We may here inquire what are the circumstances which have determined the arrangement of these minerals into stratified masses in the one instance, and into unstratified masses in the other. This question is by no means of easy solution in the present state of our knowledge : but while we wait for information, it may be observed that the conditions, under which the two classes of rocks were pro- duced, must, to a certain extent, have been very distinct. Yet we find, still viewing the subject in the mass, that the same elementary substances have produced the same minerals in both, the only difference between them being their general difference of arrangement relatively to each other, so that they should constitute a stratified compound in the one case, and not in the other. Looking into the structure of gneiss, mica slate, chlorite slate, talc slate, &c. we find, if we except the thick-bedded gneiss or stratified granite, that it is the arrange- ment of the mica, chlorite, or talc in certain general planes which has produced the fissile and schistose structure. This, however, has not been the only cause of stratification, (if it may be so termed, the lines of fissility not being necessarily those of stratification,) for we find, in the thick-bedded gneiss, the hornblende rock, the quartz rock, the eurite, and the sac- charine limestone, that other causes must have produced thick beds of confusedly crystallized substances. There is, nevertheless, so much apparent mineralogical re- semblance between these two classes of rocks, that we can scarcely refrain from conjecturing the remote origin of the one and of the other to be in some manner connected, modi- fying circumstances having impressed certain characters on each. It must be confessed this is a mere hypothesis, and the student must be careful only to consider it in that light; but it may be asked, what essential difference there is between thick-bedded gneiss, particularly that with imbedded crystals of felspar, and granite, between some hornblende rocks and greenstone, except that the one occurs quietly interstratified in beds, while the other is unstratified, even sometimes cutting through stratified and similar compounds ? We may here also notice serpentine and diallage rock, of which there is often good evidence (as will be seen in the next section) for consi- dering igneous and injected rocks, cutting strata in the man- ner of granite and greenstone. I have never myself observed these rocks stratified, but Dr. Macculloch appears to be cer- tain that they are so in the Scottish Isles. A priori* we should imagine that there was as much probability in finding strati- Inferior Stratified Rocks. 439 fied rocks, whose mineralogical composition should render them serpentine, and its common associate diallage rock, as that we should find stratified rocks mineralogically the same with granite and greenstone : therefore we should be disposed to admit them into the catalogue of inferior stratified rocks, even if we had not the direct opinions of Dr. Macculloch and some other geologists on the subject. As the question is one of some interest, it should be stated that the localities where the stratification may be observed, and which are pointed out by this author, are ; for diallage rock, Unst, Balta, and Fetlar; and for serpentine, also the Shetland Islands. The stratification is described as often obscure; but the dial- lage rock is stated to be associated with gneiss, mica slate, chlorite slate, and argillaceous slate, alternating with them ; and when occurring distinct, presenting the same dip and di- rection as the neighbouring rocks. According to Dr. Mac- culloch, there can be no doubt that serpentine is stratified in Unst ; as also appears to be the case in Fetlar, though the strata are not there so regular. Let us now cast a glance at the substances which enter into the composition of some of the more marked inferior stratified rocks, and see in what respect such rocks differ chemically from each other. To do this we must search for the best ex- isting analyses of those minerals which enter into their com- position, and then calculate the relative proportions of the constituent substances in one hundred parts of each rock. These calculations will necessarily be little else than approxi- mations to the truth, more particularly as we shall take the mean of several analyses of the same mineral, and consequently the mean of the losses in each; moreover we shall be com- pelled to suppose definite compounds of those things which vary much in nature ; but it is hoped that the calculations will be sufficiently accurate to answer the purpose for which they are intended. If we assume that quartz, as it occurs in these rocks, is pure silica, we shall commit no great error as far as regards the present inquiry. With mica, however, we shall have far more difficulty? inasmuch as two substances, of much the same ex- ternal characters, pass by that name, the one containing lithia, the other fluoric acid. How far the one may extensively pre- vail over the other is not well known, but probably the fluoric acid mica is most common in the inferior stratified rocks. As- suming this, for the sake of our inquiry, we may proceed. The mean of fifteen analyses of mica from various parts of the world, by Klaproth, Vauquelin, Rose, and Beudant, gives : Silica 46*14, alumina 26'16, potash 10'12, magnesia 4'99 V 440 Inferior Stratified Rocks. lime 0-35, peroxide of iron 8'17, oxide of manganese 0-61, fluoric acid 1*09, and water 2. Seven analyses of felspar by Klaproth, Vauquelin, Bucholz, Rose, Berthier, and Beudant, give, for the mean composition of that mineral, Silica 64-04, alumina 18-94, potash 13-66, lime 0-76, and oxide of iron 0'74. A gneiss, therefore, composed of equal parts of quartz, fel- spar, and mica, would contain, Silica 70-06 Alumina 15'03 Magnesia 1-66 Lime 0'37 Potash . 7-92 Oxide of iron 2-97 Oxide of manganese ... 0-20 Fluoric acid 0-36 Water 0-66 We should not forget, that instead of common felspar, albite sometimes enters into the composition of gneiss, and other of the inferior stratified rocks. The mean of four analyses of Albite from Finland, Fimbo, Arendal, and Chesterfield (United States), by Tengstrom, Eggertz, Rose, and Stromeyer, gives for the composition of that mineral, Silica 69*45, alu- mina 19-44, soda 9-95, lime 0-22, magnesia 0-13, and the oxides of iron and manganese 0*27. Hence a gneiss composed of equal parts of quartz, albite, and mica, would contain, Silica 71-86 Alumina 15-20 Potash 3-37 Soda .. 3-31 Magnesia 1'70 Lime 0'25 Fluoric acid 0-36 Oxides of iron and man- ganese 3-01 Water . 0-35 The composition of the gneiss with imbedded crystals of fel- spar, by no means an uncommon rock, would of course differ from the varieties of gneiss above noticed, in proportion to the abundance of such crystals. A mica slate, composed of equal parts of quartz and mica, would contain, Silica 73-07 Alumina 13-08 Magnesia 2-49 Lime 0-17 Potash . 5-06 Oxide of iron 4-08 Oxide of manganese ... 0-30 Fluoric acid 0-54 Water 1-00 A mica slate, composed of equal parts of quartz, mica, and garnet, by no means an uncommon mixture, would contain (taking the mean of several analyses of garnet by Vauquelin, Hisinger, and Wachtmeister, to be, Silica 39-69, alumina 20-19, protoxide of iron 35'99, protoxide of manganese 3'09, and lime T01), Inferior Stratified Hocks. 441 Silica 61-94 Alumina 15-45 Potash 3-37 Magnesia 1'66 Lime ., 0-45 jOxideofiron 1472 Oxide of manganese 1-23 Fluoric acid 0-36 Water , , 0-66 The chief difference in this compound from the gneiss first noticed, would consist in a less proportion of silica (8'12) and potash (4-55), and in a larger proportion of the oxides ofiron (11-75). Hornblende appears to differ much in the quantity of iron it contains. We shall probably commit no great error if we take Bonsdorff 's analysis of a hornblende from Pargas as affording a fair view of the substances contained in this mineral, more particularly as it approaches the mean of several analyses of hornblende from different places. The hornblende in ques- tion contained, silica 45'69, alumina 12-18, lime 13'83, magnesia 18*79, protoxide ofiron 7'32, protoxide of manga- nese 0-22, and fluoric acid 1*50. That variety of hornblende rock which is almost entirely composed of confused crystals of hornblende, will necessarily consist of little else than the constituent parts of the mineral. A variety of hornblende rock composed of equal parts of horn- blende and felspar, would contain, Silica 54-86 Alumina 15-56 Lime 7-29 Potash 6-83 Magnesia 9-39 Oxide of iron 4-03 Oxide of manganese ... 0-11 Fluoric acid .. 0-75 The mean of three analyses of chlorite by Vauquelin, Ber- thier, and Gruner, gives for the composition of that mineral, Silica 27*43, alumina 17*90, oxide and protoxide of iron 30-63, magnesia 14-56, potash 1-56, lime 0'50, and water 6' 92. Assuming this to be a fair estimate of the substances forming chlorite, a chlorite slate composed of equal parts of chlorite and quartz would contain, Silica 63-71 Ahunina 8'95 Magnesia 7-28 Potash 0-78 Oxide of iron 15-31 Lime 0-25 Water , 3-46 A talcose slate, composed of equal parts of quartz and talc, would contain, assuming the mean of two analyses of talc by Berthier from St. Bernard and St. Foix to afford a fair estimate of the constituent parts of this mineral (viz. silica 56-9, alu- mina 0-8, lime 4*0, magnesia 26*4, protoxide of iron 8'1, water 3'0), Silica 78-45 Magnesia 13-20 Oxide of iron 4-05 Lime 2-00 Alumina 0-40 Water 1-50 442 Inferior Stratified Rocks. Protogine is a rock which occurs extensively in the Alps, and differs only from gneiss in containing, as before stated, talc or steatite instead of mica. Talc and steatite do not differ materially in their chemical contents ; both probably occur in protogine, but as steatite generally prevails, we will suppose a compound of equal parts of quartz, felspar, and steatite. This protogine would contain (taking the mean of three ana- lyses of steatite by Vauquelin, Bucholz, and Brandes, at si- lica 61*68, magnesia 27*80, oxide and protoxide of iron 2-50, lime 0*25, alumina 0*83, and water 6*00), Silica 75-24 Alumina 6-59 Potash 4-55 Magnesia 9*26 Lime 0-33 Oxide of iron 1-08 Water 2-00 It is by no means easy to determine the chemical composi- tion of Eurite. If we consider it the same with the compact felspar rock of Dr. Macculloch, the analysis of compact fel- spar ought to afford us the requisite information. There would appear little doubt that substances not precisely the same pass under the name of compact felspar. The author last cited states that compact felspar contains both potash and soda at the same time. According to Bucholz, compact felspar from Passau is composed of Silica 60-00 Alumina 22-00 Potash . 14-00 Lime Loss 0-75 3-25 According to Klaproth, compact felspar from Siebenlen contains, Silica 51-00 Alumina 30-00 Soda 4-00 Lime 11-25 Oxide of iron 1-75 Loss 2-00 The petrosilex of some authors is sometimes also classed under the head of compact felspar. According to Berthier the petrosilex of Nantes is formed of Lime ... Silica 75-20 Alumina 15-00 Potash . , 3-40 1-20 Magnesia 2-40 Water 1-50 Upon the whole we can scarcely avoid agreeing with M. Beudant, who observes, that, although there are certainly vari- eties of compact felspar, many substances, which cannot be considered as felspars, are so called, often because it is not known what else to do with them*. Be this as it may, eurite is sometimes a mixture of the, so called, compact felspar with mica, and at others with quartz. Beudant, Traite de Mint-ralogie, 2me Edition, 1832, t. ii. p. 106. Inferior Stratified Rocks. 443 Quartz rock^ as its name implies, is, when pure, composed of little else than silica. When it is formed of equal parts of mica and quartz, it contains the same substances as has been noticed under the head of mica slate, which in fact it then is. Quartz rock, composed of equal parts of quartz and felspar, would contain, Silica 82-02 Alumina 9-47 Potash 6-83 Oxide of iron . 0-37 Lime 0-38 The limestones of this age are commonly saccharine, though varieties are not wanting which are merely compact. The carbonate of lime is not always pure, but often becomes mixed with carbonate of magnesia, constituting dolomite. As pre- viously observed, these rocks frequently contain disseminated minerals. There can be little doubt that the argillaceous slates vary materially in their chemical composition, though they present nearly the same external appearance. Silica and alumina ap- pear, however, to enter largely into their composition. The reader will have observed that silica constitutes the principal ingredient of all these rocks ; for the limestones and dolomites form such an insignificant part of the whole that they may be readily omitted ; indeed, the mass of the inferior strati- fied rocks must be composed of more than one half of this substance. Alumina is the next abundant substance; then follow potash, magnesia, and soda. Lime, though by no means scarce, occurs for the most part in very small quantities. Fluoric acid also occurs extensively in small proportions. The oxides of iron and manganese are also common, the former being the most abundant. It would thus appear that the enor- mous mass of matter constituting the inferior stratified rocks is essentially composed of a few simple substances, variously combined ; and that the greater proportion of them exists in the state of silicates. It must not be inferred, from the small space here dedicated to the inferior stratified rocks, that they are of little importance; for they are found to occupy a large portion of the earth's sur- face, wherever, from denudations and disruptions of strata, or from the original absence of superincumbent rocks, they are exposed to our observation. As whenever they are observed, whether in Asia, North America, or Europe, they appear with constant general characters, we may assume that common causes have produced them over the surface of the globe, and that these common causes are principally chemical, inasmuch as the prevalent mineralogical character of the mass is con- fusedly crystalline. We may therefore infer, from finding these rocks with con- 444- Inferior Stratified Rocks. slant general characters, whenever circumstances permit us to observe them emerging from beneath the mass of strata in which organic remains are entombed, that general chemical laws have been in operation contemporaneously over the sur- face of our planet, and previously to the existence of animal and vegetable life upon it, producing rocks of great collective thickness. Hence the student may always consider, that, whatever may be the nature of the deposits on which he stands, such strata exist beneath them, unless in cases where masses of igneous rocks have, by protrusion, forced them asunder, and left no stratified substances intermediate between the sur- face and the interior of the globe. It would be tedious to -enumerate the various situations where these rocks may be found ; it will suffice to state that there is scarcely any very large extent of country, where from some accident or other they are not exposed on the surface. They abound in Norway, Sweden, and Northern Russia; they are common in the North of Scotland, whence they stretch over into Ireland. In the Alps and some other mountains they occupy the central lines of elevation, as if brought to light bv the movements which have thrown up the different chains. They abound in the Brazils, and occur extensively in the United States. Our navigators have shown that they are suf- ficiently common in the various remote parts of North America visited by them. They occupy a considerable area in central India, and are found extensively in the great range of the Himalah. Ceylon is in a great measure composed of them ; and they do not appear to be scarce in various other parts of Asia. In Africa also we know that they are not wanting, though but so small a part of that continent has been yet ex- plored with scientific views. '!?,: Un stratified Rocks. 44-5 SECTION XL UNSTRATIFIED ROCKS. THE rocks constituting this natural group are widely distri- buted over the surface of the world, are found mixed with al- most all the stratified rocks, and bear every mark of having been ejected from beneath. They commonly occur either as protruded masses, as overlapping masses, resulting from the spread of matter after ejection, or as veinstones filling fissures, apparently consequent on some violence to which the strata have been subjected. The aspect of the unstratified rocks is exceedingly various as far as respects their texture, and the absence or presence of the few minerals which essentially enter into their composition. These variations would however in general appear the result of the circumstances to which they have been exposed; and not unfrequently the same mass, if of tolerable extent, will present a great variety of compounds, to which separate names might be (and indeed have been) assigned, if, instead of directing at- tention to the mass, the small changes in mineralogical struc- ture are alone observed. In the earlier days of geology, granite was considered the fundamental rock on which all others were accumulated ; but this opinion, like many others, has now given way before facts; for, as will be seen in the sequel, we have examples of granite resting upon stratified and fossiliferous rocks of no very great comparative antiquity. It must however be confessed, that granite appears sometimes to alternate in considerable thick- ness with the inferior stratified rocks, and that the separation of it from gneiss, particularly thick-bedded gneiss, is very am- biguous. Before, however, we proceed further with the con- sideration of the unstratified rocks, it will be necessary to pre- mise a sketch of their mineralogical characters, omitting those of the rocks usually termed volcanic, which have been already noticed. Granite Is a confusedly crystalline compound of quartz, felspar, mica, and hornblende. It is not essential that all these four minerals should be present; on the contrary, rocks have been termed granite when only felspar and mica, felspar and quartz, felspar and hornblende, and quartz and hornblende, have been the 446 Unstratified Rocks. constituent minerals. Such an employment of the term gra- nite must be used with much caution, as for instance in the case of the compound of felspar and hornblende, which in fact is mineralogical greenstone, and shoufd not be named granite unless it constitutes a very subordinate portion of a mass to which the term may be more properly applied, and results from the accidental absence of one or two of the above-named minerals for a limited space. The most prevalent compound is one with quartz, felspar, and mica; when hornblende re- places the mica, it is sometimes termed sienite. Other minerals, such as chlorite, talc, steatite, &c. are sometimes arranged with those above enumerated in various ways and proportions; but such compounds can only be considered as accidental varieties. When the quartz and felspar occur alone, and the crystalliza- tion is such that the former appears disseminated in the latter, it is termed graphic granite, from the supposed resemblance it bears to antique characters. Granite is occasionally porphy- ritic, as is the case in Cornwall and Devonshire, large crystals of felspar being disseminated through the mass, showing that however confused the general crystallization may have been, circumstances were such as to permit the production of distinct crystals of felspar. Diallage Hock (Euphotide, Haiiy ; Schiller/els, Germ.). Ser- pentine (Ophiolite, Al. Brong. ; Serpentin, Germ.). These are so intimately connected, that to separate them seems impossible, passing, as they sometimes do, in all direc- tions into each other. Diallage rock when pure is composed of diallage and felspar. Serpentine when pure is generally considered as a simple mineral substance, and forms large masses in that state, but seldom prevails to any extent without acquiring diallage. These rocks are sometimes blended with compounds of the greenstone class, and apparently pass so insensibly into them that they can only be considered as parts of a common mass, though the serpentine and diallage rock generally prevail in such cases. Greenstone (Griinstein, Germ.; Diabase, Al. Brong.), and the other Roc/cs usually termed Trappean. These also so pass one into the other, that frequently in a mass of inconsiderable extent a great variety may readily be obtained. They vary in texture from an apparently simple rock to a confusedly crystalline compound, in which crystals of felspar are disseminated. It has long since been observed by Dr. Macculloch that "the predominant substance in the members of this family is a simple rock, of which indurated Unstratified Rocks. 447 clay or wacke may be placed at one extreme, and compact fel- spar at the other ; the intermediate member being claystone and clinkstone. In some cases it forms the whole mass ; in others it is mixed with other minerals, in various proportions and in various manners; thus producing great diversities of aspect, without any material variations in the fundamental character*." As may be readily imagined, no exact definition can be given of that which is constantly changing in nature. Claystone, as its name implies, resembles clay under different degrees of induration ; and not unfrequently, when in mass, acquires a columnar structure. Clinkstone appears an inter- mediate step to compact felspar, which, according to Dr. Mac- culloch, contains both potash and soda, while common felspar contains potash only. I have elsewhere t applied the term cornean to designate some of the more simple forms of that kind of rock known as hornstone, which would appear in some cases to be nothing else than compact felspar ; in others, how- ever, it partakes of the characters of other minerals. Thus, in Pembrokeshire, where there is a remarkable variety of trap- pean rocks, the corneans may be divided into felspathic, quart- zose, and hornblendic, as those minerals appear to prevail in the mass ; the quartzose variety, which is the most rare, even appearing like some kinds of quartz rock, with the exception that it is unstratified. These more simple forms of trap-rock very frequently become porphyritic by the admixture of either quartz or felspar crystals, and sometimes of both in the same mass, as in the red quartziferous porphyries, rocks which not unfrequently pass into granite. Porphyries are generally known by the name of the base or paste which includes the disseminated crystals ; thus, we have claystone porphyry (Thonstein porphyr. Germ.; Argillophyre, Brongniart) ; fel- spathic porphyry ( True porphyry of Brongniart ; Porphyre Eu- ritique, Fr. ; Hornstein porphyr, Felspath porphyr, Germ.); and clinkstone porphyry (Klingstein porphyr). It very frequently appears as if the elements of quartz, fel- spar, and hornblende composed the mass, and various circum- stances determined their union in such a manner as to produce a large proportion of the various compounds known as trap- rocks, sometimes the hornblende being in mass, at others the felspar, while the quartz rarely predominates. In other situ- ations confusedly crystalline compounds have been the result; quartz, felspar, and hornblende united, form sienite; or felspar and hornblende without the quartz, constitute greenstone. The * Macculloch. Geological Classification of Rocks, 1821, p. 480. t Geology of Southern Pembrokeshire; Geol. Trans. 2nd Series, vol. ii. 4-4-8 Unstratified Rocks. granular structure of these compounds varies materially, and finally becomes somewhat imaginary; at least this texture is rather inferred than seen. The compounds occasionally con- tain disseminated crystals of felspar, and thus become what are commonly known as greenstone porphyries (Diabase por- phyroide, Fr. ; Gntnstein porphyr, Germ.). A paste of green hornblende cornean containing crystals of felspar constitutes the ophite of Brongniart, the antique green porphyry. Some of the rocks of this family are not unfrequently vesi- cular, in the manner of modern lavas, the vesicles however being generally filled up by some mineral substances which have since been infiltrated into them. Such substances are not unfrequently agates, and those employed in the arts are principally thus derived. From these cavities being frequently of an almond shape, or rather from the appearance of their solid contents resembling almonds in form, the term Amygda- loid, has been applied to rocks of this class. It will be readily understood that the base or paste of the amygdaloids is not constantly the same, but varies materially. A trappean rock is sometimes both amygdaloidal and porphyritic at the same time (Devonshire, Scotland, &c.). The amygdaloidal cavities afford the mineralogist a great abundance of siliceous, calca- reous, zeolitic, and other minerals. Other minerals than those above enumerated occur in the trappean rocks, but cannot be considered as forming an essen- tial part of them, with the exception of augite and hypersthene, which with the mixture of either common, compact, or glassy felspar, constitute the augite and hypersthene rocks of Dr. Mac- culloch. It would be endless to attempt a notice of the various aspects under which these rocks present themselves ; it should however be remarked that the term basalt is applied to sub- stances which are not precisely the same, being sometimes given to a fine compound of augite and compact felspar, at others to a minute mixture of hornblende and compact felspar, sometimes to dark indurated claystones, and finally to a com- pound of felspar, augite, and titaniferous iron. The last mix- ture seems that now most commonly termed basalt. Let us now consider how far some of the more marked of the unstratified rocks differ chemically from each other. A granite composed of equal parts of quartz, felspar, and mica, would contain precisely the same substances, and in the same proportion, as the gneiss previously noticed (p. 440.). Mica, however, rarely constitutes a third part of a large mass of granite; it is usually in smaller proportions. A granite com- posed of two fifths of quartz, two fifths of felspar, and one fifth of mica, would appear much more common. Such a rock Unstratified Rocks. 449 would contain (taking the composition of felspar and mica to be the same as noticed under the head of the inferior stratified rocks), Silica 74-84 Alumina 12-80 Potash 7-48 Magnesia 0-99 Lime 0-37 Oxide of iron 1*93 Oxide of manganese 0-12 Fluoric acid .. , 0-21 When granite contains disseminated crystals of felspar, by no means a rare circumstance, felspar may be considered as constituting at least one half of the compound. Assuming that in such a rock the felspar = -J, the quartz = ^, and the mica = ^, the chemical composition would be, Silica 73-04 Alumina l 13-83 Potash 8-51 Magnesia 0-83 Lime 0'44 Oxide of iron 1-73 Oxide of manganese 0-10 Fluoric acid .. 0-18 In this estimate of the contents of such a granite, we must not forget that the disseminated crystals are sometimes those of albite : if we assume that such disseminated crystals are = ^, the quartz = |, the common felspar = |, and the mica = , we should have for the chemical composition of such a rock (taking the constituent parts of albite to be the same as those noticed under the head of the inferior stratified rocks), Silica 73-94 Lime 0-35 Alumina 13-92 Potash 6-24 Soda 1-66 Magnesia 0-85 A granite composed of quartz, felspar, and hornblende, in that case usually termed sienite, would contain (supposing the minerals to be in equal parts, and the chemical constituents of hornblende to be as noticed under the head of the inferior stratified rocks), Oxide of iron 1-64 Oxide of manganese 0-11 Fluoric acid .. 0-18 Silica 69-91 Alumina 10-37 Potash 4-55 Lime... , 4-86 Magnesia 6-26 Oxide of iron 2-69 Oxide of manganese 0'07 Fluoric acid ........ 0-50 A granite composed of quartz, felspar, mica, and hornblende is by no means a very common variety ; it does, however, oc- casionally occur, even in considerable masses. Such a rock, supposing the four minerals to be in equal parts, would con- tain, Silica 63-96 Alumina 14-32 Potash , r >-94 Lime 3-73 Magnesia 5-94 Oxide of iron 4-06 Oxide of manganese 0-21 Fluoric acid 0'65 A granite formed of quartz and felspar in equal parts would 2 G 450 Vnstratijied Rocks. afford, upon analysis, the same result as the substance noticed under the head of quartz rock, and supposed to be similar!}' composed. A rock, formed of schorl and quartz, and hence termed schorl rock, is often found in granite districts ; and as it some- times constitutes important masses, should be noticed. Schorl is now generally admitted to be only one of the black varieties of tourmaline. These vary in their constituent parts, but agree in containing silica and alumina in larger propor- tions than any other integral substance, and in the presence of boracic acid. The mean of six analyses by Gmelin of black tourmalines from Rabenstein, Saint Gothard, Greenland, Bo- vey, Eibenstock, and Karingbrika, gives as the composition of this mineral, Silica 36*03, alumina 35'82, potash 0'71, soda T96, lime 0*28, magnesia 4*44, oxide of iron 13'71, oxide of manganese 1*62, and boracic acid 3-49. The relative proportion of the minerals in this rock, like all those now under consideration, varies materially ; but sup- posing, for the sake of our inquiry, that the quartz and schorl occur in equal parts, the rock would contain, Silica 68-01 Alumina 17-91 Potash 0-35 Soda 0-98 Lime 0-14 Magnesia 2-22 Oxide of iron 6-85 Oxide of manganese 0-81 Boracic acid .. 1-79 A greenstone, composed of equal parts of felspar and horn- blende, would present the. same results as the hornblende rock previously noticed, and similarly constituted. For the sake of comparison it may be useful to repeat the calculation, which is as follows : Silica 54-86 Alumina 15-56 Potash 6-83 Lime... 7'29 Magnesia 9-39 Oxide of iron 4-03 Oxide of manganese 0-11 Fluoric acid . 75 Kporphyritic greenstone, assuming the disseminated cry- stals of felspar to form one third of the mass, the confusedly crystallized felspar one third, and the hornblende one third, would contain, Silica 57-92 Alumina 16-69 Potash 9-10 Lime... .... 5-11 Magnesia 6-26 Oxide of iron 2-93 Oxide of manganese 0-07 Fluoric acid .. 0-50 The exact composition of hypersthene being little known, that of hypersthene rock, assuming any proportions of the con- stituent minerals which we may consider a fair average view of its chemical contents, can necessarily be but imperfectly calculated. The hypersthene from Labrador contains, accord- Unstratified Rocks. 451 ing to Klaproth, Silica 54'25, magnesia 14'00, oxide of iron 24-50, lime 1-50, alumina 2'25, water TOO. If we suppose the rock to be composed of equal parts of hypersthene, thus con- stituted, and common felspar, it would contain, Silica 59-14 Alumina 10-59 Potash 6-83 Lime .. 1-13 Magnesia 7*00 Oxide of iron 12-62 Water . 0-50 If we assume that the same rock is composed of equal parts of hypersthene and albite, it would contain, Silica 61-85 Alumina 10-84 Soda 4-97 Lime.., 0-86 Magnesia 7*06 Oxide of iron 12-32 Oxide of manganese 0-06 Water . 0-50 When quartz enters into the composition of hypersthene rock, the proportion of silica is necessarily increased, while that of the other constituent substances is diminished, accord- ing to the relative amount of quartz. From the want of good analyses, it becomes exceedingly difficult to arrive at any approximative estimate of the chemi- cal contents of a great variety of porphyries. In the first place, the base or paste in which the crystals are imbedded varies most materially; and in the second, the disseminated crystals, are those of different minerals, though they principally consist of common felspar, albite, hornblende, and augite. Some of the bases of porphyries have the character of that rock which is commonly termed compact felspar; and we might proceed to estimate the chemical contents of porphyries with this base, if it were clear that several substances, differing from each other in composition, were not known by the name of compact felspar. Porphyries, for the most part, seem to consist prin- cipally of variable mixtures of silica, alumina, potash, soda, lime, iron, and manganese. The imbedded crystals are most frequently those of either common felspar or albite; and, ac- cordingly, as the one or the other prevails in the mass, should we expect to find potash or soda most common. In like manner the chemical contents of those various trap- pean compounds, which I have elsewhere termed corneans, are exceedingly difficult to estimate. Silica certainly some- times prevails in them to a great extent ; so that, probably, if the circumstances under which they have been produced had been such as to allow of confused crystallization, the rock would be one in which quartz would form a principal consti- tuent mineral. That they are silicates of various substances of the same kind as those contained in granites, greenstones, and the like, can be scarcely doubted. Very frequently we 2 G 2 4-52 Unstratified Rocks. find circumstances have been such, in various parts of their mass, that crystallization has taken place to a greater or less extent; the products being porphyries of different kinds, green- stones, or sienites. Various dark coloured rocks of a close fine-grained texture having been termed basalts, there is again difficulty under this head. Certain of these basalts are but exceedingly fine-grained greenstones, and would therefore have the same chemical com- position. If, as it is supposed by some, true basalt is a com- pound of augite, felspar, and titaniferous iron, we can scarcely attempt a calculation, inasmuch as the analyses of minerals which have been termed augite differ much from each other ; and we scarcely know what proportions to assume as a fair estimate of the contained minerals. According to Phillips, a basalt from Saxony was composed of Silica 44-50 Alumina 16-75 Lime 9-50 Magnesia 2-25 Soda 2-60 Oxide of iron 20-00 Oxide of manganese 0-12 Water . 2-00 A basalt from Baulieu afforded, on analysis, to M. Beudant, Silica 59-5 Alumina 11-5 Peroxide of iron 0*5 Protoxide of iron .. 19-7 Lime 1-3 Soda '. 5-9 Potash . 1-6 Trachyte is again another igneous rock, the chemical com- position of which is most difficult to estimate. Silica certainly sometimes prevails more than at others. Both the potash and soda felspars apparently enter largely into it. Our difficulties are by no means diminished when we arrive at the lavas. They appear of a very variable composition. Still, however, the mass of the substances of which they are formed consists chiefly of silicates. Potash or soda commonly constitutes a portion of their general composition. The analyses of pitchstone from Newry, by Knox, and of the same rock of Meissen, by Dumenil, do not, as will be seen be- neath, differ very materially from each other ; and are remark- able not only for exhibiting this general similarity in their chemical contents, but also for showing that bitumen may enter somewhat largely into the composition of an igneous rock. Newry Meissen Pitchstone. Pitchstone. Silica 72-80 7,3-00 Alumina 11-50 10-84 Soda 2-86 1-48 Lime 1-12 1-14 Oxide of iron 3-03 1-90 Bituminous matter . 8-50 9-40 Unstratified Rocks. 4:53 It would be useless to estimate the contents of obsidian. It appears little else than the vitreous condition of various melted rocks, as is indeed shown by the analyses which have been made of it ; for these differ very considerably. The analyses of serpentine, though they vary as to the num- bers and proportions of the substances which enter into the composition of this rock, always afford silica, magnesia, and water as the principal constituents, as will readily be seen beneath (a. Serpentine from Germantown, by Nutall ; b. from Skyttgrufa, by Hisinger; c. by John*, from a place not men- tioned) : a. b. c. Silica 42-00 43-07 42-50 Magnesia 33-00 40-37 38-63 Alumina 0-00 0-25 1-00 Lime 3-50 0-50 0-25 Oxide of iron 7'00 1-17 1'50 Oxide of manganese... 0-00 0-00 1'62 Oxide of chrome 0-00 O'OO 1-25 Water 13*00 12-45 15-20 We may here remark, that, though chrome does not always constitute one of the ingredients of serpentine, chromate of iron is exceedingly common in it and some diallage rocks. The chrome of commerce, so extensively used in the preparation of certain paints, seems almost entirely to be thus derived. It is by no means easy to calculate the approximative com- position of diallage rock, from the variable nature both of the diallage and felspar of which it is composed. Minerals much resembling each other in appearance, but differing in the re- lative proportions of their constituent substances, have been named diallage. The felspar is sometimes compact, at others common, and is not unfrequently an albite. According to Berthier, the diallage from La Spezia contains, Silica 4-7% magnesia 24'4, lime 13*1, protoxide of iron 7'4-, alumina 3'7, and water 3 '2. Albite is frequently mixed with the diallage in the same country, forming diallage rock. Assuming that the rock is composed of two fifths of diallage and three fifths of albite, which is not an uncommon proportion in the diallage rock of La Spezia, we should have for the chemical contents of this rock, Lime 5-37 Oxides of iron .. 3-03 Silica 60-55 Alumina 13-14 Soda 5-97 Magnesia 9*83 Oxides of manganese ... 0-07 Water . 1-28 If, instead of albite, we take common felspar as mixed with diallage in diallage rock, in the proportion of two Mohs's Mineralogy, by Haidingcr. 454 Unstratified Rocks. thirds of the former mineral to one third of the latter, we obtain, Silica 58-42 Alumina 13'86 Potash 9-10 Lime ... 4-87 Magnesia 8'13 Oxide of iron 2-96 Water . 1-06 Though the foregoing calculations are merely approxima- tive and incomplete, they nevertheless afford data for many important inferences. To enter at length into this subject would not accord with the plan of the present work. It may however be remarked, that, like the inferior stratified rocks, the unstratified rocks are composed of a few simple substances, variously combined. The silicates of alumina, potash, soda, lime, arid magnesia, with the oxides of iron and manganese, constitute the chief ingredients of the rocks under considera- tion. In the granites without hornblende the silicates of alu- mina, potash, and soda prevail, but principally the two former. When hornblende enters into the composition of granite, the silicates of lime and magnesia are not without their importance, particularly when hornblende constitutes at least one third of the mass. The importance of the silicates of lime and mag- nesia is increased in the greenstones, and in a large portion of the basalts. The silicate of magnesia is in excess in the serpentines. The relative fusibility of the rock is necessarily determined by the relative proportions of the above noticed ingredients contained in it. Thus, when silica greatly prevails, it is re- fractory ; and the like seems to occur when silicate of mag- nesia is in excess (as takes place in the serpentines) without the presence of a sufficient quantity of any substance, such as silicate of lime, which may act as a flux. When silicate of lime is in tolerable quantity, the rock is then readily fusible. This takes place when either hornblende or augite are suffi- ciently abundant; indeed the augitic rocks, which contain a large amount of silicate of lime*, may be considered as the most fusible. Although a large proportion of the unstratified rocks may be considered as having been ejected directly from beneath all stratified rocks, it by no means follows that some of them do not consist of the stratified rocks fused and thrown up. In fact, the composition of some lavas would not be opposed to this inference. That the inferior stratified rocks could readily be converted, by fusion, into many of the unstratified rocks, is sufficiently apparent, their general composition being the same. The conversion of some of the fossiliferous rocks * The proportion of lime in the minerals known under the common name of augite, varies in the different analyses from 13 to 24 per cent. Unstratified Recks. 455 into lavas and substances of that character is not less simple. If we abstract a large proportion of carbon and lime, the fos- sil iferous rocks appear, in a great measure, derived from the degradation or chemical destruction of the inferior stratified or unstratified rocks. The unequal dispersion of the materials so derived will of course produce unequal accumulations of them ; so that a collection of grains of quartz, forming a sili- ceous sandstone, would be highly refractory. But suppose, as very often happens, that the grains of quartz are cemented by calcareous matter (no consequence whence derived), the com- pound would be exceedingly fusible. Pumice is clearly often nothing else than a schistose rock sufficiently heated to pro- duce the necessary vesicular character, the heat not being in- tense enough to cause fusion. The pumice of the volcanic district of the Rhine is probably the grauwacke schists of that country which have been thus circumstanced. Such are the rocks commonly considered unstratified. It will have been seen that they so pass into one another that distinctions are not easily established between them. A com- mon passage of diallage rock into greenstone, the reason for which does not at first sight seem apparent, will be seen, by comparing the foregoing calculations respecting these two rocks, to be produced by a very small difference in the propor- tions of the substances composing them. Mineralogicai granite passes through various stages, and graduates into the com- pounds named greenstone, and others of the trappean class*. * Dr. Hibbert notices the passage of granite into one of those compounds named basalt, (in this case formed of an intimate mixture of hornblende with a small proportion of felspar,) as taking place in the Shetland Islands. As this author's account is illustrative of such changes in general, it may ad- vantageously find a place here. The basalt extends from the Island of Mickle Voe northwards to Roeness Voe, a distance of twelve miles. On the west of this is a considerable mass of granite. The transition is thus described : " Not far from the junction we may find, dispersed through the basalt, very minute particles of quartz. This is the first indication of an approaching change in the nature of the rock. In again tracing it still nearer the granite, we find the particles of quartz dispersed through the ba- salt becoming still more distinct, more numerous, and larger, an increase of magnitude even extending to every other description of particles. The rock may now be observed to consist of separate ingredients of quartz, hornblende, felspar, and greenstone ; the latter substance (greenstone) being a homoge- neous commixture of hornblende and felspar. Again, as we approach still nearer the granite, the disseminated portions of greenstone disappear, their place being supplied by an additional quantity of felspar and quartz. The rock now consists of three ingredients, felspar, quartz, and hornblende. The last change which takes place results from the still increasing accumulation of quartz and felspar, and from the proportionate diminution of hornblende. The hornblende eventually disappears, and we have a well characterized granite, consisting of two ingredients of felspar and quartz." Hibbert, Brewster's Edin. Journal of Science, vol. i. p. 107. The same author also notices a passage of felspar porphyry into granite near Hillswick Ness. 4-56 Umtratified Rocks. Instead also of being solely mixed with rocks of the oldest date, it is found, among the Montagnes de 1'Oisans (Western Alps), cutting through and superincumbent upon deposits, referrible, according to M. Elie de Beaumont, to the oolitic series*. Observations of the same kind have been made by MM. Hugi and Studer on part of the Swiss Alps. The annexed figure is a section, by M. Hugi, of the Botzberg. Fig. 113. a, limestones and slates, referred to the lias ; b, mica slate ; c, gneiss ; d, granitic rock. Assuming the section to be correct, the super- position of the crystalline rocks, in this case, is evident. But it may have been produced in two ways : the rock (a) considered as lias may have been quietly deposited upon the mica-slate (&), which in its turn reposed on the gneiss (c), also resting on the granitic rock (d),and the whole may have been thrown over, a circum- stance not so rare as may be supposed in great mountain chains; or the granitic rock, supposing it to belong to the class now under consideration, may have been ejected from beneath, thus overflowing the fossiliferous beds, and altering those nearest to it. If we have recourse to the observations of M. Studer on this part of the Alps, we shall find that the former opinion is most probably the true one, at least in the greater number of cases. The annexed figure represents his section of the Jungfrau. a, limestones and slates, referred to rocks of the oo- Fig. 1 1 4-. litic group; b b, gneiss. This Jungfrau. gneiss is described as com- posed of talcose mica, white or brown felspar, with little quartz ; and is stated to os- cillate between gneiss and granite, appearing on the large scale to belong rather to the former than the latter. In fact it seems to belong to that great system of gneiss with steatite, talc, or talcose mica, constituting a large por- tion of the central range of the Alps, and not (infrequently known as Protogine. As a mass, the rocks equivalent to those of the oolitic group, rest unconformably on the mica slate and * Elie de Beaumont, sur les Montagnes de 1'Oisans ; Mem. de la Sor. d'Hist. Nat. de Paris, torn. v. ; as also Sections and Views illustrative of Geo- logical Phenomena, pi. 15. Roththal. Umtratified Rocks. 457 gneiss of the Alps, though of course there must be numerous sections where the planes of the beds of the two rocks are pa- rallel to each other. Such being the case, a deposit of lime- stones and slates on an unequal surface of gneiss would neces- sarily mould itself on the latter, filling up the cavities, as often happens with other rocks similarly circumstanced. It hence follows that if the gneiss be upraised by a force acting along the central range of the Alps, the limestones and slates would be tilted up at the same time, so that the whole, when viewed as it now appears, would have the aspect of gneiss and fossili- ferous limestones reciprocally protruded into each other. If the reader will place the section of the Jungfrau before him, in such a manner that the limestone and slates become hori- zontal, he will perceive how easily they may have been moulded on the gneiss. If another very interesting section by M. Stu- der, representing five horizontal and reciprocal protrusions of gneiss and limestones into each other at the Gstellihorn, be also so turned that the gneiss is beneath and the limestone above, the former rise in peaks into the latter. We must here re- mark the uncertainty of many sections of this kind, however clearly the lines of separation may be exhibited on the face of a huge precipice, as this is; for it is well known to all accus- tomed to examine disturbed districts, where unconformable rocks have been tilted up together, that natural lines of section often cause a large mass of inferior rock to appear included in the superior beds, when in fact such appearance is entirely deceptive. This arises from a portion of the older rock pro- jecting into the newer rock, having been accidentally cut through in the line of section. If fortunate sections were made through the upper uneven surface of the chalk in many parts of England and France, where this rock projects in pinnacles and ridges into the plastic clay, or beds of that character, we should have such deceptive appearances ; and if both rocks were tilted up at right angles to their former position, the result would be a section not differing in appearance, though greatly in magnitude, from the Alpine sections above noticed. M. Studer considers these appearances to have arisen from the breaks and contortions of the limestones and gneiss when the whole was upraised. This, which is also a probable expla- nation of the phenomena, might be ascertained by a careful examination of the relative stratification of the two rocks*. While, however, we take this view of the subject, we must not neglect facts which may be thought to support the hypo- thesis, that the crystalline rocks have been protruded in a * Studer, Bulletin de la Societe Geologique dc France, t. ii. p. 53. pi. 1. figs. 2 and 3. 4-58 Unstratijied Rocks. heated state among the^calcareous strata. The calcareous beds intermingled with the gneiss of the Tossenhorn are observed to bedolomitic,and in certain cases crystalline*. The beds which rest on the gneiss at the Jungfrau occur in the following as- cending order: 1. compact dolomite, 30 feet; 2. quartz rock, associated with variegated argillaceous schist, 15 feet; 3. oolitic iron ore; 4. limestone, generally black, gray, and schistose. The limestone and iron ore contain numerous organic remains, considered as referrible to those of the oolitic group. It might be supposed that the lowest of the calcareous beds were to a certain extent altered, but this can scarcely be considered as a sufficient explanation. Moreover, M. Studer remarks that in the Roththal, where the gneiss covers the limestone series, there is, though the contact of the two rocks is well exposed, no appearance whatever of their having produced any effect on each other. Whether or not granite is to be considered as having been intruded among the limestones of the Bernese Alps, it has al- ready been seen that it covers the chalk of Weinbohla, whence it may be inferred that granite was produced at the supracre- taceous epoch. Assuming therefore that the evidence is good, we should expect to find granitic rocks traversing or superin- cumbent upon beds of all ages, from the inferior stratified to the cretaceous inclusive. The superposition of granitic rocks to fossil iferous limestone has long since been remarked by Von Buch in Norway, and by Dr. Macculloch in the Isle of Sky. Similar rocks have also been noticed as incumbent on those of the age of either the oolitic or cretaceous series at Predazzo. Respecting the latter, Sir J. Herschel observes, that where the dolomite plunges beneath the granitic rock at Canzocoli, at an angle of 50 to 60, both rocks appear altered, and that there are laminae of serpentine between the tvvof. The contact of the granite with the oolitic rocks of Brora is attributed by Prof. Sedgwick and Mr. Murchison to the elevation of the granite in mass, which is supposed to have turned up the edges of the oolitic deposit;):. The same au- thors have also remarked a very curious occurrence of granite and limestone on the north coast of Caithness, near Sandside, where the granite appears thrust up among the limestones, and a breccia has been produced containing fragments of limestone and granite. The cement of this breccia is described as gene- rally granitic, though it is calcareous in some places, and ap- proaches a sandstone in others. One great block of limestone * M. von Declien states that the limestone of these protruding wedges contains laminae of talc. German Transl. of Manual, p. 552. f Herschel, Edinburgh Journal of Science, vol. iii. J Sedgwick and Murchison, Geol. Trans, vol. ii. pi. 34. Unstratified Rocks. 459 is noticed as apparently entangled in the granite. The lime- stone beds on the eastern side are stated not to be much dis- turbed, while those on the western side are in the utmost con- fusion, and, which are important circumstances, crystalline and cellular*. Thus far we have only seen granite rising through and covering other rocks in considerable masses; but we have also evidence in granite veins, that the matter of the rock was in such a state of fusion, as to penetrate into thin clefts opened in stratified and older rocks by some violence, such as probably resulted from the upburst of the igneous matter ac- companied by elastic vapours. If we imagine fractures to be suddenly produced in contact with a mass of rock in fusion, such as we may presume granite to have been, the natural re- sult would be the injection of the substance in fusion into all the crevices, in consequence of the great pressure exerted on one side; the intruding substance breaking off and including in it all loose fragments, and those projecting portions which opposed the fury of the injection. This is precisely the con- dition of granite veins, which, though much doubted during the reign of the Wernerian theory, are now known to be abundant in nature. Glen Tilt, which is reported to have produced such delight in Hutton when viewed by him for the first time, presents ex- cellent examples of the intrusion of granite veins into other and stratified rocks. The great features consist of a mass of gra- nite on the northern side of the glen, and of schist and lime- stone on the southern ; from the former, veins issue in all di- rections, disturbing and intermingling with the latter in such a complicated manner, as to render a description useless without the aid of maps and sections, for which, and for a detail of the various singular phenomena observable in Glen Tilt, the stu- dent must be referred to the memoirs of Lord Webb Seymour, Prof. Playfairf, and Dr. MaccullochJ. Granite veins traversing the stratified rocks are now known in various parts of the world. Some fine examples are to be observed in the district of the Land's End ; among them one at Cape Cornwall shows that there has been a shift or fault in the slate rocks, for a quartz vein has been cut through, and ele- vated more on one side than the other, thus proving that force has been employed . At Mousehole the veins can be seen to * Sedgwick and Murchison, Geol. Trans, vol. iii. p. 132. f Trans, of Royal Soc. of Edinburgh, vol. vii. + Geol. Transactions, 1st Series, vol. iii. Oeynhauscn and Von Dechen, Phil. Mag. and Annals of Philosophy, 1 829 ; also Sections and Views illustrative of Geological Phenomena, pi. 17. fig. L 4-60 Unstralified Rocks. proceed from the main body of the granite *. In the Alps they also proceed from masses of granite, which appear to have much influenced the present position of strata in parts of those mountains, as has been shown by M. Necker de Saussuref. They traverse gneiss in the Vallee de Vallorsine, as also at the head of the lake of Como. It is curious that in the Hartz, once considered as beautifully illustrating the Wernerian the- ory of granite, masses of granite cut through the direction of the clay slate and grauwacke nearly at right angles, sending off veins. At the Rosstrappe and at the Rehberg Graben, on the south side of the Brocken, granite veins clearly run into the clay slate and grauwacke. Few circumstances can more clearly prove how a leading theory may pervert the judge- ment, and thus cause the misrepresentation of facts. Ac- cording to M. Dufrenoy, the granite veins in the vicinity of St. Paul de Fenouillet (Pyrenees) occur in limestones refer- rible to the age of the cretaceous group. These limestones become more and more crystalline as they approach the gra- nitic masses^. Granite veins are not confined to Europe, but are found cut- ting and including portions of slate rocks at the Cape of Good Hope, as has been shown by Captain Basil Hall and Dr. Clarke Abel^. In America also they have been observed by Mr. Hitchcock traversing mica slate, hornblende slate, limestone (described as of a peculiar character), gneiss, and granite in Connecticut ; the veins frequently branching out in various di- rections || . Granite veins therefore cannot be considered as rare ; on the contrary, they would appear sufficiently common when circumstances permit good sections of the junctions of the granitic mass, and of the rocks among which they appear intruded. We should expect these veins to be of various dates, and accordingly we find that masses of granite are themselves traversed by veins, also of granite. The exact composition of the granite in these veins, must naturally vary, depending much on local circumstances; for if we suppose a substance in fusion to be injected into fissures of rocks, such injected matter will be subjected to different conditions. Where the fused substance cooled more suddenly, as was likely to be the case in the distant and smaller fissures, * Sections and Views illustrative of Geological Phenomena, fig. 5. f Necker de Saussure, sur le Vallee de Vallorsine; Mem. de la Soc. de Physique et d'Hist. Nat. de Geneve. J Dufr6noy, Bull, de la Soc. Geol. de France, t. ii. p. 71. Basil Hall, Transactions of the Royal Soc. of Edinburgh, vol. vii. ; and Clarke Abel's Voyage to China. || Hitchcock, On the Geology of Connecticut, American Journal of Science, vol. vi. Unstratified Rocks. 46 1 the result would be less crystalline ; while in the wider clefts, and near the great heated mass, the crystallization would be more perfect, and bear the greatest resemblance to the parent mass. Consequently, in a system of granite veins we should expect a great diversity in the aspect of the granitic matter, which generally appears to be the case. The trappean rocks, though there is much difficulty in se- parating them from the granitic, may for convenience be con- sidered separately from them. They also form considerable masses, and constitute dykes and veins. When considered in the mass, they may be regarded as containing much less mica than the granitic rocks, while hornblende has become much more abundant ; they also, when viewed on the large scale, appear more abundantly among the comparatively modern deposits than the granites, though it cannot be denied that they run into the latter in a remarkable manner. If this opinion of the greater prevalence of the granitic rocks over the trappean at the earliest periods be correct, it would seem to point to a certain condition of things at such periods, which subsequently became so modified that the igneous eruptions became altered. Of what that condition of things may have been, we do not as yet appear to have any very definite ideas, and we obtain little help on the subject from the phenomena of modern volcanos, gra- nite never having been known to flow from them. We however learn from this circumstance that igneous eruptions into the atmosphere are not favourable to the production of granites; and we may consequently infer, that the conditions under which granite was produced were not similar to those which we now observe on the surface of the earth, at least so far as relates to those phenomena which occur in the atmosphere. We do not exactly see why a difference of pressure beneath water, or any cause of that kind, should render mica less abundant, or increase the quantity of hornblende ; and therefore we may infer that there was something in the then condition of our planet's surface, which permitted the production of that great abundance of granite so commonly associated with the earliest stratified rocks with which we are acquainted, and which fre- quently differ from it only in being stratified, or having the component minerals arranged in lamina?. Admitting this prevalence of granitic compounds at the ear- liest periods, their production at more recent epochs shows that the conditions necessary for their formation continued up to such epochs, though they may have been infinitely more rare, having in a great measure given place to those under which the more common trappean rocks were produced. Trappean rocks, under their various modifications, are so common in nature, that to attempt a notice of localities would 462 Umtratified Eocks. be entirely useless. They occur mingled with the stratified rocks in every possible way, injected among the beds for considerable distances, so that sections are exhibited in which the igneous rock appears quietly interstratified with the aque- ous deposits; constituting caps of hills, thus appearing like a stratified and quiet deposit on other beds, the continuous parts that once connected these caps into a sheet of matter, ejected from beneath, having been removed by denudation; or as dykes or veins filling fissures, previously produced, in some instances showing that the igneous matter entered the fissure with such force as to tear away portions of the sides, while at others it seems to have risen more slowly, gradually filling the rent. There would appear no more convenient place for observing all these modes of occurrence, or indeed the various mineral aspects of the rocks themselves, than the coasts and islands of Scotland, which have been described by Dr. Macculloch* and other geologists, and where the student possesses the great ad- vantage of innumerable coast sections, those invaluable aids in all geological investigations. Apparent interstratifications of igneous rocks with beds which have had a different origin may be observed in many places, but may be well studied in High Teesdale, where the igneous matter has been injected among strata of limestone, sandstone, and shale, forming part of the carboniferous lime- stone series, in such a manner that a great apparent bed, com- monly known as the Great Whin Sill, was considered as con- stituting a regularly stratified portion of a common whole, be- fore the investigations of Prof. Sedgwick showed that it had evidently been injected among the aqueous deposits, and was connected with a mass of igneous matter which had disturbed and altered a continuation of the same rocks f. In Derbyshire, trappean rocks, generally known by the provincial term toad- stones, from the aspect of the prevailing amygdaloid, are apparently interstratified with the carboniferous limestone. These we may, from all analogy, consider as injected among the limestones, the strata of which would easily be separated by the application of the proper force, in the manner already noticed under the head of Volcanic Rocks (p. 140). The stu- dent will find ample details of this association of trap-rocks and limestones in Mr. Conybeare's account of the rocks of Derby shire J. * Macculloch's Western Islands. f Sedgwick, Trans, of the Cambridge Phil. Soc. vol. ii. p. 139 ; and Sec- tions and Views illustrative of Geological Phenomena, pi. 13. In some situ- ations the limestone and slate have been turned up by the trap, and the former has become granular, and the latter indurated. * Outlines of the Geology of England and Wales, Book III. chap. v. Umtratified Rocks. 463 According to Mr. Aikin, a good example of the apparent interstratification of greenstone with the coal-measures is ob- servable at Birch Hill colliery, Staffordshire. The bed seems to be connected with a mass of trap on one side, whence it has been injected among the coal strata, altering the coal where it covers it, by depriving it of its bitumen*. The connexion of trap rocks with coal-measures has often been insisted on, and certainly in some countries they are much associated ; but when the facts are narrowly examined, it ge- nerally appears that the igneous rocks have been introduced among the sandstones, shales, and coal, subsequently to the deposit, and even consolidation, of the latter. It does not however necessarily follow that igneous eruptions and coal de- posits may not have been contemporaneous ; on the contrary, we may inquire if the violent movements of land, which pro- bably accompanied such igneous eruptions, did not assist in the destruction of the vegetation, by depressing it beneath the waters; and even if accompanying and violent agitations of the atmosphere, similar to that which happened during the great eruption of Sumbawa, might not also contribute something to- wards the transport of the different parts of plants in parti- cular cases f. Difficulties very frequently arise from the want of good natural sections ; for it is clear that a mass of injected trap may be so spread over, or injected among, the coal strata, in a district wanting such sections, and only explored by means of the miner's galleries, that ambiguous appearances abound, more particularly when the whole mass has been traversed by faults, as is often the case. Among the various trap dykes noticed by Mr. Winch as traversing the coal-measures in the vicinity of Newcastle, there is one described by Mr. Hill as occurring at Walker colliery, which has converted the coal contiguous to it into coke. This dyke is stated not to alter the level of the coal-measures, though it cuts through them; but in the plan which accompanies the memoir there is a fault marked on the south side of the dyke, parallel to it, and on the eastern part, producing a dislocation to the amount of nine feet, so that the fracture does not appear to have been quite simple J. Trap dykes are to be found in all parts of the world, the * Aikin, Geol. Trans, vol. iii. In the section which illustrates this me- moir, a fault is seen to have traversed the beds after the injection of the trap, for it is dislocated with the rest ; a fact also observable in Prof. Sedgwick's sections of High Teesdale, where dislocations are represented to have affected all the rocks equally. f During tropical hurricanes among islands, such as those in the West Indies, plants, more particularly their lighter parts, are carried abundantly out to sea. I Geol. Trans. 1st Series, vol. iv. Unstratified Rocks. composition of the rock varying materially, even in the dyke itself, as we might expect from differences in the cooling and pressure, so that the central parts are not unfrequently more crystalline than the sides *. There is good evidence of the great mechanical force which has been exerted on the stratified rocks, as has been already pointed out in the North of Ireland, where large disrupted masses have been caught up in the igneous matter ; similar phaenomena have also been noticed by Dr. Macculloch and Mr. Murchison in the Western Islands of Scotland, the dis- rupted and included rocks being in the latter cases of an oldet date than those fractured in Northern Ireland f. The annexed figure will illus- ^. trate a considerable fracture and alteration in the limestones at the Black Head, Babbacombe Bay, Devon, effected by the eruption of greenstone, which, though it overlies the limestones in this section, occurs beneath them not a b f c e f b far distant. , argillo-calcareous slate, traversed by veins of calcareous spar, and occasionally indurated, b b, limestones which have become semi-crystalline; they have also, judging from lines of colour, been once more fissile than at present. , a slate, with a thin bed of reddish limestone (e). This slate is apparently much altered, d d> greenstone and its varieties, constituting the mass of the hill, and traversed by calcareous veins near the limestones, fj, lines of fracture which divide the limestones and slates into three masses. The slates and limestones have evidently suffered, not only from the mecha- nical action of the erupted greenstone, but also chemically * One of the longest dykes with which we are acquainted is that described by Prof. Sedgwick in his memoir on those of Yorkshire and Durham. It is very probably continued from "High Teesdale to the confines of the eastern coast, a distance of more than sixty miles." During this traverse it cuts the coal-measures, red sandstone, and lias. Cambridge Phil. Trans., vol. ii. p. 31 ; and Sections and Views illustrative of Geological Phaenomena, pi. 14. Archdeacon Verschoyle notices many east and west trap dykes in the counties of Sligo and Mayo, one of which can be traced for sixty or seventy miles, and probably extends to greater length. These dykes cut through all the stratified rocks in that part of Ireland, from the gneiss to the carbonife- rous limestone inclusive. Proceedings of the Geol. Soc., Nov. 21, 1832. f Macculloch, On the Western Islands of Scotland. Several of the sec- tions contained in this work are copied into Sections and Views illustrative of Geological Phaenomena, for the purpose of exhibiting the various modes in which trappean are associated with stratified rocks in the Hebrides. Mr. Murchison has represented fragments of the oolitic series of the same islands as caught up in the trap of the southern side of Mull. Geol. Trans. 2nd Series, vol. ii. pi. 35. Unstratificd Rocks. . 465 from the proximity of the mass in a state of igneous fusion. Notwithstanding the general pressure preventing the disen- gagement of the carbonic acid contained in the limestones, some elementary substances have been given off, and filled crevices and cracks in the trappean mass above with carbonate of lime. The alterations of limestone in contact with trappean rocks is sufficiently common, producing a greater or less amount of crystallization, in accordance with the well-known experiments of Sir James Hall *, who has proved that carbo- nate of lime when subjected to great heat beneath sufficient pressure, does not part with its carbonic acid, but that it is fused and is rendered crystalline, a fact previously doubted. The alteration of limestones and trap at their contact would not always appear to be confined to a crystalline arrangement in the parts of the former ; for Dr. Macculloch has observed trap to become changed into serpentine at the point of contact with limestone, at Clunie in Perthshire. A trap vein traverses limestone, and the rock is noticed as a kind of greenstone, the greater part of which is moderately coarse, passing into lamel- lar towards the sides. This (lamellar) structure "gradually becomes more distinct towards the edges of the vein, where it frequently splits off by the process of decomposition into la- minae, resembling, on a cursory view, black slate. These laminas are often intersected by other cross fissures, dividing the whole into cuboidal masses, which sometimes decompose still further into spheroidal forms, during the approximation to the calcareous boundary ; and whether the laminae are pre- sent or not, the texture becomes gradually finer and softer, the rock still retaining its black colour, or sometimes assuming a greenish cast. At length the observer finds the vein converted under his hand into serpentine, without being at first aware that any change has been brought about-j-." Thus the transi- tion from greenstone to serpentine can gradually be traced, but only where the vein traverses the limestone ; for where the continuation of the same vein cuts through schist and con- glomerate, no such change is observable. The trap is much entangled, in the small scale, with the limestone, and the mi- nuter ramifications from the vein are entirely composed of ser- pentine. The limestone does not pass into the serpentine ; on the contrary, the line of separation is described as well defined. Veins of green asbestos and steatite occur in the serpentine. Dr. Macculloch also states that the trap veins traversing the calcareous sandstone at Strathaird abound in steatite, tound in the outer parts of the vein, and approaching the calcareous * Sir James Hall, Trans, of the Royal Soc. of Edinburgh, vol. vi. f Macculloch, Brewsters Edin. Journal of Science, vol. i. p. 1. 2H 466 Unstratified Rocks. rock. The same author observes, that the trap vein which traverses the white marble of Strath passes into serpentine at its outer edges, as at Clunie. " At the line of contact, a zone of transparent serpentine of a fine oil-green colour, is found intermixed with the limestone *." The above is sufficient to show that trap under certain con- ditions may pass into serpentine. We have now to consider dykes and masses of serpentine and diallage rock which occur under circumstances analogous to those of the trap rocks. Mr. Lyell has described a serpentine dyke which cuts through a sandstone (equivalent either to grauwacke or the old red sand- stone), near West Balloch Farm, in Forfarshire. The phseno- mena can be well observed where the dyke traverses the Carity. The serpentine dyke is ninety feet thick, nearly vertical, and ranges east and west. It is stated to be flanked in part by a hard compact rock, about three yards thick, standing vertically and forming a parting wall between the sandstone and the ser- pentine. "This rock consists of equal parts of green serpen- tine and an indurated brick-coloured rock, harder than ser- pentine, and sometimes passing into jasper." The serpentine is also described as bounded, on the left bank of the Carity, by " a vertical mass of sandstone conglomerate, evidently much altered, about five yards thick. Some parts of this rock ap- proach jasper in hardness and appearance." But the most interesting fact connected with this altered conglomerate is, that the quartz pebbles contained in it have been fractured and reunited, a circumstance also noticed by Mr. Lyell in a con- glomerate flanking a greenstone dyke on the Isla, also in For- farshire. This fracture of the quartz pebbles is precisely what we should expect from a sudden application of heat, and would speak strongly in favour of the once igneous fusion of the ser- pentine in the dyke, if any evidence were wanting. That com- mon association of serpentine with greenstone is here also ob- servable, the dyke being bordered on the right side of the Ca- rity by a fine-grained rock of that kind. The dyke can be traced at intervals for at least fourteen miles, stretching in a straight line from Cortachie to Banfff. The serpentine and diallage rocks of Liguria are particularly instructive, as they occur under a variety of forms, and appear connected with the disturbed condition of the strata in that country. These rocks pass into each other in all directions, and into those of a trappean character (Levanto). Between Braco and Matanara the student may observe them with perfect ease, on the high road from Genoa to Florence, cut- * Macculloch, Brewster's Edin. Journ. of Science, vol. i. f Lyell, Brewster's Edin. Journ. of Science, vol. iii. Unstratified Rocks. 467 ting through the limestone and slate as a dyke, insinuated be- tween the strata, so as to seem interstratified, and constituting an enormous mass, apparently thrust upwards. The whole district is full of interesting facts of this nature. If it has been correctly determined that the limestones of La Spezia are of the age of the oolitic groups of England, France, and Germany, the serpentine and diallage rocks of southern Liguria have been erupted since that period ; for the La Spezia limestones and their associated deposits have been upheaved, contorted, and cut through by them. Possibly also the date of their intrusion may even be later, and of the supra- cretaceous period, for the lignite deposits of Caniparola, near Sarzana, are thrown into a vertical position, and I did not de- tect any serpentine or diallage rock pebbles among their asso- ciated conglomerates ; this latter date, however, must be con- sidered uncertain, for the serpentine rocks are not observed to be actually intruded among the supracretaceous deposits. At Capo Mesco, between Levanto and Monte Rosso, gray schist and a compact calcareo-siliceous sandstone (one of the Italian macignos) are broken into faults by a mass of serpen- tine and diallage rock, which branches from a larger mass at Levanto. The valley of Rochetta, near Borghetto, has at- tracted much attention since it was noticed by M. Brongniart*. It shows the intrusion of the serpentine and diallage rocks (which here also pass into each other in various ways) among stratified rocks, similar to those which are observed at Capo Mesco. At the entrance into the valley, the sandstone is seen dipping at a considerable angle and resting upon gray lime- stone and schist, which are supported by serpentine. The ser- pentine then passes over contorted gray limestone and schist, and occupies a considerable portion of the valley, mixed with diallage rock, until the latter predominating to the exclusion of the serpentine, the mass rests on beds of red and green jas- per, having the same dip as the sandstones at the entrance of the valley. These jasper beds repose on contorted gray lime- stone and schist, on the left bank of the river and opposite Rochetta. The jasper beds have sometimes been considered as a subordinate part of the serpentine: that it may be an al- tered rock is very possible ; but I do not imagine it can be re- garded as a portion of the un stratified mass of the diallage rock and serpentine, more particularly as similar jaspers occur among the limestones in the gulf of La Spezia, not far from Lerici, interstratified with them and distant from either ser- pentine or diallage rock. The mass of serpentine and diallage rock which constitutes * Annales des Mines, 1821. 2 H2 468 Unstratijied Rocks. the Monte Ferrato, north of Prato, in Tuscany, rests also upon stratified jasper on the west, and this again upon a schistose rock based on limestone: this also appears an accidental cir- cumstance, for jasper is interstratified with brown. shale at Paciana on the opposite side of the mountain, where it is not in contact with the serpentine. The diallage rock and ser- pentine here also pass into each other in all directions, and one variety of the former is worked for millstones. The whole seems a mass ejected from beneath, which has overflowed the stratified rocks, appearing to cut through them on the north- ward, beyond the north-west knoll, where there is a good sec- tion of the serpentinous mass resting on the jaspers, slates, and limestones. At the Lizard, Cornwall, there is a well known mass of ser- pentine, which seems intimately connected with greenstones : unfortunately, however, from its position, we do not obtain any clear idea of its relative date*. The volcanic rocks, at least such as have been considered the products of what are commonly termed modern and ex- tinct volcanos, have already been noticed ; therefore a state- ment of their general characters need not be repeated. If we regard these various igneous products as a mass of matter which has successively, and during the lapse of all that time comprehended between the earliest formation of the stra- tified rocks and the present day, been ejected from the interior of the earth, we shall be struck with certain differences of these rocks on the great scale, which has led to their practical ar- rangement under the heads of granitic, trappean, serpentinous, and volcanic products, as above noticed. The two former and the last occur most abundantly, whilst the third is comparatively more scarce, though sufficiently common in nature. It has been generally considered that the mineralogical cha- racter of igneous rocks has changed during the deposit of the stratified rocks, through which they have more or less forced their way ; that is, we do not find granite and serpentine flow- ing from modern volcanos, nor trachite nor leucitic lavas inti- mately associated with the oldest strata in such a manner, that their relative differences of age could not be very considerable. Admitting that true mineralogical granite may be reckoned among the products of the supracretaceous period, the mass of granite is associated with the oldest rocks, even omitting all consideration of the gneiss, composed of the same minerals, and probably of exactly the same amount of elementary sub- stances. The same with those igneous compounds into which * For descriptions of this district, consult the memoirs of Prof. Sedgwick, Cambridge Phil. Trans, vol. i. ; of Mr. Magendie, Trans. Geol. Soc. of Corn- wall, vol. i. ; and of Mr. Rogers, same work, vol. ii. Unstrdtified Rocks. 469 augite largely enters, which abound in the more recent pro- ducts, while certainly they are scarce, if they be not altogether absent, among the older rocks of an igneous origin ; and we have no stratified rocks of similar mineralogical composition constituting extensive districts, as is the case with gneiss. We are compelled therefore to admit, that the conditions, under which the two kinds of igneous rocks have been formed, have not been the same. What those conditions may have been is a separate question, and one, as above noticed, requiring in- vestigation ; but it will be at once obvious, that the ejection of a mass, in a state of fusion, into the atmosphere, would be likely to have its constituent parts arranged in a different manner from those in a similar mass forced out beneath great pressure, such as we may consider to exist beneath deep seas or a great mass of superincumbent rock. Independently, however, of this consideration, there appears to have been something in the condition of the world at the earliest times, causing certain compounds to be formed in great abundance, which does not now continue in such force as to permit the production of similar compounds. We cannot conclude this sketch of the unstratified rocks without adverting to the concretionary and columnar structure which they frequently assume. The most familiar examples of the columnar structure are those of the basalt in the Giant's Causeway and at Staffa, in the latter place constituting the sides of the justly celebrated FingaPs Cave *. The concretionary or globular structure is often visible in the decomposition of trap- pean and volcanic rocks, and is remarkable in a solid rock named the orbicular granite of Corsica (diorite orbiculaire, Al. Brong.), in which balls or spheroids of concentric and alternate coats of hornblende and compact felspar are disse- minated in the mass of the rock. We are indebted to Mr. Gregory Watt for our first great advance towards a knowledge of the circumstances which have produced this structure. This author fused seven hundred weight of an amorphous basalt named Rowley Rag, described as fine-grained and of a confused crystalline texture ; the fire was maintained for more than six hours, and the fused mass was suffered to cool very gradually, so that eight days elapsed before it was removed from the furnace. The fused mass was then three feet and a half long, two feet and a half wide, about four inches thick at one end, and above eighteen inches at the other. This irregularity of form, resulting from the shape of the furnace, was highly advantageous, showing the arrange- * Sec Macculloch's Western Islands of Scotland; and Sections and Views illustrative of Geological Phenomena, pi. 11. and 19. 470 Unstratifad Rocks. ment of the bodies passing from a vitreous to a stony state. A portion taken out while the basalt was in fusion became perfect glass. The most important result observed was the formation of spheroids, sometimes extending to a diameter of two inches. They were radiated with distinct fibres, the latter also forming concentric coats, when circumstances were only favourable to such an arrangement; but this structure gradually disappeared when the temperature was sufficiently continued, the centres of most of the spheroids becoming compact before they at- tained the diameter of half an inch. This structure gradually pervaded the whole body of the spheroid. " A continuation of the temperature favourable to arrangement speedily induces another change. The texture of the mass becomes more gra- nular, its colour rather more gray, and the brilliant points larger and more numerous; nor is it long before these brilliant molecules arrange themselves in regular forms, and finally the whole mass becomes pervaded by thin crystalline laminae which intersect it in every direction, and form projecting crystals in the cavities." Mr. Gregory Watt applied the facts here noticed in expla- nation of the globular structure of many decomposing basaltic rocks, in which, after a certain stage of disintegration, the in- cluded balls resist decomposition with great obstinacy. He moreover extended his remarks to the columnar structure, and observed, that when in his experiments " two spheroids came into contact, no penetration ensued, but the two bodies became mutually compressed and separated by a plane, well defined and invested with a rusty colour ;" and when several met, they formed prisms. His inferences from this arrange- ment were as follows : " In a stratum composed of an indefinite number in super- ficial extent, but only one in height, of impenetrable, sphe- roids, with nearly equidistant centres, if their peripheries should come in contact on the same plane, it seems obvious that their mutual action would form them into hexagons ; and if these were resisted below and there was no opposing cause above them, it seems equally clear that they would extend their dimensions upwards, and thus form hexagonal prisms, whose length might be indefinitely greater than their diame- ters. The further the extremities of the radii were removed from the centre, the nearer would be their approach to paral- lelism ; and the structure would be finally propagated by nearly parallel fibres, still keeping within the limits of the hexagonal prism with which their incipient formation com- menced ; and the prisms might thus shoot to an indefinite length into the undisturbed central mass of the fluid, till their quently art duced by Unstratified RocJcs. 471 structure was deranged by the superior influence of a counter- acting cause*. Basaltic columns are often curved, and sometimes there is a somewhat confused arrangement of them, so that the dis- turbing causes have been considerable. They are also fre- articulated, which Mr. Watt considered might be pro- by the same cause which determined the concentric fractures of the fibres of the spheroids. Supposing the gene- ral theory of the formation of columns correct, the irregulari- ties of the prisms would obviously depend upon the unequal distances of the centres of the spheroids, and the consequent unequal pressure. Mr. Watt accounts for the horizontal ar- rangement of basaltic columns in some perpendicular dykes, such for example as those at the Giant's Causeway, from the refrigerating or absorbing cause operating on each side of the vein, so that columns should strike out from it, but would not coincide so as to form continuous prisms across the vein, as there would be confusion where the two sets of columns met, if indeed circumstances were sufficiently favourable to produce a meeting. The columnar arrangement is not confined to basalt, but is more or less observed in all the trappean rocks, the magni- tude of the columns being often very considerable. Granite also assumes a prismatic form, as has already been remarked by Mr. Came respecting that of the western part of Corn- wall f, where it is well seen near the Land's End; but instead of assuming an hexagonal arrangement, such as we might presume to be the figure, if the theory respecting basalt was altogether applicable to it, it is quadrangular, and so divided into joints that the resulting solids are parallelepipeds and even cubes. If the student should pass round the Land's End, Cornwall, in a boat, he will be particularly struck with the general arrangement of granite into columns, and the pic- turesque effect is considerably heightened by the varied disin- tegration of the blocks from the united action of the sea and ft , atmosphere. While on the subject of the columnar structure of rocks, it jnay be remarked, that some of the stratified rocks, more par- ticularly the arenaceous, are rendered columnar by the action of heat upon them. Some sandstones, when kept in our fur- naces at a heat insufficient to fuse them, take this structure. * Gregory Watt, Observations on Basalt, and on the Transition from the vitreous to the stony Texture which occurs in the gradual Refrigeration of melted Basalt; Phil. Trans. 1804. f Carne on the Granite on the western part of Cornwall; Geol. Trans, of Cornwall, vol. iii. p. 208. 472 Unstratified Rocks. From the want of a good material for road-making between Halifax and Huddersfield, a coal-measure sandstone is thrown into kilns and burnt, when it frequently becomes columnar, the columns varying in the number of their sides. They are generally curved, and about half an inch in diameter. Dr. Macculloch was, I believe, the first to connect this al- tered structure of a sandstone in a furnace with the columnar character of masses of sandstone in nature. He observed in a hearth-stone, taken down from a blast-furnace, at the Old Park Iron Works, near Schiffnall, and which had been in constant work for sixteen or eighteen years, that its structure was prismatic. The prisms sometimes extended through the whole thickness of the stone, about ten inches, while in other instances they did not penetrate so deep. The prismatic structure was considered to have been produced by the long- continued action of considerable heat upon the slab of sand- stone. Reasoning from this fact, the same author explains the occurrence of the columnar sandstone beneath basaltic rock at the hill of Scuirmore, in Rum, as also the columnar rocks at Dunbar, by the action of heat *. Mr. Yates informs me that at about a mile N.E. of Bad- Ems, in the duchy of Nassau, where clay ironstone is in con- tact with a mass of trap, the former exhibits the red colour and columnar structure of the Staffordshire clay ironstone after it has been roasted, and previously to smelting f. Capt. Belcher notices a columnar sandstone on the Rio Nunez, west coast of Africa, which appears, from the circumstances connected with it, to have assumed its columnar structure from the effect of heat acting on it after consolidation:}:. * Macculloch, Quarterly Journal of Science, 1 829. f Yates, MSS. J Belcher, Journal of the Geographical Society, vol. ii. p. 282. On the Mineralogical Differences of Rocks* 473 SECTION XII. On the Mineralogical Differences in Contemporaneous flocks, either Original, or resulting from Alteration after Deposi- tion. AMONG the variety of stratified rocks which have been noticed above, it will have been observed that there was much dif- ference as well in the mineralogical composition as in the zoo- logical character of the deposits. Some rocks are evidently formed from the destruction of others; some are chemical; while others appear as if they had suffered alteration subse- quently to deposition. The rocks which have been produced by deposit from water, in which mud, sand, gravel, and great blocks were for a time mechanically suspended, have already been sufficiently discussed ; and that they should not precisely resemble each other over considerable areas is only what would have been expected, as we cannot imagine any detritus so uni- form as to be the same over considerable spaces, for it as- sumes a perfect uniformity in the transporting power, a con- stant, equal, and uniform supply of detritus, and a surface over which the transported substances were carried, so con- stituted as to offer an equal resistance throughout. When a stratified rock is crystalline, it has evidently been chemically and not mechanically produced; but it remains to inquire whether such structure be original, or consequent on circumstances which have permitted an alteration of the rock. This investigation is one of considerable difficulty, as we can- not always obtain the data necessary for decision, since it is obvious that the same substance may often be obtained in dif- ferent ways: thus crystalline carbonate of lime may either be produced directly by deposition from an aqueous solution of that substance, or common limestone may be fused by heat under pressure, and the results be similar. The same with many other substances. It therefore becomes a very difficult though always interesting question to discover, whether such stratified and crystalline substances have been produced in the one way or the other. There are certain generalities on which we may base our in- vestigations. If crystalline and stratified substances occur as sheets of matter, included among beds evidently of mechani- cal origin, and igneous rocks are not intimately connected with the whole, and there has been no violent disturbance of 474? On the Mineraloglcal Differences of Rocks* strata, permitting the possible influence of gaseous com- pounds on the beds, we may fairly infer that the crystalline rock was formed by aqueous chemical deposition, and that its occurrence among decidedly mechanical compounds, merely shows a difference in the condition of the medium from which they have both resulted; there being solution in the one case, and mere mechanical suspension in the other. When we find uncrystalline strata assuming a crystalline structure in the immediate vicinity of igneous rocks, so that the crystalline and uncrystalline portions constitute different parts of a common whole, the question assumes another cha- racter, and we have to inquire if this difference arises from an alteration of a part of the whole, subsequently to deposition, or whether it is the result of certain causes which have ope- rated only on parts of the same whole during deposition. It has been seen that dolomite, a crystalline compound of carbonate of lime and carbonate of magnesia, occurs in the oolitic series of Poland and Germany; therefore we should not be surprised that it occurred in the same series in the Alps, and apparently among the same rocks in Dalmatia and Greece. From this presence of a particular crystalline com- pound in a given rock, and over a considerable area, we should be led to consider, that circumstances existed during the for- mation of the rock which produced this compound over the area, and consequently, that it was original, and did not re- sult from the subsequent application of heat, or any other chemical agent. Supposing compounds of this nature to have resulted from an aqueous solution of the carbonates of lime and magnesia, we should not be surprised at the absence of organic remains; for it would scarcely be a mixture in which animals would flourish. Organic remains are however not absent from dolo- mite, though they are rare in it, for I have seen them in the dolomite of Nice, and they have been noticed elsewhere. The occurrence of organic remains in dolomite does not, it must be confessed, well accord with the supposition that it has been a limestone on which chemical agents have subsequently so acted that it became crystalline and charged with magnesia ; for we cannot well understand in the new arrangement of par- ticles how the form of the organic remains could be preserved, more particularly when they are of the same substance with the rock, or solely carbonate of lime. The fossiliferous do- lomites would therefore appear to be excluded from the al- tered rocks, and reduced to those of original and chemical deposition. There are however masses of dolomite not so easily reconcileable with the supposition of aqueous deposi- tion, which occur in patches among limestones, and not far On the Mineralogical Differences of Rocks. 4-75 removed from igneous rocks, in such a manner that Von Buch considers them to have resulted from the action of che- mical agents on the limestones, subsequently to the deposition and consolidation of the latter, and at the time when igneous rocks were intruded among the stratified mass. To convert a series of beds into a crystalline mass to a certain distance from a rock in a state of fusion, provided there was suf- ficient pressure to prevent the escape of the carbonic acid, would, we know, not be difficult : but it is difficult to obtain the magnesia necessary to produce the dolomite, unless it was insinuated into the altered mass at the time when the various particles were arranging themselves conformably to the laws of crystallization ; in fact, when all the elementary substances were so circumstanced that they could freely unite according to their proper affinities. Von Buch considers this was ef- fected by the escape of magnesia from the augite porphyries or melaphyres*, at the period when such porphyries were protruded through limestones, as in the Tyrol and other places. He is of opinion that the gas evolved at the time these igneous rocks were upheaved, entered among the fissures of the limestone, and converted a considerable proportion of it into dolomite. This celebrated author adduces the mountain of San Salvador, on the lake of Lugano, as confirming the truth of his theory. A red conglomerate, of a similar charac- ter to that which occurs on the lake of Como, separates the mica slate on which Lugano is situated from the limestones and dolomite. " These beds dip rapidly at 70 to the south, and form a promontory on which the chapel of San Martino is built. This rock appears in place for about ten minutes walk, the dip of the beds diminishing to 60. It is then co- vered by a compact smoke-gray limestone, in beds about a foot thick. These dip as the beds on which they rest, and have the same inclination on the side of the mountain ; but in their prolongation towards the lake the dip continually dimi- nishes, until, at its level, it is scarcely 20. The beds as they rise describe a curve that somewhat resembles a parabola. The further we advance on the road, the more we find these beds traversed by small veins, the sides of which are covered by rhombs of dolomite. Similar crystals are also observable in small cavities of the rocks. As we advance, the rock ap- pears divided into fissures, and the stratification ceases to be * If we consider with M. Rose, that augite and hornblende constitute one mineral, we shall find, judging at least from published analyses, that the magnesia contained in it varies materially. The extremes are the horn- blende from Pargas, which, according toBonsdorff, contains 18'79 per cent, of magnesia, and the augite of Taberg, containing, according to H. Rose, -1-99 per cent, of the same substance. 476 On the Miner alogical Differences of Hocks. distinct. Lastty, where the face of the mountain becomes nearly perpendicular, it is found to be entirely formed of do- lomite. There is no marked separation between the limestone and the latter rock. By the increase of the veins and geodes the calcareous rock entirely disappears, and pure dolomite occurs in its place. * * * * As we advance along the high road, the purer we find the dolomite, and at the same time the more white and granular. * * * * From hence to beyond Melide the mountains are composed of dark augite porphyry mixed with epidote, the same as it appears at Campione, Bis- sone, and Rovio *." This is undoubtedly a remarkable case, as the mass of au- gitic rock is on the side of the dolomite, and as crystals of dolomite are found in the cracks of the limestones, because the latter circumstance shows that such crystals were not con- temporaneous with the deposition of the limestone, but were formed subsequently, after cracks had been produced in it, while the former circumstance is precisely in accordance with the theory. According to Von Buch's sections, however, a small quantity of red porphyry and mica slate intervenes be- tween the mass of the augite porphyry and the dolomite : it certainly does not follow that they should constantly intervene, because they may do so in one situation and not in another, therefore this is no great objection ; indeed, according to the map, they do not always separate the dolomite from the igneous rock. Other masses of dolomite occur round a large patch of granite extending westward from the south-western branch of the Lago di Lugano, on which are situated Casco al Monte and Porto. One of these masses at Monte Schieri is connected with tufaceous augite rock ; while others are only in contact with the granite, as far at least as regards the surface : but this proves little, for the augite porphyry may be beneath them, as it is seen to cut through the granite at Brincio. From its vicinity to these places the dolomites of the lakes of Como and Lecco acquire considerable interest, although augite rocks have not yet been detected among them. They certainly occur intermingled with the limestones, which are compact and gray, while at other times they also appear the prolongation of limestone beds which have gradually lost their compact texture, and at the same time have acquired magnesia and become crystalline. In countries like these, where so much confusion prevails, and where we may expect to find extensive faults, a perfect continuance of a given series Von Buch, Ann. des Sci. Nat. 1827, where there are sections and a map of the district of the lakes Orta, Maggiore, and Lugano ; as also in Sections and Views illustrative of Geological Phamoigena, pi. 8. fig. 2. and pi. 30. On the Miner alogical Differences of Rocks. 477 of beds is most difficult to trace ; but with ample allowance for these difficulties, there seems every probability that the con- tinuations of some of the limestone beds become dolomite*. The north part of the lake of Como is composed of gneiss and mica slate, which correspond on either shore, and dip southwards; the lake of Lecco and the southern part of that of Como are formed of limestones and dolomite. Between the two masses of rock are conglomerates and sandstones which have the same dip and direction as the gneiss and mica slate. On the south of these latter rocks the sides of the lake cease to correspond. Thus, on the eastern shore, after passing a small portion of dolomite, we find limestones, among which are the black marbles of Varenna, and these continue as far as opposite to Bellaggio Point, while on the other and western side dolomite prevails during the whole corresponding di- stance, with the exception of a few limestone beds on the south of Menaggio. Here then is no correspondence ; on the con- trary we have limestones on the one side and dolomite on the other, the latter containing a mass of gypsum at Nobiallo. If we proceed down the lake of Como, from Bellaggio to Como, we find nothing but limestones, producing the fine scenery for which this lake is so celebrated, after having passed the pro- montory of Dosso d'Albido and the opposite shores of the Croci Galle : but if we go down the lake of Lecco to the town of the same name, also from Bellaggio, there is scarcely any- thing to be seen but dolomite, if we except a mass of gypsum included in it at Limonta, and a long strip of limestone be- tween Lierna and Mandello. Here also we have no corre- spondence, though the general direction of the beds in both lakes would lead us to suspect, that those in the one were continued from those in the other. And this view is strength- ened if we ascend the Monte San Primo, a mountain already noticed as strewed over with thousands of erratic blocks, for the highest crest is composed of limestone ranging W.N.W. and E.S.E., with a dip to the S.S.W. If we follow the di- rection of these beds to the lake of Lecco on the east, we find dolomite, so that the change in this place appears somewhat sudden. Notwithstanding this apparent conversion of limestone into dolomite in the direction of the beds, which might lead us to suppose that some cause had produced a change in the rock after its consolidation, it must be confessed that there is also an interstratification of the dolomite with the limestone, and moreover, the dolomite rests upon limestone in the lake of * See Geological Map and Sections of the shores of the lake of Como, in Sections and Views illustrative of Geological Phenomena, pi. 31. and 32. 478 On the Mineralogical Differences of Rocks. Lecco ; facts which are at variance with the supposition that all the dolomites of this part of Italy have been altered rocks. Some of them at least appear to have been original deposits ; and this supposition, as far as it affects the rocks which are apparently limestones in one place and dolomites in another, involves a curious question ; for if we admit a contemporane- ous deposition of the two to have taken place, it follows, that carbonate of lime was thrown down in one place, while a mixture of the carbonates of lime and magnesia was deposited close to it in another, and that the two depositions were so far influenced by circumstances, that the one was compact while the other was either wholly or semicrystalline. These obser- vations do not apply to the alternations, as there we have to suppose a change of circumstances over the same place, and this by no means sudden; for the calcareous beds seem gra- dually to acquire magnesia, as may be seen on the western shores of the lake of Lecco. Some good examples of a mixture of dolomite and lime- stone may be observed near Nice, where again the continuation of limestone beds becomes dolomitic, the dolomite here, as elsewhere, generally losing its stratification when most pure, while the less pure semicrystalline compounds are distinctly stratified. This is however not a general rule, for I have seen some nearly pure beds stratified ; and supposing such beds to be original deposits, their division into strata is no more re- markable, than that the saccharine marble of Carrara should be stratified. In the vicinity of Nice, gypsum also accompanies the dolo- mite, and the connexion of the two is so intimate that the gypsum of Sospello contains rhombs of dolomite, a circum- stance also observed in the gypsum accompanying the dolo- mites in the Tyrol. This frequent association of gypsum with dolomite has not yet been satisfactorily explained. Gypsum is not a rare accompaniment of rocks, even those of a decidedly mechanical origin, yet it must be considered either as a chemical deposit, or an altered rock ; its occurrence therefore in such situations shows that other causes were in force than a mere drift of detritus ; and when gypsum is to a certain extent characteristic of a deposit over a considerable area, it proves that the operation of such causes has not been local, but that during such period, and over such area, the circumstances permitting those deposits prevailed extensively. Gypsum has been considered characteristic of the upper part of the red sandstone series, generally known as red or varie- gated marls. Without however asserting that it is a necessary part of the rock, or that it is constantly present, it is remark- able how very frequently it is discovered in this deposit, in On the Mineralogical Differences of Rocks. 479 England, France, and Germany, proving that circumstances were then favourable to its production, if it proves nothing more. When we recollect that the intrusion of igneous rocks has been sufficient to convert chalk into granular limestone in the north of Ireland, we need not be surprised that other rocks have been altered by the intrusion of similar substances. The slates for instance in many parts of the country surrounding the granite of Dartmoor, Devon, have suffered from its intru- sion, some being simply micaceous, others more indurated and with the characters of mica slate and gneiss, while others again appear converted into a hard zoned rock strongly impregnated with felspar. The alteration of the rocks in this case is of very easy ex- planation. The grauwacke, which is for the most part the altered rock, is, when taken in the mass, only the consolidated detritus of more ancient and crystalline rocks, composed of a few simple substances. These substances have necessarily been variously accumulated in different beds, so that their relative proportions would also vary. If long-continued heat, insufficient to produce fusion, be now applied to the ends of these beds, thus differently constituted, the results will neces- sarily be dissimilar. The various substances would have a tendency to resume their original state, at least that state in which they existed in the crystalline rock, whence they have been derived ; and consequently we should have compounds resembling various crystalline stratified rocks. The granite of Dartmoor has evidently been intruded among the grauwacke of the same district, sending veins into it. It to a certain ex- tent cuts through the line of direction of the grauwacke, twisting and contorting the strata, which come more imme- diately in contact with it, to various distances. Such strata must have been exposed to long-continued heat, insufficient to fuse them, and hence their altered character. M. von Dechen remarks, that the grauwacke of the Hartz is in like manner altered by granite, which converts the former into flinty slate, quartz rock, greenstone, &c. Transitions of this kind, even into coarse-grained grauwacke, may be ob- served. The slates of the Vosges and Swartzwald are also altered by granite. It might at first sight appear necessary that complete fusion should take place before the rock could be converted into sub- stances like hornblende rocks, &c., and that therefore the stra- tified character of the rock would be destroyed. The common experiment of throwing a fragment of green bottle glass into a fire, and keeping it for a long time at a heat insufficient to fuse it, shows that the internal particles of a body may, under 480 On the Mineralogical Differences of Rocks. certain circumstances, arrange themselves in a crystalline form without any alteration of the external appearance of the same body. In like manner a bed of grauwacke composed of com- minuted hornblende and felspar, probably resulting from the destruction of some hornblende rocks, would, when exposed to a long-continued heat, insufficient to produce fusion, become converted into a greenstone or hornblende rock, arranged in beds. Cases of induration and alteration of rocks in contact with igneous products are so common that it would be useless to enumerate them ; but the student must carefully distinguish between igneous rocks which have evidently been intruded among the others, and those which are the older rocks, upon which the others have been deposited ; for it may happen that the older rocks have been disintegrated previous to deposition of the superincumbent substance: in which case, if the latter be arenaceous, there may be an apparent passage of the are- naceous into the igneous rock, producing the false appearance of an altered substance. There can be little doubt that many rocks resulting from the deposit of detritus have, more particularly when such de- tritus has been in a highly comminuted state, suffered altera- tion by the chemical action of various substances on each other. Changes have also been produced by the percolation of water, charged with various substances, through rocks. Matter so- luble in water is thus conveyed from a superior to an inferior rock. The cavities or vesicles of igneous rocks filled with car- bonate of lime and other minerals, constituting an amygdaloid, afford a well-known illustration of this percolation. The change in the mineralogical character of certain calca- reous rocks, at different points of the area which they may happen to cover, has been previously adverted to, and it has been shown that in the oolitic group there was a probability of a portion having been produced at the bottom of a deep sea, while other parts were formed in shallower waters. The phy- sical circumstances under which the different parts of the de- posit would be placed, could scarcely do otherwise than influ- ence the product; but what that precise influence may have been we are as yet ignorant: it may, however, be anticipated that the differences of pressure, and the liability to be disturbed by currents in one situation, while the latter might be scarcely felt, if not entirely absent, in another, would alone cause a great variation in the mineralogical texture. It might also happen, that in a deep part of an ocean suc- cessive depositions were effected during periods when frequent changes were produced in other and remote situations, so that though contemporaneous there should be no mineralogical On t7ie Elevation of Mountains. 481 agreement between them; and if in the course of events, the continuous and quiet deposits were upheaved, and a continent be the result, the difficulty of identifying clear divisions in the one place with the mass in the other would be insurmountable. It is more than probable that this supposition has been realized on the surface of our planet, and that eventually geologists will show less determination in identifying deposits, more par- ticularly those of moderate comparative antiquity, over very considerable distances. It is much more desirable, for in- stance, that India should be described with reference to itself, so that when its geology shall have been sufficiently advanced, Europe may be fairly compared with it, than that there should be a determination to find nothing but European equivalents in that quarter of the world. On the Elevation of Mountains. Although the direction of the various chains of mountains has long engaged the attention of geologists and geographers, and although the direction of upturned strata has also long been noticed by the former, and found generally to coincide with that of the mountain chains which they constitute, it has only been recently and since the labours of M. Elie de Beau- mont, that the subject has acquired a new interest, and will henceforth form an important branch of geological investiga- tion, whether the theory of this distinguished geologist shall eventually be found tenable to the extent supposed, or require very material modifications. Von Buch appears some time since to have discovered that the mountain-systems of Germany were not contemporaneous, but were of distinct dates; and geologists had long been in the habit of noticing the unconformable position of strata, an older and inferior rock having been upheaved, while a newer rock rested upon the edges of the upturned strata. Here, however, the subject seemed to rest, until M. Eliede Beaumont, from a series of very exact observations in certain parts of France and the Alps, remarked, that the dislocations of the strata were not only referrible to distinct epochs, but that there was a pa- rallelism between dislocations and upheaved mountains of the same date; and he further considered that these events pro- duced breaks in the rocks then in the course of deposition, so that those subsequently formed rested unconformably on the disturbed strata of the older rocks. To determine the general unconformable position of two rocks sometimes requires very great care, though at first sight it may appear extremely easy to observe whether one rock rests on the upturned edges of another, or not; and so it undoubc- 2 i 482 On the Elevation of Mountains. edly is in many cases; but when they meet at small angles*, or the one rests on the contortions of the other, the inquiry becomes more difficult, and it requires numerous observations to be certain of the general fact. The difficulty in the case of contortions will be seen in the annexed cut. If a section be obtained on the left or right, it will very & . . g ; ! 6> easily be seen that the beds a a rest upon the contorted strata b b; but if one were b c b only observed at c, the unconformable position of the two rocks might be doubtful. The student may perhaps consider, that from a little research in the same place he would soon find evidence of the disturbed condition of the strata beneath; and if the contortions were always on the small scale, and na- tural sections common, such would be the case: but when the latter are scarce, or the undulations and contortions on the large scale, the great bends being measured by miles instead of fathoms, the subject is not so easy. It may be stated as an example, that the mass of the calcareous Alps is considered to rest unconformably on the mass of those composed of pro- togine, gneiss, &c.; but the situations where the contrary opinion may be formed are very numerous, the sections there exposing perfect conformability. It also requires great care in tracing strata up to a mountain range, for the purpose of ascertaining its relative antiquity, to distinguish between those beds which have been decidedly upturned subsequently to de- position, and those which may have originally taken a small angle during their formation on the flanks of a chain pre- viously elevated to a certain extent. Perhaps the annexed diagram may assist the student in com- prehending the manner in which the relative age of mountain ranges is determined from the position of strata. If the rocks a a are found resting quietly on the upturned strata b b, it is inferred that b b have been disturbed previous Fig. 117. * A general unconformability does not always prove a movement in the inferior rocks prior to the deposition of the superior ; for supposing a given series to be so produced that the newer rocks may be formed within succes- sively diminishing areas, and another deposit to cover the whole, it is evi- dent that the higher mass will so far rest unconformably on the inferior rocks that it will cover them all in succession. Now this is what has hap- On the Elevation of Mountains. 483 to the deposition of a a ; and consequently, if a a be a known rock occupying a certain place in the series, we obtain a rela- tive date for the elevation of b b, so far as relates to a a; but if it should chance that there are no commonly known deposits absent between them, we obtain the exact relative date of the elevation of so much of the mountains as do not exhibit any other unconformability of strata, and of the whole range, if no such unconformability can be detected. When however this does occur, as in the above diagram, we learn, that more than one elevation of strata has taken place in the same chain, for b b resting in discordant stratification on c, proves that c was tilted up prior to the deposition of b b ; so that in this case there would be evidence of two disturbances in the same chain, and the relative dates of both would be obtained on the same principles. It will be obvious if two lines of elevated strata cross each other, there would be much confusion where they traverse; and should great violence have been employed, so that the strata be even thrown over, it will require much caution to determine the relative age of the fractures at those points. From the obliging communications of M. Elie de Beaumont, it appears that he now considers he can distinguish twelve sy- stems of mountains in Europe, each regarded as characterized by the relative direction and elevation of its strata, and each elevation as corresponding with a solution observed in the continuity of the sedimentary deposits, also of Europe*. These various systems are as follows, commencing with that of the greatest relative antiquity : I. System of Westmoreland and the Hundsruck. The direc- tion of the slate rocks in Westmoreland is, according to Prof. Sedgwick, N.E. by E. and S. W. by W. ; and that of the slates and grauwacke of the Eifel, the Hundsruck, and of the Nassau Mountains, about N.E. and S. W. The slates, grauwacke, and grauwacke limestones of the northern and central part of the Vosges have the same direction. The carboniferous rocks rest on the upturned rocks of the North of England; coal-measures repose upon the edges of those of the Vosges ; and the carbo- pened with the chalk and oolite groups in England, the former overlapping the various members of the latter as they successively fine off- See Sections and Views illustrative of Geological Phenomena, for an overlap of the chalk on the coasts of Dorset and Devon. The angles at which the creta- ceous and other rocks meet is there so small, that their unconformability could scarcely be determined at any particular point, though in the mass it is evident. * M. Elie de Beaumont's communication to me, of which the accompa- nying sketch is a brief notice, will be found in the Phil. Mag. and Annals of Philosophy, for October 1831. 2 i 2 484 On the Elevation of Mountains. niferous rocks of Belgium and Saarbruck were probably de- posited at the foot of the Eifel, Hundsruck, &c. II. System of the Ballons (Fosges), and of the Hills of the Bocage (Calvados}. It is observed under this head, that the first system only shows that the slates and grauwacke of West- moreland and the Hundsruck have been elevated before the deposition of the carboniferous series ; but it would also ap- pear that there has been an elevation of strata before the more recent transition strata were deposited, so that the latter have not been elevated in a N.E. and S. W. direction, but would on the contrary seem to have been formed on upturned beds of the former. Such are the calcareous, marly, and arenaceous deposits, with Orthoceratites, Trilobites, &c., in Podolia, of the environs of St. Petersburg, and of Sweden, where they are but slightly removad from their original horizontal position. Such are also the transition beds of Dudley and Gloucestershire ; and possibly also the transition beds of the South of Ireland may be included in the same list, which M. Elie de Beaumont considers may also contain certain slate and grauwacke beds with anthracite, forming the south-east angle of the Vosges. When these beds are not horizontal, they are dislocated in directions the most marked of which is comprised between an E. and W. line and one E. 15 S. and W. 15 N. III. System of the North of England. This is the north and south range of the carboniferous series, noticed by Prof. Sedg- wick. It is considered to have been produced immediately previous to the deposit of the red conglomerate (rothliegendes). M. Elie de Beaumont remarks ; " The elevation of the chain in the North of England is not probably an isolated pheno- menon; but if we glance at the geological map of England by Mr. Greenough, or that accompanying the memoir of Prof. Buckland and Mr. Conybeare on the environs of Bristol, we are naturally led to remark, that the problematical rocks which penetrate and dislocate the coal deposits of Shrewsbury and Colebrooke Dale, and those which form the Malvern Hills, appear to be connected with a series of fractures which run nearly N. and S., and are prolonged across the more recent transition and the carboniferous rocks to the environs of Bristol." It is also considered probable that the form of the west coast of the department of La Manche, which runs nearly N. and S., may be due to a fracture of this age. IV. System of the Netherlands and South Wales. This is the great E. and W. range of the carboniferous rocks from the en- virons of Aix-la-Chapelle to St. Bride's Bay, Pembrokeshire, which, whenever visible from beneath other deposits, exhibits this general direction ibr about 400 miles. It is considered that the beds of the (new) red sandstone series which repose On the Elevation of Mountains. 485 on this dislocation are not so ancient as those noticed in the previous group*. V. System of the Rhine. The Vosges and the Swarzwald terminate opposite one another in two long cliffs parallel to each other and to the course of the Rhine. These are appa- rently due to great faults, having a direction S. 15 W. and N. 15 E. These fractures preceded the deposit of the rocks in the basin of Alsace, among which are the red or variegated sandstone (gres bigarre), the muschelkalk, and the variegated marls. VI. System of the S.IV. coast of Brittany and of La Vendee, ofMorvan, of the Bohmerwaldgebirge, and of the Thwhigerwald. The general direction of this system is from N.W. to S.E. ; and while the red or variegated marls (marnes irisees), the muschelkalk, and all older strata have been thrown out of their original positions, the oolitic series, comprehending the lias and its lower sandstone, have remained undisturbed in these vari- ous situations. VII. System of the Pilas 9 the Cote d'Or, and the Erzgebirge. In this system, which also contains a portion of the Cevennes, the strata are disturbed up to the oolitic rocks inclusive, while the cretaceous series (green sand and chalk) remain apparently in the position in which they were deposited. The direction of this system is considered to be N.E, and S.W. VIII. System of Mont Viso. "The French Alps and the south-west extremity of the Jura, from the environs of Antibes to those of Pont d'Ain, and Lons le Saulnier, present a series of crests and dislocations with a direction towards the N.N. W., in which the older rocks of the Wealden formation, the green sand, and chalk, are found upheaved, as well as those of the oolitic series. The pyramid of primaeval rocks composing Mont Viso is traversed by enormous faults, which, from their direction, evidently belong to this system of fractures. The eastern crests of the Devolny, on the north of Gap, are formed of the oldest beds of the green sand and chalk thrown up in the direction in question, and raised more than 4700 English feet above the sea. At the feet of these enormous escarp- ments are, horizontally deposited and at more than 2000 feet * It should be here noticed that some recent observations among the Mendip Hills have shown me that they have been dislocated in N. and S. lines, subsequently to their E. and W. elevation. These faults are the cross courses so well known to the miners of the district, and exhibit disturbances parallel to the lines of dislocation in System III. Faults with a general N. and S. direction are also observable in the Blackdown Hills, to the south of the Western Mendips. These latter were clearly produced after the deposit of the chalk, and perhaps of certain supracretaceous rocks : whether they may or may not be contemporaneous with the N. and S. Mendip faults is not yet determined. 4-86 On the Elevation of Mountains. lower down, those upper beds of the cretaceous system which are distinguished from the rest by the presence of Nummulites, Cerithia, Ampullarice^ and other shells, the genera of which were long considered as exclusively found in the tertiary (su- pracretaceous) rocks. Thus it was between the two portions of that which is commonly termed the series of the Wealden formation, green sand, and chalk, that the beds of the Mont Viso system have been thrown up." IX. Pyreneo-Apennine System. " This includes the whole chain of the Pyrenees, the northern and some other ridges of the Apennines, the calcareous chains on the north-east of the Adriatic, those of the Morea, nearly the whole Carpathian chain, and a great series of inequalities continued from that chain through the north-east escarpment of the Hartz Moun- tains to northern Germany." The general direction of this system is about W.N.W. and E.S.E. X. System of the Islands of Corsica and Sardinia. This elevation is considered to have taken place during the supra- cretaceous period, and it is remarked that the north and south direction of the system in Corsica and Sardinia is observed "in many small valleys and ridges of mountains in the Apen- nines, and in Istria, and in the disposition of many volcanic masses and metalliferous sites of Hungary." " It is worthy of remark that the directions of the system of the Pilas and the Cote d'Or, of that of the Pyrenees, and that of the islands of Corsica and Sardinia, are respectively nearly parallel to those of the system of Westmoreland and the Hundsruck, of the system of the Ballons and of the Bocage, and of the system of the North of England. The correspond- ing directions differ but a small number of degrees, and the corresponding systems of the two series have succeeded each other in the same order ; leading to the supposition that there has been a kind of periodical recurrence of the same, or nearly the same, directions of elevation." XI. System of the Western Alps. The mean direction of this system is about N.N.E. and S.S. W., and the elevation is con- sidered to have taken place after the deposit of those recent Mipracretaceous beds named Shelly Molasse (molasse coquil- liere\ beds contemporaneous with thefahluns of Touraine. The direction of strata is of a complicated character where this system and that to be next mentioned cross each other, as they do around the Mont Blanc, Mont Rose, and Finsteraar- horn, at about an angle of 45 or 50. XII. System of the principal Chain of the Alps (from the Valais into Austria], comprising also the Chains of the Ventoux^ the Lebaron and the Sainte Baumc (Provence}. The direction of this system is about E. 3- N.E. and W. ^ S.W., and the On the Elevation of Mountains. 487 strata are considered to have been elevated previous to the dispersion of the erratic blocks and those gravels which have been termed diluvium, but which in the vicinity of the Alps are found to have been deposited upon other gravels, often of considerable thickness. M. Elie de Beaumont concludes with the following observa- tions: " The independence of successive sedimentary forma- tions is the most important result obtained from the study of the superficial beds of our globe ; and one of the principal objects of my researches has been to show, that this great fact is the consequence and even a proof of the independence of mountain-systems having different directions. " The fact of a general uniformity in the direction of all beds upheaved at the same epoch, and consequently in the crests formed by these beds, is perhaps as important in the study of mountains, as the independence of successive forma- tions is in the study of superimposed beds. The sudden change of direction in passing from one group to another has permitted the division of European chains into a certain num- ber of distinct systems, which penetrate and sometimes cross each other without becoming confounded. I have recognised from various examples, of which the number now amounts to twelve, that there is a coincidence between the sudden changes established by the lines of demarcation observable in certain consecutive stages of the sedimentary rocks, and the elevation of the beds of the same number of mountain-systems. " Pursuing the subject as far as my means of observation and induction will permit, it has appeared to me, that the dif- ferent systems, at least those which are at the same time the most striking and recent, are composed of a certain number of small chains, ranged parallel to the semi-circumference of the earth's surface, and occupying a zone of much greater length than breadth ; and of which the length embraces a con- siderable fraction of one of the great circles of the terrestrial sphere. It may be observed respecting the hypothesis of each of these mountain-systems being the product of a single epoch of dislocation, that it is easier geometrically to conceive the manner in which the solid crust of the globe may be elevated into ridges along a considerable portion of one of its great circles, than that a similar effect may have been produced in a more restricted space. " However well established it may be by facts, the assem- blage of which constitutes positive geology, that the surface of the globe has presented a long series of tranquil periods, each separated from that which followed it by a sudden and violent convulsion, in which a portion of the earth's crust was dislo- cated, that, in a word, this surface was ridged at intervals 4-88 On the Elevation of Mountains. in different directions ; the mind would not rest satisfied if it did not perceive, among those causes now in action, an ele- ment, fitted from time to time to produce disturbances dif- ferent from the ordinary march of the phenomena which we now witness. "The idea of volcanic action naturally presents itself when we search, in the existing state of things, for a term of com- parison with these great phaenomena. They nevertheless do not appear susceptible of being referred to volcanic action, unless we define it, with M. Humboldt, as being the influence exercised by the interior of a planet on its exterior covering during its different stages of refrigeration. " Volcanos are frequently arranged in lines following frac- tures parallel to mountain chains, and which originate in the elevation of such chains ; but it does not appear to me that we can thence regard the elevation of the chains themselves as due to the action of volcanic foci, taking the words in their ordinary and restricted sense. We can easily conceive how a volcanic focus may produce accidents circularly and in the form of rays from a central point, but we cannot conceive how even many united foci could produce those ridges which follow a common direction through several degrees. " Volcanic action, such as it is commonly understood, could not therefore be itself the first cause of these great phenomena ; but volcanic action appears to be related (and this is a subject which has long occupied M. Cordier, though he has considered it under another point of view,) with the high temperature now existing in the interior of the globe. " Now the secular refrigeration, that is to say, the slow diffusion of the primitive heat to which the planets owe their spheroidal forms and the generally regular disposition of these beds from the centre to the circumference, in the order of spe- cific gravity, the secular refrigeration, on the march of which M. Fourier has thrown so much light, does offer an element to which these extraordinary effects may be referred. This element is the relation which a refrigeration so advanced as that of the planetary bodies establishes between the capacity of their solid crusts and the volume of their internal masses. In a given time, the temperature of the interior of the planets is lowered by a much greater quantity than that on their sur- faces, of which the refrigeration is now nearly insensible. We are, undoubtedly, ignorant of the physical properties of the matter composing the interior of these bodies ; but analogy leads us to consider, that the inequality of cooling above no- ticed would place their crusts under the necessity of continually diminishing their capacities, notwithstanding the nearly rigo- rous constancy of their temperature, in order that they should On the Elevation of Mountains. 489 not cease to embrace their internal masses exactly, the tem- perature of which diminishes sensibly. They must therefore depart in a slight and progressive manner from the spheroidal figure proper to them, and corresponding to a maximum of capacity; and the gradually increasing tendency to revert to that figure, whether it acts alone, or whether it combines with other internal causes of change which the planets may contain, may, with great probability, completely account for the ridges and protuberances which have been formed at intervals on the external crust of the earth, and probably also of all the other planets*." From this sketch of M. Elie de Beaumont's theory, in which his views respecting the connexion of distant mountains with those of Europe have been omitted, it will be evident that it will require much time and very exact observation in various parts of the world before we can fairly ascertain what are ex- ceptions and what the general rules. It will have been observed that M. Elie de Beaumont has already remarked on the near parallelism of three particular systems respectively with three other particular systems of European mountains, leading to the presumption that parallelism is alone insufficient to deter- mine the relative age of an elevated range of strata ; a con- clusion that may be still further strengthened by observing certain lines of disturbed strata in the British Isles, which, when we regard the general surface of the world, are not far distant from each other. The disturbed strata in the Isle of Wight range east and west, as do those also in the Weymouth district, in the Mendip Hills, in a large part of Devonshire, and in South Wales. The date of the elevation of the Isle of Wight beds was certainly pos- terior to the deposition of the London clay, and there would appear little reason to doubt that the disturbed and fractured condition of the Weymouth district was effected at the same time. But when we continue our researches into Devonshire, we find that the east and west arrangement of a large propor- tion of the grauwacke in that country was produced anterior to the deposit of the (new) red sandstone series, since the latter rests upon the upturned edges of the former f . If we now proceed northwards to the carboniferous rocks of the Mendips and South Wales, we find they also have suffered an elevation in an east and west direction, prior to the formation of the (new) red sandstone; so that in the southern parts of England the strata have been twice elevated in a given direction at dif- * Elie de Beaumont, MSS. f Both the one and the other have been subsequently fractured, and many of the faults have somewhat of east and west direction. 490 On the Elevation of Mountains. ferent dates ; and if we continue our researches, we find, from the observations of Mr. Weaver, that the grauwacke in the South of Ireland was elevated previous to the deposition of the old red sandstone, also in an east and west direction; thus affording three elevations of strata (not far distant from each other) in the same direction, but at different dates *. In offering these observations it is by no means intended to combat the general principle of the possibly contemporaneous elevation of strata at various and distant places, resulting from the gradual refrigeration of the globe ; beds of various kinds having been subsequently and quietly deposited over large areas, on such disturbed strata ; but simply to remark, that, though parallelism may frequently exist, it may not be a neces- sary condition of contemporaneously elevated strata ; for per- haps by laying too much stress upon this point, we not only are in danger, in the present state of our knowledge, of permit- ting theory to take the lead of facts, but of shutting ourselves out from a consideration of other possible lines which contem- poraneously elevated strata may follow. And should it even- tually be discovered that contemporaneously disturbed beds are by no means parallel though still in straight lines, it does not appear that the main principle of M. Elie de Beaumont's theory would be affected by it. What lines may eventually be found to prevail, will, as previously remarked, require much time and great patience to discover; but let the event be as it may, geologists will not the less have reason to feel thankful to M. Elie de Beaumont for having rescued the subject from the state in which he found it; it being impossible but that the investigations, to which this theory will necessarily give rise, must end in the most important additions to geological know- ledge f. It has already been noticed by Prof, Sedgwick, that the change in the zoological character of deposits has not always coincided with disruptions of strata ; and the student will have collected from the foregoing pages, as indeed is also remarked by Prof. Sedgwick, that there was, in Europe, no important change in the general zoological character of deposits up to the zechstein inclusive ; the first great alteration, as far as we * It has been considered that the north and south line of the carboni- ferous rocks in Northern England was elevated at a different epoch from the east and west line of South Wales and Somersetshire, but it must be confessed that this point is far from being proved. f It is but right to inform the student that various objections have been made to this theory by different geologists, more particularly as relates to some of the lines of elevation and their relative dates. Dr. Boue, who was among the first to notice that mountain-masses were elevated at different dates, has inserted a series of objections in the Journal de Geologie, torn. iii. p. 338. Occurrence of Metals in Hocks. 491 can at present see our way, being observed in the remains en- tombed in the variegated sandstone (gresbigarre), and muschel- kalk. It has already been observed, but may be conveniently repeated here, that the effect produced on animal and vegetable life by an upburst of a line of rocks sufficient to produce a range of mountains, might destroy all terrestrial animals and even a large proportion, if not the whole, of the vegetation within the influence of the disturbing cause, not only by pro- ducing a deluge over the land, which might wash off the animals and carry away a great proportion of the vegetation, but by elevating such vegetation into colder regions of the at- mosphere where it could no longer exist. In this case we suppose land so situated as to produce plants and to support terrestrial creatures ; but it will be evident that if we admit a contemporaneous disruption of strata at different points, it would take place under various conditions. In one place it may be effected in the atmosphere, in another in shallow seas, and in a third beneath a great depth and pressure of the ocean ; consequently the resulting phaenomena would be as various as the conditions under which the disruption and elevation of strata were produced : but it will be obvious that the destruction of marine life would be very difficult, and we can scarcely, from known facts, consider that there has been a disruption of the rocks composing the earth's surface so general as to annihi- late all marine creatures at a given time, even with every allow- ance for the operations of powerful and destructive currents ; though we can conceive that near the centres of every great disturbance there might be a very great, and as far as related to certain areas complete destruction of marine creatures. On the Occurrence of Metals in Rocks. To enter fully into this subject would require a volume; the following notice is therefore solely intended to call the attention of the student to a few circumstances which may be generally interesting. Metals occur in rocks either disseminated ; in bunches ; in a net-work of strings or small veins ; in beds ; or in veins filling fissures, which traverse beds or masses of rock. When metals are disseminated through a rock, as tin often is in granite, and iron pyrites in many trap rocks and clay slates, there can be little doubt that they constituted original portions of the rock, and that they were chemically separated from the mass during consolidation. When metals occur in bunches, as the copper at Ecton, Staffordshire, or the lead in the Sierra Nevada, in Spain, there is a difficulty in considering them otherwise than contemporaneous with the rocks in which they are included. 492 Occurrence of Metals in Rocks. The occurrence also of metals in strings or small veins crossing each other in all directions, so that in a section they appear like net-work, reminds us strongly of the small strings or veins of carbonate of lime in many limestones, as has already been observed by Mr. Weaver respecting those of copper in Ross Island, Lake of Killarney ; so that if riot precisely contempo- raneous with the original formation of the including rock, they were, like the calcareous veins in the limestone, secreted from the rock into small cracks possibly produced during consolida- tion. The occurrence of metals in beds has been much dis- puted or commented on, but it must be admitted that iron ore frequently occurs in beds, and we must regard the copper slate of Thuringia and other adjacent countries as to a certain extent a metallic bed, though it does not strictly come under the head of a bed of solid ore. The appearance of metals in beds is often deceptive, being nothing more than a continuation of a vein laterally between strata ; thus in the rich copper mine of Allihies, in the South of Ireland, " the ore occurs in a large quartz vein, which generally intersects the slaty rocks of the country from north to south, but in some cases runs parallel to the stratification *." Mr. Taylor informs me that the lead at Nent Head in Alston Moor, Cumberland, shoots out later- ally among the strata, and that the same fact is observable in different mines in Yorkshire and Flintshire. The most common occurrence of metals is however in vein? ; or, as they are termed in Cornwall, lodes. These are in part filled up, but in various proportions, with metallic substances, and have the general appearance of fissures. They dip at various angles, not unfrequently approaching a vertical position. It was at one time much disputed whether these fissures had been filled from above or beneath ; but from facts that have been noticed within a few years, more particularly by Mr. Taylor and Mr. Carne, there is much difficulty in considering that either hypothesis is generally correct. It now appears that the mineral character of a metalliferous vein greatly de- pends upon the rock which it traverses, that is, when a vein traverses two rocks, as for instance granite and slate, the con- tents of the vein are not generally the same in the two rocks, but will be different in the one and the other. Mr. Carne has observed respecting the metalliferous veins of Cornwall, that it is a rare circumstance when a vein which has been productive in one rock continues rich long after it has entered into another. The same author has also remarked that a similar change will be observed even in the same rock, should such rock become harder or softer, more slaty or more * Weaver, Proceedings of the Geol, Soc. June 4, 1830. Occurrence of Metals in Rocks. 493 compact. He admits that such changes are sometimes small, but states that the general fact is sufficiently apparent, and often very striking *. Such facts are not confined to Cornwall, but have been ob- served elsewhere ; thus the lead veins traversing the carboni- ferous limestone of Derbyshire, which is in some places much associated with trap rocks, are found to be so altered in their passage through the trap, which, from the mode of association, presents the appearance of interstratification, that it was once considered the trap cut off the lead veins ; this is however now well known not to be the case. This fact of the alteration of metallic veins in their passage from one kind of rock to another, or in the same rock, should that become changed, would lead us to consider, with Mr. Fox, that their formation has been in a great measure due to the silent though powerful influence of electricity. This inquiry may yet be considered in its infancy ; but the experiments of Mr. Fox on the electro-magnetic properties of the metalliferous veins of Cornwall will be read with great interest f. That many of these veins are fissures produced by disloca- tions similar to those which are commonly found in various countries, and are supposed to abound more in the coal- measures only because opportunities of detecting them are there more frequent, seems highly probable ; indeed if veins are of different ages, and by cutting one another shift each other, as has been shown to be frequently the case in Cornwall, we can scarcely doubt it. The following is, according to Mr. Carne, the relative ages of the veins in Cornwall : 1 . oldest tin lodes ; 2. the more recent tin lodes ; 3. the oldest east and west copper lodes ; 4. the contra copper lodes ; 5. cross courses ; 6. the more recent copper lodes ; 7. the cross flukans (clay veins) ; and 8. the slides (faults with clay in the fissures) J. Now if this relative antiquity of veins be generally correct as far as respects Cornwall, it becomes a curious question, why, if similar causes have produced them, similar results should * Carne, Trans. Geol. Soc. of Cornwall, vol. iii. p. 81. f Fox, Philosophical Transactions, 1830, p. 399. This author considers that the relative power of conducting galvanic electricity is in the following order in some of the metalliferous minerals. Conductors : Copper nickel, purple copper, yellow sulphuret of copper, vitreous copper, sulphuret of irpn, arsenical pyrites, sulphuret of lead, arsenical cobalt, crystallized black oxide of manganese, Tennantite, Fahlerz. Very imperfect conductors : Sulphuret of molybdenum, sulphuret of tin, or rather bell-metal ore. Non-conductors : Sulphuret of silver, sulphuret of mercury, sulphuret of antimony, sulphuret of bismuth, cupriferous bismuth, realgar, sulphuret of manganese, sulphuret of zinc, and mineral combinations of metals with oxygen, and with acids. I On the relative Age of the Veins in Cornwall ; Carne, Geol. Trans, of Cornwall, vol. ii. 494? Occurrence of Metals in Rocks. not be the consequence. If we admit the possibility of secre- ting the contents of veins from the rocks by electrical means, we cannot so readily understand why different metals should fill the veins in the same rocks, though the direction of the veins might have considerable influence on the conditions and mineralogical combinations of the same metal. While again if we consider them ejected from beneath, we are at a loss to understand why the metallic veins should be so much altered in their passage through different rocks. We are certainly not prepared to say what effect may have been produced on the vein, and on the including rocks, from the continued pass- age of electricity through the vein during an immense lapse of time, or from the arrangement of rocks on the large scale, producing, when properly connected, the effects of a grand galvanic battery ; but as the information at present stands, the history of metalliferous veins is anything but clear. It is quite certain from the dissemination of metals in rocks, that they may constitute an original portion of them ; the small strings also which cross each other, and are unconnected with great veins, have all the appearance of chemical separations from the including rock ; therefore a given rock may contain the necessary elements for secreting substances into a fissure, in the same manner that carbonate of lime frequently fills fissures in limestones, and quartzose veins are common in rocks where silica is abundant. If the theory of internal heat be well founded, it will be evident that the two ends of a metallic vein will be differently heated, and therefore we should have a thermo-electrical ap- paratus on the large scale, producing effects which, though slow, might be very considerable. How far such really exist in nature remains questionable ; but it may be observed that the experiments of Mr. Fox show the possibility of their occurrence ; and should further researches in this highly in- teresting subject merely so divide it, that some of its present apparent complexity may disappear, a great advance will be made in this now obscure branch of geological inquiry. Organic Remains in Blue Marls of the S. of France. 495 SECTION XIII. ORGANIC REMAINS. Organic Remains of the Supracretaceous Group. [THE reader will find a great number of organic exuviae noticed, in con- nexion with the various rocks which contain them, under the head of the Supracretaceous Group. The following are lists which were considered too long to be inserted with the descriptions of the beds in which they are stated to have been discovered.] A. Organic Remains in the Supracretaceous Blue Marls of the South of France, according to M. Marcel de Serres*. LENTICULITES complanata, Defr., Italy, Bordeaux, and C. VAGINELLA depressa, Bast., Bordeaux; BULLA ampulla, Lam., Italy; B. striata? Lam., Italy, Bord. ; B. hydatis? Lam., Italy; B. truncatula, Broc., Italy, Bord. ; B. lignaria, Lam., Bord., Italy, Paris, England ; TESTACELLA haliotidea, Draparnaud, an analogue. PLANORBIS minutus, Faujas de St. Fond. AURICULA Pisum, M. de S., Italy ; Au. (species resembling Voluta myo- tis, Broc.), Italy ; Au. myosotis, Draparnaud, Italy. TORNATELLA fasciata, Lam., analogous with the existing species, T. alle- gata, Desk., Paris ; T. inflata, Ferussac, Bord., Paris. PALUDINA Brardii. AMPULLARIA Faujasii. MELANOPSIS laevigata, Lam. ; M. deperdita, M. de S. MELANIA ventricosa, Fauj. de St. Fond; M. pyramidata, Fauj. de St. Fond. NERITA Plutonis, Bast., Bordeaux. NATICA epiglotina, Al. Brong., Italy and C. ; N. patula, Italy, England ; N. cruentata? Lam., Italy; N. vitellus ? Lam., Italy; N. Guilleminii, Pay- randeau, analogous to the living species, N. Olla, M. de S., Italy ; N. helicina, Broc., Italy. DELPHINULA Solaris (Trochus Solaris, Broc.), Italy. TURBO rugosus, Broc., Italy ; T. tuberculatus, M. de S. ; numerous oper- cula of the Turbo. TROCHUS cingulatus, Broc., Italy ; T. striatus, Broc., Italy ; T. magus, Lam. ; T. conulus, Lam. ; T. Matonii, Payrandeau (analogous with the spe- cies now living in the Mediterranean) ; T. Fermonii, Payrandeau ; T. zizy- phinus, Lam., an analogue ; Trochus, resembling T. moniliferus, Lam. ; T. patulus, Broc., Bord., Italy ; T. agglutinans, Broc., Italy ; T. granulatus, M. de S\ PHASIANELLA pulla, Payrandeau (analogous with the existing species) ; Ph. leevis, M. de S. SOLARIUM sulcatum, Lam., Paris ; Solarium, very near S. Isevigatum, Lam. SCALARIA Textorii, M. de S. (Turbo pseudo-scalaris, Broc.), Italy and C. ; Sc. cancellata (Turbo cancellatus, Broc.), Italy ; Sc. lamellosa (Turbo la- mellosus, Broc.), Italy. * The names which follow those of the authors who have named the species, point out the other localities, or Supracretaceous basins as they are termed, in which the same fossil is considered to be found. When the letter C. is appended, it shows that it is also discovered in the oalcaire moellon of the South of France. 496 Organic Remains in Blue Marls of the S. of France. TURRITELLA rotifcra, Lam., in the marine sands, the calcareous marls, and the calcaire moellon ; T. terebralis, Lam., Bord., Italy and C. ; T. tere- bra, Lam., (analogue of the existing species), Italy and C. ; T. turris, Bast., Italy and C. ; T. tricarinata (Turho tricarinatus, Broc.), Italy ; T. varricosa (Turbo varricosus, Broc.), Italy; T. cathedralis, Al. Brong., Italy ; T. coch- leata (Turbo cochleatus, Broc.), Italy; T. Archimedis, Al. Brong., Italy; T. serrata (Trochus serratus, Broc.), Italy; T. marginalis (Turbo marginalis, Broc.), Italy; T. muricata (Turbo muricatus, Broc.), Italy; T. imbricata? Lam., Paris and C.; T. duplicata (Turbo duplicates, Broc.), Italy; T. per- forata? Lam., Paris; T. acutangula (Turbo acutangulus, Broc.), Italy; T. triplicata (Turbo triplicatus, Broc.), Italy and C. ; T. vermicularis, Al. Brong., Italy and C. ; T. fuscata, Lam., (analogous to the species existing in the Mediterranean and Ocean) ; T. Proto, Bast., Bordeaux, Italy, and C. ; T. replicata (Turbo replicatus, Broc.), Italy; T. quadriplicata, Bast., Bordeaux ; T. lata, M. de S. ; T. corona, M. de S. CERITHIUM marginatum, Bruguiere, Italy and C. ; C. prismaticum, AL Brong., Italy ; C. cinctum, Ba.it. (not C. cinctum, Bruguiere), Bordeaux ; C. cinctum, Bruguiere (C. lemniscatum, Al. Brong.), Italy ; C. pictum, Bast., Bord., Italy; C. sulcatum, Bruguiere (C. plicatum, Bast.), Bord., Italy ; C. doliolum (Murex doliolum, Broc.), Italy ; C. plicatum, Bruguiere (not Bast.), Bordeaux; C. papaveraceum, Bast., Bordeaux; C. subgranosum, Lam., Paris; C. tuberculosum ? Lam., Paris; C. umbilicatum, Lam., Paris; C. Castellini, Al. Brong., Italy; C. Lima, Bruguiere (Murex scaber, Broc.), Italy; C. mutabile, Lam., Paris; C. bicarinatum, Lam., Paris; C. turbinatum (Murex turbinatus, Broc.), Italy ; C. vulgatum antiquum, Italy ; C. multi- sulcatum, Al. Brong., Italy; C. calcaratum, AL Brong., Italy; C. multi- granulatum, M. de S. ; C. alucaster (Murex alucaster, Broc.), Italy ; C. baccatum, AL Brong., Italy; C. ampullosum? AL Brong., Italy. PLEUROTOMA turricula (Murex turricula, Broc.), Italy; P. dimidiata (Murex dimidiata, Broc.), Italy; P. muricata, M. de S., Italy; P. auricula (Murex auricula, Broc.), Italy ; P. textile (Murex textile, Broc.), Italy ; P. oblonga (Murex oblongus, Broc.), Italy ; P. contigua (Murex contiguus, Broc.), Italy; P. mitrseformis (Murex mitrasformis, Broc.), Italy; P. mul- tinoda, Lam., Paris; P. spiralis, M. de S.; P. subulata (Murex subulatus, Broc.), Italy ; P. Farinensis, M. de S. ; P. harpula (Murex harpula, Broc.), Italy; P. clathrata, M. de S.', P. Pannus, Bast., Bordeaux. Fusus lignarius, Payrandeau (analogue to the existing species, common in the Mediterranean), Italy; F. subcarinatus, Al. Brong., Italy; F. subulatus (Murex subulatus, Broc.), Italy ; Fusus, a species between F. Syracusanus of Lamarck, and another and imdescribed species of the Mediterranean ; F. polygonus, AL Brong., Italy ; F. rugosus, Lam., Paris ; F. longirostris (Murex longirostris, Broc.), Italy ; F. uniplicatus, Lam., Paris. CANCELLARIA clathrata, Lam., Paris. PYRULA transversalis, M. de S. ; P. ficoides, Lam., (analogue of a living species), Italy ; P. clathrata, Lam., Italy ; P. clathroides, M. de S. RANELLA marginata, AL Brong. (Buccinum marginatum, Broc.), Bor- deaux, Italy ; R. ranina, Lam. (an analogue). MUREX brandaris, Lam., Italy ; M. anguliferus. Lam. (apparently an analogue of the living species), Italy ; Murex Motacilla, Lam. (an analogue of the living species), Italy; M. craticulatus, Broc., Italy; Murex approaching M. trunculus, Italy ; M. intermedius, Broc., Italy ; M. calcitrapoides, Lam., Paris ; M. Blainvillii, Payrandeau (so like the living species in the Medi- terranean, that it cannot be distinguished from it) ; Murex cornutus, Lam. (apparently the analogue of the existing species), Italy; M. Haustellum, Lam. (resembles the living species) ; M. brevispina, Lam. (an analogue of an existing species); M. tenuispina, Lam. (an analogue), Bordeaux, Italy; M. crassispina, Lam. (analogous to a living species), Italy ; M. rarispina, Lam. (a complete analogue), Italy ; Murex, approaching M. heptagonus of Organic Remains in Blue Marls of the S. of France. 4-97 Brocchi, Italy; M. tripterus, Lam. (var.) ; M. cristatus, Broc., Italy; M. decussatus, Broc., Italy ; M. transversalis, M. de S. ; M. rostratus, Broc., Italy ; M. oblongus, Broc., Italy. TURBINELLA infundibulum ? Lam., analogous to the existing species. TRITON corrugatum, Lam. (an analogue), Italy ; T. pileare, Lam. (analo- gous to a species now living in the Mediterranean), Italy ; T. doliare, Broc., Bordeaux, Italy ; T. personatum, M. de S. ; T. intermedium (Murex in- termedium, Broc.), Italy; T. Chlorostoma, Lam., (an analogue.) ROSTELLARIA Pes Pelicani (Strombus Pes Pelicani), Italy, Bordeaux. STROMBUS pugilis, Lam. (a species completely -analogous with that now existing in the Mediterranean) ; S. tuberculiferus, M. de S. CASSIDARIA echinophora (Buccinum echinophorum, Broc.) (an analogue), Italy. CASSIS Rondeleti, Bast., Bordeaux ; C. marginatus, M. de S. ; C. diluvii, M. de S. ; C. striatus, M. de S. ; C. inflatus, M. de S. DOLIUM, casts of, NASSA gibba (Buccinum gibbum, Broc.), Italy ; N. Caronis, Al. Brong., Italy ; N. semi-striata, AL Brong., Italy. BUCCINUM asperulum, Broc., Italy ; B. semi-striatum, Broc., Italy ; B. transversale, M. de S. ; B. corrugatum, Broc., Italy ; B. semi-costatum, Broc., Italy ; B. Calmeilii, Payr andean (altogether resembling the species so common in the Mediterranean) ; B. prismaticum, Broc., Italy ; B. Lace- pedii, Payrandeau, C. ; Buccinum, apparently approaching B. gemmulatum of Lamarck, C. ; B. polygonum, Broc., Italy ; B. flexuosum, Broc., Italy ; B. clathratum, Lam., Italy; B. gibbum, Broc., Italy; B. Miga, Lam. (closely approaches the living species) ; B. angulatum, Broc., Italy ; B. re- ticulatum, Lam. (analogous to the existing species), Bordeaux, Italy; B. oli- vaceum, Lam. (apparently an analogue of the living species) ; B. Turbinellus, Broc., Italy ; B. politum, Bast., Bordeaux, Italy ; B. mutabile (completely analogous to the living species), Italy ; B. crenulatum, Lam. (apparently an analogue of the existing species), Italy ; B. Carcassonii, M. de S. ; B. cos- tulatum, Broc., Italy; B. parvulum, M. de S. ; B. gibbosulum, Broc., Italy; B. pusillum, M. de S., Italy. TEREBRA duplicata, Lam. (an analogue of an existing species), Bordeaux, Italy ; T. Vulcani, Al. Brong., Italy ; T. pertusa, Bast., Bordeaux, Italy ; T. dimidiata (an analogue of the existing species) ; T. plicaria. Bast., Bor- deaux. MITRA scrobiculata (Voluta scrobiculata, Broc.), Italy; M. Brocchii, M. de S., Italy ; M. Gervilii, Payrandeau, C. ; M. pyramidella (Voluta py- ramidella, Broc.), Italy. PURPURA Lassaignei, Bast., Bordeaux ; P. bicostalis, Lam. (analogue of the living species) ; P. undata, Lam. (also an analogue of an existing spe- cies.) VOLUTA varricosa, Broc., Italy ; V. piscatoria, Broc., Italy ; V. citha- rella, Al. Brong., Italy ; V. buccinea, Broc., Italy ; V. pyramidella, Broc., Italy; V. tornatilis, Broc., Italy. RISSOA Cimex, Bast., Bordeaux, Italy ; R. cancellata, Lam. ; R. pusilla, (Turbo pusillus, Broc.), Italy; R. cochlearella, Lam., Bordeaux, Italy, Paris. MARGINELLA cyprseola (Voluta cyprseola, Broc.), Italy; M. buccinea (Voluta buccinea, Broc.), Italy. CYPR^A Amygdalum, Broc., Italy ; C. Mus, Lam. (analogous to the living species), C. ; C. Coccinella, Bast., Bordeaux, C. elongata, Broc., Italy, C. ANOPLAX inflata, Al Brong., Italy. OVULA carnea, Lam. (an analogue to the existing species), C. CONUS betulinoides, Lam., Paris ; C. virginalis, Broc., Italy ; C. Pyrula, Broc., Italy; C. avellana, Lam., Italy; C. turricula, Broc., 'Italy ; C. Al- 2 K 498 Organic Remains in Blue Marls of the S. of France. drovandi, Broc., Italy ; C. Pelagicus, J3roc., Italy ; C. Mercati, Bast., Bord., Italy ; C. canaliculatus, Broc., Italy and C. ; C. deperditus, Broc., Bor- deaux, Italy, Paris ; C. mediterraneus, Lam. (analogous to the existing species), C. SIGARETUS costatus (Nerita costata, Broc.), Italy, C. ; S. striatus, M. de S. PILEOPSIS Paretti, M. de S. CALYPTR-EAlaevigata, Desk., Paris; C. muricata (Patella muricata, Broc.), Italy, C. CREPIDULA unguiformis, Bast., Bordeaux, Italy. PATELLA vulgata? Lam. ; P. Bonardii, Payrandeau (analogous to the existing species), C. ; P. Umbella, Lam. (also an analogue), C. ; P. glabra, Desk., Paris. FISSURELLA graeca, Desk. (Patella graeca, Broc.), Italy, Paris. EMARGINULA, a species closely approaching E. fissura of Lamarck, and E. reticulata of Sowerby. AVICULA, species not determined. PERNA mytiloides, Lam., Bordeaux, Italy. LIMA bullata, Payrandeau (analogue) ; L. Breislaki, Bast., Bordeaux, Italy ; L. mutica, Lam., Italy ; L. nivea, (Ostrea nivea, Renieri, Broc.), Italy. PECTEN laticostatus, Lam., Italy and C. ; P. benedictus, Lam., Bordeaux, Italy and C.; P. Plica (Ostrea Plica, Broc.), Italy; P. scabrellus, Bast., Bord., Italy and C. ; P. dubius (O. dubia, Broc.), Italy and C. ; P. multi- radiatus, Bord., Italy ; P. plebeius (O. plebeia, Broc.), Italy ; P. arcuatus (O. arcuata, Broc.), Italy ; P. turgidus, Lam., apparently approaches the species found in the American seas ; P. lepidolaris, Lam., Italy and C. ; P. striatulus, Lam., Italy and C. ; P. striatus (O. striata, Broc.), Italy ; P. in- aequicostalis ? Lam., Italy ; P. Pusio, Lam., Italy and C. ; P. scutularis ? Lam., Italy and C. ; P. unicolor, Lam., Italy and C. ; P. flabelliformis (O. flabelliformis, Broc.), Italy; P. palmatus, Lam., Bordeaux; P. solarium, Lam., Italy; P. terebratulaeformis, M. de S., Italy and C. ; P. Tournalii, M. de S. ; P. Phaseolus ? Lam., Italy ; P. seniensis, Lam., Italy and C. ; P. jacobseoides, M. de S., Italy ; P. pusioides, M. de S., Italy. SPONDYLUS gagderopus, Broc., Bord., Italy ; S. rastellum, Lam., Italy and C. HINNITES Brussonii, M. de S. ; H. Leufroyi, M. de S. PLICATULA, species not determined. OSTREA canalis, Lam., Paris and C. ; O. crassissima, Lam., C. ; O. undata, Lam., Bord., Italy and C.; O. virginica, Al. Brong., Italy; O. edulina, Lam., Italy and C. ; O. colubrina, Lam., Paris ; O. scabrella, M. de S., Italy ; O. anomialis, Lam., Italy, Paris ; O. flabellula, Lam., Bord., Italy, Paris ; O. frondosa, M. de S. ; O. crenulatoides, M. de S. ; O. cristata, Lam., appa- rently analogous to the existing species, Italy ; O. corrugata, Broc., Italy. ANOMIA Ephippium, Broc., analogous to the existing species, Italy ; A. cos- tata, Broc., Bord., Italy ; A., sulcata, Broc., analogous to the species now living in the Mediterranean, Italy; A. radiata, Broc., Italy; A. cepa, Lam., analogue of the existing species, Italy ; A. sinistrorsa, M. de S. ; A. elec- trica, Lam., analogue, Italy and C. ; A. Lens, Lam., closely approaches the living species, Italy and C. ; A. Pellis Ser^entis, Broc., Italy and C. PINNA subquadrivalvis, Italy; P. augustana? Lam. ; P. tetragona, Broc., Italy ; P. pectinata, Lam., approaches the living species. ARCA barbata, Lam., analogue ; A. Gaymardi, Payrandeau, apparently analogous to the living species ; A. antiquata, Lam., analogue of the exist- ing species, Italy ; A. Diluvii, Bast., Bord., Italy and C. ; A. aurita, Broc. t Italy ; A. biangula, Lam., Italy ; A. lactea, Lam., analogue of the living species, Italy; A. Quoyi, Payrandeau, analogous to the living species; A. cardiiformis, Bast., Bordeaux, Italy ; A. Breislaki, Bast., Bord., Italy ; A, pectinata, Broc., Italy ; A. clathrata, Bast., Bordeaux. Organic Remams in Blue Marls of the S. of France. 499 PECTUNCULUS violacescens, Lam., analogous to the living species; P.num- marius (Area nummaria, Broc.), C. ; P. pygmaeus, Lam., Paris; P. sub- concentricus, Lam., Paris; P. pulvinatus, Bord., Italy, Paris, England, and C. NUCULA minuta (Area minuta, Broc.), Italy ; N. pella, Lam., analogue of the living species, Italy ; N. nicobarica, Lam., Italy ; N. rostrata, Lam. (an analogue), Italy; N. margaritacea, Lam., analogous to the existing species, Bord., Italy. MODIOLA discrepans, Lam., C. ; M. Semen, analogous to the existing species; M. subcarinata? Lam.; Mytilus edulis, Bast, (Broc.), Bord., Italy. UNIO, species undetermined. ANODONTA, perhaps many species. CYPRICARDIA, many species, not determined. CARDITA ajar, Lam. ; C. trapezia, Lam., an analogue ; C. sinuata, Pay- ran deau, an analogue. CRASSATELLA latissima, Lam. ISOCARDIA Cor, Lam., exactly resembling the living species, Bord., Italy. CHAMA intermedia, Broc., approaches Cardita of Lamarck, Italy; C. pectinata, Broc., Italy ; C. gryphoides, Lam., analogous to the existing species, Bordeaux, Italy. CARDIUM hians, Broc., Italy ; C. punctatum, Broc., Italy ; C. ciliare, Broc., Italy; C. oblongum? Broc., Italy; C. serratum? Lam., Italy; C. rusticum, Broc., Italy; Cardium, approaching C. tuberculatum, Lam., Italy; C. rhomboides ? Lam., Italy and C. ; C. scrobinatum, Lam., C. ; C. distans ? Lam., Italy ; C. laevigatum, analogous to the existing species, Italy ; C. edule, Broc. (Bast.), an analogue, widely spread, from Antibes to the Py- renees, Italy and C. ; C. glaucum, Bruguiere, an analogue; C. fragile, Broc., Italy ; C. striatulum, Broc., Italy ; C. planatum, Broc., Italy ; C. echi- natum, Broc., Bord, Italy. TELLINA stricta, Broc., Italy ; T. carinulata, Desh., Paris and C. ; T. zonaria, Lam., Bordeaux; T. tenui-stria, Desh., Italy, Paris; T. pellucida, Broc., Italy ; T. rudis, Lam., Paris ; T. subrotunda, Desh., Paris ; T. elegans, Bast., Bordeaux ; T. depressa, Lam., analogous to the existing species, Italy ; T. elliptica, Broc., Italy ; T. strigosa, Lam., analogous to the exist- ing species, Italy; T. compressa, Italy and C. ; T. pulchella, C. ; T. planata, Lam., C. ; T. striatella, Broc., Italy and C. ; T. rostralina, De*h. ; T. nitida, Lam., analogous to the existing species. LUCINA lactea, Lam., analogous to the existing species, C. ; L. Scopu- lorum, Bast., Italy and C. ; L. Saxorum, Desh., Paris ; L. concentrica, Lam., Paris and C. CORBIS lamellosa, Lam., Paris ; C. ventricosa, M. de S. CYRENA, many species, not determined. CYCLAS, perhaps many species. CYPRINA islandicoides, Lam., in the marine sands, the calcaire moellon, the blue marls, and in the supracretaceous basins of Bordeaux, Italy, Paris, and England. CYTHEREA exoleta, Lam., analogous to the existing species ; C. eryci- noides, Bast., Bordeaux ; C. Lincta, Lam., an analogue, Bordeaux, Italy ; C. Chione (Lam.) (Broc.), Italy ; C. elegans, Lam. (Desh.), Paris ; C. eryci- noides, Lam., Bordeaux, Italy ; C. mactroides, Lam. ; C. Cypria ? (Venus Cypria, Broc.), Italy ; C. Deshayesiana, Bast., Bordeaux ; C.nitidula, Lam., Paris ; C. Aphrodite, M. de S., Italy ; C. undata, Bast., Bordeaux ; C. se- misulcata, Lam., Paris ; C. incrassata (Venus incrassata, Broc.), Italy ; C. globulosa?? Dssh., Paris. VENERICARDIA Jouannetii, Bast., Bord.; V. Laurae, Al.Broqg., Italy; V. planicosta, Lam., Paris ; V. pinnula, Bast., Bordeaux. VENUS impressa, M. de S. ; V. angula, M. de S. ; V. senilis, Broc,, Italy ; V. Pullastra, Lam., an analogue, Italy ; V. Dysera, Broc., Bord., Italy and C. ; V. gallina, Lam., an analogue ; V. rugosa, Broc., Italy ; V. cassinoides, 2 K2 500 Organic Remains in Blue Marls of the S. of France. Bast., Bordeaux ; V. Pectunculus, Broc., Italy ; V. radiata, Broc., Italy ; V. circinnata, Broc., Italy ; V. Lupinus, Broc., Italy. DONAX nitida, Lam., Paris ; D. Basterotina, Desk., Paris ; D. Fabagella, Payrandeau, an analogue. MYA conglobata, Broc., Italy. CORBULA revoluta, Bast., Bord., Italy. PETRICOLA, striata, Lam. LUTRARIA elliptica? Lam., Italy; L. piperatra, Lam.', L. solenoides? Lam., Italy. MACTRA triangula, Bast., also in the fahluns of Touraine ; M. crassatella, Lam., England ; M. lactea, C. SOLEN Vagina, Lam., Italy; S. siliqua? Lam., Italy and C. ; S. strigil- latus, Lam., Bord., Italy ; S. candidus, Broc., Bord., Italy ; S. coarctatus, Broc., Italy ; all these species of Solen have their existing analogues. PSAMMOBIA Labordei, Bast., Bordeaux ; Ps. pulchella, Lam., Italy ; Ps. vespertina, Lam., an analogue. PANOP^EA Faujasii, Menard de la Groye, Bord., Italy and C. SANGUINOLARIA, species not determined. GASTROCH^ENA cuneiformis, Lam., analogous to the existing species, C. TEREBRATULA ampulla, M. de S. (Anomia ampulla, Broc.), Italy. PHOLAS Branderi, Bast., Bordeaux. CIRRIPEDA. BALANUS Tintinnabulum, Lam., also in the marine sands and C. ; B. miser? Lam., also in marine sands and C. ; B. semiplicatus ? Lam., also in marine sands and C. ; B. perforatus, Lam., also in marine sands and C. ; B. patellaris, Lam., also in marine sands, C., and in Italy ; all the above Balani are analogues ; B. pustularis, Lam., marine sands, C., Italy ; B. crispatus, Lam., marine sands, C., Italy. ANNUL ATA. SERPULA quadrangularis ? Lam.', S. arenaria, Lam., Italy; S. contortuplicata, Lam.', S. spirorbis? Broc., analogous to the ex- isting species, Italy; S. spirulaea, Lam. ; S. ammonoides, Broc., Italy ; S. annulata ? Lam. ; S. protensa, Lam., analogous to the existing species in the Mediterranean, Italy. DENTALIUM elephantinum, Lam., apparently analogous to the existing species ; D. sexangulum, Broc. ; D. triquetrum, Broc. ; D. entalis, Lam., apparently analogous to the existing species, Italy ; D. coarctatum, Lam., Italy ; D. Tarentinum, Lam., Italy ; D. striatum, Lam., Italy, Paris. RADIARIA. Species of the Echinite family are not stated to occur in the blue marls ; but the following are found, according to M. Marcel de Serres, in the calcaire moellon, or calcareous marls. ECHINUS miliaris, Lam., calc. moel. ; E. granularis ? Lam., perhaps ana- logous to the existing species, calc. moel. SCUTELLA striatula, M. de S., calc. moel. ; S. gibercula, M. de S., calc. moel. GALERITES pustulata, M. de S., calcareous marls. CLYPEASTER altus, Lam., calc. moel., and in Italy; C. marginatus, Lam. r calc. moel., and also Bord., Italy ; C. politus ? Lam., calc. moel., also Italy ; Clypeaster, closely approaching C. oviformis of Lamarck, calc. moel. ; C. excentricus, Lam., calc. moel. ; C. hemisphericus, Lam., calc. moel., also Italy; C. stelliferus, Lam., calc. moel., also Italy; C. gibbosus, M. de S.,^, calc. moel. ; C. scutellatus, M. de S., calc. moel. SPATANGUS canaliferus, Lam., calc. moel. : the specimens of this fossil found highly preserved in the calc. moel. of Barcelona appear quite anacHjp gous to the existing species ; Sp. kevis ? Deluc, calc. moel. ; Sp. arcuarius, Lam., analogous to the existing species, calc. moel. ; Sp. retusus ? Lam., calc. moel. CRUSTACEA. PODOPHTHALMUS Defrancii, Desmarest. (This is the only species noticed by M. Marcel de Serres in the blue marls ; but he states that the Atelecylus rugosus, Desmarest, is found in the calcaire moellon, near Montpellier. Remains of the genus Portunus are also mentioned.) Organic Remains of Bordeaux and Dax. 501 B. Fossil Shells contained in the Supracretaceous Rocks of Bordeaux and Dax, and enumerated by M. De Basterot . NAUTILUS Aturi, Bast., Dax., Houdan ; not considered identical with the N. Pompilius existing in the Eastern seas. LENTICULITES complanata, Defr., Dax, Leognan, Antwerp, Pontoise, Montpellier, and Italy; common at Saucats. NUMMULITES laevigata, Lam., common in many supracretaceous deposits ; N. complanata, Lam., Dax, Soissons ? LYCOPHORIS lenticularis, Defr., common near Bordeaux, Claudiopolis in Transylvania. VAGINELLA depressa, Bosc. Leognan, Saucats. BULLA lignaria, Linn, (var.), analogous to the existing species, Dax, Leognan, Piedmont, England, Paris ; B. cylindrica, Brug., Grignon, Pied- mont, Vienna, Dax, Bordeaux ; B. Utriculus, Broc., analogous to the ex- isting species, Piacenza, Bordeaux, Dax ; B. Lahrella, Per., Dax ; B. cla- thrata, Defr., Dax ; B. truncatula, Brug., analogous to the existing species, environs of Paris, Sienna, Riluogo, Dax. BULLINA Lajonkaireana, Bast., abundant at Saucats, Leognan, and Me"- rignac. HELEX nemoralis, analogous to the existing species, fresh-water limestone at Saucats ; H. variabilis, Drap., analogous to the existing species, fresh- water limestone, Saucats. BULIMUS? terebellatus, Lam., analogous to the existing species, Grignon, Placentine, Dax. PLANORBIS corneus, Drap., analogous to the existing species, Saucats. LYMNEA palustris, Drap., analogous to the existing species, in many su- pracretaceous deposits, Saucats. AURICULA ringens, Lam., analogous to the existing species, Paris, Nice, Italy, Touraine, Bordeaux, Dax ; A. hordeola, Lam., Grignon, Leognan. TORNATELLA sulcata (Auricula sulcata, Lam.), Grignon, Dax, Bordeaux ; T. inflata, Fer., Champagne, Dax ; T. semistriata, Defr., Leognan ; T. punc- tulata, Fer., Leognan, Saucats, Dax ; T. papyracea, Bast., Dax ; T. Dar- gelasi, Bast., Leognan, Saucats. PYRAMIDELLA Mitrula, Fer., Leognan, Me>ignac ; P. terebellata (Auri- cula terebellata, Lam.), Grignon, Volterra, Bordeaux, Dax. TURBO Parkinsoni, Bast., Dax; T. Fittoni, Bast., Dax; T. Lachesis, Bast., common at Bordeaux and Dax. DELPHINULA marginata, Lam., Grignon, Dax ; D. Scobina (Turbo Sco- bina, Al. Brong.), Castelgomberto, Dax, and near Valognes ; D. sulcata, Lam., Grignon, var. at Leognan ; D. trigonostoma, Bast., Dax. TURRITELLA terebralis, Lam., common at Dax, Leognan, and Saucats ; T. Archimedis, Al. Brong. (var. Burdigalensis, Bast.), Ronca, var. at, Bas- sano, var. /3 Anjou, var. y Bordeaux ; T. asperula, Al. Brong., Ronca, Dax ; T. Tun-is, Bast., analogous to the existing species, Dax ; T. quadriplicata, Bast., above the fresh-water limestone at Saucats ; T. cathedralis, Al. Brong., Turin, Leognan, Saucats; T. Proto, Bast., Saucats; T. Desma- restina, Bast., Dax. SCALARIA communis (var.), analogous to the existing species, Placentine, Volterra, Bramerton, Dax ; S. acuta, Sow., Barton, Dax ; S. multilamella, Bast., Parnes, Leognan. CYCLOSTOMA Lemani, Bast., fresh-water limestone, Saucats, Dax, and Tongres, near Maastricht. PALUDINA pusilla (Bulimus pusillus, AL Brong.), analogous to the existing species, Paris, Bordeaux. MONODONTA elegans, Faujas de St. F., L6ognan, rare at Bordeaux ; M. * Description GSologique du Bassin Tertiaire du Sud-Ouest de la France: Mem, dc la Soc. d Hist. Nat. de Paris, t. ii. 502 Organic Remains of Bordeaux and Dax. Modulus, Lam., analogous to the existing species, Dax ; M. Araonis, Bast., analogous to the existing species ? Merignac, Touraine, Dax. TROCHUS Benetti, Sow., Stubbington, Turin, LSognan, Saucats ; T. patn- lus, Broc., Placentine, Bologna, Vienna, Bordeaux, Dax ; T. Boscianus, Al. .Browg., Castelgomberto, Dax ; T. Labarum, Bast., Dax ; T. turgidulus ? Broc., Italy, Merignac ; T. Bucklandi, Bast., above the fresh-water lime- stone, Saucats ; T. Audebardi, Bast., L6ognan. ROTELLA Defrancii, Bast., L6ognan. SOLARIUM, carocollatum, Lam., LSognan, Dax. AMPULLARIA compressa, Bast., Dax; A. crassatina, Lam., Poritchartrain, var. at Dax. MELANIA costellata, Lam., Grignon, Ronca, Sangonini, Dax ; M. subu- lata, Volterra, Leognan, Dax ; M. hordeacea, Lam., Houdan, Pierrelaye, Beauchamp, Isle of Wight, var. at Saucats ; M. clathrata, Bast., Dax; M. nitida, Lam., Grignon, Placentine, Parnes, Dax ; M. distorta (Turbo po- litus, Montagu), analogous to the existing species, Thorigne, Bordeaux. MELANOPSIS Dufourii, Per., Dax. RISSOA Cochlearella (Melania Cochlearella, Lam.}, Grignon, Merignac, var. at Dax ; R. Cimex (Turbo Cimex, Broc.), analogous to the existing species, Bologna, Isle of Ischia, Merigriac, var. at Dax ; R. varicosa, Bast., Merignac ; R. ? Grateloupi, Bast., Merignac. PHASIANELLA turbinoides, Lam., Grignon, M6rignac, Dax; P. Prevostina, Bast., Leognan, Saucats. NATICA Canrena, Broc., analogous to the existing species, Italy, England, Leognan, Saucats, Dax ; N. glaucina, Broc., analogous to the existing spe- cies, Italy, Leognan, Dax. NERITA Plutonis, Bast., Merignac. NERITINA fluviatiKs, Lam., analogous to the existing species, Tuscany, Merignac, Dax (often preserves its colours). CONUS deperditus, Lam. (analogous to the existing species at Owhyhee), Grignon, Ronca, Turin, Bordeaux, Dax ; C. alsiosus, Al. Brong., Ronca, Dax, Bordeaux ; C. Mercati, Vienna, San Miniato, Saucats. CYPR^EA Coccinella, Lam., Grignon, Suffolk, Angers, Nantes, Dax ; Cy. annulus, Broc., analogous to the existing species, Piedmont, Ronca, Bor- deaux ; Cy. annularia, AL Brong., Turin, Bordeaux ; Cy. leporina, Lam., Dax ; Cy. lyncoides, AL Brong., Turin, Bordeaux ; Cy. Duclosiana, Bast., Dax. OLIVA plicaria; Lam., Merignac, L6ognan, Dax, Saucats; O. Clavula, Lam., Merignac, Dax ; O. Dufresnii, Bast., M6rignac, Dax, Saucats. ANCILLARIA canalifera, Lam. (A. turrellata, Sow.), Grignon, Barton, Dax, Bordeaux; A. inflata (Anolax inflata, AL Brong.), Turin, Vienna, Le"ognan, Me"rignac, Dax, Saucats. VOLUTA Lamberti, Sow., analogous to the existing species, Suffolk, Anjou, LSognan ; V. rarispina, Lam., Dax, Bordeaux ; V. affinis, Broc., Ronca, Turin, Leognan. MARGINELLA cyprseola, Placentine, Touraine, Dax. MITRA Dufresnii, Bast., rare at L6ognan ; M. scrobiculata, Placentine, Piedmont, Sienna, var. at Bordeaux ; M. incognita, Bast., Merignac, Dax. CANCELLARIA acutangula (C. acutangularis, Lam.), L6ognau, Saucats ; C. trochlearis, Lam., Leognan, Saucats ; C. doliolaris, Bast., rare, Leognan, C. Geslinii, Bast., Leognan, Saucats ; C. buccinula, Lam., Vienna, Cr6py inValois, Bordeaux; C. contorta, Bast., Italy, Saucats; C. cancellata, Lam., anal, exist, species ; Piedmont, Placentine, Sienna, Bordeaux. BUCCINUM Veneris, Faujas de St. F., Leognan, Saucats ; B. baccatuin, Bast., Saucats, Leognan, Me'rignac, var. et Dax. var. /3 Saucats, var. y Vi- enna ; B. politum, Bast., Piedmont, Saucats. EBURNA spirata (Buccinum spirata, Brug.), anal, exist, species, Rennes, Dax, Saucats. Organic Remains of Bordeaux and Dax. 503 NASSA reticulata (Buccinum reticulatum, Broc.), anal, exist, species, San Miniato, Castel-Arquato, Sienna, var. a, Dax, var. ft Saucats and LSognan ; N. asperula, Broc., Placentine, Sienna, var. Dax, var. ft Lognan and Saucats ; N. angulata, Volterra, Saucats ; N. columbelloides, Bast., Vienna, Angers, Touraine, Dax, Leognan, Saucats (approaches a living species) ; N. Desnoyersi, Bast., Dax, Saucats ; N. cancellaroides, Bast., Dax ; N. An- drei, Bast., Bordeaux. PURPURA costata (Nerita costata, Broc.}, Placentine, Dax, Bordeaux ; P. Lassaignei, Bast., LSognan. CASSIS Saburon (Cassidea Saburon, Brug.}, analogous to the existing spe- cies, Calabria, Placentine, Vienna, L6ognan, Saucats, Dax ; C. Rondel eti, Bast., Leognan, Dax. CASSIDARIA Cythara, Italy, Bordeaux. TEREBRA plicaria, analogous to the existing species, Saucats; T.plicatula, Lam., Grignon, Saucats, Leognan, Dax ; T. cinerea (T. aciculina, Lam.) t analogous to the existing species, Piedmont, L6ognan, Saucats ; T. striata, analogous to the existing species, Saucats ; T. duplicata, Lam., analogous to the existing species ? Sienna, Piedmont, Rome ; T. pertusa (var.), ana- logous to the existing species, Saucats ; T. murina, Dax. CERITHIUM margaritaceum, Al. Brong., Sienna, Mayence; C. corrugatum, Al. Brong., Ronca, Saucats ; C. inconstans, Bast., Saucats ; C. ampullosum, Al. Brong., Castelgomberto, Vienna, M6rignac, Dax ; C. plicatum, Lam., Montpellier, Pontchar train, Mayence, Castelgomberto, Saucats ; C. cinctum, Lamb., Montpellier, Pontchartrain, Beynes, Houdan, Saucats ; C. Charpen- tieri, Bast., Dax; C. papaveraceum, Bast., Touraine, Merignan ; C. lem- niscatum, Al. Brong., Ronca, Dax ; C. Salmo, Bast., L6ognan, Me'rignac ; C. pictum, Bast., Vienna, M6rignac, Saucats ; C. lamellosum, Lam., Cour- tagnon, Grignon, var. Dax; C. angulosum, Grignon, Saucats; C. Diaboli, AL Brong,, the Diablerets, Switzerland, Dax ; C. resectum, Defr., Haute- ville, Dax, Merignac ; C. calculosum, Bast., Dax, Leognan ; C. pupaeforme, Bast., rare, Merignac; C. granulosum (Murex granulosus, Broc.), analo- gous to the existing species, Volterra, Merignac; C. scaber, analogous to the existing species, Italy, Merignac, Leognan. MUREX Pomum, Linn.,, analogous to the existing species, Placentine, Saucats, Merignac; M. sublavatus, Bast., rare, Merignac, Leognan, Saucats; M. Lingua-Bovis, Bast., Saucats, Leognan; M. suberinaceus, Bast., Bor- deaux. TYPHIS tubifer (Murex tubifer, Lam.), analogous to the existing species, Grignon, Barton, Highgate, Leognan. TRITON doliare (Murex doliaris, Broc.), Placentine, Pisa, Sienna, L6ognan. RANELLA marginata, Al. Brong., Piedmont, Pisa, Placentine, Volterra, Turin, Leognan, Merignac ; R. leucostoma, Lam., analogous to the existing species, Placentine, Bordeaux. Fusus lavatus, Sow., Barton, Paris, Leognan, Saucats, Dax ; F. bucci- noides, Bast. (Buccinum subulatum, Broc.), Placentine, Saucats, Merig- nac (a Mediterranean shell approaches this species); F. rugosus, Lam., Grignon, Valognes, Dax (different from the Fusus rugosus of Sowerby) ; F. clavatus, Placentine, var. Bordeaux. PLEUROTOMA tuberculosa, Bast., Vienna, Saucats, Leognan ; P. Pannus, Bast., Leognan, Saucats, Dax ; P. denticulata, Bast., Touraine, Saucats, Leognan, Merignac, Dax ; P. ramosa, Bast., Tnorigne", Angers, Leognan, Saucats; P. Borsoni, Bast., Saucats, Leognan, Merignac; P. plicata, Lam., Grignon, Dax ; P. undata, Lam., Grignon, Epernay, Dax ; P. multinoda, Lam., Bordeaux ; P. Turrella, Lam., Grignon, var. Dax ; P. crenulata, Lam., Grignon, var. Leognan ; P. cataphracta, Placentine, Sienna, Bologna, Bordeaux ; P. purpurea, Bast., analogous to the existing species, Leognan ; P. terebra, Bast., Leognan, Saucats, Dax ; P. costellata, Lam., Grignon, Leognan, Dax ; P. cheilotoma, Bast., Bordeaux. 504? Organic Remains of Bordeaux and Dax. FASCIOLARIA Burdigalensis, Defr., Leognan, Saucats, Merignac; F. uni- plicata (Fusus uniplicatus, Lam.), Grignon, Epernay, Dax. PYRULA condita, Al. Brong., Turin, Leognan, Saucats; P. Clava, Bast., Dax, Bordeaux ; P. Lainei, Bast., Saucats, Leognan, Merignac, Dax ; P. Melongena, analogous to the existing species, Courtagnon, Saucats, Dax, Merignac ; P. rusticula, Bast., Dax, Bordeaux. TURBINELLA Lynchi, Bast., rare, Leognan. STROMBUS decussatus, Dax ; S. Bonelli, AL Brong., Turin, Dax. ROSTELLARIA Pcs-Pelicam, analogous to the existing species, common in many supracretaceous deposits, Leognan, Dax ; R. curvirostris, Lam., ana- logous to the existing species, Dax. SIGARETUS canaliculatus, Sow. Hordwell, Paris, Bordeaux, Dax. CAPULUS (Pileopsis) sulcosus (Nerita sulcosa, Broc.), Placentine, Merig- nac. CREPIDULA unguiformis (C. Italica, Defr.), analogous to the existing spe- cies, Placentine, Sienna, Vienna, Saucats ; C. cochlearia, Bast., analogous to the existing species, Merignac. FISSURELLA costaria, Desk,, Grignon? Dax. CALYPTR^EA deformis, Lam., Dax, Bordeaux ; C. depressa, Lam., abun- dant at Bordeaux ; C. muricata (Patella muricata, Broc.}, analogous to the existing species, Piedmont, Placentine, Castel-Arquato, Leognan, Saucats ; C. ornata, Bast., Dax. HIPPONYX granulatus, Bast., Dax. OSTREA flabellula, Lam., Grignon, Hordwell, Barton, Brussels, Saucats, L6ognan ; O. undata, Lam., Dax, Bordeaux; O. Cyrnbula, Lam., Grignon, Barton, Saucats. PECTEN scabrellus, Lam. (Ostrea dubia, Broc.), Val Andone, Piedmont, Saucats ; P. Burdigalensis, Lam., Saucats, approaches P. Pleuronectes, P. obliteratus, and P. Laurenti ; P. multiradiatus, Lam., Italy, Saucats. SPONDYLUS, fragments. PERNA Ephippium, Lam., analogous to the existing species, Bordeaux. AVICULA phalaenacea, Lam., Leognan. PINNA, fragments. ARCA biangula, Lam., Grignon, Le"ognan ; A. scapulina, Lam., Grignon, Merignac ; A. clathrata, Defr., analogous to the existing species, Angers, Thorigne, Nice, Merignac ; A. Diluvii, Lam. (A. Pectinata, Broc.), Hou- dan, Touraine, Placentine, Sienna, Turin, Vienna, Bordeaux ; A. Breislaki, Bast., Dax. PECTUNCULUS Cor, Lam., Saucats, Merignac, Leognan ; P. pulvinatus, Lam., Paris, Touraine, var. Dax and Bordeaux, var. /3 Leognan ;M.de Basterot considers this shell the same with that found at Walton. NUCULA emarginata, Lam., Leognan, Saucats ; N. margaritacea, Lam. (Area Nucleus, Broc.), analogous to the existing species, Grignon, Placen- tine, Barton, Highgate, Leognan, Dax. MYTILUS antiquorum, Sow., Suffolk, var. Saucats, Merignac ; M. Brardii, AL Brong., Mayence, Dax, Merignac ; M. edulis, Linn., analogous to the existing species, Piedmont, Placentine, Sienna, Volterra, Saucats. MODIOLA cordata, Lam., Paris, Domfront, Saucats. CARDITA hippopea, Bast., Saucats. VENERICARDIA Pinnula, Bast., beds above the fresh-water limestone, Saucats, Dax ; V. Jouanneti, Bast., Italy, Vienna, Bordeaux ; V. inter- media (Cardita intermedia, Lam.), Placentine, Sienna, Dax. ERYCINA elliptica, Lam., Ecouen, Senlis, Saucats. CHAM A gryphoides, Broc., analogous to the existing species, Piedmont, Placentine, Sienna, Dax, Leognan, Saucats, Merignac. CARDIUM edule, Linn., analogous to the existing species, Placentine, Piedmont, Sienna, Bramerton, Ipswich, Dax; C. Burgalinum, Lam., Dax, Bordeaux ; C. serrigerum, Lam., Grignon, Bordeaux ; C. echinatum, Organic Remains of Bordeaux and Dax. 505 Brug., analogous to the existing species, Placentine, Touraine, Bordeaux, var. Vienna ; C. Pallassianum, Bast., Dax ; C. multicostatum, Broc., Pla- centine, var. Leognan ; C. discrepans, Bast., bed above the fresh-water limestone, Saucats, Dax. DONAX anatinum, Lam., analogous to the existing species ; var. Dax, Bordeaux ; D. elongata, Lam., analogous to the existing species, Merignac ; D. triangularis, Bast., approaches an existing species, Saucats; D. irregularis, Bast., Dax; D.? difficilis, Bast., Dax. CYRENA Brongniartii, Bast., Ronca, Merignac, Saucats; C. Sowerbii, Bast., Paris, Saucats. TELLINA zonaria, Lam., Dax, Saucats, Leognan, Merignac (preserves its colours) ; T. elegans, Desk., Hauteville, Grignon, above the fresh-water beds, Saucats j T. bipartita, Bast., Saucats ; T. biangularis, Desk., Paris, var. Dax. LUCINA Columbella, Lam., Touraine, Leognan, Saucats, Dax, Merignac ; L. divaricata, Lam., analogous to the existing species, Grignon, Leognan, Merignac, common at Hordwell and Saucats ; L. scopulorum, Al. Brong., Ronca, Turin, Merignac, Saucats ; L. dentata, Bast., Dax, Saucats ; L. di- gitalis, Lam., analogous to the existing species, rare, Saucats; L. hiatelloides, Bast., rare, Leognan ; L. gibbosula, Lam., analogous to the existing species ; Grignon, Dax ; L. renulata, Lam., analogous to the existing species, Grig- non, Bordeaux ; L. neglecta, Bast., Dax, Bordeaux. VENUS Dysera, Linn., analogous to the existing species, Piedmont, Pla- centine, Dax, Saucats; V. casinoides, Lam., Vienna, L6ognan, Saucats; V. vetula, Bast., Saucats, Leognan ; V. radiata, Broc., analogous to the existing species, Italy, Saucats, Leognan, Dax. CYTHEREA erycinoides, Lam., analogous to the existing species, Paris, Turin, Rome, Saucats, Leognan, Dax; C. Deshayesiana, Bast., Saucats, Leognan ; C. tincta, Lam., perfect resemblance to existing species, Saucats, C. leonina, Bast., Leognan, Saucats ; C. undata, Bast., Saucats, abundant at Merignac ; C. nitidula, Lam. (Venus transversa, Sow.), Grignon, Barton, Saucats. CYPRINA Islandicoides, Lam. (Venus aequalis, Sow.), Suffolk, Placentine, Antwerp, Dax, Bordeaux. VENERUPIS Faujasii, Bast., Bordeaux. PETRICOLA peregrina, Bast., in large madrepores, Merignac. SAXICAVA anatina, Bast., in the holes which it has bored in the fresh-water limestone, when the latter was covered by the waters of an ancient sea, Saucats. CLOTHO? unguiformis, Bast., in holes which it has pierced in the marine and fresh-water limestones, Saucats. CORBULA revoluta. Sow., Barton, Italy, Dax, Leognan, Merignac, Saucats; C. striata, Lam., Grignon, var. Angers, var. Bordeaux. MACTRA striatella. Lam., analogous to the existing species ? Saucats ; M. deltoides, Lam., Grignon, Saucats; M. triangula, Broc., analogous to the existing species, Placentine, Saucats. LUTRARIA Sanna, Bast., Saucats. MYA ornata, Bast., Dax. PANOP^EA Faujasii, Mesnard de la Groye, Parma, Sienna, Pisa, San Mi- niato (Reggio), Placentine, Piedmont, Leognan. PSAMMOBIA Labordei, Bast., approaches Ps. vespertina, Leach, a living species, Saucats. SOI.EN strigillatus, Linn., analogous to the existing species, Placentine, Piedmont, Vienna, Grignon, Leognan, Dax ; S. Vagina, Linn., analogous to the existing species, Placentine, Grignon, Saucats; S. Legumen, Linn., Saucats. PHOLAS Branderi, Bast., in the rolled stones and corals, Touraine and Merignac. CLAVAGELLA coronata, Desk., Meaux, Pauliac, nine leagues from Bordeaux. 506 Gosau Fossils. C. Gosau Fossils. The following is a list, according to Prof. Sedgwick and Mr. Murchison, of the organic remains detected by them either in the Gosau deposit, or its equivalents in the Alps. (G., Gosau; Z., Zlam; W., Flanks of the Wand; M., Marzoll; R., HinterReutter; T., Bavarian Traunstein.) Zoophyta. Tragos. Nullipora. Madrepora. Cellepora, G. Lithodendron granulosum, Gold/., G. (also Castell' Arquato, Supracretaceous). Fungia radiata, Gold/., G. Fung, discoidea, G. Fung, poiymorpha, Gold/., G., Z. (also Bassano and Dauphine, Supracretaceous). Fung, undulata, Gold/., G. Diploctenium cordatum, Gold/., G. (also Maestricht.) Turbinolia compla- nata, Gold/., G. Turb. duodecimcostata, G. (also Castell' Arquato, Supra- cretaceous). Turb. lineata, Goldf.,G. Turb. cuneata, Gold/., G. (also Castell' Arquato, Supracretaceous). Turb. aspera, Sow., G. Cyathophyllum rude, Sow., G. Cy. compositum, Sow., G. Meandrina agaricites, Gold/., G. Astrea striata, Gold/., G. Ast. formosa, Gold/., G. Ast. reticulata. Gold/., G. Ast. agaricites, Gold/., G. Ast. grandis, Sow., G. Ast. media, Sow., G. Ast. formosissima, Sow., G. Ast. ambigua, Sow., G. Ast. tenera, Sow., G. Ast. ramosa, Sow., G. Annulata. Serpula. Conchifera. Teredo, G. Solen, G. Panopaea plicata? Sow., G. (also Sandgate, Lower Green Sand). Anatina, G. Crassatella impressa, Sow., G. Corbula angustata, Sow., G. Sanguinolavia Hollowaysii ? ? Sow., G. (also Bracklesham Bay, London Clay). Lucina, G. Astarte macrodonta, Sow., G. Cyclas cuneiformis? Sow., G., W. (also Woolwich, Plastic Clay). Cytherea laevigata, Lam., G. (also Grignon, Calcaire Grossier). Venus, G. Vene- ricardia, G. Cardium productum, Sow., G., M. Isocardia, G. Cucullaea carinata, Sow., G. (Blackdown, Green Sand). Area, G. Pectunculus Plumsteadiensis, Sow., G. (also Plumstead, near Woolwich, Plastic Clay). Pect. brevirostris, Sow., G. (also Bognor, London Clay). Pect. pulvinatus, Lam., G. (also Grignon, Bracklesham, Calcaire Grossier, and London Clay). Pect. calvus, Sow., G., M., W. Nucula amygdaloides, Sow., G. (also South- end, London Clay). Nuc. concinna, Sow., G., R. Trigonia aleeformis, Sow., G. (also Parham Park and Black Down, Green Sand). Modiola, G. Ino- ceramus Cripsii, Mant., G., W. (also Hamsey, Chalk Marl). Avicula, G. Pecten quinquecostatus, Sow., G. (common cretaceous fossil). Plicatula aspera, Sow., G., W. Gryphsea elongata, Sow., G. Gryphsea expansa, Sow., G. Exogyra, G. Ostrea, G. Terebratula dimidiata? Soiv., G. (also Haldon Hill, Green Sand). Axinus? G., W. Trigonellites, G., W. Mollusca. Dentalium grande? Desk., G., M. (also Calcaire Grossier). Calyptrsea? G. Auricula decurtata, Sow., G. Aur. simulata, Sow., G., M. (also Barton Cliff, London Clay). Melania, G. Melanopsis, G. Natica Ambulacrum, Sow., G. (also Barton Cliff, London Clay). Nat. lyrata, Sow., G. Nat. angulata, Sow., G. Nat. bulbiformis, Sow., G., Z. Nerita, G. Solarium quadratum, Sow., G. Trochus spiniger, Sow. Turbo arenosus, Sow., G. Turritella angusta, Desk., G. Tur. biformis, Desk. G., T. Tur. rigida, Desk., G. Tur. laeviuscula, Desk., G. Tornatella gigantea, Desk., G., Z., Meyersdorf, &c. Torn. Lamarckii, Desk., Gams-Gebirge. Nerinea flexuosa, Desk., G. Cerithium reticosum, Desk., G. Cer. corioideum, Desk., G., Z., T. Cer. pustulosum, Desk., G. Pleurotoma prisca, Sow., G., M. (also Barton Cliffj London Clay). Pleur. fusiforme, Sow., G. Pleur. spi- nosum, Sow., G. Fasciolaria elongata, Sow., G. Fusus intortus, Lam., G. (also Grignon and Ronca, Supracretaceous). Fus. heptagonus, Sow., G. Fus. carinella, Sow., G. Fus. muricatus, Sow., G. Fus. abbreviatus, Sow., G. Fus. cingulatus, Sow., G. Rostellaria plicata, Sow., G. Rost. costata, Sow., G. Rost. granulata, Sow., G., M. Rost. laevigata, Sow., G. Nassa carinata, Sow., G. Nas. affinis, Sow., G. Mitra pyramidella? Broc., G. (Supracretaceous). Mit. cancellata, Sow., G: Voluta coronata? Broc., G. Organic Remains of the Cretaceous Group. 507 (also Supracretaceous). Vol. citharella? Al. Brong., G. (also Turin, Supra- cretaceous). Vol. acuta, Sow., G. Terebra coronata, Sow., G. Volvaria laevis, Sow., G. Baculites or Hamites, G. Organic Remains of the Cretaceous Group. PLANTJE. Conferva. 1. Confervites fasciculata, Ad. Brong. Arnager, Bornholm, Ad, Brong. Chalk, Sussex, Mant. 2. aegagropiloides, Ad. Brong. Arnager, Bornholm, Ad. Brong. , species not determined. Chalk, Sussex, Mant. Algce. 1. Fucoides Orbignianus, Ad. Brong. Isle d'Aix, Rochelle, Ad. Brong. 2. strictus, Ad. Brong. Isle d'Aix, Rochelle, Ad. Brong. 3. tuberculosus, Ad. Brong. Isle d'Aix, Rochelle, Ad. Brong. 4. difformis, Ad. Brong. Bidache, Bayonne, Ad. Brong. 5. intricatus, Ad. Brong. Bidache, Ad. Brong. 6. Lyngbianus, Ad. Brong. Arnager, Bornholm, Ad. Brong. 7. Brongniarti, Mant. Chalk, Sussex, Mant. 8. Targioni, Ad. Brong. Chalk, Sussex, Mant. 9. canaliculatus, Ad. Brong., Env. of Bayonne ; Bidache ; Green Sand, Rochefort, Dufr. , species not determined. Chalk, Gault, Sussex, Mant. Na'iades. 1. Zosterites cauliniasfolia, Ad. Brong. Isle d'Aix, Ad. Brong. 2. lineata, Ad. Brong. Isle d'Aix, Ad. Brong. Bellovisana, Ad. Brong. Isle d'Aix, Ad. Brong. 4. elongata, Ad. Brong. Isle d'Aix, Ad. Brong. Cycadeee. 1. Cycadites Nilssonii, Ad. Brong., Chalk, Scania. 1. Thuites aliena, Sternb. Smetschna (Rakonitzer Kreis), G. T. Dicotyledonous wood, perforated by some boring shell; Chalk, Sussex, Mant. ; Green Sand, Lyme Regis, De la B. Cones of Coniferae, Green Sand, Lyme Regis, De la B. Green Sand? Kb'pinge, Scania, Nils. Ferns? Green Sand, Lyme Regis, De la B. leaves, between Platanus and Lyriodendron, Sternb. Green Sand, Tetschen; Blankenburg ; Wernigerode; Quedlinburg, G. T. ZOOPHYTA. 1. Achilleum glomeratum, Goldf. Maestricht, Goldf. 2. fungiforme, Goldf. Maestricht, Goldf. 3. Morchella, Goldf. Cretaceous Rocks, Essen, Westphalia, Sack. 1 . Manon capitatum, Goldf. Maestricht, Goldf. 2. tubulifemm, Goldf. Maestricht, Goldf. 3. pulvinarium, Goldf. Maestricht; Essen, Westphalia, Goldf. 4. Peziza, Goldf. Maestricht ; Cretaceous Hocks, Essen, West- phalia, Goldf. 5- stellatum, Goldf. Cretaceous Rocks, Essen, Goldf. pyriforme, Goldf. Chalk, Coesfeld, Goldf. 7- verticillites, . Maestricht, G. T. 1. Scyphia mammillaris, Goldf. Essen, Westphalia, Goldf. 2. furcata, Goldf. Cretaceous Rocks, Essen, Goldf. 508 Organic Remains of the Cretaceous Group. 3. Scyphia infundibuliformis, Goldf. Essen, Goldf. 4. foraminosa, Goldf. Cretaceous Rocks, Essen, Goldf. 5. Sackii, Goldf. Essen, Westphalia, Sack. 6. tetragona, Goldf. Essen, Goldf. 7. fungiformis, Goldf. Green Sand, Coesfeld, Westphalia, Goldf. 8. Mantellii, Goldf. Coesfeld, Goldf. 9. Dechenii, Goldf. Coesfeld, Goldf. 10. Oeynhausii, Goldf. Green Sand, Darup, Westphalia, Goldf. 11. Murchisonii, Goldf. Darup, Goldf. 12. verticillites, Goldf. Maestricht; Chalk, Nehou, Goldf. 1. Spongia ramosa, Mant. Chalk, Sussex, Mant. ; Chalk? Yorkshire, Phil. ; Noirmoutier, Al. Brong. 2. lohata, Flem. Chalk, Sussex, Mant. 3. plana, Phil. Chalk, Yorkshire, Phil. 4. capitata, Phil. Chalk, Yorkshire, Phil. 5. osculifera, Phil. Chalk, Yorkshire, Phil. 6. convoluta, Phil. Chalk, Yorkshire, Phil. 7. marginata, Phil. Chalk, Yorkshire, Phil. 8. radiciformis, Phil. Chalk, Yorkshire, Phil. 9. terehrata, Phil. Chalk, Yorkshire, Phil. 10. Isevis, Phil. Chalk, Yorkshire, Phil. 11. porosa, Phil. Chalk, Yorkshire, Phil. 12. cribrosa, Phil. Chalk, Yorkshire, Phil. 1. Spongus Townsendi, Mant. Chalk, Sussex, Mant.; Chalk, Rouen, Pas. 2. lahyrinthicus, Mant. Chalk, Sussex, Mant. ; Chalk, Rouen, Pas. 1. Tragos Hippocastanum, Goldf. Maestricht, Goldf. 2. deforme, Goldf. Cretaceous Rocks, Essen, Goldf. 3. rugosum, Goldf. Cretaceous Rocks, Essen, Westphalia, Sack. 4. pisiforme, Goldf. Cretaceous Rocks, Essen, Westphalia, Goldf. 5. stellatum, Goldf. Cretaceous Rocks, Essen, Goldf. 1. Alcyonium globulosum, Defr. Chalk, Beauvais; Meudonj Amiens; Tours ; Gien ; Baculite Limest., Normandy, Desn. ? pyriforme, Mant. Chalk, Sussex, Mant. , species not determined. Chalk, Sussex, Mant. ; Upper Green Sand, Warminster, Lons. 1. Choanites subrotundus, Mant. Chalk, Sussex, Mant. 2. Konigi, Mant. Chalk, Sussex ; Warminster, Mant. 3. flexuosus, Mant. Chalk, Sussex, Mant. 1. Ventriculites radiatus, Mant. Chalk, Sussex, Mant.; Chalk, Moen, Al. Brong. 2. alcyonoides, Mant. Chalk, Sussex ; Warminster, Mant. 3. Benettiae, Mant. Chalk, Sussex, Mant. ; Chalk, Yorkshire, Phil, 1. Siphonia Websteri, Mant. Chalk, Sussex, Mant. 2. cervicornis, Goldf. Chalk, Haldern, Westphalia, Goldf. 3. Ficus, Goldf. Green Sand, Quedlinburg, Goldf. 4. punctata, Goldf. Green Sand, Goslar, G. T. 1. Hallirhoa costata, Lam*. Green Sand, Normandy, De la B. ; Upper Green Sand, Warminster, Lons. 1. Serea pyriformis, Lam. Green Sand, Normandy, Al. Brong. 1. Gorgonia bacillaris, Goldf. Maestricht, Goldf. 1. Nullipora racemosa, Goldf. Maestricht, Goldf. 1. Millepora Fittoni, Mant. Chalk, Sussex, Mant. 2. Gilberti, Mant. Chalk, Sussex, Mant. 3. antiqua? Defr. Baculite Limest., Normandy, Desn. 4. madreporacea, Goldf. Maestricht, Goldf. 5. compressa, Goldf. Maestricht, Goldf. species not determined. Chalk, Meudon, Al. Brong. 1. Eschara cyclostoma, Goldf. Maestricht, Goldf. Kjuge, Sweden, His. Organic Remains of the Cretaceous Group. 509 2. Eschara pyriformis, Goldf. Maastricht, Goldf. 3. stigmatophora, Goldf. Maastricht, Goldf. 4. sexangularis, Goldf. Maastricht, Goldf. 5. cancellata, Goldf. Maastricht, Goldf. 6. arachnoidea, Goldf. Maastricht, Goldf. 7. dichotoma, Goldf. Maastricht, Goldf. 8. striata, Goldf. Maastricht, Goldf. 9. filograna, Goldf. Maastricht, Goldf. 10. disticha, Goldf. Meudon, Goldf. 1. Cellepora ornata, Goldf. Maastricht, Goldf. 2. Hippocrepis, Goldf. Maastricht, Goldf. 3. Velamen, Goldf. Maastricht, Goldf. 4. dentata, Goldf. Maastricht, Goldf. 5. crustulenta, Goldf. Maastricht, Goldf. 6. bipunctata, Goldf. Maastricht, Goldf. 7. escharoides, Goldf. Cretaceous Rocks, Essen, Westphalia, Goldf, 1. Coscinopora infundibuliformis, Goldf. Coesfeld, G. T. 2. macropora, Goldf. Stormede, Miinster, G. T. 1. Retepora clathrata, Goldf. Maestricht, Goldf. 2. Kchenoides, Goldf. Maestricht, Goldf. 3. truncata, Goldf. Maestricht, Goldf. 4. disticha, Goldf. Maestricht, Goldf. 5. cancellata, Goldf. Maestricht, Goldf. 1. Flustra utricularis, Lam. Chalk, Sussex, Mant. 2. ? reticulata, Desm. Baculite Limestone, Normandy, Desn. 3. flabelliformis, Lam. Baculite Limestone, Normandy, Desn. , species not determined. Chalk, Sussex, Mant. 1. Cceloptychium lobatum, Goldf. Chalk, Coesfeld, Goldf. 2. ' acaule, Goldf. Maestricht ; Munster, Goldf. 3. agaracoides, Goldf. Coesfeld, G. T. \. Ceriopora micropora, Goldf. Maestricht, Goldf.; Essen, G. T. 2. cryptopora, Goldf. Maestricht, Goldf. 3. anomalopora, Goldf. Maestricht, Goldf. 4. dichotoma, Goldf. Maestricht, Goldf. 5. milleporacea, Goldf. Maestricht, Goldf. ; Morby ; Kjuge, &c. Sweden, His. 6. madreporacea, Goldf. Maestricht, Goldf. 7. tubiporacea, Goldf. Maestricht, Goldf. ; Kjuge, &c., Sweden, His. 8. verticillata, Goldf. Maestricht, Goldf. 9. spiralis, Goldf. Maestricht, Goldf. 10. pustulosa, Goldf. Maestricht, Goldf. 11. compressa, Goldf. Maestricht, Goldf. 12. stellata, Goldf. Maestricht; Cretaceous Rocks, Essen, Goldf. 13. Diadema, Goldf. Maestricht, Goldf. 14. polymorpha, Goldf. Cretaceous Rocks, Essen, Westphalia, 15. gracilis, Goldf. Cretaceous Rocks, Essen, Goldf. 16. spongites, Goldf. Cretaceous Rocks, Essen, Goldf. 17. clavata, Goldf. Essen, Westphalia, Goldf. 18. trigona, Goldf. Cretaceous Rocks, Essen, Goldf. 19. Mitra, Goldf. Cretaceous Rocks, Essen, Goldf. 20. venosa, Goldf. Cretaceous Rocks, Essen, Goldf. 21. cribrosa, Goldf. Cretaceous rocks, Essen, Goldf. 1. Lunulites cretacea, Defr. Maestricht; Tours; Baculite Limestone, Normandy, Desn. 1. Orbulites lenticulata, Lam. Chalk, Sussex, Mant.; Green Sand, Perte du Rhone, Al. Brong. ; Chalk, Bray r Normandy, Pas. 510 Organic Remains of the Cretaceous Group. 1. Lithodendron- gibbosum, Munst. Green Sand, Bochum, Westphalia, Munst. 2. gracile, Goldf. Green Sand, Quedlinburg, Goldf. 1. Caryophyllia centralis, Mant. Chalk, Sussex, Mant. ; Chalk, Yorkshire, Phil.; Baculite Limestone, Normandy, Desn. ; Chalk, Du- clair; Dieppe, Pas. 2. Conulus, Phil. Speeton Clay, Yorkshire, Phil. 1. Antophyllum proliferum, Goldf. Faxoe, G. T. 1. Turbinolia mitrata, Goldf. Aix-la-Chapelle, Goldf. 2. Kcenigi, Mant. Gault, Sussex, Mant. 1. Fungia radiata, Goldf. Cretaceous Sand, Aix-la-Chapelle, Goldf. 2. cancellata, Goldf. Maestricht, Goldf. 3 coronula, Goldf. Cretaceous Rocks, Essen, Westphalia, Goldf. 1. Chenendopora fungiformis, Lam. Upper Green Sand, Warminster, Lons. ; Havre ; Rouen, Pas. 1. Hippalimus fungoides, Lam. Upper Green Sand, Warminster, Lons. 1. Diploctenium cordatum, Goldf. Maestricht, Goldf. 2. Pluma, Goldf. Maestricht, Goldf. 1. Meandrina reticulata, Goldf. Maestricht, Goldf. 1. Astrea flexuosa, Goldf. Maestricht, Goldf. 2. geometrica, Goldf. Maestricht, Goldf. 3. clathrata, Goldf. Maestricht, Goldf. 4. escharoides, Goldf. Maestricht, Goldf. 5. textilis, Goldf. Maestricht, Goldf. 6. velamentosa, Goldf. Maestricht, Goldf. 7. gyrosa, Goldf. Maestricht, Goldf. 8. elegans, Goldf. Maestricht, Goldf. 9. angulosa, Goldf. Maestricht, Goldf. 10. geminata, Goldf. Maestricht, Goldf. 11. arachnoides, Schroter. Maestricht, Goldf. 12. Rotula, Goldf. Maestricht, Goldf. 13. macrophthalma, Goldf. Maestricht, Goldf. 14. muricata, Goldf. Chalk, Meudon, Goldf. 15. stylophora, Goldf. Meudon, Goldf. 1. Pagrus Proteus, Defr. Meudon; Tours; Baculite Limestone, Nor- mandy, Desn. Polypifers, genera not determined. Green Sand, Grand Chartreuse, Beaum.; Green Sand, Maritime Alps, De la B. ; Lower Green Sand, Isle of Wight, Seda. ; Gourdon, S. of France, Dufr. RADIARIA. 1. Apiocrinites ellipticus, Miller. Chalk, Sussex, Mant. ; Chalk, Yorkshire, Phil. ; Chalk, Touraine ; Baculite Limestone, Normandy, Desn. ; Westphalia ; Maestricht, Goldf. ; Chalk, Dieppe, Pas. 1. Pen tacrinites, species not determined. Chalk, Sussex, Mant. ; Speeton Clay, Yorkshire, Phil. 1. Marsupites ornatus, Miller. Chalk, Sussex, Mant. ; Chalk, Yorkshire, Phil. 1. Glenotremites paradoxus, Goldf. Marly Chalk, Speldorf, between Duis- berg and Muhlheim, Goldf. 1. Asterias quinqueloba, Goldf. Chalk, North Fleet ; Chalk, Maestricht ; Rinkerode near Miinster, Goldf. , species not determined. Chalk, Paris ; Rouen ; Al. Brong. ; Baculite Limstone, Normandy, Desn. ; Chalk, England. 1. Cidaris cretosa, Mant. Chalk, Sussex, Mant. 2. variolaris, Al. Brong, Chalk, Sussex, Mant. ; Green Sand, Organic Remains of the Cretaceous Group. 511 H&vre; Green Sand, Perte du Rhone, Al. Brong. ; Cretaceous Rocks, Koesfeld and Essen, Westphalia ; Cretaceous Rocks, Saxony, Goldf. 3. Cidaris claviger, Konig. Chalk, Sussex, Mant. ; Rouen, Pas. 4. vulgaris. Chalk, Poland, Al. Brong. 5. regalis, Goldf. Maastricht, Goldf. 6. vesiculosa, Goldf. Cretaceous Rocks, Essen, Westphalia, Goldf. 7. scutiger, Munst. Cretaceous Rocks, Kelheim, Bavaria, Goldf. 8. crenularis, Lam. Chalk, France, Goldf. 9. granulosa, Goldf. Chalk, Aix-la-Chapelle ; Maestricht ; Cre- taceous Rocks, Essen, Westphalia, Goldf. , species not determined. Chalk, Speeton Clay, Yorkshire, Phil. 1. Echinus regalis, Hwninghaus. Cretaceous Rocks, Essen, Westphalia, Goldf. 2. alutaceus, Goldf. Cretaceous Rocks, Essen, Goldf. 3. granulosus, Munst. Cretaceous Sandstone, Kelheim, Bavaria, Munst. 4. areolatus, Wahl. Balsberg, Scania, Nils. Green Sand, Wilts; Lyme Regis, Konig. 5. Benettiae, Konig. Green Sand, Chute, Wilts, Konig. , species not determined. Green Sand, M. de Fis, Al. Brong. ; Baculite Limestone, Normandy, Desn. ; Upper Green Sand, Warminster, Lons. 1. Galerites albo-galerus, Lam. Chalk, Sussex, Mant.; Chalk, Yorkshire, Phil ; Chalk, Dieppe, Al. Brong. ; Chalk, Quedlinberg and Aix-la-Chapelle, Goldf. Chalk, Lublin, Poland, Pusch. Chalk, Lyme Regis, De la B. 2. vulgaris, Lam. Chalk, Sussex, Mant. ; Chalk, Dreux, &c., AL Brong. ; Quedlinberg ; Aix-la-Chapelle, Goldf. Chalk, Lyme Regis, De la B. 3. subrotundus, Mant. Chalk, Sussex, Mant. ; Chalk, Yorkshire, Phil. 4. Hawkinsii, Mant. Chalk, Sussex, Mant. 5. abbreviatus, Lam. Cretaceous Rocks, Quedlinberg ; Aix-la- Chapelle, Goldf. 6. canaliculatus, Goldf. Cretaceous Rocks, Biiren and Brencken, Westphalia, Goldf. 7. Subuculus, Linnteus. Cretaceous Rocks, Koesfeld and Essen, Westphalia, Goldf.; Havre, Pas. 8. sulcato-radiatus, Goldf. Maestricht, Goldf. 9. ? depressus, Lam. Green Sand, M. de Fis, Al. Brong. , species not determined. Chalk, Upper Green Sand, Warmin- ster, Lons. Clypeus, species not determined. Upper Green Sand, Warminster, Lons. 1. Clypeaster Leskii, Goldf. White Chalk, Maestricht, Goldf. 2. fornicatus, Goldf. Cretaceous Rocks, Mimster, Westphalia, Goldf. 3. oviformis, Lam. Green Sand, Mans, Desn. 1. Echinoneus subglobosus, Goldf. Maestricht, Goldf. 2. Placenta, Goldf. Maestricht, Goldf. 3. Lampas, De la B. Green Sand, Lyme Regis, De la B. 4. peltiformis, Wahl. Balsberg, Scania, Wahl. 1. Nucleolites Ovulum, Lam. Maestricht, Goldf.; Rouen, Pas. 2. scrobicularis, Goldf. Maestricht, Goldf. 3. Rotula, Al. Brong. Chalk, Rouen ; Green Sand, M. de Fis, Al. Brong. 4. _ castanea, Al. Brong. Green Sand, M. de Fis, AL Brong. ; Rouen, Pas. 512 Organic Remains of the Cretaceous Group. 5. Nucleolites patellaris, Gold/. Maastricht, Goldf. 6. pyriformis, Goldf. White Chalk, Maastricht and Aix-la-Cha- pelle, Goldf. 7. lacunosus, Goldf. Cretaceous Rocks, Essen, Westphalia, Goldf. 8. cordatus, Goldf. Cretaceous Rocks, Essen, Goldf. 9. carinatus, Goldf. Chalk, Aix-la-Chapelle and Hildesheim ; Cretaceous Rocks, Essen, Westphalia, Goldf. 10. Lapis Cancri, Goldf. Aix-la-Chapelle ; Maestricht, Goldf. ; Upper Green Sand, Warminster, Lons. 11. depressa, Al. Brong., Rouen, Pas. ] 2. heteroclita, Defr., Chalk, Beauvais, Pas. , species not determined. Baculite Limestone, Normandy; Lower Chalk, Tours ; Rouen, Desn. 1. Ananchytes ovata, Lam. Chalk, Sussex, Mant. ; Chalk, Yorkshire, Phil. ; Chalk, Moen ; Meudon, Al. Brong. ; Baculite Lime- stone, Normandy, Desn. ; Limhamn, Sweden, Nils. ; Creta- ceous Rocks, Coesfeld, Westphalia, Goldf.; Chalk, Lublin, Poland, Pusch ; Chalk, Rouen, Pas. 2. hemisphaerica, AL Brong. ; Chalk, Yorkshire, Phil. ; Chalk, Etretat ; Duclair, Normandy, Pas. 3. intumescens, . Chalk, Yorkshire, Phil. 4. pustulosa, Lam. Chalk, Joigny ; Paris ; Rouen ; and Moen, Al. Brong. ; Chalk, Norwich, Woodward. 5. conoidea, Goldf. Cretaceous Rocks, Aubel, Belgium, Goldf. g. striata, Lam. Maestricht; Aix-la-Chapelle; Quedlinburg, Goldf. 7. sulcata, Goldf. Chalk, Aix-la-Chapelle ; Maestricht, Goldf. 8. Corculum, Goldf. Cretaceous Rocks, Coesfeld, Westphalia, Goldf. , species not determined. Chalk, Warminster, Lons. 1. Spatangus Cor-anguinum, Lam. Chalk, Sussex, Mant. ; Chalk, York- shire, Phil. ; Chalk, Meudon ; Joigny ; Dieppe ; Green Sand, M. de Fis, Al. Brong. ; Baculite Limestone, Normandy, Desn.; Torp, Scania, Nils. ; Chalk, Dorset and Devon, De la B. ; Marly Chalk, Paderborn ; Bielefeld; Minister; Coesfeld; Aix-la-Chapelle, Goldf.; Planerkalk, Saxony, Munst.; Chalk, Lublin, Poland, Pusch ; Mont-Ferrand ; Pic de Bugarach, Pyrenees, Dufr. 2. rostratus, Mant. Chalk, Sussex, Mant. ; Chalk, Joigny, AL Brong. 3. planus, Mant. Chalk, Sussex, Mant. ; Chalk, Yorkshire, Phil. 4. retusus, Park. Upper Green Sand, Wiltshire, Lons. 5. cordiformis, Mant. Chalk, Sussex, Mant. 6. suborbicularis, Defr. Green Sand, Dives, Normandy, AL Brong.; Marly Chalk, Maestricht, Goldf. ; Rouen, Pas. 7. punctatus, Lam. Upper Green Sand, Warminster, Lons. 8. granulosus, Goldf. Maestricht, Goldf. 9. subglobosus, Leske. White Chalk, Quedlinburg, Cretaceous Rocks, Bu'ren. Paderborn, Goldf. ; Rouen, Pas. 10. . nodulosus, Goldf. Cretaceous Rocks, Essen, Westphalia, Goldf. 11. radiatus, Lam. Maestricht, Goldf. 12. truncatus, Goldf. White Chalk, Maestricht, Goldf. 13. ornatus, Cuv. Chalk, Aix-la-Chapelle, Goldf.; Env. of Bayonne, Dufr. ; Chalk, Dieppe ; Rouen, Pas. 14. Bucklandii, Goldf. Cretaceous Rocks, Essen, Goldf. 15. Bufo, AL Brong. Chalk, Meudon, Havre, AL Brong. ; Chalk, Sussex, Mant* ; Baculite Limestone, Normandy, Desn. ; Chalk, Aix-la-Chapelle ; Maestricht, Goldf. * Sp. Prunella of Mantell, according to Brongniart. Organic Remains oj the Cretaceous Group. 513 16. Spatangus arcuarius, Lam. White Chalk, Maestricht, Goldf. 17. Prunella, Lam. Marly Chalk, Maestricht, Goldf. 18. Amygdala, Goldf. Chalk, Aix-la-Chapelle, Goldf. 19. gibbus, Lam. Cretaceous Rocks, Paderborn, Westphalia, Goldf. 20. Cor-testudinarium, Goldf. White Chalk, Maestricht and Qued- linburg; Cretaceous Rocks, Coesfeld, Westphalia, Goldf. 21. Bucardium, Goldf. Chalk, Aix-la-Chapelle, Goldf. 22. lacunosus, Linnaus. Chalk, Quedlinburg and Aix-la-Chapelle, Goldf. 23. Murchisonianus, Kcenig. Upper Green Sand, Sussex, Murch., Mant. 24. hemisphaericus, Phil. Chalk, Yorkshire, Phil. 25. i argillaceus, Phil. Speeton Clay, Yorkshire, Phil. 26. Isevis, Defr. Green Sand, Perte du Rhone, Al. Brong.; Havre ; Rouen, Pas. 27. acutus, Desk., S. of France ; Rouen, Desk. 28. Ambulacrum, Desk., Pyrenees, Desk. 29. crassissimus, Defr., Havre, Pas. , species not determined. Gault and Lower Green Sand, Sussex, Mant. ; Green Sand, Grande Chartreuse, Beaum. ; Chalk, Warminster, Lons. ANNULATA. 1. Serpula ampullacea, Sow. Chalk, Sussex, Mant.; Chalk, Norfolk, Barnes. 2. Plexus, Sow. Chalk, Sussex, Mant. 3. Carinella, Sow. Green Sand, Blackdown, Sow. 4. antiquata, Sow. Green Sand, Wilts, Sow. 5. rustica, Sow. Upper Green Sand, Folkstone, Goodhall. 6. articulata, Sow. Upper Green Sand, Folkstone, Sow. 7. obtusa, Sow. Chalk, Norfolk, Rose. 8. fluctuate, Sow. Chalk, Norfolk, Barnes. 9. ? macropus, Sow. Chalk, Norfolk, Leathes. 10. Trachinus, Goldf. Green Sand, Essen, Westphalia, Goldf. 11. lophioda, Goldf. Green Sand, Essen, Goldf. 12. laevis, Goldf. Green Sand, Essen, Goldf. 13. triangularis, Miinst. Gault? Rinkerode, Minister, Miinst. H. draconocephala, Goldf. Chalk Marl, Maestricht, Goldf. 15. depressa, Goldf. Green Sand, Essen, Goldf. 16. - - Rotula, Goldf. Green Sand, Regensburg, Goldf. 17. _ quadricarinata, Miinst. Green Sand, Regensburg, Goldf. 18. cincta, Goldf. Green Sand, Essen; Green Sand, Coesfeld; Aix- la-Chapelle, Goldf. 19. arcuata, Miinst. Green Sand, Regensburg, Goldf. 20. subtorquata, Miinst. Cretaceous Blue Marl, Rinkerode near Minister, Goldf. 21. sexangularis, Miinst., Rinkerode, Goldf. 22. Noggerathii, Miinst., Rinkerode, Goldf. 23. erecta, Goldf. Cretaceous Marl, Maestricht, Goldf. 24. Amphishcena, Goldf. Green Sand, Bochum, Westphalia; Cre- taceous Marl, Maestricht, Goldf. 25. spirographis, Goldf. Green Sand, Essen, Goldf. 26. parvula, Miinst. Green Sand, Essen, Goldf. 27. subrugosa, Miinst. Blue Cretaceous Marl, Baumberg, near Miinster, Goldf. 28. crenato-striata, Miinst., Baumberg, Goldf. 29. vibicata, Miinst. Blue Cretaceous Marl, Rinkerode, Goldf. 30. gordialis, Schlot., Miinster; Paderborn; Essen; Osnabriick ; 2L 514- Organic Remains of the Cretaceous Group. Maastricht ; Regensburg ; Strehla and Perna, near Dresden, Goldf. Serpula, species not determined. Red Chalk, Speeton Clay, Yorkshire, Phil. ; Chalk, Paris, Al. Brong. ; Charlottenlund ; Kbpinge, Scania, Nils. CIRRIPEDA. 1. Pollicipes sulcatus, Sow. Chalk, Sussex, Mant. 2. raaximus, Sow. Chalk, Norfolk, Barnes. CONCHIFERA. 1. Magas pumilus, Sow. Chalk, Norwich, Taylor; Chalk, Meudon, AL Brong. ; Chalk, Dieppe, Pas. 1. Thecidea radians, Defr. Chalk, Maestricht, Fauj. de St. Fond; Baculite Limestone, Normandy, Desn. ; Chalk, Dieppe, Pas. 2. ' recurvirostra, Defr. Maestricht; Baculite Limestone, Normandy, Desn. 3. hieroglyphica, Defr. Chalk, Essen, Hcen. 1. Terebratula subrotunda, Sow. Chalk, Sussex, Mant. ; Green Sand, Bochum, Hcen. 2. carnea, Sow. Chalk, Sussex, Mant. ; Chalk, Meudon, Al. Brong. ; Green Sand, Bochum, Horn. ; Chalk, Rouen ; Dieppe, Pas. 3. ovata, Sow. Chalk, Lower Green Sand, Sussex, Mant. ; Kopinge, Scania, Nils. ; Green Sand, Bochum, Hcen. ; Chalk, Rouen, Pas. 4. undata*, Sow. Chalk, Sussex, Mant. ; Chalk, Rouen, Pas. 5. elongata, Sow. Chalk, Sussex, Mant. 6. plicatilis, Sow. Chalk, Sussex, Mant. ; Chalk, Meudon, Moen ; M. de Fis, Al. Brong.; Green Sand, Grande Chartreuse, Beaum.; Chalk, Gravesend, Sow. ; Jonsac; Cognac, Dufr. 7. subplicata, Mant. Chalk, Sussex, Mant. ; Chalk, Yorkshire, Phil. ; Chalk, Maestricht ; Tours ; Beauvais ; Bac. Limestone, Normandy, Desn. 8. curvirostris, Nils. Kopinge, Scania, Nils. 9. Mantelliana f, Sow. Chalk, Sussex, Mant. 10. Martini J, Mant. Chalk, Sussex, Mant. 11. rostrata, Sow. Chalk, Sussex, Mant. 12. squamosa, Mant. Chalk, Sussex, Mant. 13. biplicata, Sow. Upper Green Sand, Sussex, Mant. ; Upper Green Sand, Cambridge, Sedg. ; Rouen ; Havre, Pas. ; Green Sand, Calvados, Her. 14. lata, Sow. Lower Green Sand, Sussex, Mant. ; Green Sand, Devizes, Sow. ; Upper Green Sand, Warminster, Lons. ; Gour- don, Dufr. 15. subundata, Sow. Chalk, Speeton Clay, Yorkshire, Phil. ; Chalk, Rouen, Al. Brong. 16. pentagonalis, Phil. Chalk, Yorkshire, Phil. 17. lineolata, Phil. Speeton Clay, Yorkshire, Phil. 18. Defrancii, Al. Brong. Chalk, Meudon, AL Brong.; Chalk, Sussex, Mant. ; Speeton Clay, Yorkshire, Phil. ; Balsberg, Morby, Sweden, Nils. ; Maestricht, Hcen. ; Chalk, Rouen, Pas. 19. intermedia, Sow. Upper Green Sand, Warminster, Sow. 20. alata, Lam. Chalk, Meudon, AL Brong. ; Kopinge ; Morby, Sweden, Nils. ; Cognac, Dufr. ; Chalk, Rouen, Pas. * T. subundata, T. intermedia, and T. semiglobosa, according to Mantell. f T. sulcata Mant. % T. Pisum of Sowerby. T. striatula of Mantell. Organic Remains of the Cretaceous Group. 515 21'. Terebratula octoplicata, Sow. Chalk, Sussex, Mant. ; Chalk, Dieppe, Al. Brong. ; Balsberg ; Ignaberga, Sweden ? Nils. ; Green Sand, Quedlinburg, Hcen. ; Jonsac ; Cognac, Dufr. ; Chalk, Rouen, Pas. 22. Gallina, AL Brong. Green Sand, Perte du Rhone, AL Brong. ; Baculite Limestone, Normandy, Desn. ; Rouen; Havre, Pas. ? 23. ornithocephala, Sow. Green Sand, Perte du Rhdne ; M. de Fis, AL Brong. 24. pectita, Sow. Baculite Limestone, Normandy, Desn. ; Ignaberga, Scania? Nils. ; Havre, AL Brong. ; Upper Green Sand, Wilts, Meade ; Maestricht, Hozn. 25. recurva, Defr. Maestricht ; Baculite Limestone, Normandy, Desn. 26. laevigata, Nils. Kbpinge, Scania, Nils. 27. triangularis, Wahl. Kbpinge, Scania, Nils. 28. longirostris, Wahl. Balsberg; Kjuge, Sweden, Nils. 29. Lyra, Sow. Upper Green Sand, Warminster, Lons. ; Havre, Pas. 30. rhomboidalis, Nils. Kjuge ; Mbrby, Sweden, Nils. 31. semiglobosa, Sow. Charlottenlund, Sweden, Nils.; Chalk, Moen, AL Brong. ; Green Sand, Bochum, Hcen. ; Chalk, Yorkshire, Phil. 32. obtusa, Sow. Upper Green Sand, Cambridge, Sedg. ; Green Sand, Quedlinburg, Hcen. 33. obesa, Sow. Chalk, Warminster, Lons. ; Chalk, Bray, Nor- mandy, Pas. 34. dimidiata, Sow. Green Sand, Haldon, Sow. 35. aperturata, Schlot. Chalk, Essen, Hcen. 36. chrysalis, Schlot. Maestricht, Hcen. 37. curvata, Schlot. Green Sand, Quedlinburg, Hcen. 38. dissimilis, Schlot. Green Sand, Bochum; Chalk, Speldorf, Hcen. 39. lacunosa, Schlot. Green Sand, Quedlinburg, Hcen. 40. microscopica, Fauj. de St. F., Maestricht. 41. nucleus, Defr. Green Sand, Bochum; Quedlinburg, Hcen. 42. ovoidea*, Sow. Green Sand, Bochum, Hcen. 43. peltata, . Maestricht, Hcen. 44. semistriata, Lam. Green Sand, Bochum, Hcen. 45. striatula, Sow. Green Sand, Bochum, Hcen. 46. varians, . Chalk, Essen, Hcen. 47. vermicularis, Schlot. Maestricht, Hcen. 48. minor, Nils. Kjuge, Nils. 49. pulchella f, Nils. Scania, Nils. 50. costata, Nils. Kjuge, Nils. 51. Lens, Nils. Charlottenlund, Sweden, Nils. 52. depressa, Lam. Gourdon, S. of France, Dufr. 53. Gibbsiana, Sow. Green Sand, Folkstone, Sow. 54. rigida, Sow. Norfolk, Sow. 1. Crania Parisiensis, Defr. Chalk, Meudon, Al. Brong.; Chalk, Brighton, Sow. ; Chalk, Dieppe, Pas. 2. antiqua, Defr. Baculite Limestone, Normandy, Desn. ; Chalk, Schlenacken, Hcen. 3. striata, Defr. Baculite Limestone, Normandy, Desn. ; Balsberg," &c. Sweden, Nils. 4, - stellata, Defr. Baculite Limestone, Normandy, Desn. * According to Von Dechen very like T. lata of Sowerby. f T. pumila, Lam. 2 L 2 516 Organic Remains of the Cretaceous Group. 5. Crania spinulosa, Nils. Kjuge; Morby, Sweden, Nils. / Maestricht, 6. tuberculata, Nils. Scania, Nils. 7. Nummulus, Lam. Balsberg ; Kjuge ; Ifo, in Scania, Nils. 8. nodulosa, Hcen. Maesti-icht, Hoen. Orbicula, species not determined. Lower Green Sand, Sussex, Martin ; Speeton Clay, Yorkshire, Phil. 1. Hippurites radiosa, Des M. Cendrieux, Perigord, Des M. 2. . Cornu Pastoris, Des M. Pyles, Perigueux, Jouannet. 3. striata, Defr. Alet, Aude ; Manbach, Berne, Des M. 4. sulcata, Defr. Alet, Aude, Des M. 5. dilatata, Defr. Alet, Aude, Des M. 6. bioculata, Lam. Alet, Aude, Des M. 7. Fistulas, Defr. Alet, Aude, Des M. 8. resecta, Defr. Marseille; Dauphine ; Regensburg, Untersberg, G. T. , species not determined. Cretaceous Rocks, South of France, Beaum. ; Pyrenees ; Jonsac (very large), Dufr. ; Western Alps, Lill von LiUienbach ; Chalk, Sussex, Mant. 1. Sphaerulites dilatata, Des M. Chalk, Royan and Talmont, mouth of the Gironde, Des M. 2. Bournonii, Des M. Royan and Talmont; Vall6e de la Couze, Dordogne, Des M. 3. ingens, Des M. Royan and Talmont, Des M. 4. Hreninghausii, Des M. Royan and Talmont ; Chalk, Languais, Dordogne, Des M. 5. foliacea, Lam. Isle d'Aix, Fleurian de Bellevue. 6. Jodamia, Des M. Mirambeau, Charente-Inferieure, Defr. 7. Jouannetti, Des M. Vallee de la Couze, Perigord, Des M. 8. crateriformis, Des M. Royan ; Languais, Dordogne, Des M. 9. cylindracea, Des M. Vallee de la Couze, Des M. 10. ventricosa, Des M. Vallee de la Couze, Des M. 11. turbinata, Des M. Vallee de la Couze, Des M. 12. cristata, Des M. Department of the Var, Des M. 13. , bioculata, Des M. Var, Des M. 14. imbricata, Des M. Var, Des M. 15. calceoloides, Des M. Vallee de la Couze, Des M. 1. Ostrea vesicularis, Lam. Chalk, Sussex, Mant. ; Chalk, Perigueux, Meudon, Al. Brong. ; Chalk, Maestricht, Fauj. de St. F. ; var. Baculite Limestone, Normandy, Desn.; Kbpinge; Kjuge, Sweden, Nils.; Chalk, Dieppe, Pas. 2. semiplana, Mant. Chalk, Sussex, Mant. 3. canaliculata, Sow. Chalk, Sussex, Mant. ; Aix-la-Chapelle, G. T. 4. carinata, Lam. Upper Green Sand, Sussex, Mant.; Green Sand, Normandy, De la B. ; Green Sand, Grasse, (Dep. of the Var,) Martin de Martigues ; Green Sand, Bochum ; Chalk, Essen, Hcen. 5. serrata, Defr. Chalk, Sweden ; Dreux, Al. Brong. ; Green Sand, Grasse, Var ; Maestricht, Hcen. ; Jonsac ; Cognac ; Angou- leme ; Coustouge, Dufr. ; Chalk, Rouen, Pas. 6. lateralis, Nils. Kopinge ; Ifo, Scania, Nils. ; Chalk, Essen, Hcen. 7. clavata, Nils. Morby, Sweden, Nils. 8. Hippopodium, Nils. Ifo ; Carlshamn, Sweden, Nils. 9. curvirostris, Nils. Ifo ; Kjuge, Scania, Nils. 10. ; acutirostris, Nils. Ifo; Scania, Nils. 11. - flabelliformis, Nils. Kjuge, Morby, Sweden, Nils.; Chalk, Essen, Hcen. 12. pusilla, Nils. Kopinge, Scania, Nils. Organic Remains of the Cretaceous Group. 517 13. Ostreadiluviana?*Zw. Balsberg; Kjuge; Morby/ Carlshamn, Sweden, Nils. ; Orcher, Rouen, Pas. 14. - lunata, Nils. Ahus, Yngsjb, Scania, Nils. 15. - truncata, Goldf. Green Sand, Griesenbeck, Hcen. 16. -- incurva, Nils. Kjuge ; Oppmanna, Nils. 17. - ? plicata, Nils. Kjuge, Sweden, Nils. 18. - biauricularis, . Jonsac ; Cognac; Angouleme, Dvfr. 19. - pectinata, Lam. Green Sand, Orcher, Rouen, Pas. ; Bochum, G. T. 20. - Rotomagensis, Defr. Rouen, Pas. 21. - pectinoides, Defr. Rouen, Pas. 22. - auriculata, Defr. Rouen, Pas. 1. Exogyra digitata, Sow. Green Sand, Lyme Regis, De la B. 2. - conica, Sow. Green Sand, Sussex ; Upper Green Sand, Wilts ; Green Sand, Blackdown, Sow. ; Kopinge, Nils. ; Green Sand, 3. - undata, Sow. Green Sand, Blackdown, Goodhall. 4. - haliotoidea, Sow. Upper Green Sand, Warminster, Lons.; Chalk, Essen, Hcen. ; Kjuge ; Balsberg ; Morby, Nils. ; Lillebonne, Pas. 5. - laevigata, Sow. Green Sand, N. of Ireland, Sow. 1. Gryphaea vesiculosa, Sow. Upper Green Sand, Sussex, Mant. ; Green Sand, Warminster, Bennet ; Green Sand, Bouches du Rhone, Hcen. ; Bourg St. Andre" ol, Env. of Pont St. Esprit; Gourdon, Duff. 2. - sinuata, Sow. Speeton Clay, Yorks., Phil. ; Green Sand, Grande Chartreuse, Beaum. ; Lower Green Sand, Isle of Wight, Sedg.; Pic de Bugarach; Bourg St. Andre"ol,Z)w/r.; Bray, Norm., Pas. 3. - auricularis, Al. Brong. Chalk, Perigueux, Al. Brong. ; Green Sand, Grande Chartreuse, Beaum. ; Chalk, Kazimirz, Poland, Pusch; Green Sand, Apt, Vaucluse, Hcen.; Jonsac; Cognac, Dufr. 4. - Aquila, Al. Brong. Green Sand, Perte du Rhone, Al. Brong. ; Pic de Bugarach, Pyrenees; Bourg St. Andreol; Jonsac; Cognac, Dufr.; Rouen, Pas. 5. - Columba, Lam. Green Sand, Normandy ; Green Sand, Mari- time Alps, De la B. ; Chalk, Kazimirz, Poland, Pusch ; Re- genburg; Pirna; Konigstein, Holl ; Chalk, Saumur; Mans, Hcen.; Env. of Pont St. Esprit; Angouleme, Dufr.; Plessis- Grimoult, Calvados, Desl. 6. - truncata, Goldf. Maestricht, Hcen. 7. - secunda, . Env. of Pont St. Esprit ; Jonsac ; Cognac ; Gourdon ; Pic de Bugarach, Pyrenees, Dufr. 8. - canaliculata, Sow. Upper Green Sand, Wilts, Sow. - , a small species in the baculite limestone and chalk of other parts of France, Desn. 1. Sphaera corrugata, Sow. Lower Green Sand, Isle of Wight, Sedg. 1. Podopsis obliqua, Mant. Chalk, Sussex, Mant. 2. - striata, Soiv. Chalk, Yorks., Phil. ; Chalk, Havre, Al. Brong.; Chalk, Essen; Bochum, Hcen.; Chalk, Dieppe, Pas.; Chalk, Sussex, Mant. 3. - truncata, Lam. Chalk, Normandy, Touraine, AL Brong. ; Balsberg and other places in Sweden, Nils.; Lyme Regis, De la B. 4. lamellata, Nils. Kjuge, Morby, Sweden, Nils. * M. Brongniart considers that this shell, cited by M. Nilsson as 0. diluviana, may be the O. serrata of Defrance. 518 Organic Remains of the Cretaceous Group. 5. Podopsis spinosa, . Coustouge, Dufr. , species not determined. Gourdon, Dnfr. 1. Spondylus? strigilis, AL Brong. ; Green Sand, Perte du Rhone, AL Brong. 1. Plicatula inflata, Sow. Chalk, Sussex, Mant. ; Chalk, Cambridge, Sedg. 2. pectinoides, Sow. Chalk, Sussex, Mant.; Gault, Cambridge, Sedg. 3. radiata, Goldf. Coesfeld, G. T. 4. spinosa, Mant. Orcher, Norm., G. T. 1. Pecten quinquecostatus, Sow. Chalk, Sussex, Mant. ; Chalk, Meudon, AL Brong. ; Green Sand, Perte du Rhone, AL Brong. ; Ba- culite Limestone, Normandy, Desn. ; Kopinge, and other places in Sweden, Nils. ; Green Sand, Blackdown, Sow. ; Green Sand, Lyme Regis, De la B. ; Upper Green Sand, Warminster, Lons. ; Green Sand, Coesfeld, Osterfeld ; Chalk, Saumur, Hcen. ; Env. of Pont St. Esprit ; Cognac ; Mont- Ferrand; Pic de Bugarach, Pyrenees; Env. of Bayonne,Z)/r.; Green Sand, Calvados, Her. ; Maestricht, G. T. 2. Beaveri, Sow. Chalk, Sussex, Mant. ; Biiren ; Quedlinburg, G. T.; Lillebonne, Pas. 3. - triplicatus, Mant. Chalk, Sussex, Mant. 4. orbicularis, Sow. Chalk, Gault, Lower Green Sand, Sussex, Mant. ; Kopinge, Sweden ? Nils. ; Green Sand, Aix-la-Chapelle,//o2M.; Lillebonne, Pas. 5. quadricostatus, Sow. Lower Green Sand, Sussex, Mant.; Chalk, Maestricht ; Baculite Limestone, Normandy, Desn. ; Green Sand, Grande Chartreuse, Beaum. ; Green Sand, Haldon, Baker ; Upper Green Sand, Warminster, Lons. 6. obliquus, Sow. Lower Green Sand, Sussex, Mant. ; Green Sand, Calvados, Her. ; Lillebonne, Pas. 7. cretosus, Defr. Chalk, Meudon, AL Brong. ; Chalk, Lublin, Po- land, Pusch; Chalk, Angers. 8. arachnoides, Defr. Chalk, Meudon and Normandy, AL Brong. ; Chalk, Lublin, Poland, Pusch. 9. extextus *, AL Brong. Chalk, Havre ; Baculite Limestone, Nor- mandy, Desn. ; Chalk, Angers, Hcen. 10. serratus, Nils. Balsberg ; Kopinge, Sweden, Nils. 11. septemplicatus, Nils. Balsberg, Kjuge, Sweden, Nils.; Maestricht, G. T. 12. multicostatus, Nils. Balsberg, Sweden, Nils. 13. undulatus, Nils. Kopinge; Kaserberga, Scania, Nib. 14. subaratus, Nib. Balsberg; Kjuge, Sweden, Nib. 15. pulchellus, Nib. Kopinge ; Balsberg, Sweden, Nils. 16. lineatus, Nils. Kopinge; Morby, Sweden, Nib. 17. virgatus, Nib. Balsberg; Morby, Nib. 18. membranaceus, Nib. Kopinge, and other places, Sweden, Nils. 19. Isevis, Nib. Kopinge; Yngsjoe, Sweden, Nils.; Aix-la-Chapelle, Hcen. 20. inversus, Nib. Kopinge, Sweden, Nils. 21. asper, Lam. Upper Green Sand, Warminster, Lons. ; Chalk, Lublin, Poland, Pusch ; Green Sand, Bochum ; Chalk, Hat- teren, Hcen. ; Green Sand, Calvados, Her. ; Lillebonne, Pas. ; Maestricht, G. T. 22. asperrimus, Hcen. Green Sand, Hardt, Hcen. 23. gryphaeatus, . Green Sand, Aix-la-Chapelle, Hcen. 24. nitidus, Sow. Chalk, Sussex, Mant. ; Green Sand, Aix-la-Chapelle, Hcen. ; Chalk, Rouen ; Dieppe, Pas. * M. Hccninghaus considers this shell the same with P. serratus t Nilsson. Organic Remains of the Cretaceous Group. 519 25. Pecten versicostatus, . Green Sand, Aix-la-Chapelle ; Green Sand, Minden, Haen. ? 26. corneus, Sow. Kbpinge ? Nils. 27. dentatus, Nils. Balsberg, Nils. 28. dubius, Defr. Chalk, Rouen ; Dieppe, Pas. , species not determined. Chalk, Sussex, Mant. ; Speeton Clay, Yorks., Phil. ; Green Sand, Maritime Alps, De la B. 1. Lima pectinoides, Hoen. Maestricht, Hoen. 2. - striata, Goldf. Maestricht, G. T. - muricata, Goldf. Maestricht, G. T. 1. Plagiostoma spinosum*, Sow. Chalk, Sussex, Mant.; Chalk, Meudon, Dieppe, Rouen, Perigueux, Poland, Al. Brong, ; Kbpinge, Sweden, Nils. ; Chalk, Dorset and Devon, De la B. ; Chalk, Weinhohla, Saxony, Weiss ; Quedlinburg, If oil ; Osterfeld, Hoen.; Env. of Pont St. Esprit; Coustouge, Dufr. 2. Hoperi, Mant. ; Chalk, Sussex, Mant. ; Orcher ; Rouen, Pas. - Brightoniense, Mant. ; Chalk, Sussex, Mant. elongatum, Sow. ; Chalk, Sussex, Mant. asperum, Mant. ; Chalk, Sussex, Mant. ; Coustouge, Dufr. ovatum, Nils. Balsberg and Kjuge, Sweden, Nils. 7- semisulcatum, Nils. Balsberg and other places, Sweden, Nils. ; Chalk, Ktinder, Saumur, Hoen. 8. Mantelli, Al. Brona. Chalk, Dover ; Moen, Denmark, Al. Brong. ; Chalk, Dieppe, Pas. 9- granulatum, Nik. Kbpinge, Kjuge, Sweden, Nik. 10. elegans, Nik. Balsberg, Mbrby, Sweden, Nik. 1 1 pusillum, Nik. Balsberg, Kbpinge, Sweden, Nik. 12. . turgidum, Lam. Chalk, Saintes; Green Sand, Osterfeld, Hoen. 13. denticulatum, Nik. Ignaberga, Kjuge, Nik. squamatum, Goldf. Maestricht, G. T. 15. Juliobonse, Pas. Lillebonne, Pas. , species not determined : Upper Green Sand, Sussex, Mant. 1. Avicula ccerulescens, Nils. Kbpinge, Kaseberga, Sweden, Nik. triptera, Bronn. Maestricht, G. T. , species not determined. Chalk, Sussex, Mant. ; Maestricht? Hoen. ; G our don, Dufr. 1. Inoceramus Cuvieri, Sow. Chalk, Sussex, Mant.; Chalk, Yorks., Phil.; Chalk, Meudon, Al. Brong. ; Balsberg ; Ignaberga, Kjuge, Sweden, Nils. ; Jonsac ; Cognac ; Gourdon, Dufr. ; Chalk, Rouen ; Dieppe, Pas. 2. Brongniarti, Mant. Chalk, Sussex, Mant. Chalk, Yorks., Phil. ; Kaseberga, Kbpinge, Sweden, Nils. ; Chalk, Czarkow, Poland, Pusch ; Quedlinburg, Hoen. Lamarckiif, Mant. Chalk, Sussex, Mant.; Rouen, Pas. 4. mytiloides, Mant. Chalk, Sussex, Mant. ; Chalk, Warminster, Lons. ; Quedlinburg ; Pirn a, Kbnigstein, Holl; Env. of Pont St. Esprit, Dufr. ; Chalk, Dieppe ; Duclair, Pas. 5 - - cordiformis, Sow. Chalk, Sussex, Mant.; Chalk, Gravesend, Sow. G - latus, Mant. Chalk, Sussex, Mant. ; Rouen ; Meulers, Pas. Pachites spinosa of Defrance. According to M. Deshayes, the species of Pla- giostoma which have been named Pachites by M. Defrance, are referrible to the genus Spondylus, while the remaining species of the same supposed genus belong to the genus Lima. t According to M. Deshayes, Inoceramus (Catillus) Lamarclrii and /. Brongniarti are --the same shells. All the Inoccrami of the chalk are Calilli according to Des- hayes. 520 Organic Remains of the Cretaceous Group. 7. Inoceramus Websteri, Mant. Chalk, Sussex, Mant. 8. striatus, Mant. Chalk, Sussex, Mant. ; Quedlinburg ; Dresden. G. T. 9. undulatus, Mant. Chalk, Sussex, Mant. 10. involutus, Sow. Chalk, Sussex, Mant. ; Chalk, Norfolk, Rose. 11. tenuis, Mant. Chalk, Sussex, Mant. 12. Cripsii, Mant. Chalk, Sussex, Mant.; Chalk, Dieppe, Pas. 13. concentricus, Park. Gault, Sussex, Mant. ; Green Sand, Perte du Rhone, M. de Fis, Al. Brong. ; Chalk, Warminster, Lons. ; Green Sand, Quedlinburg, Bocnum, and Essen, Hoen. ; Chalk, Rouen ; Meulers, Pas. 14. sulcatus, Park. Gault, Sussex, Mant. ; Green Sand, Perte du Rhone ; M. de Fis, Al. Brony. ; Kbpinge, Scania, Nils. ; Green Sand ? Nice, De la B. ; Rouen ; Meulers, Pas. 15. gryphaeoides, Sow. Gault, Sussex, Mant. ; Green Sand, Lyme Regis, De la B. 16. pictus, Sow. Chalk, Surrey, Murch. 17. rugosus, . Quedlinburg, Hoen. 18. fornicatus, Goldf. Westphalia, G. T. 19. cardissoides, Goldf. Quedlinburg, G. T. -, species not determined. Lower Green Sand, Sussex, Martin ; Baculite Limestone, Normandy, Desn. 1. Pachymya Gigas, Sow. Lower Chalk, Lyme Regis, De la B. 1. Meleagrina approximata, Bronn. Maestricht, G. T. 1. Gervillia aviculoides, Sow. Lower Green Sand, Sussex, Mant.; Green Sand, Lyme Regis, De la B. ; Quedlinburg, Holl ; Lower Green Sand ? Isle of Wight, Sedg. 2. solenoides, Defr. Lower Green Sand, Sussex, Mant. ; Bacu- lite Limestone, Normandy, Desn. ; Green Sand, Lyme Regis, Dela B.; Upper Green Sand, Warminster, Lons. ; Maestricht, Hoen. ; Upper Green Sand, Aix-la-Chapelle, Dum. 3. acuta, Sow. Lower Green Sand, Sussex, Mant. 1. Pinna gracilis, Phil. Speeton Clay, Yorks., Phil. 2. tetragona, Sow. Upper Green Sand, Devizes, Gent; Maestricht; Bochum ; Aix-la-Chapelle ; Green Sand, Pirna, G. T. 3. , restituta, . Chalk, Valkenburg, Hoen. ? 4. subquadrivalvis, Lam. Cotentin ; Saumur, Hoen. 1. Mytilus lanceolatus, Sow. Lower Green Sand, Sussex, Mant. ; Green Sand, Blackdown, Sow. 2. laevis, Defr. Chalk, Bougival, Al. Brong. ; Rouen, Pas. 3. edentulus, Sow. Green Sand, Blackdown. Sow. 4. problematicus, . Green Sand, Bochum, Hoen. 5. simplex, Pas. Chalk, Rouen, Pas. 1. Modiola sequalis, Sow. Lower Green Sand, Sussex, Mant. 2. bipartita, Sow. Lower Green Sand, Sussex, Mant. ; Env. of Pont St. Esprit, Dufr. 1. Chama Cornu Arietis, Nils. Kjuge; Morby, Sweden, Nils. 2. laciniata, Nik. Kjuge ; Balsberg ; Morby, Sweden, Nils. , species not determined. Chalk, Sussex, Mant. 1. Trigonia Daedalea, Park. Lower Green Sand, Sussex, Mant.- Green Sand, Haldon? Baker; Lower Green Sand, Isle of Wight Sedg. ; Env. of Pont St. Esprit, Dufr. 2. aliformis, Sow. Lower Green Sand, Sussex, Mant.; Blackdown, De la B.; Upper Green Sand? Eddington, Lons.; Lower Green Sand, Isle of Wight, Sedg.; Gourdon, Dufr.; Aix-la- Chapelle; Quedlinburg, G. T. ; Orcher, Normandy, Pas. 3. spinosa, Sow. Lower Green Sand, Sussex, Martin; Green Sand, Blackdown, Steinhauer ; Lillebonne, Pas. Organic Remains of the Cretaceous Group. 521 4. Trigonia rugosa, Lam. Green Sand, Perte du Rhone, Al. Brong. Rouen, Pas. 5. scabra, Lam. Green Sand, Perte du Rhone, AL Brong.; Bacu- lite Limestone? Normandy, Dem.; Plessis-Grimoult, Calva- dos, Her. ; Lillebonne ; Rouen, Pas. 6. . pumila, Nils. Kopinge, Scania, Nils. 7. eccentrica, Sow. Green Sand, Blackdown, Steinhauer. 8. . nodosa, Sow. Lower Green Sand, Hythe, Kent, Sow. 9. spectabilis, Sow. Green Sand, Blackdown, Goodhall. 10. arcuata, Lam. Aix-la-Chapelle, Horn. 1 1 . alata, . Env. of Pont St. Esprit ; Pic de Bugarach, Py- renees, Dufr. -', species not determined. Lower Green Sand, Wiltshire, Lons. 1. Nucula pectinata, Mant. Gault, Sussex, Mant.] Blue Marl, Bray, Normandy, Pas. 2. ovata, Mant. Gault, Sussex, Mant.; Speeton Clay, Yorkshire, Phil. 3. impressa. Sow. Lower Green Sand, Sussex, Mant. ; Green Sand, Blackdown, Sow. 4. subrecurva, Phil. Speeton Clay, Yorkshire, Phil. 5. ovata, Nils. Kopinge; Kaseberga, Scania, Nils. 6. - ' truncata, Nils. Kaseberga, Scania, Nils. 7. panda, Nils. Kaseberga, Scania, Nils. 8. producta, Nils. Kaseberga, Scania, Nils. 9. antiquata, Sow. Green Sand, Blackdown, Sow. 10. angulata, Sow. Green Sand, Blackdown, Sow. 1 1 . undulata, Soiv. Gault, Folkestone, Sow. 12. - siliqua, Goldf. Maestricht, G. T. 1. Pectunculus lens, Nils. Balsberg; Kopinge, Sweden, Nils. 2. sublaevis, Sow. Green Sand, Blackdown, Sow.; Paderborn ; Quedlinburg, G. T. 3. umbonatus, Sow. Green Sand, Blackdown, Sow.; Rouen, Pas. 1. Area carinata, Sow. Upper Green Sand, Sussex, Mant. 2. - exaltata, Nik. Carlshamn, Sweden, Nik. ; Green Sand? Aix-la- Chapelle, JHoen. 3. rhombea, Nils. Balsberg, Sweden, Nik. ; Aix-la-Chapelle, G. T. 4. clathrata, Lam. Chalk, Angers ; Saumur, Hoen. 5. ovalis, Nils. Kopinge, Scania, Nils.; Aix-la-Chapelle, G. T. 6. subacuta, Hoen. Maestricht, Hoen. , species not determined. Chalk, Gault, Sussex, Mant. 1. Cucullsea decussata, Sow. Lower Green Sand, Sussex, Mant. ; Chalk, Rouen, Al. Brong. 2. glabra, Sow. Green Sand, Blackdown, Sow. ; Upper Green Sand, Warminster, Lons. ; Rouen, Pas. 3. carinata, Sow. Green Sand, Blackdown, Sow. ; Aix-la-Chapelle, G. T.; Rouen, Pas. 4, fibrosa, Sow. Green Sand, Blackdown, Hill. 5. costellata, Sow. Green Sand, Blackdown, Sow. 6. - crassatina, Lam. Chalk, Beauvais, Hoen. , species not determined. Chalk, Sussex, Mant.; Speeton Clay, Yorkshire, Phil. ; Gourdon, Dufr. 1. Cardita Esmarkii, Nils. Kopinge, Scania, Nik. 2. Modiolus, Nils. Kaseberga, Scania, Nils. tuberculata, Sow. Upper Green Sand, Devizes, Gent. 4. crassa, Lam. Chalk, Doue, Hoen. , species not determined. Upper Green Sand, Sus&ex, 1. Cardium decussatum, Sow. Chalk, Sussex, Mant. 522 Organic Remains of the Cretaceous Group. 2. Cardium Hillanum, Sow. Green Sand, Blackdown, Hill; Env. of Pont St. Esprit; Gourdon, Dufr. 3. proboscideum, Sow. Green Sand, Blackdown, Hill. Venericardia, species not determined. Chalk, Sussex, Mant. 1 . Astarte striata, Sow. Green Sand, Blackdown, Sow. ; Upper Green Sand, Devizes, Lons. , species not determined. Chalk, Sussex, Mant. ; Lower Green Sand, Wilts, Lons. 1. Thetis minor, Sow. Lower Green Sand, Sussex, Mant.; Green Sand, Lyme Regis, De la B. 2. major, Sow. Upper Green Sand, Devizes, Gent; Green Sand, Blackdown, Hill. 1. Venus Ringmeriensis, Mant. Chalk, Sussex, Mant. *2. parva, Sow. Lower Green Sand, Sussex, Mant. / Green Sand, Lyme Regis, De la B. ; Green Sand, Isle of Wight, Sow. *3. angulata, Sow. Lower Green Sand, Sussex, Mant. ; Green Sand, Blackdown, Hill. *4. Faba, Sow. Lower Green Sand, Sussex, Mant.; Green Sand, Blackdown ; Green Sand, Isle of Wight, Sow. 5. ovalis, Sow. Lower Green Sand, Sussex, Mant. *6. lineolata, Sow. Green Sand, Blackdown, Hill; Green Sand, Bo- chum, Hcen. 7. plana, Sow. Green Sand, Blackdown, Hill. 8. caperata, Sow. Green Sand, Lyme Regis, De la B.; Green Sand, Blackdown, Hill. 9. ? exerta, Nils. Kopinge, Nils. 1. Lucina sculpta, Phil. Speeton Clay, Yorkshire, Phil. 1. Tellina aequalis, Mant. Lower Green Sand, Sussex, Mant. 2. insequalis, Sow. Lower Green Sand, Sussex, Mant.; Green Sand, Blackdown, Sow. 3. striatula, Sow. Green Sand, Blackdown, Sow. , species not determined. Speeton Clay, Yorkshire, Phil. 1. Corbula striatula, Sow. Lower Green Sand, Sussex, Mant. 2. Punctum, Phil. Speeton Clay, Yorkshire, Phil. 3. gigantea, Sow. Green Sand, Blackdown, Hill. 4. laevigata, Sow. Green Sand, Blackdown, Hill. 5. ovalis, Nils. Kopinge, Nils. 6. caudata, Nils. Kopinge, Nils. , species not determined, Rouen, Pas. 1. Crassitella latissima, Hcen. Maestricht, Hcen. 2. . tumida, . Coustouge, Dufr. 1. Cythere subdeltoidea, Munst. Chalk, Rinkerode; Strehla; Haldem; Maestricht, G. T. 2. compressa, Munst.; Haldem, G. T. 1. Lutraria Gurgitis, Al. Brong.; Green Sand, Perte du Rhone, Al. Brong.; Kopinge, Morby, Sweden, Nils. 2. ? carinifera, Sow. Chalk, Lyme Regis, De la B.; Rouen, Pas. , species not determined. Speeton Clay? Yorkshire, Phil. 1. Panopa?a plicata, Sow. Green Sand, Osterfeld, Hcen.; (var.?) Lower Green Sand, Sussex, Mant. ; Coustouge, Dufr. 1. Mya mandibula, Soiv. Lower Green Sand, Sussex, Martin; Gault, Isle of Wight, Fitton ; Gourdon, Dufr. ; Rouen, Pas. 2. phaseolina, Phil. Speeton Clay, Yorkshire, Phil. ? Chalk, near Calne, Lons. * Cylhcrea of Lamarck, according to Lonsdale. Organic Remains of the Cretaceous Group. 523 Teredo, species not determined. Maestricht, Hoen. 1. Pholas? constricta, Phil. Speeton Clay, Yorkshire, Phil. 1. Fistulana pyriformis, Mant. Gault, Sussex, Mant. MOLLUSC A. 1. Dentalium striatum, Sow. Gault, Sussex, Mant. 2. - ellipticum, Sow. Gault, Sussex, Mant. ; Rouen, Pas. 3. decussatum, Sow. Gault, Sussex, Mant. 4. nitens, Hcen. Maestricht, Hcen. , species not determined. Lower Green Sand, Sussex, Mant. 1. Patella ovalis, Nils. Balsberg, Scania, Nils. , species not determined. Lower Green Sand, Sussex, Mant. ; Lower Green Sand, Wiltshire, Lons. I. Emarginula Sanctse Catherines, Pas.; Rouen, Pas. 2. pelagica, Pas. ; Rouen, Pas. Pileopsis, species not determined. Lower Green Sand, Sussex, Mant. 1. Helix Gentii, Sow. Upper Green Sand, Devizes, Gent. 1 . Auricula incrassata, Sow. Chalk, Sussex, Mant. ; Green Sand, Black- down, Hill. 2. obsoleta, Phil. Speeton Clay, Yorkshire, Phil. 3. avellana, Mant.; Rouen, Pas. Melania, species not determined. Speeton Clay? Yorkshire, PhiL 1. Paludina extensa, Sow. Green Sand, Blackdown, Hill. 1. Ampullaria canaliculata. Gault, Sussex, Mant.; Rouen, Pas. 2. spirata, Hcen. Maestricht, Hcen. , species not determined. Green Sand, M. de Fis, Al. Brony. 1 . Nerita rugosa, Hcen. Maestricht, Hcen. 1 . Natica canrena, Park. Lower Green Sand, Sussex, Mant. 2. spirata, . Green Sand, Aix-la-Chapelle, Hcen. , species not determined. Gault, Sussex, Mant. ; Lower Green Sand, Wiltshire, Lons. ; Env. of Pont St. Esprit, Dufr. 1. Vermetus polygonalis, Sow. Lower Green Sand, Hythe, Kent, Lord Green oc/c. umbonatus, Mant. Chalk, Sussex, Mant. Sowerbii, Mant. Chalk, Sussex, Mant. ; Speeton Clay, York- shire, Phil. 4. concavus, Sow. Lower Green Sand, Sussex, Mant. ; Upper Green Sand, Wilts, Lons. , species not determined. Lower Green Sand, Isle of Wight, Sedg. Delphinula, species not determined. Speeton Clay, Yorkshire, PhiL 1. Solarium tabulatum? Phil. Speeton Clay, Yorkshire, Phil. 1. Cirrus depressus, Mant. Chalk, Sussex, Mant. 2. - perspectivus, Mant. Chalk, Sussex, Mant. 3. granulatus, Mant. Chalk, Sussex, Mant. 4. plicatus, Sow. Gault, Sussex, Mant. 1. Pleurotomaria Rhodani, Al. Brong. Rouen, Pas. 2. depressa, Sow. Rouen, Pas. 3. perspectiva, Sow. Chalk, Rouen, Pas. , species not determined. Maestricht, Hcen.; Gourdon ; Bourg St. Andreol, Dufr. 1. Trochus Basteroti, AL Brong. Chalk, Sussex, Mant.; Kopinge, Scania, Nils. ; Rouen, Pas. 2. linearis, Mant. Chalk, Sussex, Mant. 3. Rhodani, AL Brong. Upper Green Sand, Sussex, Mant. ; Green Sand, Perte du Rhone, Al. Brong. ; Lower Chalk. Lyme Regis, DelaB.; Green Sand, Essen ; Green Sand, Osterfeld, Hosn. 4. - bicarinatus, Sotv. Upper Green Sand ? Sussex, Mant. 524? Organic Remains of the Cretaceous Group. 5. Trochus Gurgitis, AL Brong. Green Sand, Perte du Rhone, Al. Brong. ; Green Sand, Bochum, Hcen.; Rouen, Pas. 6. ? Cirroides, Al. Brong. Green Sand, Perte du Rhone, Al. Brong. ; Rouen, Pas. 7. laevis, Nils. Kopinge, Scania, Nils. 8. onustus, Nils. Kopinge, Scania, Nils. , species not determined. Green Sand, M. de Fis, Al. Brong. 1. Turbo pulcherrimus, Bean. Speeton Clay, Yorkshire, Phil. 2. _ sulcatus, Nils. Chalk, Kopinge, Scania, Nils. 3. moniliferus, Sow. Green Sand, Blackdown, Sow. 4. carinatus, Sow. Green Sand, Coesfeld, Hcen. ; Rouen, Pas. 1. Turritella terebra, Broc. Green Sand, Weddersleben, Hcen. -, species not determined. Speeton Clay? Yorkshire, Phil. 1. Cerithium excavatum, Al. Brong. Green Sand, Perte du Rhone, Al. Brong.; Green Sand, Aix-la-Chapelle, Hcen. . , species not determined. Green Sand, M. de Fis, Al. Brong. 1. Pyrula planulata, Nils. Chalk, Kopinge, Scania, Nils. 2. minima, Hcen. Green Sand, Aix-la-Chapelle, Hosn. ] . Fusus quadratus, Sow. Green Sand, Blackdown, Sow. 1. Murex Calcar, Sow. Green Sand, Blackdown, Sow. 1. Pterocera maxima, Hcen. Martigues, Hcen. 1. Rostellaria Parkinsoni, Mant. Chalk, Lower Green Sand, Sussex, Mant. ; Green Sand, Bochum ; Coesfeld, Hcen. Lillebonne, Pas. 2. carinata, Mant. Gault, Sussex, Mant. ; Rouen, Pas. 3. . calcarata, Sow. Lower Green Sand, Sussex, Mant. ; Green Sand, Blackdown, Sow. 4. anserina, Nils. Chalk, Kopinge, Scania, Nils. 5. . inflata, Pas. ; Rouen, Pas. , species not determined. Lower Green Sand, Isle of Wight, Sedg. 1. Strombus papilionatus, . Chalk, Maestricht, Aix-la-Chapelle, Hcen. 1. Cassis avellana, Al. Brong. Chalk, Sussex, Mant.; Chalk, Rouen; M. de Fis, AL Brong. 1. Dolium nodosum, Sow. Chalk, Sussex, Mant. Eburna, species not determined. Green Sand, Perte du Rh6ne, Al. Brong.; Chalk? Sussex, Mant. 1. Nummulites, species not determined. Green Sand, Alps of Savoy, Dauphiny, and Provence, Beaum. ; Maritime Alps, De la B. ; Chalk, Weinbohla, Saxony, Klipstein; Cretaceous rocks, South of France ; Pyrenees, Dnfr. 1. Lenticulites Comptoni, Sow. Green Sand, Earlstoke, Wilts, Sow. ; Green Sand, Scania, Nils. 2. cristella, Nils. Chalk, Charlottenlund, Sweden, Nils. 1. Lituolites nautiloidea, Lam. Chalk, Paris, Al. Brong.; Rouen, Pas. 2. difformis, Lam. Chalk, Paris, AL Brong. Miliolites, . S. of France ; Pyrenees, Dttfr. 1. Planularia elliptica, Nils. Charlottenlund, Sweden, Nils. 2. angusta, Nils. Kopinge, Scania, Nils. 1. Nodosaria sulcata, Nils. Chalk and Green Sand, Scania, Nils. 2. Isevigata, Nils. Scania, Nils. 1. Belemnites mucronatus, Schlot. Chalk, Sussex, Mant.; Chalk, York- shire, Phil. ; Green Sand, Sweden, Nils. ; Chalk, Meudon, &c., AL Brong. ; Baculite limestone, Normandy, Desn. ; Chalk, Lublin, Poland, Pusch ; Maestricht, Aix-la-Chapelle, Schlot. ; Haldem ; Rinkerode, Miinst. ; Dieppe, Pas. 2. granulatus, Defr. Chalk, Sussex, Mant. Organic Remains of the Cretaceous Group. 525 3. Belemnites lanceolatus *, Schlot. Chalk, Sussex, Mant. ; Quedlinburg, Roll. 4. minimus, Lister. Gault, Sussex, Mant.; Red Chalk, Yorkshire, Phil. 5. attenuates, Sow. Gault, Sussex, Mant. 6. mamillatus f, Nils. Chalk, Scania, Nils. , species not determined. Speeton Clay, Yorkshire, Phil. ; Green Sand, Perte du Rhone, Al. Brony. 1. Actinocamax verus, Miller. Chalk, Kent, Miller. 1. Nautilus elegans, Sow. Chalk, Sussex, Mant.; Chalk, Rouen, Al Brong. 2. expansus, Sow. Chalk, Sussex, Mant. 3. insequalis, Sow. Gault, Sussex, Mant. ; Rouen, Pas. 4. obscurus, Nils. Chalk, Scania, Nils. 5. simplex, Sow. Upper Green Sand, Warminster, Berrett. Lyme Regis, De la B. Rouen, Al. Brong.; Green Sand? Aix-la Chapelle, Hoen. 6. Listeri, Mant., Gault, Sussex, Mant. ; Quedlinburg, G. T. ; Green Sand, Havre, Pas. 7. undulatus, Sow. Upper Green Sand, Nutfield, Sow. Green Sand, Griesenbruch, near Bochum, Hcen. , species not determined. Lower Green Sand, Sussex, Martin ; Speeton Clay, Yorkshire, Phil. ; Green Sand, M. de Fis, AL Brong.; Baculite limestone, Normandy, Desn. 1. Scaphites striatus, Mant. Chalk, Sussex, Mant.; Chalk, Rouen; Mont de Fis, Al. Brong. 2. costatus, Mant. Chalk, Sussex, Mant. ; Chalk, Rouen, AL Brong. species not determined. Baculite limestone, Normandy, Desn. ; Kopinge, Nils. 1 . Ammonites varians, Sow. Chalk, Sussex, Mant. ; Chalk, Rouen ; M. de Fis, Al. Brong.; Baculite limestone, Normandy, Desn.; Chalk and Upper Green Sand, Wiltshire, Lons.; Green Sand, Bo- chum, Hcen. 2. Woollgari, Mant. Chalk, Sussex, Mant. ; Rouen, Pas. 3. navicularis, Mant. Chalk, Sussex, Mant. ; Rouen, Pas. 4. catinus, Mant. Chalk, Sussex, Mant. 5. Lewesiensis, Mant. Chalk, Sussex, Mant. ; Chalk, Essen, Hoen. ; Rouen, Pas. 6. peramplus, Mant. Chalk, Sussex, Mant. 7. rusticus, Sow. Chalk, Lyme Regis, Buckl. ; Chalk, Sussex, Mant. ; Green Sand, Bochum, Hoen. 8. undatus, Sow. Chalk, Sussex, Mant. 9. Mantelli, Soiv. Chalk, Sussex, Mant.; Hanover, Holl ; Green Sand, Bochum ; Chalk, Saumur, Hcen. 10. RhotomagensisJ, Al. Brong. Chalk, Sussex, Mant.; Baculite limestone, Normandy, Desn.; Rouen, AL Brong.; Chalk, Wilts, Sow. 11. cinctus, Mant. Chalk, Sussex, Mant. 12. falcatus, Mant. Chalk, Sussex, Mant.; Chalk, Rouen, AL Brong. 13. curvatus, Mant. Chalk, Sussex, Mant. 14. complanatus, Mant. Chalk, Sussex, Mant. 15. rostratus, Sow. Chalk, Sussex, Mant.; Chalk, Oxfordshire, Buckl. * B. semicanaliculatus, Blainv. f B. Scanise, Blainv. According to Sowerby, Am. Rhotomagensis and Am. Sussexiensis are the same shell. 526 Organic Remains of the Cretaceous Group. 16. Ammonites tetrammatus, Sow. Chalk, Sussex, Mant. 17. planulatus, Sow. Upper Green Sand, Sussex, Mant. 18. Catillus, Sow. Upper Green Sand, Sussex, Mant. 19. , splendens, Sow. Gault, Sussex, Mant.; Green Sand, Senefon- taine, Norm., Pas. 20. auritus, Sow. Upper Green Sand, Devizes, Gent . ; Gault, Sus- sex, Mant. ; Havre, Pas. 21. planus, Mant. Gault, Sussex, Mant.; Speeton Clay? York- shire, Phil. 22. lautus, Park. Gault, Sussex, Mant. 23. tuberculatus, Sow. Gault, Sussex, Mant. 24. Goodhalli, Sow. Lower Green Sand, Sussex, Mant. ; Green Sand, Blackdown, Goodhall; Green Sand, Lyme Regis, De laB. 25. venustus, Phil. Speeton Clay, Yorkshire, Phil. 26. concinnus, Phil. Speeton Clay, Yorkshire, Phil. 27. . . Rotula, Sow. Speeton Clay, Yorkshire, Phil. 28. trisulcosus, Phil. Speeton Clay, Yorkshire, Phil. 29. . marginatus, Phil. Speeton Clay, Yorkshire, Phil. 30. parvus, Sow. Speeton Clay? Yorkshire, Phil. 31. hystrix, Phil. Speeton Clay, Yorkshire, Phil. 32. fissicostatus, Phil. Speeton Clay, Yorkshire, Phil. 33. curvinodus, Phil. Speeton Clay, Yorkshire, Phil. 34. inflatus, Sow. Green Sand, I. of Wight, Buckl. ; Green Sand, Perte du Rhone ; Rouen ; Havre ; M. de Fis, Al. Brong. ; Upper Green Sand, Wilts, Lons. 35. Deluci, Al. Brong. Green Sand, Perte du Rhone ; M. de Fis, Al. Brong.; Puits de Meulers, Norm., Pas. 36. subcristatus, De Luc Green Sand, Perte du Rhone, Al. Brong. 37. Beudanti*, Al. Brong. Green Sand, Perte du Rhone ; M. de Fis, Al. Brong. 38. clavatus, De Luc. Green Sand, M. de Fis, Al. Brong.; Rouen, Pas. 39. Selliguinus, Al. Brong. Green Sand, M. de Fis, Al. Brong. ; Chalk, Lublin, Poland, Pusch; Chalk, Essen, Ham.; Gault, Sussex, Mant. ; Rouen, Pas. 40. Gentoni, Defr. Baculite Limestone, Normandy, Desn.; Gault, Sussex, Mant. ; Chalk, Rouen, Al. Brong. 41. constrictus, Sow. Baculite Limestone, Normandy, Desn.; Chalk, Lublin, Poland, Pusch. 42. Stobaei, Nils. Chalk, Scania, Nils. 43. varicosus, Sow. Green Sand, Blackdown, 'Sow. 44. Hippocastanum, Sow. Chalk with quartz grains, Lyme Regis, De la B. ; Green Marl, Puits de Meulers, Norm., Pas. 45. Benettianus, Sow. Gault, Warminster, Lons. 46. denarius, Sow. Green Sand, Blackdown, Goodhall. 47. Nutfieldiensis, Sow. Chalk, near Calne, Lons. 48. virgatus, Goldf., G. S. Moskau, G. T. 49. Coupei, Al. Brong ; Rouen, Pas. 50. canteriatus, Al. Brong. ; Rouen, Pas. 1. Turrilites costatus, Sow. Chalk, Sussex, Mant.; Chalk, Rouen; Havre, Al. Brong. ; Chalk, near Calne, Lons. 2. undulatus, Sow. Chalk, Sussex, Mant.; Rouen, Pas. 3. tuberculatus, Sow. Chalk, Sussex, Mant. * According to Sowerby, Am. Beudanti, Am. Selliguinus, and Am, lavigatns are the same shell. Organic Remains of the Cretaceous Group. 527 4. Turrilites Bergeri, AL Brong.; Green Sand, Perte du Rhone; M. de Fis, AL Brong. 5. ? Babeli, AL Brong. Green Sand, M. de Fis, AL Brong.; Rouen, Pas. 6. acutus, Pas. ; Rouen, Pas. , species not determined. Green Sand, Maritime Alps, Risso. 1. Baculites Faujasii, Lam. Chalk, Sussex, Mant.; Chalk, Norfolk, Rose ; Maestricht, Desm. ; Chalk, Sweden, Nils. ; Boclium, Aix-la- Chapelle, Hoen. 2. obliquatus, Sow. Chalk, Sussex, Mant. ; Scania, Nils. ; Rouen, Pas. 3. vertebralis, Defr. Chalk, Maestricht, Fauj. de St. Fond; Ba- culite Limestone, Normandy, Desm. 4. anceps, Lam. Chalk, Scania, Nils. 5. triangularis, Desm. Maestricht, Desm. 1. Hamites armatus, Sow. Chalk, Sussex, Mant.; Chalk, Oxfordshire, Buckl. ; Rouen, Pas. 2. plicatilis, Mant. Chalk, Sussex, Mant. Speeton Clay? York- shire, Phil.; Rouen, Pas. 3. alternatus, Mant. Chalk, Sussex, Mant. ; Speeton Clay, York- shire, Phil. 4. ellipticus, Mant. Chalk, Sussex, Mant. ; Baculite Limestone ? Normandy, Desn. 5. attenuatus, Sow. Chalk, Gault, Sussex, Mant. ; Speeton Clay, Yorkshire, Phil. ; Rouen, Pas. 6. maximus, Sew. Gault, Sussex, Mant. ; Speeton Clay, York- shire, Phil. 7. intermedius, Sotv. Gault, Sussex, Mant. ; Speeton Clay, York- shire, Phil.; Green Sand, Aix-la-Chapelle, Hoen.; Rouen, Pas. 8. tenuis, Sow. Gault, Sussex, Mant.; Rouen, Pas. 9. rotundus, Sow. Gault, Sussex, Mant.; Speeton Clay, Yorkshire, Phil. ; Green Sand, Perte du Rhone, AL Brong. ; Green Sand, Aix-la-Chapelle, Hccn. ; Rouen, Pas. 10. compressus, Sow. Gault, Sussex, Mant. ; Green Sand, Nice, Risso. 1 1 . raricostatus, Phil. Speeton Clay, Yorkshire, Phil. 12. - Beanii, Y. $ B. Speeton Clay, Yorkshire, Phil. 13. Phillipsii, Bean. Speeton Clay, Yorkshire, Phil. 14. funatus, AL Brong. Green Sand, Perte du Rhone; M. de Fis, AL Brong. ; Rouen, Pas. 15. canteriatus, AL Brong. Green Sand, Perte de Rhone, AL Brong. 16. virgulatus, AL Brong. Green Sand, M. de Fis, AL Brong. 17. cylindricus, Defr. Baculite Limestone, Normandy, Desn. 18. spinulosus, Sow. Green Sand, Blackdown, Miller. 19. grandis, Sow. Lower Green Sand, Kent, Buckl. 20. Gigas, Sow. Lower Green Sand, Hythe, Kent, G. E. Smith. 21. spiniger, Sow. Gault, Folkestone, Gibbs. CRUSTACEA. 1. Astacus Leachii, Mant. Chalk, Sussex, Mant. 2. Sussexiensis, Mant. Chalk, Sussex, Mant. ornatus, Phil. Speeton Clay, Yorkshire, Phil. 4. longimanus, Sow. Green Sand, Lyme Regis, De la B. , species not determined. Gault, Sussex, Mant. 1. Pagurus Faujasii, Desm. Chalk? Sussex, Mant. ; Maestricht. 1. Scyllarus Mantelli, Desm. Chalk, Sussex, Mant. 528 Organic Remains of the Wealden Rocks of England. 1 . Orythia Labechii, DesL ; Green Sand, Calvados, DesL Eryon, species not determined. Chalk, Sussex, Mant. Arcania, species not determined. Gault, Sussex, Mant. Etysea, species not determined. Gault, Sussex, Mant. Coryster, species not determined. Gault, Sussex, Mant. PISCES. Squalus Mustelus? Chalk, Sussex, Mant. Galeus? Chalk, Sussex, Mant. pristodontes, Bronn. Aix-la-Chapelle, G. T. Muraena, Lewesiensis, Mant. Chalk, Sussex, Mant. Zeus Lewesiensis, Mant. Chalk, Sussex, Mant. Salmo? Lewesiensis, Mant. Chalk, Sussex, Mant. Esox Lewesiensis, Mant. Chalk, Sussex, Mant. Amia ? Lewesiensis, Mant. Chalk, Sussex, Mant. Fish, genera not determined. Speeton Clay, Yorkshire, Phil. ; Chalk, Paris, Al. Brong.; Chalk, Lyme Regis, De la B.; Upper Green Sand, Wilts, Lorn. Gault, Isle of Wight, Fitton ; Chalk, Troyes, Clement-Mullet. teeth and palates; common in England and France, var. authors; Bochum ; Aix-la-Chapelle, Hcen. ; Scania, Nik. REPTILIA. 1. Mososaurus Hoffmanni, Maestricht, Fauj. de St. Fond; Chalk, Sussex, Mant. 1. Crocodile of Meudon, Cuv. ; Chalk, Meudon, AL Brong. Reptiles, genera not determined. Speeton Clay, Yorkshire, Phil. Organic Remains of the Wealden Rocks of England. PLANTS. 1. Sphenopteris Mantelli, Ad. Brong. Hastings Sands, Sussex, Mant. 1. Lonchopteris Mantelli, Ad. Brong. Hastings Sands, Sussex, Mant. Lycopodites ? species not determined. Hastings Sands, Sussex, Mant. 1. Clatharia Lyellii, Mant. Hastings Sands, Sussex, Mant. 1. Carpolithus Mantelli, Ad. Brong. Hastings Sands, Sussex, Mant. Lignite, and undescribed vegetables. Hastings Sands, Sussex, Mant. CONCHIFERA AND MoLLUSCA. Corbula, species not determined. Pounceford, Fitton. Tellina, species not determined. Pounceford, Fitton. Mytilus, species not determined. Pounceford, Fitton. Ostrea, species not determined. Weald Clay, Isle of Wight, Sedg. ; Purbeck Beds, near Weymouth, BucM. and De la B. 1 . Cyclas membranacea, Sow. Weald Clay, Hastings Sands, Ashburnham Beds, Sussex, Mant. ; Weald Clay ? Swanage Bay, Fitton. 2. media, Sow. Weald Clay, Hastings Sands, Ashburnham Beds r Sussex, Mant. ; Weald Clay, Isle of Wight, Swanage Bay Hastings Sands, Isle of Wight ; Sussex, Fitton. species not determined. Weald Clay, Isle of Wight; Swanage Bay, Fitton. 1. Unio porrectus, Sow. Hastings Sands, Sussex, Mant. 2. compressus, Sow. Hastings Sands, Sussex, Mant. 3. antiquus, Sow. Hastings Sands, Ashburnham Beds, Sussex, Mant. 4. aduncus, Sow. Hastings Sands, Sussex, Mant. 5. cordiformis, Sow. Hastings Sands, Sussex, Mant. Bulla, small species. Tilgate beds, Fitton. Organic Remains of the Oolitic Group. 529 Melanopsis, species not determined. Pounceford, Fitton. 1. Paludina vivipara, Lam. Weald Clay, Hastings Sands, Ashburnham Beds, Sussex, Mant. ; Purbeck Beds, Purbeck, Conyb. 2. elongata, Sow. Weald Clay, Hastings Sands, Ashburnham Beds, Sussex, Mant. ; Weald Clay, Isle of Wight ; Swanage Bay, Fitton. 3. carinifera, Sow. Weald Clay, Sussex, Mant. 4. Sussexiensis, - . Wealden rocks, Fitton. Potamides, species not determined. Weald Clay, Sussex, Mant. Neritina, species not determined. Tilgate Beds, Fitton. PISCES. Lepisosteus, species not determined. Hastings Sands, Sussex, Mant. Silurus, species not determined. Hastings Sands, Sussex, Mant. Remains of Fish, genera not determined. Weald Clay, Ashburnham Beds, Sussex, Mant. ; Purbeck Beds, Purbeck, De la B> ; Hastings Sands, Isle of Wight, Fitton. CRUSTACEA. 1. Cypris faba, Desm. Weald Clay, Isle of Wight; Swanage Bay, &c. Fitton; Weald Clay, Hastings Sands, Sussex, Mant. REPTILIA. 1. Crocodilus priscus, . Hastings Sands, Sussex, Mant. , species not determined. Ashburnham Beds, Sussex, Mant. ; Purbeck Beds, Purbeck, Conyb. ; Weald Clay, Swanage Bay, Fitton. 1. Iguanodon Mantelli, v. Meyer. Hastings Sands, Sussex, Mant. Megalosaurus . Hastings Sands, Ashburnham Beds, Sussex, Mant. Hylaeosaurus . Hastings Sands, Sussex, Mant. Reptiles of the genera Trionyx, Emys, Chelonia, Plesiosaurus, and Ptero- dactylus ? Hastings Sands, Sussex, Mant. Tortoise, Purbeck Beds, Purbeck, Conyb* Organic Remains of the Oolitic Group. PLANTS. Alga. 1. Fucoides furcatus, Ad. Brong. Stonesfield slate, Ad. Brong. 2. Stockii, Ad. Brong. Solenhofen, Ad. Brong. 3. encelioides, Ad. Brong. Solenhofen, Ad. Brong. Equisetaceat. 1. Equisetum columnare, Ad. Brong. Lower carbonaceous series, York- shire, Phil. ; Brora, Murch. Filices. 1. Pachypteris lanceolata, Ad. Brong. Coal, shale, &c. between inferior and great oolite, Yorkshire, Phil. 2. ovata, Ad. Brong. Coal, shale, &c. between inferior and great oolite, Yorkshire, Phil. I. Pecopteris Reglei, Ad. Brong. Forest marble, Mamers, Dcsn. * In this list, the sands, sandstones, and clays, grouped by Mr. Mantell under me head of Tilgate Beds, are given as Hastings Sands, although this an'aiigpmetrt. thay perhaps clash with one or two local divisions. 2 M 530 Organic Remains of the Oolitic Group. 2. Pecopteris Desnoyersii, Ad. Brong. Forest marble, Mamers, Dem. 3. polypodioides, Ad. Brong. Coal, shale, &c. between cornbrash and great oolite, Yorkshire, Phil. 4. denticulata, Ad. Brong. Coal, shale, &c. between cornbrash and great oolite, Yorkshire, Phil. 5. Phillipsii, Ad. Brong. Coal, &c. of the oolitic series, Yorkshire, Ad. Brong. 6. Whitbiensis, Ad. Brong. Coal, shale, &c. between cornbrash and great oolite, Yorkshire, Phil. 1. Sphsenopteris hymenophylloides, Ad. Brong. Stonesfield slate, Buckl.; Coal, shale, &c. between great and inferior oolite. Yorkshire. Phil. 2. ? macrophylla, Ad. Brong. Stonesfield slate, Buckl. 3. Williamsonis, Ad. Brong. Coal, &c. of the oolitic series, York- shire, Ad. Brong. 4. crenulata, Ad. Brong. Coal, &c. of the oolitic series, Yorkshire, Ad. Brong. 5. denticulata, Ad. Brong. Coal, &c. of the oolitic series, York- shire, Ad. Brong. 1. Tseniopteris latifolia, Ad. Brong. Coal, shale, &c. between cornbrash and great oolite, Yorkshire, Phil. 2. vittata, Ad. Brong. Coal, shale, &c. between cornbrash and great oolite, Yorkshire, Phil. 1. Cyclopteris Beanii, L.&H. Coal, shale, &c. of the oolitic series, York- shire, Williamson. 2. digitata, L. & H. Sandstone, near Scarborough, Geol. Soc. Mus. 1. Glossopteris Phillipsii, Ad. Brong. Shale, near Scarborough, Bean. 1. Neuropteris recentior, L. & H. Gristhorpe, Scarborough, Bean. 2. ligata, L. & H. Gristhorpe, Scarborough, Bean. Lycopodiacece. 1. Lycopodites falcatus, L. & H. Clougton, Yorkshire, Bean. Cycadece. 1. Pterophyllum Williamsonis. Coal, shale, &c. between cornbrash and great oolite, Yorkshire, Phil. 2. i comptum, L. & H. Gristhorpe, Scarborough, Bean. 3. minus, L. & H. Near Scarborough, Mus. Geol. Soc. 4. Nilsoni, L. & H. Near Scarborough, Bean. 1. Zamia pectinata, Ad. Brong. Stonesfield slate, Buckl. 2. patens, Ad. Brong. Stonesfield slate, Ad. Brong. 3. longifolia, Ad. Brong. Coal, shale, &c. between cornbrash and great oolite, Yorkshire, Phil. 4. pennseformis, Ad. Brong. Coal, shale, &c. between great and inferior oolite, Yorkshire, Phil. 5. elegans, Ad. Brong. Coal, shale, &c. between great and inferior oolite, Yorkshire, Phil. 6. Goldiaei, Ad. Brong. Coal, &c. of the oolitic series, Yorkshire, Ad. Brong. 7. acuta, Ad. Brong. Coal, &c. of the oolitic series, Yorkshire, Ad. Brong. 8. laevis, Ad. Brong. Coal, &c. of the oolitic series, Yorkshire, Ad. Brong. 9. Youngii, Ad. Brong. Coal, shale, &c. between great and inferior oolite, Yorkshire, Phil. 10. Feneonis, Ad. Brong. Coal, &c. of the oolitic series, Yorkshire, Ad. Brong. Organic Remains of the Oolitic Group. 531 1 1 . Zamia Mantelli, A d. Brong. Coal, shale, &c. between great and inferior oolite, Yorkshire, Phil. 1. Zamites Bechii, Ad. Brong. Forest marble, Mamers, Desn. ; Lias, Lyme Regis, De la B. 2. Bucklandii, Ad. Brong. Forest marble, Mamers, Desn. ; Lias, Lyme Regis, De la B. 3. Lagotis, Ad. Brong. Forest marble, Mamers, Desn. 4. hastata, Ad. Brong. Forest marble, Mamers, Desn. ConifertB. 1. Thuytes divaricata, Sternb. Stonesfield slate, BuckL; Solenhofen, G. T. 2. expansa, Sternb. Stonesfield slate, Buckl. 3. acutifolia, Ad. Brong. Stonesfield slate, BucJcl. 4. cupressiformis, Sternb. Stonesfield slate, Buckl. 1. Taxites podocarpoides, Ad. Brong. Stonesfield state, Buckl. Lilia. 1 . Bucklandia squamosa, Ad, Brong. Stonesfield, Buckl. Class uncertain. 1. Mamillaria Desnoyersii, Ad. Brong. Mamers, Desn. Many undescribed vegetables. Lias, Lyme Regis, De la B. ZOOPHYTA. 1. Achilleum dubium, Goldf. Solenhofen, Gold/. 2. cheirotonum, Goldf. Oolitic rocks, Baireuth, Munst. 3. muricatum, Goldf. Streitberg, Munst. 4. tuberosum, Munst. Hattheim, Munst. 5. cancellatum, Munst. Hattheim, Munst. 6. costatum, Munst. Streitberg, Munst. 1. Manon Peziza, Goldf. Streitberg; Hattheim: Giengen ; Regensberg, Goldf. 2. marginatum, Munst. Streitberg : Muggendorf, Munst. 3. impressum, Munst. Muggendorf, Munst. 1. Scyphia cylindrica, Goldf. Muggendorf, Munst. 2. elegans, Goldf. Thurnau; Baireuth, Goldf. 3. calopora, Goldf. Thurnau; Baireuth, Goldf. 4. pertusa, Goldf. Streitberg; Baireuth, Goldf. 5. texturata, Goldf. Giengen, Wurtemberg, Goldf. 6. texata, Goldf. Legerberg, Switzerland ; Streitberg, Goldf. Calc. Grit, Bernese Jura, Thur. 7. polyommata, Goldf. Baireuth & Switzerland, Goldf. 8. clathrata, Goldf. Streitberg ; Baireuth, Goldf. 9. milleporata, Goldf. Baireuth, Goldf. 10. parallela, Goldf. Streitberg, Munst. 11. psilopora, Goldf. Muggendorf, Goldf. 12. obliqua, Goldf. Muggendorf, Munst.; Oxford Clay, Bernese Jura, Thur. 13. rugosa, Goldf. Streitberg, Munst. 14. articulata, Goldf. Muggendorf, Goldf. 15. pyriformis, Goldf. Streitberg, Munst. 16. radiciformis, Goldf. Streitberg, Goldf. 17. punctata, Goldf. Streitberg, Munst. 18. reticulata, Goldf. Streitberg, Goldf. 19. dictyota, Goldf. Streitberg, Munst. 20. procumbens, Goldf. Baireuth, Goldf. 21. paradoxa, Munst. Streitberg & Amberg, Munst. 22. empleura, Munst. Streitberg, Munst. 2 M 2 532 Organic Remains of the Oolitic Group. 23. Scyphia striata, Munst. Streitberg & Muggendorf, Munst. 24. Buchii, Munst. Streitberg, Munst. 25. Munsteri, Goldf. Regensberg ; Streitberg, Goldf. 26. propinqua, Munst. Streitberg ; Muggendorf, Munst. 27. cancellata, Munst. Streitberg ; Muggendorf, Munst. 28. decorata, Munst. Muggendorf, Munst. 29. Humboldtii, Munst. Muggendorf, Munst. 30. ' Sternbergii, Munst. Streitberg, Munst. 31. Schlotheimii, Munst. Thurnau; Streitberg, Munst. 32. Scbweiggeri, Goldf. Baireutb, Goldf. 33. secunda, Munst. Heiligenstadt; Streitberg, Munst.; Calc. Grit, Bernese Jura, Thur. 34. verrucosa, Goldf. Streitberg & Wurgau, Goldf. 35. Bronnii, Munst. Wurtemberg & Baireuth, Munst. ; Calc. Grit, Bernese Jura, Thur. 36. milleporacea, Munst. Thurnau ; Aufsees ; Streitberg, Munst. 37. pertusa, Goldf. Streitbeig & Amberg, Goldf. 38. intermedia, Munst. Hattheim ; Streitberg, Munst. 9. Neesii, Goldf. Streitberg, Goldf. 40. turbinata, Goldf. Streitberg, Goldf. 41. tenuistriata, Goldf. Streitberg, Goldf. 1. Tragos pezizoides, Goldf. Muggendorf, Goldf. 2. Palella, Goldf. Wurtemberg & Switzerland ; Rabenstein ; Heili- genstadt, Goldf. 3. sphaerioides, Goldf. Sigmaringen, Wurtemberg, Goldf. 4. tuberosum*, Goldf. Inferior Oolite, Rabenstein; Streitberg, Munst. 5. acetabulum, Goldf. Streitberg ; Randen, Goldf. 6. radiatum, Munst. Streitberg, Munst. 7. rugosum, Munst. Streitberg, Munst. 8. reticulatum, Munst. Streitberg, Munst. 9. verrucosum, Munst. Streitberg, Munst. 1. Spongia floriceps, Phil. Coral Oolite, Yorkshire, Phil. 2. clavaroides. Lam. Great Oolite, Wiltshire, Lons. , species not determined. Lower Calcareous Grit, Yorkshire, Phil.; Inferior Oolite, Middle and South of England, Conyb. ; Fo- rest Marble, Wiltshire, Lons. Alcyonium, species not determined. Forest Marble, Norman dy,De Cau.; Great Oolite ? Wilts, Lons. 1. Cnemidium lamellosum, Goldf. Randen, Switzerland, Goldf. 2. stellatum, Goldf. Randen, Switzerland, Goldf. 3. striato-punctatum, Goldf. Randen, Goldf. 4. rimulosum, Goldf. Randen, Goldf. 5. mammillare, Goldf. Streitberg, Goldf. 6. Rotula, Goldf. Thurnau, Goldf. 7. granulosum, Munst. Streitberg, Munst. 8. astrophorum, Munst. Hattheim ; Regensberg, Munst. 9. capitatum, Munst. Amberg, Munst. 1. Limnorea rnammillaris f, Lam x . Forest Marble, Normandy, De Can. 1. Siphonia pyriformis, Goldf. Streitberg, Goldf. 1. Myrmecium hemisphsericum, Goldf. Thurnau, Goldf. I. Gorgonia dubia, Goldf. Gliicksbrunn ; Thuringia, Goldf. 1. Millepora dumetosa, Lam*. Forest Marble, Normandy, De Can. 2. corymbosa, Lam*. Forest Marble, Normandy, De Cau. 3. conifera, Lam*. Forest Marble, Normandy, De Cau. * Limnorea lamellosa of Lamouroux according to M. Goldfuss. f Is this Limnorea mammillosa, Lam. ? If it be, it is the Cnemidium tuberosum of Goldfuss. Organic Remains of the Oolitic Group. 533 4. Millepora pyriformis, Lam*. Forest Marble, Normandy, De Can. 5. macrocaule, Lam'. Forest Marble, Normandy, De Can. 6. straminea, Phil. Great Oolite and Cornbrasb, Yorkshire, Phil. , species not determined. Cornbrash and Forest Marble, North of France, Bobl. ; Forest Marble, Mamers, Normandy, Desn.; Forest Marble and Great Oolite, Wiltshire, Lons. 1. Madrepora limbata, Golclf. Heidenheim, G. T. 1. Cellepora orbiculata, Gold/. Streitberg, Munst. ; Oxford Clay, Haute Saone, Thir. 2. echinata, Goldf. Inferior Oolite, Haute Saone, Thir. , species not determined. Inferior Oolite, Midland and Southern England, Conyb. Retepora? . Great Oolite, Yorkshire, Phil Flustra, species not determined. Great Oolite, Wiltshire, Lont. 1 . Ceriopora radiciformis, Goldf. Thurnau, Baireuth, Goldf. 2. striata, Goldf. Streitberg ; Thurnau, Munst. 3. angulosa, Goldf. Thurnau, Munst. 4. alata, Goldf. Thurnau, Munst. 5. crispa, Goldf. Thurnau, Munst. 6. favosa, Goldf. Streitberg ; Thurnau, Munst. 7. radiata, Goldf. Thurnau, Munst. 8. compressa, Munst. Thurnau, Munst. 9. orbiculata, Goldf. Inferior Oolite, Haute Saone, Thir. ; Calc. Grit, Inferior Oolite, Bernese Jura, Thur. 1. Agaricia rotata, Goldf. Randenberg, Switzerland, Goldf. 2. crassa, Goldf. Randen, Switzerland, Goldf. 3. granulata, Munst. Bale ; Hattheim, Munst. 1. Lithodendron elegans, Munst. Wurtemberg, Munst. 2. compressum, Munst. Heidenheim, Wurtemberg, Munst. 3. Raaracum, Thur. Coral Rag, Bernese Jura, Thur. 1. Caryophyllia cylindrica, Phil. Coralline Oolite, Yorks., Phil. 2. truncata, Lam*. Forest Marble, Normandy, De Can. 3. Brebissonii, Lam*. Forest Marble, Normandy, De Can. 4. convexa, Phil. Inferior Oolite, Yorkshire, Phil. 5. like C. cespitosa, Ellis. Coral Oolite, Yorks., Phil. ; Great Oolite, Mid. and S. of England, Conyb. 6. like C. flexuosa, Ellis. Coral Oolite, Yorkshire, Phil. ; Great Oolite, Midland and Southern England, Conyb. 7. - approaching C. Carduus, Park. Coral Rag, Great Oolite, Mid- dle and South of England, Conyb. , species not determined. Inferior Oolite, North of France, Bobl. ; Rochelle Beds, Dufr. ; Forest Marble, Mamers, Normandy, Desn.; Forest Marble, Bradford Clay, and Great Oolite, Wiltshire, Lons. 1. Anthophyllum turbinatum, Munst. Hattheim; Heidenheim, Munst. 2. obconicum, Munst. Hattheim ; Heidenheim, Munst. ; Calc. Grit, Bernese Jura, Thur. 3. decipiens, Goldf. Alsace, Goldf. 1. Fungia orbiculites, Lam . Forest Marble, Normandy, De Cau. ; Corn- brash, Wiltshire, Lons. 2. Isevis, Goldf. Calc. Grit, Bernese Jura, Thur. , species not determined. Inferior Oolite, Midland and Southern England, Conyb. 1 . Turbinolia dispar, Phil. Coral Oolite, Yorkshire, Phil. 2. dydyma, Goldf. Coral Rag, Bernese Jura, Thur. , species not determined. Inferior Oolite and Lias, North of France, Bobl. 1. Turbinolopsis ochracea, Lam x . Forest Marble, Normandy, De Cau. 534? Organic Remains of the Oolitic Group. 1 . Cyathophyllum Tintinnabulum, Goldf. Banz ; Staffelstein ; Bamberg, Goldf. 2. Mactra, Goldf. Banz ; Bamberg, Goldf. 3. quadrigeminum, Goldf. Coral Rag, Bernese Jura, Thur. 4. ceratites, Goldf. Calc. Grit, Bernese Jura, Thur. 5. plicatum, Goldf. Calc. Grit, Bernese Jura, Thur. 6. vermiculare, Goldf. Calc. Grit, Bernese Jura, Thur. 1. Meandrina Soemmeringii, Munst. Hattheim; Heidenheim, Munst. 2. astroides, Goldf. Coral Rag, Haute Saone, Thir. ; Giengen, Goldf. 3. tenella, Goldf. Giengen, Goldf. ; Coral Rag, Bernese Jura, Thur. 4. magna, Thur. Coral Rag, Bernese Jura, Thur. 5. foliacea, Thur. Coral Rag, Bernese Jura, Thur. , species not determined. Inferior Oolite and Coral Oolite, Yorks., Phil. ; Inferior Oolite ? Midi, and Southern England, Conyb.; Kimmeridge Clay, Haute Saone, Thir.; Great Oolite, Wilts, Lons. 1. Astrea Microconos, Goldf. Biberbach, near Muggendorf, Goldf. 2. limbata, Goldf. Giengen, Goldf. 3. concinna, Goldf. Giengen, Goldf. 4. pentagonalis, Munst. Hattheim; Heidenheim, Munst. 5. gracilis, Munst. Boll, Wurtemberg, Munst. 6. explanata, Munst. Wurtemberg, Munst. 7. tubulosa, Goldf. Wurtemberg, Goldf. ; Coral Rag, Haute Saone, Thir. ; Coral Rag, Bernese Jura, Thur. 8, oculata, Goldf. Giengen, Goldf. ; Coral Rag, Haute Saone, Thir. 9. alveolata, Goldf. Heidenheim, Wurtemberg, Goldf. 10. helianthoides, Goldf. Heidenheim; Giengen, Goldf.; Inferior Oolite, Coral Rag, Haute Saone, Thir. ; Coral Rag, Inferior Oolite, Bernese Jura, Thur. 11. confmens, Goldf. Heidenheim; Giengen, Goldf.; Coral Rag, Haute Saone, Thir. ; Coral Rag, Bernese Jura, Thur. 12. caryophylloides, Goldf. Giengen, Goldf. ; Coral Rag, Haute Saone, Thir. ; Coral Rag, Bernese Jura, Thur. 13. cristata, Goldf. Giengen ; Heidenheim, Goldf. ; Coral Rag, Bernese Jura, Thur. 14. sexradiata, Goldf. Giengen, Goldf. 15. favosioides, Smith. Coral Oolite, Yorkshire, Phil. ; Coral Rag and Great Oolite, Midland and Southern England, Conyb. 16. insequalis, Phil. Coral Oolite, Yorkshire, Phil. 17. micastron, Phil. Coral Oolite, Yorkshire, Phil. 18. arachnoides, Flem. Coral Oolite, Yorkshire, Phil. 19. tubulifera, Phil. Coral Oolite, Yorkshire, Phil. 20. macropthalma, Goldf. Kim. Clay, Porrentruy, Bernese Jura, Thur. 21. textilis, Goldf. Coral Rag, Bernese Jura, Thur. 22. geminata, Goldf. Coral Rag, Bernese Jura, Thur. 23. velamentosa, Goldf. Coral Rag, Bernese Jura, Thur. 24. geometrica, Goldf. Coral Rag, Bernese Jura, Thur. , species not determined. Coral Rag, Normandy, numerous, De Can. ; Great Oolite, Midland and Southern England, Conyb. ; Lias, Hebrides, Murch. ; Great Oolite, Wiltshire, Lons. 1. Thamnasteria Lamourouxii, Le Sauvage. Coral Rag, Norm., De Can. 1 . Aulopora compressa, Goldf. Rabenstein ; Grafenberg, Munst. 2. dichotoma, Goldf., Streitberg, Goldf. 3. intermedia, Miinst. Streitberg, Munst. Organic Remains of the Oolitic Group. 535 1. Entalophora cellarioides, Lam*. Forest Marble, Normandy, De Can. Favosites, species not determined. Forest Marble, Mamers, Normandy, 1 . Spiropora tetragona, Lam*. Forest Marble, Normandy, De Can. 2. caespitosa, Lam*. Forest Marble, Normandy, De Can.; Great Oolite, Wiltshire, Lons. 3. elegans, Lam x . Forest Marble, Normandy, De Cau. 4. intricata, Lam x . Forest Marble, Normandy, De Cau. 1. Eunomia radiata, Lam*. Forest Marble, Normandy, De Cau. ; Great Oolite, Wiltshire, Lons. 1. Chrysaora damsecornis, Lam x . Forest Marble, Normandy, De Can.; Great Oolite, Wiltshire, Lons. 2. spinosa, Lamx. Forest Marble, Normandy, De Cau. 1 . Theonoa clathrata, Lam*. Forest Marble, Normandy, De Cau. ; Great Oolite, Wiltshire, Lons. 1. Idmonea triquetra, Lam*. Forest Marble, Normandy, De Cau. ; Great Oolite, Wiltshire, Lons. 1. Alecto dichotoma, Lam*. Great Oolite, Wiltshire, Lons. ; Forest Mar- ble, Normandy, De Cau. , species not determined. Inferior Oolite, Midland and Southern England, Conyb. 1. Berenicea diluviana, Lam*. Great Oolite, Wiltshire, Lons.; Forest Marble, Normandy, De Cau. , species not determined. Great Oolite, Haute Saone, Thir. ; Forest Marble, Wiltshire, Lons. 1. Terebellaria ramosissima, Lam*. Forest Marble and Great Oolite, So- merset, Lons. ; Forest Marble, Normandy, De Cau. 2. Antilope, Lam*. Forest Marble, Normandy, De Cau. 1. Cellaria Smithii, Phil. Cornbrash, Yorkshire. Phil. 1. Sarcinula astroites, Goldf. Coral Rag, Bernese Jura, Thur. 1. Intricaria Bajocensis, Defr. Inferior Oolite, Bernese Jura, Thur. Explanaria, species not determined. Great Oolite, Wilts, Lons. Polypifers, genera not determined. Lias (rare), Lyme Regis, De la B. ; Lias (rare), Yorkshire, Phil.; Lias (rare), Normandy, De Cau. ; Coral Rag (numerous), North of France, Bobl. ; Coral Rag (abundant), Burgundy, Beaum. ; Coral Rag (abundant), South of France, Dufr.; Inferior Oolite, Calvados, Her. RADIARIA. 1. Cidaris florigemma, Phil. Coral Oolite, Yorkshire, Phil. 2. intermedia, Park. Coral Oolite, Yorkshire, Phil. 3. - monilipora, Y. fy B. Coral Oolite, Yorkshire, Phil. 4. vagans, Phil. Calcareous Grit, Cornbrash, and Great Oolite, Yorkshire, Phil. 5. crenularis, Lam. Coral Rag, Midland and Southern England, Conyb. ; Calc. Grit, Bernese Jura, Thur. (3. ornata, . Bradford Clay, North of France, Bobl. 7. globata, Schlot. Coral Rag, North of France, Bobl. 8. maxima, Munst. Baireuth ; Hohenstein, Saxony, Munst. 9. Blumenbachii, Munst. Thurnau, Muggendorf, Pretzfeld and Theta, Goldf.; Calc. Grit, Bernese Jura, Thur. 10. nobilis, Munst. Baireuth, Munst. 11. - elegans, Munst. Baireuth, Munst.; Kelloway Rock, Haute Saone, Thir. 12. marginata, Goldf. Regensburg, Heidenheim, Goldf. 13. coronata, Goldf. Coral Rag, Midland and Southern England, Conyb.; Streitberg, Thurnau, Stafielstein, Heidenheim, Ran- den, Goldf. ; Calc. Grit, Bernese Jura, Thur. 536 Organic Remains of the Oolitic Group. 14. Cidaris propinqua, Munst. Streitberg, Munst. ; Kim. Clay, Calc. Grit, Bernese Jura, Thur. 15. glandifera, Gold/. Altdorf, Bavaria; Wurtemberg; Randen, Goldf. ; Calc. Grit, Bernese Jura, Thur. 16. Schmidelii, Munst. Dischingen, Switzerland, Munst. 17. subangularis, Goldf. Thurnau ; Muggendorf, Gold/.; Kim. Clay, Bernese Jura, Thur. 18. variolaris, Al. Brony. Streitberg, Regensberg, Heidenheim, Goldf. , species not determined. Inf. Oolite, Yorkshire, Phil. ; Lias, Lyme Regis, De la B. ; Cprnbrash, Bradford Clay, Great Oolite, Inferior Oolite and Lias, Midland and Southern En- gland, Conyb. ; Coral Rag, Forest Marble, Normandy, De Can. ; Forest Marble, Great Oolite, Wiltshire, Lons. , spines of. Great Oolite and Lias, Yorkshire, Phil. ; Lias, Mid. and South of England, Conyb. ; Oolite beds, Lower System, South of France, Bobl.; Coral Rag, Normandy, Desn. ; Coral Rag, Haute Saone, Thir. 1. Echinus germinans, Phil. Coral Oolite, Calcareous Grit, and Great Oolite, Yorkshire, Phil. 2. lineatus, Goldf. Regensburg, Bale, Goldf. ; Calc. Grit, Bernese Jura, Tliur. 3. excavatus, Leske. Regensburg, Goldf. ; Calc. Grit, Bernese Jura, Thur. 4. nodulosus, Munst. Baireuth, Munst. 5. hieroglyphicus, Goldf. Regensburg; Thurnau, Goldf.; Calc. Grit, Bernese Jura, Thur. 6. sulcatus, Goldf. Thurnau; Streitberg; Muggendorf; Heiden- heim, Goldf. , species not determined. Coral Rag, North of France, Bobl. 1. Galerites depressus, Lam. Wurtemberg; Bavaria, Goldf.; Coral Oolite, Calcareous Grit, Cornbrash, Yorkshire, Phil. ; Oxford Clay, Normandy, Desn. ; Oxford Clay, Haute Saone, Thir. ; Ho- henstein, Saxony, Munst. ; Oxford Clay, Compound Great Oolite, Bernese Jura, Thur. 2. speciosus, Munst. Heidenheim, Wurtemberg, Munst. 3. Patella, . Oxford Clay, Normandy, Desn. 1 . Clypeaster pentagonalis, Phil. Calcareous Grit, Yorks., Phil. , species not determined : Coral Rag, Normandy, De Can. Kimmeridge Clay, Haute Saone, Thir. 1. Nucleolites scutatus, Lam. Oxford Clay, Normandy, Desn. ; Oxford Clay, Haute Saone, Thir. ; Oxford Clay, Compound Great Oolite, Bernese Jura, Thur. 2. columbarius, . Cornbrash, Forest Marble, North of France, Bobl. 3. granulosus, Munst. Amberg ; Streitberg ; Wiirgau, Munst. 4. semiglobus, Munst. Pappenheim ; Monheim ; Bavaria, Munst. 5. excentricus, Munst. Kehlheim, Bavaria, Munst. 6. canaliculatus, Munst. Blaubeuren, Wurtemberg, Munst. , species not determined : Oxford Clay, North of France, Bobl. 1. Ananchytes bicordatus, Lam. Oxford Clay, Normandy, Desn. ; Calc. Grit, Bernese Jura, Thur. 1 . Spajtangus ovalis, Park. Coral Oolite, Calcareous Grit, Kelloway Rock, Yorkshire, Phil. 2. - intermedius, Munst. Blaubeuren, W'urtemberg, Munst. 3. carinatus, Goldf. Baireuth, Wurtemberg, Goldf. 4, capistratus, Goldf. Baireuth, Goldf.; Oxford Clay, Haute Saone, Thir.; Oxford Clay, Bernese Jura, Thur. Organic Remains of the Oolitic Group. 537 Spatangus, species not determined : Cornbrash, Forest Marble, North of France, Bobl. 1 . Clypeus sinuatus, Park. Coral Oolite, Yorkshire, Phil. ; Coral Rag, Cornbrash, Great Oolite, Inferior Oolite, Mid. and Southern England, Conyb. ; Forest Marble, Normandy, De Can. 2. emarginatus, Phil. Coralline Oolite, Yorkshire, Phil. 3. clunicularis, Smith. Coral Oolite, Cornbrash, Yorkshire, Phil.; Coral Rag, Cornbrash, Great Oolite, Inferior Oolite, Midland and Southern England, Conyb. ; Forest Marble, Normandy, De Can. ; Coral Rag, Weymouth, Sedg. 4. dimidiatus, Phil. Coral Oolite, Yorkshire, Phil. 5. - semisulcatus, Phil. Coralline Oolite, Yorkshire, Phil. 6. orbicularis, Phil. Cornbrash, Yorkshire, Phil. , species not determined : Cornbrash, Great Oolite, Wiltshire, Lons. Echinites, genera not determined. Inferior Oolite, Normandy, De Can. , spines of. Coral Rag, Burgundy, Beaum. ; Coral Rag, North of France, Bobl. ; Forest Marble, Mamers, Desn. ; Mauriac beds, South of France, Dufr. 1. Eugeniacrinites caryophyllatus, Goldf. Baireuth; Wurtemberg; Swit- zerland, Gold}. ; Calc. Grit, Bernese Jura, Thur. 2. mutans, Goldf. Streitberg ; Muggendorf, Goldf. ; Calc. Grit, Bernese Jura, Thur. 3. pyriformis, Munst. Randen, Goldf. 4. - moniliformis, Munst. Thurnau; Streitberg; Switzerland, Goldf. 5. Hoferi, Munst. Switzerland ; Streitberg, Goldf. 6. compressus, Munst. Baireuth ; Wurtemberg, Munst. 1 . Apiocrinites rotundus, Miller. Forest Marble, Normandy, De Can. ; Bradford Clay, Great Oolite, Mid. and S. England, Conyb. ; Forest Marble, Buckl. ; Great Oolite, Alsace, Al. Brona. ; Forest Marble, Normandy, De Can. ; Forest Marble, Wilt- shire ; Great Oolite, Somerset, Lons. ; Germany ; Alsace, Goldf. ; Calc. Grit, Bernese Jura, Thur. 2. Prattii, Gray. Great Oolite, Somerset, Lons. 3. elongatus, Miller. Bale ; Soleure ; Elsas, near Befort, Alsace ; Forest Marble, Normandy, Goldf. 4. rosaceus, Schlot. Canton Soleure; Elsas; Muggendorf, Goldf.; Calc. Grit, Bernese Jura, Thur. 5. - mespiliformis, Schlot. Heidenheim ; Giengen, Goldf. ; Kim. Clay, Haute Saone, Thur. 6. - - Milleri, Schlot. Wurtemberg, Goldf.; Calc. Grit, Oxford Clay, Bernese Jura, Thur. 7. - flexuosus, Goldf. Wurtemberg, Goldf. 8. - subconicus, Goldf. Bath, Goldf. 1. Pentacrinites vulgaris, Schlot. Cornbrash, Coral Oolite, and Lias, Yorks., Phil. ; Inf. Oolite, and Lias, Midi, and S. England, Conyb. ; Lias, Alsace, Gundershofen, Figeac, Al. Brong. 2. subangularis, Miller. Inferior Oolite and Lias, Midland and Southern England, Conyb. ; Lias, Banz ; Boll, Goldf. 3. Briareus, Miller. Lias, Midland and Southern England, Conyb.; Lias, Yorkshire, Phil. ; Lias, Banz ; Boll, Goldf. 4. - basaltiformis, Miller. Lias, Midland and Southern England, Conyb. ; Lias, Alsace, Voltz ; Baireuth ; Banz ; Boll, Goldf. 5. - tuberculatus, Miller. Lias, Midland and Southern England, Conyb. ; Lias, Alsace, Voltz. - subteres, Goldf. (Var.) Oxford Clay, Haute Saone, Thir. 7. - Jurensis, Munst. Coral Rag, Haute Saone, Thir. 8. scalaris, Goldf. Baireuth ; Banz ; Boll, Goldf. 538 Organic Remains of the Oolitic Group. 9. Pentacrinites cingulatus, Munst. Streitberg ; Thurnau, Goldf. 10. pentagonalis, Goldf. Streitberg; Thurnau; Boll, Goldf.; Ox- ford Clay, Bernese Jura, Thur. 11. moniliferus, Munst. Lias, Baireuth, Goldf. 12. subsulcatus, Munst. Lias, Baireuth, Goldf. 13. subteres, Munst. Streitberg, Goldf. 14. ?paradoxus, Goldf. Baireuth; Wurtemberg, Goldf. , species not determined. Forest Marble, Normandy, De Can. ; Bradford Clay, North of France, Bobl. ; Cornbrash, Forest Marble, Great Oolite, Midland and Southern England, Conyb.; Inferior Oolite, Wotton-under-Edge, Forest Marble, Great Oolite, Somerset, Lons. 1. Solanocrinites costatus, Goldf. Giengen; Heidenheim, Wurtemberg, Goldf. 2. scrobiculatus, Munst. Streitberg ; Thurnau, Goldf. 3. Jaegeri, Goldf. Baireuth, Goldf. 1. Rhodocrinites echinatus, Schlot. Amberg; Wurtemberg; Switzerland; Berrach, Goldf. 1. Comatula pinnata, Goldf. Solenhofen, Goldf. 2. tenella, Goldf. Solenhofen, Goldf. 3. pectinata, Goldf. Solenhofen, Goldf. 4. filiformis, Goldf. Solenhofen, Goldf. 1. Ophiura Milleri, Phil. Lias, Yorkshire, Phil.; Inferior Oolite sands, Bridport, De la B. 2. speciosa, Munst. Solenhofen, Goldf. 3. carinata, Munst. Solenhofen, Goldf. 1. Asterias lumbricalis, Schlot. Walzendorf, Coburg; Lichtenfels, Bam- berg, Goldf. 2. lanceolata, Goldf. Walzendorf ; Lichtenfels, Goldf. 3. arenicola, Goldf. Porta Westphalica, Goldf. 4. Jurensis, Munst. Hattheim, Wurtemberg ; Baireuth, Goldf. 5. tabulata, Goldf. Streitberg, Goldf. 6. scutata, Goldf. Streitberg ; Heiligenstadt, Goldf. 7. stellifera, Goldf. Streitberg, Goldf. 8. prisca, Goldf. Wasseralfingen, Schiibber. , species not determined. Coral Rag, Bernese Jura, Thur. ANNULATA. 1. Lumbricaria Intestinum, Munst. Solenhofen, Goldf. 2. Colon, Munst. Solenhofen, Goldf. 3. recta, Munst. Solenhofen, Goldf. 4. gordialis, Munst. Solenhofen, Goldf. 5. coniugata, Munst. Solenhofen, Goldf. 6. Filaria, Munst. Solenhofen, Goldf. 1. Serpula squamosa, Bean. Coral Oolite, Yorkshire, Phil. 2. lacerata, Phil. Calcareous Grit, and Great Oolite, Yorkshire, Phil. ; Calc. Grit, Bernese Jura, Thur. 3. intestinalis, Phil. Oxford Clay, and Cornbrash, Yorkshire, Phil. 4. deplexa, Bean. Inferior Oolite, Yorkshire, Phil. 5. capitata, Phil. Lias, Yorkshire, Phil. 6. quadrangularis, Lam. Oxford Clay, Normandy, Dem. ; Calc. Grit, Bernese Jura, Thur. 7. sulcata, Sow. Calcareous Grit, Oxford, Sow. g. ,.. tricarinata, Sow. Calcareous Grit, Oxford ; Coral Rag, Steeple Ashton, Wilts, Sow. ; Oxford Clay, Haute Saone, Thir. 9. . triangulata, Sow. Bradford Clay or Great Oolite, Bradford, Sow. 10. . runcinata, Soiv. Coral Rag, Oxford, Sow. Organic Remains of the Oolitic Group. 539 11. Serpula tricristata, Gold/. Lias, Banz, Goldf. 12. quinque-cristata, Munst. Lias, Banz, Goldf. 13. quinque-sulcata, Munst. Lias, Theta, Baireuth, Goldf. 14. circinnalis, Munst. Lias, Banz, Goldf. 15. complanata, Goldf. Lias, Theta, Munst. 16. grandis, Goldf. Ferrug. Oolite, Baireuth; Wurtemberg ; Coral Oolite, Haute Saone ; Upper Jura Limestone, Heidenheim, Goldf. 17. Limax, Goldf. Ferrug. Oolite, Baireuth, Goldf. 18. conformis, Goldf. Alsace, Goldf.; Kim. Clay, Bernese Jura, Thur. 19. convoluta, Goldf. Ferrug. Oolite, Wasseralfingen ; Baireuth, Goldf. 20. lituiformis, Munst. Ferrug. Oolite, Grafenberg, Baireuth, Goldf. 21. Delphinula, Gcldf. Thurnau; .Streitberg, Goldf. 22. capitata, Goldf. Streitberg, Goldf. ; Calc. Grit, Bernese Jura, Thur. 23. limata, Munst. Streitberg, Goldf. 24. plicatilis, Munst. Grafenberg ; Streitberg, Goldf. 25. gibbosa, Goldf. Muggendorf, Goldf. 26. nodulosa, Goldf. Streitberg, Goldf. 27. Spirolinites, Munst. Streitberg, Goldf. 28. tricarinata, Goldf. Ferrug. Oolite, Rabenstein ; Baireuth ; Alsace, Goldf. 29. pentagona, Goldf. Streitberg, Goldf. 30. quinquangularis, Goldf. Kim. Clay, Largue, Sundgau; Nor- mandy, Goldf.; Calc. Grit, Bernese Jura, Thur. 31. quadrilatera, Goldf. Ferrug. Oolite, Rabenstein; Buxweiler, Goldf. 32. vertebralis, Sow. Buxweiler, Goldf. 33. _ prolifera, Goldf. Streitberg, Goldf. 34. - planorbifonnis, Munst. Thurnau ; Streitberg, Goldf. 35. trochleata, Munst. Streitberg, Goldf. 36. macrocephala, Goldf. Thurnau, Goldf. 37. heliciformis, Goldf. Neuburg; Doubs, Goldf. 38. quadristriata, Goldf. Berrach, Burgundy ; Amberg, Goldf. 39. convoluta, Munst. Streitberg, Goldf. 40. canaliculata, Munst. Streitberg, Goldf. 41. Deshayesii, Munst. Streitberg, Goldf. 42. volubilis, Munst. Ferrug. Oolite, Rabenstein, Goldf. 43. spiralis, Munst. Muggendorf; Hattheim ; Heidenheim, Goldf. 44. cingulata, Munst. Streitberg, Goldf. 45. Flagellum, Munst. Streitberg, Goldf. 46. substriata, Munst. Ferrug. Oolite, Rabenstein, Goldf. 47. flaccida, Munst. Ferrug. Oolite, Rabenstein; Bale; Elsas, Goldf. 48. gordialis, Schlot. Streitberg; Heidenheim; Buxweiler, Goldf.; Coral Rag, Bernese Jura, Thur. 49. intercepta, Goldf. Streitberg ; Culmbach, Goldf. 50. Ilium, Goldf. Thurnau ; Streitberg, Goldf. ; Calc. Grit, Ber- nese Jura, Thur. 51. Filaria, Goldf. Ferrug. Oolite, Grafenberg; Streitberg, Goldf. 52. socialis, Goldf. Bavaria ; Swabia ; Burgundy, Goldf. ; Calc. Grit, Bernese Jura, Thur. 53. problematica, Munst. Solenhofen, Goldf. , species undetermined. Coral Rag, Oxford Clay, Cornbrash, Forest Marble, Bradford Clay, Great Oolite, Mid. and South of England, Conyb.; Oxford Clay, Inferior Oolite, Haute Saone, Thir. ; Cornbrash, Forest Marble, Bradford Clgv, Great Oolite, Fuller's Earth, Wiltshire, Lorn. 540 Organic Remains of the Oolitic Group. CONCHIFERA. 1. Spirifer Walcotii, Sow. Lias, Yorkshire, Phil. ; Lias, Bath, Lyme Regis, De la B. ; Lias, Normandy, De Cau. ; Lias, South o France, Dufr. ; Lias, Western Islands, Scotland, Murch. 1. Delthyris* verrucosa, Von Buck. Lias, Bahlingen, Wurtemberg, Von Buck. 2. rostrata, Schlot. Lias, Wurtemberg, Von Buck. ; Oxford Clay, Bernese Jura, Thur. 1. Terebratula intermedia, Sow. Coral Oolite, and Great Oolite, Yorks., Phil. ; Cornbrash, Mid. and S. England ; Inferior Oolite, Dundry, Conyb. ; Kim. Clay, Beniese Jura, Thur. 2. globata, Sow. Coral Oolite? Great Oolite, Yorkshire, Phil. ; Forest Marble, Normandy, De Cau. ; Oolite, Env. of Bath ; Sow. ; Fuller's Earth, Env. of Bath, Great Oolite, Haute Saone, Thir. 3. ornithocephala, Sow. Coralline Oolite, and Kelloway Rock, Yorkshire, Phil. ; Kelloway Rock, Cornbrash, Lias ? Mid. and South of England ; Inferior Oolite, Dundry, Conyb. ; Oxford Clay and Lias, Normandy, De Cau. ; Inferior Oolite, Uzer, South of France, Dufr. ; Kimmeridge Clay, Great Oolite, Haute Saone, Thir.; Inferior Oolite, Wiltshire, Lons.; Soleure, Buxweiler, Hoen.; Oxford Clay, Bernese Jura, Thur. ? 4. ovata, Sow. Coralline Oolite? Yorkshire, Phil.; Inferior Oolite, Mid. and South of England, Conyb.; Coral Rag, Haute Soane, Thir. 5. obsoleta, Sow. Coralline Oolite ? Inferior Oolite, Yorkshire, Phil. ; Cornbrash, Bradford Clay, Great Oolite, and Inferior Oolite, Mid. and South of England, Conyb. ; Great Oolite, Normandy, De Cau ; Lias and Inferior Oolite, South of France, Dufr. ; Forest Marble, Wiltshire, Lons. ; Oxford Clay, Bernese Jura, Thur. 6. socialis, Phil. Calcareous Grit, and Kelloway Rock, York- shire, Phil. 7. , ovoides, Sow. Cornbrash ? Yorkshire, Phil. ; Inferior Oolite, Normandy, De Cau. ; Rubbly Limestone, &c. Braambury Hill, Brora, Murch.; Inf. Oolite, Calvados, Her.; Leisacker; Neuburg; Neresheim, G. T. 8. digona, Sow. Cornbrash, Yorks., Phil. ; Cornbrash and Brad- ford Clay, Mid. and S. England ; Inferior Oolite, Dundry, Conyb.; Forest Marble, Normandy, De Cau.; Bradford Clay and Coral Rag? North of France, Bobl. ; Forest Marble, Bradford Clay, Great Oolite, Wilts, Lons. 9. spinosa, Townsend and Smith. Great Oolite, Yorkshire, Phil.; Inf. Oolite, Bath, Lons. ; Oxford Clay, Bernese Jura, Thur. *9. spinosa, Schlot. Inferior Oolite, Southern Germany, Munst. 10. trilineata, Y. 8f B. Inferior Oolite and Lias, Yorkshire, Phil. 11. bidens, Phil. Inferior Oolite and Lias, Yorks., Phil. ; Lias, Boll, G. T. 12. punctata, Sow. Lias, Yorkshire, Phil. ; Inferior Oolite, Mid. and South of England, Conyb. ; Lias, Western Islands, Scot- land, Murch. ; Inf. Oolite, Southern Germany, Munst. 13. resupinata, Sow. Lias, Yorkshire, Phil. ; Inferior Oolite, Mid. and South of England, Conyb.; Inferior Oolite, Barendprf; Thurnau, Munst. * The genus Delthyris, Dalman, is the same with the genus Spirifer, Sowerby ; both names have been retained above for the purpose of more easy reference. Organic Remains of the Oolitic Group. 541 14. Terebratula acuta, Sow. Lias, Yorkshire, Phil.; Inferior Oolite and Lias, Mid. and South of England, Conyb. ; Lias, Normandy, De Can.; Fuller's Earth, Frome, Lons.; Lias, Wurtemberg, Von Buck. 15. triplicata, Phil. Lias, Yorkshire, Phil. ; Lias, Wurtemberg, Von Buch. 16. tetraedra, Sow. Lias, Yorkshire, Phil. ; Inferior Oolite, Mid. and South of England, Conyb.; Lias, South of France, Dufr.; Forest Marble ? Mauriac, South of France, Dufr. ; Lias and Micaceous Sandstone, Western Islands, Scotland, Murch. ; Echterdingen, Buxweiler, Hcen. ; Lias, Waldhausen, Tu- bingen, G. T. 17. subrotunda, Sow. Cornbrash and Inferior Oolite, Mid. and South of England, Conyb. ; Cornbrash and Forest Marble, North of France, Bobl. ; Forest Marble? Mauriac, South of France, Dufr. ; Inferior Oolite, Env. of Bath, Lons. 18. obovata, Sow. Cornbrash, Mid. and South of England, Conyb.; Inferior Oolite, Env. of Bath, Lons. 19. reticulata, Sow. Bradford Clay, Mid. and South of England, Conyb. ; Forest Marble, Normandy, De Cau. 20. media, Sow. Inferior Oolite, Dundry, Conyb.; Inferior Oolite, Great Oolite, and Bradford Clay, North of France, Bobl. ; Dunrobin Oolite, Scotland, Murch. ; Fuller's Earth, Inferior Oolite, Env. of Bath, Lons. 21. crumena, Sow. Inf. Oolite and Lias? Mid. and S. England, Conyb. ; Echterdingen, Hcen. 22. concinna, Sow. Fuller's Earth, Mid. and South of England, Conyb. ; Inferior Oolite, Normandy, De C. ; Forest Marble ? Mauriac, South of France, Dufr. ; Fuller's Earth, Frome ; Inferior Oolite, Env. of Bath, Lons. ? 23. biplicata, Sow. Oxford Clay, Forest Marble, Great Oolite, Nor- mandy, Her. ; Soleure, Hcen. ; Kim. Clay, Bernese Jura, Thur. ; Papenheim, G. T. 24. tetrandra, . Forest Marble, Normandy, De C. 25. coarctata, Park. Forest Marble, Normandy, De C. ; Bradford Clay, North of France, Bobl. ; Bradford Clay, Bath, Loscombe. 26. - plicatella, Sow. Inf. Oolite, Bridport, De la B.; Forest Marble, Normandy, De C. ; Bavaria ; Hohenstein, Saxony, Munst. 27. serrata, Sow. Lias, Lyme Regis, De la B. ; Lias, Waldhaen- serhof, G. T. 28. - bullata, Sow. Inferior Oolite, Normandy, De C. ; Inferior Oolite, Bridport, Dorset, Sow.; Cornbrash, Wiltshire; Fuller's Earth, Env. of Bath, Lons. 29. sphseroidalis, Sow. Inferior Oolite, Normandy, De C. ; Inferior Oolite, Dundry, Braikenridge. 30. - emarginata, Sow. Inferior Oolite, Normandy, De C. ; Inferior Oolite, Env. of Bath, Lons. 31. - quadrifida, . Inf. Oolite, Calvados, Desl. 32. numismalis, Lam. Lias, Bahlingen ; Gonningen, Von Buch. ; Inf. Oolite, Calvados, Desl. 33. perovalis, Smv. * Inf. Oolite, Dundry, Braikenridge ; Forest Marble? Mauriac, and Kim. Clay, Cahors, South of France; Rochelle Limestone, Dufr. ; Oxford Clay, Kell. Rock, Haute Saone, Thir. ; Inf. Oolite, Calvados, Desl. 34. - maxillata, Sow. Inf. Oolite, Env. of Bath, Sow.; Forest Marble, Wilts, Lons. * T. bistffarcinata) Schlot., according to Count Munster. 542 Organic Remains of the Oolitic Group. 35. Terebratula flabellula, Saw. Great Oolite, Ancliff, near Bradford, Wilts, Cookson. 36. furcata, Sow. Great Oolite, Ancliff, Cookson. 37. orbicularis, Sow. Lias, Bath, Sow. 38. hemisphserica, Sow. Great Oolite, Ancliff, Cookson. 39. inconstans, Sow. Shelly Limestone and Calc. Grit, Portgower, &c. N. of Scotl., and Shell Limestone, Beal, Isle of Sky, Murch. ; Coral Rag, Weymouth, Sedg. ; Bavaria ; Wurtemberg ; Porta Westphalica ; Hohenstein, Saxony, Munst. 40. avicularis, Munst. Inf. Oolite, Southern Germany, Munst. 41. loricata, Schlot. Baireuth, Hcen. ; Thurnau; Streitberg, G. T. 42. plicata, Lam. Thurnau, G. T. 43. spinosa, Schlot. Compound Great Oolite, Bernese Jura, Thur. ; Blomberg ; Wurtemberg ; Donaueschingen, G. T. 44. vulgaris, Schlot. Porta Westphalica, Hcen. 45. Defrancii, Al. Brong. Amberg, Hcen. 46. Hceninghausii, Blain. Baireuth, Hcen. 47. rimosa, Von Buck. Lias, Bahlingen, Wurtemberg, Von Such. 48. bicanaliculata, Sow. Hohenstein, Saxony ; Ferruginous Oolite, Bavaria and Wurtemberg, Munst. 49. cornuta, Sow. Inf. Oolite, Ilminster, Sow. ; Bavaria ; Hohen- stein, Munst. 50. trilobata, Munst. Bavaria ; Porta Westphalica ; Hohenstein, Munst. 51. lacunosa, Schlot. Calc. Grit, Oxford Clay, Bernese Jura, Thur. 52. varians, Schlot. Oxford Clay, Compound Great Oolite, Bernese Jura, Thur. 53. depressa, Sow. Oxford Clay, Bernese Jura, Thur. 54. variabilis, Schlot. Oxford Clay, Bernese Jura, Thur. 55. personata, Her. Inf. Oolite, Calvados, Desl. 56. sella, Sow. Lochen ; Bahlingen, G. T. 57. impressa, . Hohenzollern ; Stufenberg ; Thurnau, G. T. 58. nucleolata, Schlot. Streitberg, G. T. 59. tegularis, Schlot. Kelheim ; Heidenheim ; Amberg, G. T. 60. ? alata, Lam. Bahlingen; Locherberg ; Hohenzollern, G. T. 1. Orbicula? radiata, Phil. Coral. Oolite, Yorks., Phil. 2. granulata, Sow. Great Oolite, Ancliff, Wilts, Cookson. , species not determined. Inferior Oolite, Yorks., Phil. 1. Lingula Beanii, Phil. Inferior Oolite, Yorks., Phil. 1. Ostrea gregarea, Sow.* Coral Rag, Yorks., Wilts, &c.; Calc. Grit and Great Oolite ? Yorks., Phil. ; Coral Rag, Mid. and S. of Eng. ; Inf. Oolite, Dundry, Conyb. ; Coral Rag and Oxford Clay, Norm., De C. ; Oxford Clay and Coral Rag, N. of Fr., Bobl; Kim. Clay, Havre, Phil.; Coral Rag, Weymouth, Sedg.; Great Oolite, Calvados, Desl. 2. solitaria, Sow. Coral Rag and Inf. Oolite, Yorks., Oxon, &c. Phil. ; Kim. Clay, Haute Saone, Thir. ; Coral Rag, Wey- mouth, Sedg. ; Kim. Clay, Bernese Jura, Thur. 3. duriuscula, Bean. Coralline Oolite, Yorks., Phil. 4. inaequalis, Phil. Oxford Clay, Yorks., Phil. 5. undosa, Bean. Kell. Rock, Yorks., Phil. 6. archetypa, Phil. Kell. Rock, Yorks., Phil. 7. Marshiif, Sow. Kell. Rock, Cornb., and Great Oolite, Yorks., Phil. ; Cornb. and Fuller's E., Mid. and S. of Eng., Conyb. ; Oxford Clay, Forest Marb., and Inf. Oolite, Norm., De C. ; Cornb., Wilts, Lons. ; Coral Rag, Weymouth, Sedg. ; Oxford * Query, Ostrea Crista-Galli, Smith, f Ostrea flabelloides, Lam. Organic Remains of the Oolitic Group. 543 Clay, N. of France, Bobl. ; Wiesgoldingen ; Fiirstenberg; Stufenberg ; Baireuth, G. T. 8. Ostrea sulcifera, Phil. Great Oolite, Yorks., Phil ; Inf. Oolite, Haute Saone, Thir. 9. deltoidea, Sow. and Smith. Kim. Clay, Yorks., Phil. ; Oxford Clay, N. of Fr. BoU. ; Kim. Clay, S. and Mid. of England, Conyb. ; Shell Limestone and Calc. Grit? Portgower, &c. Scot!., Murch. ; Kim. Clay, Havre, Phil ; Sandst., Limest., and Shale, Inverbrora, Scotl., Murch. ; upper part of Coral Rag, Weymouth, Sedg. 10. expansa, Sow. Portland Stone, Conyb. 11. palmetta, Sow. Oxford Clay, Mid. and S. of Eng., Conyb.; Oxf. Clay and For. Marb., Norm., De C. ; Baireuth ; Wur- temberg, G. T. 12. acuminata, Sow. Bradford Clay and Inf. Oolite, Mid. and S. of Eng., Conyb. ; Great Oolite and Brad. Clay, N. of Fr., Bobl. ; Great Oolite, Haute Saone, Thir. ; Fuller's E., Inf. Oolite, Environs of Bath, Lons.; Comp. Great Oolite, Bernese Jura, Thur.; Inf. Oolite? Calvados, Desl. 13. rugosa, Sow. Inf. Oolite, Mid. and S. of Eng., Conyb. 14. - minima, Desl. Coral Rag and Oxford Clay, Norm., De C. 15. plicatilis. Oxford Clay, Norm., De C. 16. costata*, Sow. Brad. Clay, N. of Fr., Bobl. ; Great Oolite, Ancliff, near Bath, Cookson. ; Comp. Great Oolite, Bernese Jura, Thur. 17. pectinata. Oxford Clay, N. of Fr., Bobl. 18. pennaria. Oxford Clay, N. of Fr., Bobl. 19. laeviuscula, Sow. Lias, Eng., Sow. 20. obscura, Sow. Great Oolite, Ancliff, Wilts, Cookson. 21. Meadii, Sow. Inf. Oolite, Env. of Bath, Lons. 22. colubrina, Lam. Nattheim, G. T. 23. carinata, Lam. Calc. Grit, Bernese Jura, Thur. 24. irregularis, Munst. Lias, Amberg, G. T. 25. Ungula, Munst. Lias, Amberg; Banz, G. T. 26. laeviuscula, Munst. Lias, Amberg, G. T. 27. Synama, Munst. Lias, Baireuth, G. T. 28. - semiplicata, Munst. Lias, Baireuth, G. T. 1. Exogyra Bruntrutana, Thur. Portland Beds, Kim. Clay, Porrentruy, Bernese Jura, Thur. 2. spiralis, Goldf. Elligser Brink, G. T. 3. reniformis, Goldf. Westphalia, G. T. , species not determined. Kim. Clay, Haute Saone, Thir. Forest Marble ? Wilts, Lons. 1. Gryphaea chamseformis, Phil. Calc. Grit, Yorks.; and Oolite, Suther- land, Phil. 2. bullata, Sow. Coral. Oolite? Calc. Grit? Phil. ; Oxford Clay, Lincolnshire, Sow. ; Ooiite of Braambury Hill, Brora, Murch. 3. inhserens, Phil Calc. Grit, Yorks., Phil 4. dilatata, Sow. Kell. Rock, Yorks., Phil. ; Oxford Clay, Mid. and S. of Eng., Conyb. ; Oxford Clay and Lias, Norm., De C.; Oxford Clay, N. of Fr., Bobl. ; Oxford Clay, Burgundy, Beaum. ; Great Arenaceous Formation, Western Islands, Scot!. Murch. ; Oxford Clay, Haute Saone, Thir. ; Lower part of Coral Rag, Weymouth, Sedg.; Oxford Clay, Beggingen, Schafhausen, Von Buck. 5. incurva, Sow. Lias, Yorks., Phil; Lias, Mid. and S. Eng. * Ostrea Knorrii, Voltz. 544? Organic Remains of the Oolitic Group. Conyb. ; Lias, Norm., De C. ; Lias and Inf. Oolite, N. of Fr., BobL ; Lias, S. of Fr., Dufr. ; Lias, Metz, Salins, Amberg, AL Brong. ; Lias, Western Islands, Scotl. ; Lias, Ross and Cromarty, Scotl., Murch. ; Goppingen, Bahlingen, Hcen. 6. Gryphaea nana, Sow. Kim. Clay, Oxford, Sow. ; Shale and Grit, Dun- robin Reefs, Scotl., Murch. ; Lias and Oxford Clay ? N. of Fr., BobL 7. Maccullochii, Sow. ; Lias, Western Islands, Scotl., Murch. ; Lias, Yorks., Phil. ; Oxford Clay, Norm., De C. ; Lias, S. of Fr., Dufr ; Lias, Env. of Bath, Lons. 8. depressa, Phil. Lias, Yorks., Phil. 9. obliquata, Sow. Lias, Mid. and S. Eng., Conyb. ; Lias, S. of Fr., Dufr.; Lias, Western Islands, Scotl., Murch.; Lias, Env. of Bath, Lons. 10. . Cymbium, Lam. Inf. Oolite, N. of Fr., BobL; Lias, S. of France; Inf. Oolite, Villefranche, S. of France, Dufr. ; Inf. Oolite, Haute Saone, Thir. ; Lias, Bahlingen, Von Buch ; Inf. Oolite, Calvados, DesL; Lias, Boll; Wasseralfingen ; Ellwangen, G.T. 11. lituola, Lam. Brad. Clay, Comb., and For. Marb., N. of Fr., BobL 12. gigantea, Sow. Lias, S. of Fr., Dufr. ; Lias, Ross and Cromarty, Scotl.; Great Arenaceous Formation, Western Islands, Scotl., Murch. ; Porta Westphalica ; Hohenstein, Saxony, Munst. 13. minuta, Sow. Great Oolite, Ancliff, Wilts, Cookson. 14. virgula*, Defrance. Kim. Clay, Havre, AL Brong. ; Kim. Clay, Burgundy, Beaum. ; Kim. Clay, S. of Fr. Dufr. ; Kim. Clay, Weymouth, BucJd. $ De la B. ; Kim. Clay, Haute Saone, Thir. ; Portland Beds, Kim. Clay, Bernese Jura, Thur. 15. Isevis, Schlot. Lias, Malsch, near Heidelberg, Bronn. 1. Plicatula spinosa, Sow. Lias, Yorks., Phil. ; Lias, Mid. and S. Eng., Conyb. ; Lias, Norm., De C. ; Inf. Oolite, N. of Fr. BobL ; Great Arenaceous Formation, Western Islands, Scotl., Murch.; Lias, Gundershoffen, Volte. ; Lias, Wittberg, G. T. 2. tubifera, . Calc. Grit, Bernese Jura, Thur. 3. pectinoides, DesL Inf. Oolite, Calvados, DesL 4. squamosa, Goldf. Elligser Brink, G. T. 1. Pecten abjectus, Phil. Coral Rag, Yorks. and Oxon; Calc. Grit, Great Oolite, and Inf. Oolite, Yorks., Phil. 2. inasquicostatus, Phil. Coralline Oolite, Yorks. ; Calc. Grit, Oxon, Phil. ; Coral Rag, Bernese Jura, Thur. 3. cancellatus, Bean. Coralline Oolite, Yorks. ; Oolite, Suther- land? Phil. 4. demissus, Phil. Coralline Oolite, Kell. Rock, Cornbrash, and Great Oolite, Yorks., Phil. 5. Lens, Sow. Coralline Oolite, Kell. Rock, Great Oolite, Inf. Oolite, and Lias, Yorks., Phil.; Coral Rag, Mid. and S. Eng.; Inf. Oolite, Dundry, Conyb. ; Coral Rag and Oxford Clay, Norm., De C. ; Comb, and For. Marb., N. of Fr., BobL ; Inf. Oolite, Alsace, and Stranen near Luxembourg, AL Brong. ; Sandst., Limest, and Shale, Inverbrora, Scotl., Murch. ; Inf. Oolite, Haute Saone, Thir. ; Elligser Brink, G. T. ; Comp. Great Oolite, Bernese Jura, Thur. 6. * vagans, Sow. Coral Rag, Yorks. and Oxford; Calc. Grit, Yorks., Phil.; For. Marb., Norm., De C.; Sandst. and Rubbly Limest., Braambury Hill, Brora, Murch. ; For. Marb., Wilts, Lons. ; Oxf. Clay, Comp. Great Oolite, Bernese Jura, Thur. * Exogyra virgula, Foltz. Organic Remains of the Oolitic Group. 545 7. Pecten fibrosus, Sow. Kell. Rock and Cornbrash, Yorks., Phil.; Coral Rag, Kell. Rock, Comb., For. Marb., Brad. Clay, and Inf; Oolite, Mid. and S. Eng., Conyb. ; Coral Rag, Norm.? De C.; Comb, and For. Marb., N. of Fr., Bobl. ; For. Marb. ? Mau- riac, S. of Fr., Dufr. ; Rubbly Limestone, &c., Braambury Hill, Brora, Murch. ; For. Marb., Wilts, Lons. ; Soleure, Hoen. ; Culmbach, G. T. 8. virguliferus, Phil. ; Inferior Oolite, Yorks., Phil 9. sublsevis, Y. fy B. Lias, Yorks., Phil. 10. aequivalvis, Sow. Lias, Yorks., Phil. ; Inf. Oolite, Mid. and S. Eng., Conyb. ; Lias, S. of Fr., Dufr. ; Lias, Western Islands, Scotl., March. ; Inf. Oolite, Env. of Bath, Lons. ; Inf. Oolite, Calvados, Desl. 11. lamellosus, Soiv. Portland Stone, Conyb. 12. arcuatus, Sow. Coral Rag, Mid. and S. Eng., Conyb. ; Portland Beds, Kim. Clay, Haute Saone, Thir. ; Kim. Clay, Bernese Jura, Thur. 13. similis, Sow. Coral Rag, Mid. and S. Eng., Conyb.; Coral Rag, Norm. ? De C. ; Great Oolite, Haute Saone, Thir. 14. laminatus, Sow. Comb., Mid. and S. Eng., Conyb. 15. barbatus, Sow. Inf. Oolite, Dundry, Conyb.; Lias, Norm., De C. ; Inf. Oolite, Lias, Env. of Bath, Lons. 16. vimineus, Sow. Oxford Clay, For. Marb., and Inf. Oolite, Norm., De C. ; Forest Marble, Malton, Sow. ; Rubbly Lime- stone, &c., Braambury Hill, Brora, Murch.; Coral Rag, Haute Saone, Thir. ; Coral Rag, Yorksh. and Oxon, Phil. ; Calc. Grit, Oxf. Clay, Bernese Jura, Thur. 17. obscurus, Sow. Stonesfield, Sow. ; For. Marb.? Mauriac, S. of Fr., Dufr. 18. annulatus, Sow. Cornb., Felmersham, Marsh. 19. marginatus, . Wasseralfingen, Hoen. 20. squamosus, Von Buch. Lias, Weissenburg, G. T. 21. Phillipsii, Foltz. Comp. Great Oolite, Bernese Jura, Thur. 22. paradoxus, Munst. Inf. Oolite, Bernese Juva, Thur. 23. striatus, Sow. Inf. Oolite, Bernese Jura, Thur. 24. corneus, . Great Oolite ; Inf. Oolite, Calvados, Desl. 25. personatus, Goldf. Mistelgau ; Wasseralfingen, G. T. 26. dentatus, Sow. Lias, Bollerbad, G. T. 27. contrarius, Von Buch. Lias, Wittberg; Metzingen, G. T. 28. canaliculatus, Goldf. Lias, Culmbach, G. T. 1. Monotis salinaria, Bronn. Regensberg, G. T. 2. similis, Munst. Pappenheim, G. T. 3. decussata, Munst. Hildesheim ; Biikkeberg ; Suntel ; Wett- bergen, G. T. 4. concinnus, Goldf. Minden ; Wurtemberg, G. T. 1. Plagiostoma, laeviusculum, Sow. Coralline Oolite, Yorks.; Coral Rag and Calcareous Grit, Oxon, Phil. ; Coral Rag, Marthon, S. df Fr., Dufr. 2. rigidum, Sow. Coralline Oolite, Yorks. ; Coral Rag, Oxon, Phil. ; Inf. Oolite, Dundry, Conyb. ; Coral Rag, N. of Fr., Bobl. ; Coral Rag, Haute Saone, Thir. 3. rusticum, Sow. Coralline Oolite, Yorks.; Calc. Grit, Oxon, Phil 4. duplication, Sow. Coralline Oolite. Oxford Clay, and Kell. Rock, Yorks., Phil. ; Inf. Oolite, Norm., De C. ; Dunrobin Oolite, Scotl., Murch. ; Lias, Env. of Bath, Lons. 5. rigidulum, Phil. Cornbrash, Yorks., Phil. 6- interstinctum, Phil. Cornb. and Great Oolite, Yorks., Phil. 2 N 546 Organic Remains of the Oolitic Group. 7. Plagiostoma cardiiforme, Sow. Petty France, Gloucestershire, Stein- hauer ; Great Oolite, Yorks., Phil. ; Cornb. and For. Marb., N. of France, Bobl. 8. giganteum, Sow* Inf. Oolite and Lias, Yorks., Phil.; Inf. Oolite, Dundry? Lias, Mid. and S. Eng., Conyb. ; Lias, Norm., De. C. ; Lias, N. of Fr., Bobl. ; Lias, Western Is- lands, Scot!., Murch. ; Inf. Oolite, Haute Saone, Thir. ; Bah- lingen, Hoen. ; Lias, Malsch, near Heidelberg, Bronn. 9. obscurum, Sow. Kell. Rock, Mid. and S. Eng., Conyb. ] o. pectinoide, Sow. Lias, Yorks., Phil. ; Shale and Grit, Reefs at Dunrobin, Scotl., Murch. ; Lias, Vachingen ; Waldhaenser- hof, G. T. 1 1 . punctatum, Sow. Inf. Oolite, Dundry. Lias, Mid. and S. En- gland, Conyb. ; For. Marb. and Inf. Oolite, Norm., De C. ; Lias, N. of France, Bobl. ; Lias, S. of France, Dufr. ; Lias, Western Islands, Scotl., Murch. ; Inf. Oolite, Barendorf; Thurnau, Munst. 12. sulcatum. Lias, S. of France, Dufr. 13. ovale, Sow. For. Marb.? Mauriac, S. of France, Dufr. 14. Hermanni, Volte. Lias, Alsace, Voltz; Lias, Env. of Bath, Lons. ; Lias, Lyme Regis, De la B. 15. obliquatum, Sow. Sandstone and Limestone, Braambury Hill, Brora, Sandst., Limest., and Shale, Inverbrora, Scotl., Murch. 16. acuticostatum, Sow. Sandst., Limest., and Shale, Inverbrora, Scotl., Murch. 17. concentricum, Sow. Lias, Ross and Cromarty, Scotl., Murch. 18. transversum, Von Buck. Nipf, Bopfingen ; Stufenberg, G. T. , species not determined. Bradford Clay and Great Oolite, Mid. and S. Eng., Conyb.; Lias, Gundershofen, Volte] Kim. Clay, Bernese Jura, G. T. 1. Posidonia Bronni, Goldf. Lias, Ubstadt, near Bruchsal, Hoen. 1. Lima rudis, Sow. Coralline Oolite, Calc. Grit, Kell. Rock, and Great Oolite, Yorks., Phil; Coral Rag, Mid. and S. Eng., Conyb.; Coral Rag, N. of Fr., Bobl.; Rubbly Limestone, &c., Braam- bury Hill, Brora, Murch. 2. proboscidea, Sow. Inf. Oolite ? Yorks., Phil. ; Inf. Oolite, Dun- dry, Conyb.; Oxford Clay, For. Marb., and Inf. Oolite, Norm., De C.; Inf. Oolite, Haute Saone, Thir.; Soleure, Bale,/fo?n.; Coral Rag, Weymouth, Sedg. ; Inf. Oolite, Barendorf; Thur- nau, Munst. ; Calc. Grit, Bernese Jura, Thur. 3. gibbosa, Sow. Cornb. and Inf. Oolite, Mid. and S. Eng., Conyb.; Great Oolite, Calvados, Her. 4. antiqua, Sow. Lias, Mid. and S. Eng., Conyb. ; Lias, S. of France, Dufr.; Inf. Oolite, Haute Saone, Thir.; Elligser Brink, G. T. 5. heteromorpha, Desl. Inf. Oolite, Calvados, Her. , species not determined. Great Oolite, Wilts, Lons. 1. Avicula expansa, Phil. Coralline Oolite, Oxford Clay? Kell. Rock and Great Oolite, Yorks., Phil. 2. ovalis, Phil. Coralline Oolite and Calc. Grit, Yorks., Phil. 3. elegantissima, Bean. Coralline Oolite, Yorks., Phil. 4. tonsipluma, Y. fy B. Coralline Oolite, Yorks., Phil. 5. Braamburiensis, Sow. Sandstone, Braambury Hill, Brora, Murch.; Kell. Rock, Great Oolite, and Inf. Oolite, Yorks., Phil.; Inf. Oolite, Bernese Jura, Thur. * Lima gigantea, Deshayes. The same author considers that the genus Plagio- stoma of Sowerby and Lamarck should be suppressed, the species composing it being referrible either to Spondylus or to Lima. Organic Remains of the Oolitic Group. 547 6. Avicula inaequivalvis, Sow. Inf. Oolite and Lias, Yorks., Phil. ; Great Oolite and Inf. Oolite, Norm., De C.; Lias, S. of Fr., Dufr.; Great Arenaceous Formation, Western Islands; and Shell Limest. and Grit, Portgower, Scotland, Murch. ; Lias, Lyme Regis, DelaB.; Bahlinghen, Hcen. ; Lias, Gandershofen, Voltz ; Full. E., Inf. Oolite, and Lias, Env. of Bath, Lons. ; Lias, Calvados, Her. ; Wisgoldingen ; Nipf, Bopfingen ; Banz ; Lias, Mogglingen ; Baireuth, G. T. 7. echinata, Sow. Lias ? Yorks., Phil. ; Cornb., Mid. and S. Eng., Conyb.; For. Marb., Norm., De C.; Brad. Clay, Cornb., and For. Marb., N. of Fr., Bobl; Great .Oolite, Haute Saone, Thir. ; Full. E., Env. of Bath, Lons. ; Inf. Oolite, Bernese Jura, Thur. 8. cygnipes, Y. fy B. Lias, Yorks., Phil. ; Lias, Western Islands, Scotl., March. 9. costata, Sow. Cornb. and Brad. Clay, Mid. and S. Eng., Inf. Oolite, Dundry, Conyb. ; For. Marb., Norm, De C. 10. lanceolata, Sow. Lias, Lyme Regis, De la B. 11. ovata, Sow. Stonesfield Slate, Sow. 12. Munsteri, Bronn. Lias, Malsch, Heidelberg, Bronn. 1. Inoceramus dubius, Soiv. Lias, Yorks., Ph il. ; Lias, Osnabriick; Gr. Gschaid, G. T. 1. Gervillia aviculoides, Sow. Coralline Oolite, Yorks., Calcareous Grit, Oxfordshire, Phil; Oxford Clay, Mid. and S. Eng., Inf. Oolite, Dundry Hill, Conyb. ; Oxford Clay, Norm., De la B.; Sandst., Limest., and Shale, Inverbrora, Scotl., Murch.; Lias, Gundershofen, Voltz ; Coral Rag, Weymouth, Sedg. ; Inf. Oolite, Barendorf; Thurnau, Munst. ; Calc. Grit, Bernese Jura, Thur. ; Boll ; Neuhausen, Germs ; Graefenberg, Niirn- berg, G. T. 2. ? acuta, Sow. Collyweston, Sow. ; Great Oolite, Yorks., Phil. 3. lata, Phil. Inf. Oolite, Yorks., Phil. 4. pernoides, Desl. Oxford Clay, For. Marb., Great Oolite, and Inf. Oolite, Norm., De C. 5. siliqua, Desl. Oxf. Clay and For. Marb., Norm., De C. 6. monotis, Desl. For. Marb., Norm., De C. 7. costellata, Desl. For. Marb., Norm., De C. -, species not stated. Coral Rag, Norm., De C. ; Kim. Clay and Inf. Oolite, Haute Saone, Thir. ; Kim. Clay, Bernese Jura, Thur. 1. Perna quadrata, Sow. Coralline Oolite, Kell. Rock, and Great Oolite, Yorks., Phil. ; Cornb., Bulwick, Sow. 2. mytiloides, Lam. Lias, Gundershofen, Voltz ; Oxford Clay, Dives, Normandy, Desh.; Neuhausen, Germs; Wisgoldfingen ; Bop- fingen ; Inf. Oolite, Metzingen ; Kahleberg, Echte, G. T. 3. plana, Thur. Kim. Clay, Bernese Jura, Thur. , species not determined. Oxford Clay, Yorks., Phil. 1. Crenatula ventricosa, Sow. Husband Bosvvorth, Leicestershire, Conyb.; Gloucestershire, Sow. ; Lias, Yorks., Phil. , species not determined. Portland Stone, Conyb. 1. Trigonellites antiquatus, Phil. Coral. Oolite, Yorks., Phil. 2. politus, Phil. Oxford Clay, Yorks., Phil. 1. Pinna lanceolata, Soiv. Coralline Oolite and Calcareous Grit, Yorks., Phil.; Inf. Oolite, Dundry, Conyb.; Lias, Norm., De C. ; Oxford Clay, N. of Fr., BobL; Coral Rag, Weymouth, Sedg.; Soleure, G. T. 2. mitis, Phil. Oxford Clay and Kell. Rock? Yorks., Phil. .'). - cuneata, Bean. Cornbrash and Great Oolite, Yorks., Phil. 2 N 2 54-8 Organic Remains of the Oolitic Group. 4. Pinna Folium, Y. $ B. Lias, Yorks., Phil 5. pinnigena, . Coral Rag, For. Marb., and Inf. Oolite, Norm., De C. 6. granulata, Sow. Kim. Clay, Weymouth, Sedg. ; Kim. Clay, Ca- hors, S. of Fr., Dufr. ; Lias, Skye, Murch. 7. diluviana, . Lias, Kaltenthal; Plieningen, Stuttgardt, G. T. , species not determined. Inf. Oolite, Env. of Bath, Lons. 1. Mytilus cuneatus, Phil. Inf. Oolite, Yorks., Phil. 2. amplus, . Great Oolite, Norm., De C. 3. pectinatus. Sow. Kim. Clay, Weymouth, Sedgwick ; Rochelle Limestone, Dufr. 4. sublsevis, Sotv. Cornb., Eng., Sow. 5. _ _ solenoides, . Kim. Clay, Cahors, S. of Fr., Dufr. 6 Jurensis, Merian. Kim. Clay, Bernese Jura, Thur. , species not determined. Coral Rag and Inf. Oolite, Mid. and S. Eng., Conyb. ; Coral Rag, Norm., De C. ; Portland beds, Haute Saone, Thir. 1. Modiola imbricata, Sow. Coralline Oolite? and Great Oolite, Yorks., Phil.; Cornb., Mid. and S. Eng., Conyb.; Cornb., Wilts, Lons. 2. ungulata, 7. # B. Coralline Oolite, Great Oolite, and Inf. Oolite, Yorks., Phil. 3. bipartita, Sow. Calc. Grit, Yorks., Phil.; Sandstone and Limestone, Braambury Hill, Brora, Murch. 4. cuneata, Sow. Oxford Clay, Kell. Rock ? and Cornb., Yorks., Phil. ; Inf. Oolite, Mid. and S. Eng., Conyb. ; Lias, Norm., De C. ; Lias, Western Islands, Scotl. ; Sandst, Limest, and Shale, Inverbrora, Scotl., Murch. ; Hohenstein ; Ferriferous Oolite, Bavaria and Wurtemberg, Munst. ; Comp. Great Oolite, Bernese Jura, Thur. .5. _ pulchra, Phil. Kell. Rock, Yorks., Phil.; Oolite, Sutherland. 6. plicata, Sow. Inf. Oolite, Yorks., Phil. ; Cornb., Mid. and S. Eng., Inf. Oolite, Dundry, Conyb.; Portland Beds, Haute Saone, Thir.; Full. E., Somerset, Lons.; Portland Beds, Kim. Clay, Bernese Jura, Thur. 7. aspera, Sow. Inf. Oolite, Yorks., Phil. ; Cornb. Mid. and S. Eng., Conyb. 8. Scalprum, Sow. Lias, Lyme Regis ; Lias, Yorks., Phil. ; Lias, S. of Fr., Dufr. 9. Hillana, Sow. Lias, Yorks., Phil. ; Lias, Mid. and S. Eng., Conyb.; Full. E.? Env. of Bath, Lons. 10. Isevis, Sow. Lias, Mid. and S. Eng., Conyb. 11. depressa, Sow. Lias, Mid. and S. Eng., Conyb. 12. minima, Sow. Lias, Mid. and S. Eng., Conyb. 13. tulipea, Lam. Oxford Clay, N. of Fr., Bobl 14. pallida, Sow. Shale and Grit, Dunrobin Reefs, &c., Scotl., Murch. 15. - gibbosa, Sow. Inf. Oolite, Env. of Bath, Lons. 10. livida, Goldf. Metz, G. T. 17. ventricosa, Goldf. Soleure, Hcen. 18. Thirriae, Foltz. Kim. Clay, Bernese Jura, Thur. 19. striolaris, Merian. Kim. Clay, Bernese Jura, Thur. 20. elegans, . Great Oolite, Calvados, Her. 21. aequiplicata, Von Slrombeck. Kahleberg, Echte, G. T. , species not determined. Lias, Gundershofen, Foltz ; Lias, Bath, Lons. 1. Lithodomus Sowerbii, Thur. Coral Rag, Bernese Jura, Thur. , species not determined. Inf. Oolite, N. of Fr., Bobl. ; Inf. Oolite, Env. of Bath, Lons. Organic Remains of the Oolitic Group. 549 1. Chama mima, or Gryphaea mima, Phil. Coral. Oolite and Calc. Grit., Yorks., Phil. 2. ? crassa, Smith. Bradford Clay, Mid. and S. Eng., Conyb. 3. Berno-jurensis, Thur. Calc. Grit, Bernese Jura, Thur. , species not determined. For. Marb., Cornb., and Brad. Clay, Wilts, Lons. 1. Unio peregrinus, Phil. Cornb., Yorks., Phil. 2. abductus, Phil. Inferior Oolite and Lias, Yorks., Phil. 3. concinnus, Sow. Lias, Yorks., PhU. ; Inf. Oolite, Mid. and S. Eng., Conyb. ; Inf. Oolite, Lias, Env. of Bath, Lons. ; Lias, Mogglingen, Gmiindt, G. T. 4. crassiusculus, Sow. Lias, Yorks., Phil. 5. Listeri, Sow. Lias, Yorks., Phil. ; Inf. Oolite, Mid. and S. Eng., Conyb. 6. crassissimus, Sow. Lias, Mid. and S. Eng., Conyb.; Lias, Norm., De C.; For. Marb.? Mauriac, and Inf. Oolite, Uzer, S. of Fr., Dufr. 1. Trigonia costata, Sow. Coralline Oolite, Great Oolite, and Inf. Oolite, Yorks., Phil. ; Cornb., For. Marb., and Brad. Clay, Mid. and S. Engl., Inf. Oolite, Dundry, Conyb.; Oxford Clay, For. Marb., and Inf. Oolite, Norm., De C. ; Oxford Clay, N. of Fr., Bobl. ; Kim. Clay and Inf. Oolite, Haute Saone, Thir. ; Lias, Gundershofen, Voltz ; Inf. Oolite, Env. of Bath, Lons. ; Coral Rag, Weymouth, Sedg. ; Porta Westphalica; Hohen- stein, Munst. ; Banz, G. T. 2. clavellata, Sow. Coralline Oolite, Kell. Rock, and Cornb., Yorks., Phil.; Portland Stone and Cornb., Mid. and S. Engl., Inf. Oolite, Dundry, Conyb.; Oxford Clay, Norm., De la B. ; Oxford Clay, N. of Fr., Bobl.; Kim. Clay? Angouleme, Dufr.; Sandst., Shale, &c., Inverbrora, Scotl., Murch.; Coral Rag and Inf. Oolite, Haute Saone, Thir. ; Coral Rag, Wey- mouth, Sedg.; Kim. Clay, Calc. Grit, Bernese Jura, Thur.; Wisgoldingen ; Stufenberg ; Ehningen Sandstone, G. T. 3. conjungens, Phil. Great Oolite, Yorks., Phil. 4. striata, Sow. Inferior Oolite, Yorks., Phil.; Inf. Oolite, Dundry, Conyb.; Inf. Oolite, Norm., De C. ; Lias, S. of Fr., Dufr. 5. angulata, Sow. Inf. Oolite, Yorks., Phil. ; Inf. Oolite, near Frome, Sow. 6. literata, Y. $ B. Lias, Yorks., Phil. 7. gibbosa, Sow. Portland Stone, Conyb.; Forest Marb., Norm., De C. 8. duplicate, Sow. Inf. Oolite, Mid. and S. Eng., Conyb. ; For. Marb., Norm., De C. 9. elongata, Sow. Oxford Clay. Norm., De C. ; Oxford Clav, Eng., Sow.; Great Oolite, Alsace, Foltz ; Cornb., Wilts, Lons. 10. imbricata, Sow. Great Oolite, Ancliff, Wilts, Cookson. 11. cuspidata, Sow. Great Oolite, Ancliff, Cookson; var. Coral Rag, Haute Saone, Thir. 12. Pullus, Sow. Great Oolite, Ancliff, Cookson. 13. navis, Lam. Lias, Gundershofen, Vollz. 14. - incurva, Benett. Portland Beds, Tisbury, Wilts, Benett. 15. nodulosa, Lam. Kim. Clay, Havre; Bray, Pas. , species not determined. Coral Rag, Midi, and S. Eng., Conyb.; Coral Rag, Norm., De C. 1. Nucula elliptica, Phil. Oxford Clay, Yorks., Phil. 2. nuda, Y. Sf B. Oxford Clay, Yorks., Phil. 3. variabilis, Sow. Great Oolite and Inf. Oolite, Yorks., Phil. ; Great Oolite, Ancliff, near Bath, Cookson. 550 Organic Remains of the Oolitic Group. 4. Nucula Lachryma, Sow. Great Oolite and Inf. Oolite, Yorks., Phil. ; Great Oolite, Ancliff, Sow. 5. axiniformis, Phil. Inferior Oolite, Yorks., Phil. 6. Ovum, Sow. Lias, Yorks., Phil. 7. pectinata, Sow. Oxford Clay, Norm., De C. ; Brad. Clay? Wilts, Lons. 8. clariformis, . Lias, S. of Fr., Dufr. 9. mucronata, Sow. Great Oolite, Ancliff, Wilts, Cookson ; Banz, G. T. 10. Stahlii, Bronn. Lias, Ubstadt, near Heidelberg, Bronn. 11. acuminata, Merian. Oxf. Clay, Bernese Jura, Thur. 12. medio-jurensis, Thur. Oxf. Clay, Bernese Jura, Thur. 13. Hammeri, Defr. Gundershofen, G. T. 14. lobata, Von Such. Metzingen ; Nipf., Bopfingen, G. T. 15. subovalis, Goldf. Wasseralfingen, G. T. 16. rostrata, Goldf. Wasseralfingen, G. T. 1 7. elongata, Goldf. Wasseralfingen, G. T. ] 8. arcacea, Goldf. Banz ; Baireuth, G. T. , species not determined. Coralline Oolite, Yorks., Phil. ; Inf. Oolite, Dundry ; Lias, Mid. and S. Eng., Conyb. ; Kim. Clay, Bernese Jura, Thur. ; Lias, Bahlingen, G. T. 1. Pectunculus minimus, Sow. Great Oolite, Ancliff, Wilts, Cookson. 2. oblongus, Sow. Great Oolite, Ancliff, Wilts, Cookson. 1. Area quadrisulcata, Soiv. Coral Rag, Malton, Sow. Coralline Oolite, Yorks., Phil. 2. semula, Phil. Coralline Oolite, Yorks., Phil. 3. pulchra, Sow. Great Oolite, Ancliff, Wilts, Cookson ; Rochelle Limestone, Dufr. 4. trigonella, . Wasseralfingen, Hosn. 5. elongata, . Wasseralfingen, Hosn. 6. rostrata, . Wasseralfingen, Hozn. 7. medio-jurensis, Thur. Oxf. Clay, Bernese Jura, Thur. , species not determined. Lias, Mid. and S. Eng., Conyb. ; Brad. Clay, Wilts; Full. E., Inf. Oolite, Env. of Bath, Lons.; Kim. Clay, Bernese Jura, Thur. ; Lias, near Heidelberg, Bronn. 1. Cucullsea oblonga, Sow. Coralline Oolite, Yorks., Phil; Inf. Oolite, Dundry, Conyb. ; Inf. Oolite, Barendorf ; Thurnau, Munst. 2. contracta*, Phil. Coralline Oolite, Yorks., Phil. 3. triangularis, Phil. Coralline Oolite, Yorks., Phil. 4. pectinata, Phil. Coralline Oolite, Yorks., Phil. 5. elongata, Sow. Coralline Oolite? and Great Oolite, Yorks., Phil.; Rochelle limestone, Dufr. Cross Hands, Gloucester- shire, Steinhauer. ; Winzingen, Wisoldingen, G. T. 6. concinna, Phil. Oxford Clay and Kell. Rock? Yorks., Phil 7. imperialis, Bean. Great Oolite, Yorks., Phil. 8. cylindrica, Phil. Great Oolite, Yorks., Phil. 9. cancellata, Phil. Great Oolite, Yorks., Phil. 10. reticulata, Bean. Inf. Oolite, Yorks., Phil. 11. minuta, Sow. Great Oolite, Ancliff, Wilts, Cookson. 12. rudis, Sow. Great Oolite, Ancliff, Wilts, Cookson. 13. parvula, Munst. Oxf. Clay, Bernese Jura, Thur. ?14. decussata, Sow. Mistelbach, Baireuth, G. T. , species not determined. Oxford Clay, Haute Saone, Thir. Lias, Yorks., Phil. ; Lias, Mid. and S. Eng., Conyb. 1. Hippopodium ponderosum, Sow. Coralline Oolite and Lias, Yorks., Phil. ; Lias, Mid. and S. Eng., Conyb. ; Inf. Oolite, Calva- dos, Desl. 1. Isocardia rhomboidalis, Phil. Coralline Oolite, Yorks., Phil. Organic Remains of the Oolitic Group. 551 2. Isocardia tumida, Phil. Calc. Grit., Yorks., Phil. 3. minima, Sow. Cornb. and Great Oolite? Yorks., Phil.; Comb., Wilts, Lons, 4. concentrica, Sow. Great Oolite and Inf. Oolite, Yorks., Phil. ; Cornb., Northamptonshire, Sow. ; Full. E., Somerset, Lons. ; Inf. Oolite, Calvados, Desl 5. angulata, Phil. Great Oolite ? Yorks., Phil. 6. rostrata, Sow. Gloucestershire, Sow. Inf. Oolite, Yorks., Phil. ; Comp. Great Oolite, Bernese Jura, Thur. 7. striata, D'Orb. Kim. Clay, Portland Beds, Haute Saone, Thir.; Portland Beds, Kim. Clay, Bernese Jura, Thur. 8. excentrica, Volte. Portland Beds, Kim. Clay, Bernese Jura, Thur. 9. inflata, Foltz. Portland Beds, Kim. Clay, Bernese Jura, Thur. 10. - carinata, Volte. Kim. Clay, Bernese Jura, Thur. 11. costulata, Volte. Kim. Clay, Bernese Jura, Thur. , species not determined. For. Marb., Norm., De C. 1. Cardita similis, Sow. Coralline Oolite, Great Oolite, and Inf. Oolite, Yorks., Phil. ; Inf. Oolite, Dundry, Conyb. 2. lunulata. Sow. Inf. Oolite, Dundry, Conyb.; Inf. Oolite, Norm., De C. 3. striata, Sow. Lias, Norm.? De C. Inf. Oolite, Bath, Sow. ' , species not determined. Portland Stone, Conyb. 1. Cardium lobatum, Phil. Coralline Oolite, Yorks., Phil. 2. dissimile, Sow. Kell. Rock, Yorks., Phil; Portland Stone, Portland, Sow. ; Rocks of the Oolite series, Braambury Hill, Brora, Murch. 3. citrinoideum, Phil. Cornb., Yorks., Phil. 4. - cognatum, Phil. Great Oolite, Yorks., Phil. 5. - acutangulum, Phil. Great Oolite and Inf. Oolite, Yorks., Phil. 6. semiglabrum, Phil. Great Oolite, Yorks., Phil. 7. incertum, Phil. Inf. Oolite, Yorks., Phil. 8. - striatulum, Sow. Sandst., Limest. and Shale, Inverbrora, Scot- land, Murch. ; Inferior Oolite, Yorks., Phil. 9. gibberulum, Phil. Inf. Oolite, Yorks., Phil. 10. truncatum, Sow. Lias, Yorks., Phil.; Sandst., Limest., &c., Inverbrora, Murch. 11. multicostatum, Bean. Lias, Yorks., Phil. 1. Myoconcha crassa, Sow. Inf. Oolite, Dundry, Brackenridge ; Inf. Oolite, Normandy, De C. 1. Astarte * cuneata, Sow. Portland Stone, S. Eng. ; Inf. Oolite? Dundry, Conyb. 2. excavata, Sow. Inf. Oolite, Dundry, Conyb. ; Inf. Oolite, Norm., De C. ; Bopfingen ; Lauchheim, Banz, G. T. ? 3. planata, Sow. Inf. Oolite, Norm., De C. ; Bradf. Clay, N. of Fr., Bobl. 4. trigonalis, Sow. Inf. Oolite, Dundry, Johnstone. 5. orbicularis, Sow. Great Oolite, Aricliff, Wilts, CooJcson. 6. pumila, Sow. Great Oolite, Ancliff, Wilts, Cookson ; Rochelle Limestone, Dufr. 7. Voltzii, Hoen. Fullon, near Vesoul, Hoen. ; Banz, G. T. 8. - medio-jurensis, Thur. Oxf. Clay, Bernese Jura, Thur. 9. modiolaris, Goldf. Normandy; Wurtemberg, G. T. 10. - ovata, Smith. Coralline Oolite, Wilts; Oxon ; Yorks., Phil. 11. elegans, Sow. Coralline Oolite and Inf. Oolite, Yorks., Phil.; Rochelle Limest., Dufr. ; Shell Limest. and Calc. Grit, Port- * Crassina of Phillips is the Astarle of Sowerby. 552 Organic Remains of the Oolitic Group. gower, &c., Murch.; Limest., Shale, Sandst., Inverbrora, Murch. 12. Astarte aliena, Phil. Coralline Oolite, Yorks., Phil. 13. extensa, Phil. Coralline Oolite, Yorks., Phil. 14. carinata, Phil. Calc. Grit, Oxford Clay, and Kell. Rock, Yorks., Phil. 15. lurida, Sow. Inf. Oolite, Dundry, Conyb.; Oxford Clay, Yorks., Phil. 16. minima, Phil. Great Oolite, Inf. Oolite, and Lias, Yorks., Phil. ; Kim. Clay, and Coral Rag, Haute Saone, Thir. , species not determined. Lias, Mid. and S. Eng., Conyb. ; Coral Rag and Kim. Clay, Haute Saone, Thir.; Cornb., Wilts, Lons. 1. Venus varicosa, Sow. Felmersham, Sow. , species not determined. Coral. Oolite, Calc. Grit and Lias, Yorks., Phil. ; Portland Stone, Smith ; Coral Rag, Norm., De C. ; Sandst., Shale, &c., Inverbrora, Scotl., Murch. 1. Cytherea dolabra, Phil. Great Oolite, Yorks., Phil. 2. trigonellaris, Voltz. Lias, Gundershofen, Voltz. 3. lucinea, Voltz. Lias, Gundershofen, Voltz. 4. cornea, Voltz. Lias, Gundershofen, Voltz. , species not determined. Coralline Oolite, Yorks., Phil. ; Lias, N. of Fr., Bobl; Kim. Clay, Bernese Jura, Thur. 1. Pullastra recondita, Phil. Great Oolite, Yorks., Phil. 2. oblita, Phil. Inf. Oolite, Yorks., Phil. , species not determined. Lias, Yorks., Phil. 1. Donax Alduini, Al. Brong. Inf. Oolite? N. of Fr., Bobl. ; Kim. Clay, Havre and the Jura, AL Brong.; Kim. Clay, Bernese Jura, Thur.; Nipf, Bopfmgen; Rautenberg, Scheppenstadt; Kahle- berg, Echte, G. T. 2. Saussurii, Al. Brong. Kim. Clay, Bernese Jura, Thur. ; Kahle- berg, Echte, G. T. 1. Corbis Isevis, Sow. Coralline Oolite? Kell. Rock? Yorks., Phil; Mar- sham Field, Oxford, Smith. 2. ovalis, Phil. Kell. Rock, Yorks., Phil. 3. uniformis, Phil. Lias, Yorks., Phil. 1. Tellina ampliata, Phil. Coralline Oolite, Yorks., Phil. 2. incerta, Thur. Kim. Clay, Bernese Jura, Thur. 1. Psammobia laevigata, Phil. Coralline Oolite, Great Oolite, and Inf. Oolite, Yorks., Phil. 1. Lucina crassa, Sow. Sandstone and Rubbly Limestone, Braambury Hill, Brora ; Great Arenaceous Formation, Western Islands, Scotl., Murch. ; Calc. Grit, Yorks., Phil. ; Lincolnshire, Sow. 2. lyrata, Phil. Kell. Rock, Yorks., Phil. 3. despecta, Phil. Great Oolite, Yorks., Phil. 4. Elsgaudiae, Thur. Kim. Clay, Bernese Jura, Thur. , species not stated. Coral Rag and For. Marb., Norm., De Cau.; Inf. Oolite, Yorks., Phil; Shale, &c. Inverbrora, Scotl., Murch. 1. Sanguinolaria undulata, Sow. Sandst., Limest, and Shale, Inverbrora, Scotl., Murch. ; Calc. Grit, Oxford Clay, and Cornbrash, Yorks., Phil. 2. elegans, Phil. Lias, Yorks., Phil. , species not determined. Lias, Ross and Cromartv, Scotl., Murch.; Lias, Yorks., Phil. 1. Corbula curtansata, Phil. Coralline Oolite and Kell. Rock, Yorks., Phil. 2. -^ depressa, Phil. Great Oolite, Yor ks., Phil. Organic Remains of the Oolitic Group. 553 3. Corbula? cardioides, Phil. Lias, Yorks., Phil ; Lias, Ofterdingen, G. T. 4. obscura, Sow. Brora, Murch. , species not determined. For. Marb., Wilts, Lons. ; Kim. Clay, Bernese Jura, Thur. 1. Mactra gibbosa, For. Marb., Norm, De C. 1. Amphidesma decurtatum, Phil. Comb, and Great Oolite, Yorks, Phil. ; Kim. Clay? and Great Oolite? Haute Saone, Thir. 2. recurvum, Phil. Coralline Oolite? and Kell. Rock, Yorks., Phil; Kim. Clay, Havre, Phil. 3. securiforme, Phil. Comb., Inf. Oolite, Yorks., Phil.; Kim. Clay, Havre, Phil 4. donaciforme, Phil. Lias, Yorks., Phil 5. rotundatum, Phil. Lias, Yorks., Phil. ; Lias, Bahlingen ; Waldhaenserhof, G. T. 1. Lutraria Jurassi, Brong. For. Marb., Ligny, Meuse, Erong. 1. Gastrochaena tortuosa, Sow. Inf. Oolite, Yorks., Phil 1. Mya literata, Sow. Coralline Oolite, Calc. Grit, Oxford Clay, Kelloway Rock, Cornb., Inf. Oolite, and Lias, Yorks., Phil; Shale, Sandstone, and Limestone, Inverbrora, Scotl., Murch. 2. depressa, Sow. Oxford Clay? Yorks., Phil ; Kim. Clay? Angou- leme, Dufr. ; Kim. Clay, Havre, Phil.; Shale, Limestone, and Sandstone, Inverbrora, Scotl., Murch. 3. calceiformis, Phil. Kell. Rock, Great Oolite, aud Inf. Oolite, Yorks., Phil. 4. dilata, Phil. Inf. Oolite, Yorks, Phil. 5. eequata, Phil. Inf. Oolite, Yorks., Phil. 6. V scripta, Sow. Inf. Oolite, Dundry, Conyb. ; Great Oolite, Al- sace, Brong. ; Micaceous Sandstone, Western Islands, Scotl., Murch. 7. Mandibula, Sow. Kim. Clay ? Env. of Angouleme, Dufr. 8. angulifera, Sow. Great Oolite, Haute Saone, Thir. ; Lias, Al- sace, Voltz ; Fuller's Eartb, Environs of Bath, Lons. ; Calc. Grit, Bernese Jura, Thur. 1. Pholadomya Murchisoni, Sow. Sandstone, Limestone, and Shale, In- verbrora, Scotl., Murch.; Coralline Oolite? and Cornbrash, Yorks., Phil; Inf. Oolite, Normandy, De Cau. ; Kim. Clay, Bernese Jura, Thur. ; Nipf, Bopfingen ; Metzingen ; Porta Westphalica, G. T. 2. simplex, Phil. Calc. Grit, Yorks., Phil 3. deltoidea, Sow. Calc. Grit, Yorks., Phil; Kell. Rock and Cornbrash, Midi, and S. Engl., Conyb. 4. obsoleta, Phil Oxford Clay and Kell. Rock, Yorks., Phil 5. ovalis, Sow. Cornbrash, Yorks., Phil ; Portland Stone, Conyb.; Oxford Clay, Normandy, DeC. ; Kim. Clay? Angouleme; Rochelle Limestone, Dufr. 6. acuticostata, Sow. Great Oolite, Yorks, Phil; Kim. Clay, Cahors, S. of Fr., Dufr. ; Kim. Clay? Angouleme, Dufr. ; Kim. Clay, Haute Saone, Thir.; Brora, Farey ; Portland Beds, Kim. Clay, Bernese Jura, Thur. 7. nana, Phil Great Oolite, Yorks., Phil 8. producta, Sow. Great Oolite ? Yorks., Phil. ; Cornb. and Inf. Oolite, Midi, and S. Engl., Conyb. ; Cornb., Wilts, Lons. 9. obliquata, Phil Great Oolite, Inf. Oolite, and Lias, Yorks., Phil 10. fidicula, Sow. Inf. Oolite, Yorks., Phil; Cornb., Mid. and S. of Eng. ; Inf. Oolite, Dundry, Conyb. ; Lias, Norm., De C. ; Cornb., Wilts; Full. E., Env. of Bath, Lons.; Soleure, Hcen.; Inf. Oolite, Haute Saone, Thir. 554? Organic Remains of the Oolitic Group. 11. Pholadomya obtusa, Sow. Inf. Oolite, Dundry, Conyb. 12. ambigua, Sow. Inf. Oolite, Dundry, Conyb.; Oxford Clay, Norm., De C.; Lias, S. of Fr., Ditfr. ; Lias, Alsace, Voltz ; Lias, Bath, Lons.; Lias, Soleure; Lias Bahlingen, Von Buck. 13. agqualis, Sow. Weymouth, Sow.; Inf. Oolite, Norm., De C. ; Lubbecke, Minden, G. T. 14. gibbosa, . Lias, Norm., De C. ; Soleure, Hcen. 15. Protei, Brong. Rochelle Limest, Dufr. ; Kim. Clay, Havre and the Jura, Brong.; Portland Beds and Kim. Clay, Haute Saone, Thir.; Kim. Clay. Bernese Jura, Thur.; Kahleberg, Echte, G. T. 16. clathrata, Munst. Bavaria; Hohenstein, Saxony, Munst. 17. angustata, Sow. Kim. Clay, Calc. Grit, Bernese Jura, Thur. 18. cardiiformis, Gold/. Hammers, G. T. 19. concentrica, Goldf. Soleure, G. T. 20. decussata, Goldf. Wurtemberg, G. T. , species not determined. Oxford Clay, Haute Saone, Thir. ; Oxford Clay, Bernese Jura, Thur. 1. Panopaea gibbosa, Sow. Great Oolite? Yorks., Phil; Inf. Oolite, Dundry, Conyb. 1. Pholas recondita, Phil. Coralline Oolite, Yorks., Phil. 2. ? compressa, Sow. Kim. Clay, Oxford, G. E. Smith. MOLLUSCA. 1. Dentalium giganteum, Phil Lias, Yorks., Phil. 2. cylindricum, Sow. Lias, Mid. and S. Eng., Conyb. , species not determined. Calc. Grit, Yorks., Phil. 1. Patella latissima, Sow. Oxford Clay, Yorks., Phil. ; Oxford Clay, Mid. and S. of Eng., Conyb. 2. rugosa, Sow. For. Marb., Mid. and S. of Eng., Conyb. ; For. Marb., Norm., De C. 3. laevis, Sow. Lias, Mid. and S. of Eng., Conyb. 4. lata, Sow. Stonesfield Slate, Sow. 5. ancylo'ides, Sow. Great Oolite, Ancliff, Wilts, Cookson. 6. nana, -Sow. Great Oolite, Ancliff, Wilts, Cookson. 7. discoides, Schlot. Lias, Gundershofen, Voltz. 8. papyracea, Goldf. Lias, Banz, G. T. 1. Emarginula scalaris, Sow. Great Oolite, Ancliff, Wilts, Cookson. 1. Pileolus plicatus, Sow. Great Oolite, Wilts, Lons. 1. Bulla elongata, Phil. Coral. Oolite, Yorks., Phil. 1. Helicina polita, Sow.* Inf. Oolite, Cropredy, Conyb. 2. expansa, Sow. Lias, Mid. and S. of Eng., Conyb. 3. solarioides, Sow. Lias, Mid. and S. of Eng., Conyb. 1. Auricula Sedgvici, Phil. Inf. Oolite, Yorks., Phil. 1. Melania Heddingtonensis, Sow. Coral. Oolite, Comb., Great Oolite and Inf. Oolite, Yorks., Phil. ; Coral Rag, Mid. and S. of Eng. ; Inf. Oolite, Dundry, Conyb. ; Coral Rag and Inf. Oolite, Norm., De C. ; Rubbly Limest., &c., Braambury Hill, Brora, Murch. ; Kim. Clay, Havre, Phil. ; Inf. Oolite ? Haute Saone, Thir.; Coral Rag, Weymouth, Sedg. ; Kel- heim ; Kahleberg, Echte, G. T. 2. striata, Sow. Coral. Oolite and Great Oolite ? Yorks., Phil. ; Coral Rag and Lias, JVIid. and S. of Eng., Conyb.; Coral Rag, N. of Fr., Bobl. ; Kim. Clay, Havre, Phil. ; Coral Rag, Weymouth, Sedg. 3. vittata, Phil. Comb., Yorks., Phil. * Turbo callosus, Deshayes. Organic Remains of the Oolitic Group. 555 4. Melania lineata, Sow. Inf. Oolite, Yorks., Phil. ; Inf. Oolite, Dundry, Conyb. ; Inf. Oolite, Norm., De C. 5. medio-jurensis, Thur. Oxf. Clay, Bernese Jura, Thur. , species not determined. Great Oolite, Mid. and S. of Eng., Conyb. Paludina, species not determined. Portland Beds, Haute Saone, Thir. 1. Ampullaria Gigas, Von Strombeck. Kahleberg, Echte, G. T. , species not determined. Coral Rag, Cornb., and Inf. Oolite, Mid. and S. Eng., Conyb. ; Coral Rag, Norm., De C. ; Brad. Clay, N. of Fr., Bobl ; Kim. Clay, Bernese Jura, Thur. 1. Nerita costata, Sow. Inf. Oolite, Yorks., Phil,; Great Oolite, Ancliff; Wilts, Cookson. 2. sinuosa, Sow. Portland Stone, Conyb. 3. laevigata, Sow. Inf. Oolite, Dundry, Conyb. ; Shell Limestone and Calc. Grit, Portgower, &c., Scotland, Murch. 4. minuta, Sow. Great Oolite, Ancliff, Wilts, Cookson. 1. Natica arguta, Smith. Coral. Oolite, Yorks., Phil. 2. - nodulata, Y. fy B. Coral. Oolite, Yorks., Phil. 3. cincta, Phil. Coral. Oolite, Yorks., Phil. 4. adducta, Phil. Great Oolite and Inf. Oolite, Yorks., Phil. 5. - tumidula, Bean. Inf. Oolite, Yorks., Phil. , species not determined. Lias, Yorks., Phil. ; Kim. Clay, Ber- nese Jura, Thur. 1. Vermetus compressus, Y. 8f B. Coral. Oolite, Inf. Oolite, Yorks., Phil. 2. Nodus, Phil. Cornb., Great Oolite, Yorks., Phil. , species not determined. Cornbrash, Wilts, Lons. Delphinula, species not determined. Coral. Oolite and Great Oolite, Yorks., Phil. 1 . Solarium Calix, Bean. Inf. Oolite, Yorks., Phil. 2. conoideum, Sow. Portland Stone, Conyb. 1 . Cirrus cingulatus, Phil. Calc. Grit, Yorks., Phil. 2. depressus, Sow. Kell. Rock, Yorks., Phil. 3. nodosus, Sow. Inf. Oolite, Dundry, Conyb. 4. Leachii, Sow. Inf. Oolite, Dundry, Conyb. 5. carinatus, Sow. Inf. Oolite, Wilts, Lons. , species undetermined. Lias, N. of Fr., Bobl. ; Oxford Clay, Haute Saone, Thir. 1. Pleurotomaria conoidea, Desh. Normandy, Desh. 2. ornata, Defr. Inf Oolite, Bayeux, Desh.; Inf. Oolite, Dundry, Conyb. ; Inf. Oolite, Norm., De C.; Lias, N. of France, Bobl. 3. decorata, Von Buch. Neuhausen, G. T. 4. compressa, Sow. Lias, Southern England, Conyb. 1. Trochus arenosus, Sow. Coral. Oolite, Calc. Grit, Cornb., and Inf. Oolite, Yorks., Phil.; Inf. Oolite, Dundry, Conyb.; Inf. Oolite, Norm., De C. 2. ? tornatilis, Phil. Coral. Oolite, Yorks., Phil. 3. Tiara, Sow. Calc. Grit, Yorks., Phil.; Coral Rag, Mid. and S. Eng., Inf. Oolite, Dundry, Conyb.; Inf. Oolite, Norm., De C. 4. guttatus, Phil. Kell. Rock, Yorks., Phil. 5. monilitectus, Phil. Great Oolite, Yorks., Phil. 6. bisertus, Phil. Inf. Oolite, Yorks., Phil. 7. pyramidatus, Bean. Inf. Oolite, Yorks., Phil. 8. . Anglicus, Sow. Lias, Yorks., Phil.; Lias, Mid. and S. Eng., Conyb. ; Inf. Oolite, Haute Saone, Thir. ; Stufenberg ; Hei- ningen, G. T. 556 Organic Remains of the Oolitic Group. 9. Trochus angulatus, Sow. Inf. Oolite, Mid. and S. Eng., Conyb. ; Inf. Oolite, Norm. 10. dimidiatus, Sow. Inf. Oolite, Mid. and S. Eng., Conyb. 11. , duplicatus, Sow. Inf. Oolite, Mid. and S. Eng., Conyb.; Inf. Oolite, Haute Saone, Thir. ; Lias, Gundershofen, Voltz. ; Banz, G. T. 12 elon^atus, Sow. Inf. Oolite, Dundry, Conyb.; For. Marb. and Inf. Oolite, Norm., De C. 13. punctatus, Sow. Inf. Oolite, Dundry, Conyb. ; Inf. Oolite, Norm., De C. 14. , abbreviatus, Sow. Inf. Oolite, Dundry, Conyb.; Inf. Oolite, Norm., De C. 15. fasciatus, Sow. Inf. Oolite, Dundry, Conyb. ; Inf. Oolite, Norm., De C. 16. prominens, Sow. Inf. Oolite, Dundry, Conyb.; Inf. Oolite, Norm., De C. 17. . imbricatus, Sow. Lias, Mid. and S. Eng., Conyb. ; Inf. Oolite, Norm., De C. ; Lias, S. of Fr., Dufr. ; Soleure, Hcen. 18. reticulatus, Sow. Inf. Oolite, Norm., De C.; Coral Rag, Wey- mouth, Sedg. 19. rugatus, Benett. Portland Beds, Tisbury, Wilts, Benett. 20. speciosus, Munst. Hohenstein, Saxony ; Inf. Oolite, Bavaria, Munst. 21. niloticiformis, Stahl. Stufenberg, G. T. , species not determined. Portland Stone and Bradford Clay, Mid. and S. Eng., Conyb. ; Coral Rag, Norm., De C. ; Oxford Clay, Coral Rag, and Great Oolite, Haute Saone, Thir. ; Kim. Clay, Coral Rag, Oxford Clay, Bernese Jura, Thw. 1. Rissoa laevis, Sow. Great Oolite, Ancliff, Wilts, Cookson. 2. acuta, Sow. Great Oolite, Ancliff, Cookson. 3. obliquata, Sow. Great Oolite, Ancliff, CooJcson, 4. duplicata, Sow. Great Oolite, Ancliff, Cookson. 1. Turbo muricatus, Sow. Coral. Oolite, Great Oolite, and Inf. Oolite, Yorks., Phil. ; Coral Rag, Mid. and S. Eng., Conyb. ; Coral Rag, Weymouth, Sedg. 2. funiculatus, Phil. Coral. Oolite, Yorks., Phil. 3. sulcostomus, Phil. Kell. Rock, Yorks., Phil. 4. unicarinatus, Bean. Inf. Oolite, Yorks., Phil. 5. Isevigatus, Phil. Inf. Oolite, Yorks., Phil. 6. undulatus, Phil. Lias, Yorks., Phil. 7. ornatus, Sow. Inf. Oolite, Mid. and S. Eng., Conyb. ; Inf. Oolite, Norm., De C. ; Lias, Gundershofen, Voltz. 8. obtusus, Sow. Great Oolite, Ancliff, Cookson. , species not determined, Cornb. and Great Oolite, Norm., De C. 1. Phasianella cincta, Phil. Great Oolite, Yorks., Phil. 2. angulosa, Sow. Porta Westphalica, G. T. 1. Turritella muricata, Soiv. Coral. Oolite, Calc. Grit, Kell. Rock, and Inf. Oolite, Yorks., Phil. ; Rochelle Limestone, Dufr. ; Shell Limestone and Grit, Portgower, &c., Scotland, Murch. 2. cingenda, Sow. Coral. Oolite ? Great Oolite, and Inf. Oolite, Yorks., Phil. 3. quadrivittata, Phil. Inf. Oolite, Yorks., Phil. 4. concava, Sow. Portland Stone, Tisbury, Benett. 5. echinata, Von Buch. Banz, Langheim, Von Buck. } species not determined. Portland Stone, Coral Rag? Cornb., For. Marb., and Brad. Clay, Mid. and S. Eng., Conyb.; Brad. Clay, N. of Fr., Bobl. ; Portland Beds and Coral Rag, Haute Saone, Thir. ; Lias, Bath, Lons. ; Oxford Clay, Bernese Jura, Thur. Organic Remains of the Oolitic Group. 557 .1. Nerinsea tuberculata, Blain. Bailly, near Auxerre, Hoen. 2. Mosee, Desk. St. Mehiel (Meuse), Desk. 3. . Bruckneri, Thur. Kim. Clay, Bernese Jura, Thur. 4. Bruntrutana, Thur. Coral Rag, Bernese Jura, Thur. 5. elegans, Thur. Coral Rag, Bernese Jura, Thur. 6. pulchella, Thur. Coral Rag, Bernese Jura, Thur. , species not determined. Coral Rag and For. Marb., Norm., De C. ; Brad. Clay, N. of Fr., Bobl. ; Coral Rag, Inf. Oolite, Haute Saone, Thir. ; Rochelle, Nancy, Desk.; Neufchatel; Kelheim; Kahleberg, Echte, Von Buch. 1. Cerithium intermedium (var.). Bohlhorst, near Minden, Hoen. 2. muricatum, . Miihlhausen, Bus Rhin, Hoen. 3. quinquangulare, Thur. Bernese Jura, Thur. , species not determined. Lias, Gundershofen, Voltz. 1. Murex Haccanensis, Phil. Coral. Oolite, Yorks., Phil. 2. rostellariformis, Von Buch. Coral Rag, Randen, Schafhausen, Von Buch. 1 . Rostellaria bispinosa, Phil. Calc. Grit ? and Kell. Rock, Yorks., Phil. 2. trifida, Bean. Oxford Clay, Yorks., Phil. 3. composita, Sow. Sandst., Limest., and Shale, Inverbrora, Scotl., Mitrch. ; Great? and Inf. Oolite, Yorks., Phil. ; Oxford Clay, Weymouth, Sow. ; Kim. Clay, Havre, Phil. , species not determined. Lias, Yorks., Phil. ; Oxford Clay, Kell. Rock, Cornb., Forest Marb., and Inf. Oolite, Mid. and S. Erig., Conyb. ; Oxford Clay, Norm., De C. ; Coral Rag, Bernese Jura, Thur. 1. Pteroceras Oceani, Al. Brong. Kim. Clay, Havre and the Jura, Al. Brong. ; Poitland Beds, Kim. Clay ? Haute Saone, Thir. ; Portland Beds, Kim. Clay, Bernese Jura, Thur. ; Kahleberg, Echte, G. T. 2. Ponti, Al. Brong. Kim. Clay, Havre and the Jura, Al. Brong. ; Kim. Clay, Haute Saone, Thir. 3. Pelagi, AL Brong. Kim. Clay, Havre and the Jura, Al. Brong. 1. Actaeon retusus, Phil. Calc. Grit, Yorks., Phil. 2. glaber, Bean. Great Oolite and Inf. Oolite, Yorks., Phil. 3. humeralis, Phil. Inf. Oolite, Yorks., Phil. 4. cuspidatus, Sow. Great Oolite, Ancliff, Wilts, Cookson. 5. acutus, Soiv. Great Oolite, Ancliff, Wilts, Cookson. , species not determined. Lias, Yorks., Phil. 1. Buccinum unilineatum, Sow. Great Oolite, Ancliff, Wilts, Cookson. , species not determined. Shale, Sandst, and Limest., Inver- brora, Scotl., Murch. 1. Terebra melanoides, Phil. Coral. Oolite, Yorks., Phil. 2. ? granulata, Phil. Coral. Oolite and Cornb., Yorks., Phil. 3. vetusta, Phil. Great Oolite and Inf. Oolite, Yorks., Phil. 4. sulcata, . Coral Rag, N. of Fr., Bobl. ; Oxford Clay, Bernese Jura, Thur. 1. Belemnites sulcatus, Mill. Coral. Oolite? Calc. Grit, Oxford Clay, and Kell. Rock, Yorks., Phil. ; Shale, Sandst., and Limest., In- verbrora, Scotl., Murch. ; Lias, S. of France, Dufr. 2. fusiformis, Mill. Coral. Oolite? Yorks., Phil. 3. gracilis, Phil. Oxford Clay, Yorks., Phil. 4. abbreviates, Mill. Great Oolite, Yorks., Phil.; Lias, Ross and Cromarty, Scotland ; and Micaceous Sandstone, Western Is- lands, Scotland, Murch. 5. elongatus, Miller. Lias, Yorks., Phil.; Lias, Ross and Cromarty, Scotl., Murch. ; Fer. Oolite, Wasseralfingen, Zieten. fi. trisulcatus, Blain. Inferior Oolite, N. of Fr., Bobl. 558 Organic Remains of the Oolitic Group. 7. Belemnites compressus, Blain. Fuller's E., N. of Fr., Bobl. ; Inf. Oolite Yorks., Sow. ; Lias, Gundershofen, Foltz ; Culmbach ; Witt- berg; Metzingen, G. T. 8. dilatatus, Blain. Fuller's E., N. of Fr., Bobl; Theta, Baireuth, G. T. 9. apicicurvatus, Blain. Lias, S. of Fr., Dufr. ; Lias, Alais, Al. Brong. 10. pistilliformis, Blain. Lias, S. of Fr., Dufr.; Lias, Gunder- shofen, Foltz. 11. brevis, Blain. Lias, Alais, Brong. ; Goppingen, G. T. 12. longissimus, Miller. Lias, Bath, Lons ; Lias, Boll, Zieten. 13. canaliculatus, Schlot. Oxford Clay and Inf. Oolite, Haute Saone, Thir. ; Inf. Oolite, Southern Germany, Miinst. ; Stufenberg, Zieten. 14. ellipticus, Miller. Inf. Oolite, Haute Saone, Thir. 15. longus, Foltz. Great Oolite, Haute Saone, Thir. 16. ferruginosus, Foltz. (var.) Oxford Clay, Haute Saone, Thir. ; Oxford Clay, Bernese Jura, Thur. ; Swabia ; Bavaria, G. T. 17. aduncatus, Miller. Lias, Bath, Lons. 18. subclavatus, Foltz. Lias, Gundershofen ; Lias, Boll, Foltz. 19. tenuis, Stahl. Lias, Gundershofen, Foltz.; Lias, Altdorf, G. T. 20. subdepressus, Foltz. Lias, Gundershofen, Foltz. 21. subaduncatus, Foltz. Lias, Gundershofen, Foltz; Lias, Boll, Zieten. 22. digitalis, Biguet. Lias, Gundershofen, Foltz. 23. breviformis, Foltz. Lias, Gundershofen, Foltz ; Lias, Boll, Zieten. 24. ventroplanus, Foltz. Lias, Befort, Haut Rhin, Foltz. 25. paxillosus, Schlot. Lias, Befort; Lias, Boll, Foltz.; Lias, Ubstadt, near Heidelberg, Bronn. 26. longisulcatus, Foltz. Lias, Wurtemberg, Foltz. 27. trifidus, Foltz. Lias, Gundershofen, Foltz. 28. comprimatus, Foltz. Lias, Bahlingen, Von Buck. 29. Aalensis, Foltz. Inf. Oolite, Nipf, Bopfingen ; Baireuth, G. T. 30. grandis, Schubler. Inf. Oolite, Stufenberg, Zieten. 31. quinquesulcatus, Blain. chlatt, Wurtemberg, Zieten; Inf. Oolite, Baireuth, G. T. 32. tumidus, Zieten. Inf. Oolite, Stufenberg, Zieten. 33. teres, Stahl. Lias, Gosbach, Wurtemberg, Zieten. 34. laevigatus, Zieten. Lias, Boll, Zieten. 35. crassus, Foltz. Lias, near Goppingen, Wurtemberg, Zieten ; Besancon, Foltz. 36. semihast'atus, Blain. Lias, Gamelshausen, Wurtemberg, Zieten; Inf. Oolite, Baireuth, G. T. 37. incurvatus, Hehl. Lias, Boll, Zieten ; Lias, Banz, G. T. 38. pyramidatus, Schubler. Lias, Gross-Eislingen, Wurtemberg, Zieten. 39. rostratus, Zieten. Lias, Boll, Zieten. 40. papillatus, Plieninger. Lias, Boll, Zieten. 41. acuminatus, Schubler. Inf. Oolite, Stufenberg. 42. subhastatus, Zieten. Inf. Oolite, Stufenberg, Zicfew. 43. oxyconus, Heyl. Lias, Boll, Zieten. ; Lias, Banz; Altdorf, G.T. 44. carinatus, Heyl. Lias, Boll, Zieten. 45. pygmaeus, Zieten. Lias, Boll, Zieten. 46. unisulcatus, Hartmann. Geisslingen, &c., Wurtemberg, Zieten. 47. bisulcatus, Hartmann. Lias, Boll, Zieten. 48. quadrisulcatus, Hartmann. Lias, Gross-Eislingen, near Gop- pingen, Zieten. Organic Remains of the Oolitic Group. 559 49. Belemnites pyramidalis, Munst. Lias, Wurtemberg, Zieten. ; Lias, Banz, G. T. 50. bipartitus, Hartmann. Gruibingen, Wurtemberg, Zleten. 51. unicanaliculatus, Hartmann. Donzdorf, Wurtemberg, Zleten. 52. bicanaliculatus, Hartmann. Ganslosen, W^urtemberg, Zleten. 53. tricanaliculatus, Hartmann. Lias, Stufenberg, Zleten. 54. quadricanaliculatus, Hartmann. Stufenberg, Zleten. 55. quinquecanaliculatus, Hartmann. Lias, near Goppingen, Zieten. 56. semisulcatus, Munst. Upper part of Oolite Group, Southern Germany, (Staffelberg ; Lichtenfels; Solenhofen, &c.), Munst. ; Oxford Clay, Bernese Jura, Thur. 57. pusillus, Munst. Streitberg, Munst. 58. acuarius, Schlot. Lias, Banz, Munst. ; Lias, Altdorf, G. T. 59. latesulcatus, Foltz. Oxford Clay, Bernese Jura, Thur. 60. deformis, Munst. Southern Germany, G. T. 61. gladius, Blain. Metzingen, Baireuth, G. T. 62. Blain villii, Foltz*. Swabia, G. T. 63. hastatus, Blain. Inf. Oolite, Baireuth; Metz; Banz, G. T. 64. tripartitus, Schlot. Lias, Altdorf, G. T. 65. - clavatus, Blain. Lias, Boll; Amberg; Banz; Lyme Regis, G. T. , species not determined. Kim. Clay and Inf. Oolite, Yorks., Phil. Kim. Clay, Coral Rag, Oxford Clay, Kell. Rock, Stonesfield Slate, Bradford Clay, and Inf. Oolite, Mid. and S. England, Conyb. ; Oxford Clay, For. Marb., Great Oolite, Inf. Oolite, and Lias, Norm., De C.; Lias, N. of Fr., Bobl. 1. Orthoceratites ? elongatus, De la B. Lias, Lyme Regis, De la B. 1. Nautilus hexagonus, Sow. Kell. Rock? Yorks., Phil. ; Calc. Grit, Ox- ford, Sow. 2. lineatus, Sow. Inf. Oolite and Lias, Yorks., Phil.; Inf. Oolite, Dundry, Conyb. ; Inf. Oolite? Haute Saone, Thir.; Lias, Bath, Lons. 3. astacoides, Y. % B. Lias, Yorks., Phil. 4. annularis, Phil. Lias, Yorks., Phil. 5. obesus, Sow. Inf. Oolite, Mid. and S. Eng., Conyb.; Inf. Oolite, Norm., De C. G. sinuatus t, Sow. Inf. Oolite, Mid. and S. Eng., Conyb. ; Oxford Clay, Norm. ? De la B. 7. intermedius, Sow. Lias, Mid. and S. Eng., Conyb. ; Wurtem- berg, G. T. 8. striatus, Sow. Lias, Mid. and S. Eng., Conyb. ; Lias, Alsace, Brong. 9. truncatus, Sow. Lias, Mid. and S. Eng., Conyb. ; For. Marb. and Lias, Norm., De Can. 10. angulosus, D'Orbigny. Portland Stone, Isle d'Aix, Brong. , species riot stated. Great Oolite, Yorks., Phil.; Kim. Clav, Coral Rag, Oxford Clay, Kell. Rock, and Stonesfield Slate, Mid. and S. Eng., Conyb. ; Coral Rag, Norm., De Can. Fuller's Earth, N. of Fr., Bobl. Hamites, species not determined. Lias, Zell, near Boll, Zieten ; Inf. Oolite, Bayeux, Desh. fy Majendie %. \. Scaphites bifurcatus, Hartmann. Lias, Goppingen, Wurtemberg, Zieten. 2. refractus , . Gamelshausen, G. T. * B. acutus and B. apiciconus, Blain. f Nautilus aganaticus, Schlot. J It should also be noticed, that M. Deshayes (Desc. des Coquilles Caracteristi- quesdes Terrains) describes and figures a Hamite, by the name of Hamites annulatus, as found in the ferruginous oolite, but unfortunately does not mention the locality. Ammonites refractus, Rein. 560 Organic Remains of the Oolitic Group. Scaphites, species not determined. Lias, S. of England, Conyb. Turrilites Babeli, Al. Brong. Coral Rag ? N. of Fr., Boll 1. Ammonites perarmatus, Sow. Coral Rag, Malton, Sow.; Coral. Oolite, Calc. Grit, and Kel. Rock, Yorks., Phil. ; Oolitic Rocks, Bra- ambury Hill, Brora, Murch. ; Coral Rag, Wilts, Lons. ; Coral Rag, Randen, Von Such. ; Oxford Clay, Bernese Jura, Thur. ; Mordberg, Niirnberg, G. T. 2. plicomphalus, Sow. Bolingbroke, Lincolnshire, Sow. ; Kim. Clay? Yorks., Phil.; Oxford Clay, Norm., De C. 3. triplicatus, Sow. Coral. Oolite, Yorks, Phil; Inf. Oolite, Norm., De C. ; Coral Rag, Randen, Von Buck. 4. plicatilis, Sow. Coral. Oolite and Kell. Rock, Yorks., Phil ; Coral Rag, Mid. and S. Eng., Conyb. ; Oxford Clay and Kell. Rock, Haute Saone, Thir. ; Coral Rag, Randen, Von Buch. 5. Williamsoni, Phil. Coral. Oolite, Yorks., Phil 6. Sutherlandiae*, Sow. Sandstone, Braambury Hill, Brora, Murch.; Coral Oolite and Calc. Grit, Yorks., Phil. ; Randen ; Thur- nau ; Staffelberg, G. T. 7. sublaavis, Sow. Coral. Oolite and Kell. Rock, Yorks., Phil. ; Full. E., Env. of Bath, Lons.; Oxford Clay, Beggingen, Schaf- hausen, Von Buch. ; Kell. Rock, Mid. and S. Eng., Conyb. ; Oxford Clay, Norm., De la B. 8. lenticularis, Phil Coral. Oolite ? Kell. Rock, and Lias, Yorks., Phil 9. vertebralis and cordatus, Sow. Coral. Oolite, Calc. Grit, and Oxford Clay, Yorks., Phil.; Coral Rag, Mid. and S. Eng., Conyb.; Oolite of Braambury Hill, Brora, Murch.; Kim. Clay and Oxford Clay, Haute Saone, Thir. ; Coral Rag, Wilts, Lons. 10. instabilis, Phil. Calc. Grit., Yorks., Phil. 11. Solaris, Phil. Calc. Grit, Yorks., Phil. 12. oculatus, Phil. Oxford Clay, Yorks., Phil. 13. Venioni, Bean. Oxford Clay, Yorks., Phil. 14. Athleta, Phil. Oxford Clay and Kell. Rock, Yorks., Phil. 15. Kcenigi, Sow. Kell. Rock, Yorks., Phil. ; Kell. Rock, Kello- ways, Wilts; Lias, Charmouth, Sow. ; Micaceous Sandst., Western Islands, Scotl., Murch.; Gammelshausen, Zieten. ; Hohenzollern, G. T. 16. bifrons, Phil, f Kell. Rock, Yorks., Phil. 17. Gowerianus, Sow. Shale, Sandst. and Limest., Inverbrora, Scotl, Murch.; Kell. Rock, Yorks., Phil. 18. CalloviensisJ, Sow. Kell. Rock, Yorks., Phil; Kell. Rock, Kelloways, Sow. 19. Duncani, Sow. St. Neots, Huntingdonshire, Duncan; Kell. Rock, Yorks., Phil. ; Oxford Clay, Mid. and S. Eng., Conyb. ; Oxford Clay, Norm., De Can. ; Oxford Clay, Haute Saone, Thir. 20. gemmatus, Phil. Kell. Rock, Yorks., Phil. 21. Herveyi, Sow. Spalden, Lincolnshire; Bradford; Knowles Hill, Somerset, Sow. ; Kell. Rock? and Cornb., Yorks., Phil. ; Inf. Oolite, Mid. and S. Eng., Conyb. ; Inf. Oolite, Wasseral- fingen, Wurtemberg, Zieten. 22. flexicostatus, Phil. Kell. Rock, Yorks., Phil. 23. funiferus, Phil. Kell. Rock, Yorks., Phil. * Am. inflatus, Rein. f This Ammonite must be distinguished from A. bifrons, Bruguiere, which is, according to M. Deshayes, the A. Walcotii of Sowerby. + Am. Jason, Rein. Organic Remains of the Oolitic Group. 561 24. Ammonites terebratus, Phil. Comb., Yorks., Phil. 25. Blagdeni, Sow. Great Oolite, Yorks., Phil. ; Inf. Oolite, Dun- dry, Conyb. ; Inf. Oolite, Norm., De Can. ; Spaichingen ; Metzingen, G. T. 26. striatulus, Sow. Inf. Oolite and Lias, Yorks., Phil. ; Lias, Was- seralfingen, Zieten. 27. heterophyllus, Sow. Lias, Yorks., Phil. ; Lias, Midland and Southern England, Conyb. 28. subcarinatus, Y. $ B. Lias, Yorks., Phil. 29. Henleii, Sow. Lias, Yorks., Phil ; Lias, Mid. and S. Eng., Conyb. 30. heterogeneus, Y. fy B. Lias, Yorks., Phil. 31. crassus, Y. % B. Lias, Yorks., Phil. 31.* crassus, Montf. Kim. Clay, H6court, Norm., Pas. 32. communis, Sow. Lias, Yorks., Phil. ; Lias, Mid. and S. Eng., Conyb.; Lias, Western Islands, Scot., Murch.; Soleure, Hcen.; Lias, Wurtemberg, Zieten. 33. angulatus, Sow. Lias, Yorks., Phil. ; Lias, Mid. and S. Eng., Conyb. 34. annulatus, Sow. Lias, Yorks., Phil. ; Inferior Oolite and Lias, Midi, and S. Eng., Conyb. ; Oxford Clay, For. Marb., and Inf. Oolite, Norm., De C. ; Inf. Oolite, Uzer, S. of Fr. ; Ro- chelle Limestone, Dufr.; Inf. Oolite and Lias, Montdor, Lyon, Al. Brong. ; Coral Rag, Inf. Oolite, Wilts, Lons. ; Coburg, If oil; Inf. Oolite, Gamelshausen, Wurtemberg, Zieten. ; Mo- ritzberg, Nurnberg, G. T. 35. fibulatus, Sow. Lias, Yorks., Phil 36. subarmatus, Sow. Lias, Yorks., Phil. 37. maculatus, Y. 8f B. Lias, Yorks., Phil. 38. gagateus, Y. 8f B. Lias, Yorks., Phil. 39. planicostatus*, -Sow. Maston Magna, Yeovil, Somerset, Sow. ; Lias, Yorks., Phil. ; Lias, Mid. and S. Eng., Conyb. ; Lias, Bath, Lons.; Kahlefeld, Hartz; Amberg, Altdorf, Hott; Lias, Bahlingen, Von Buck. 40. balteatus, Phil. Lias, Yorks., Phil. 41.- arcigerens, Phil. Lias, Yorks., Phil. 42. brevispina, Sow. Lias, Western Islands, Scotl., Murch. ; Lias, Yorks., Phil. 43. Jamesoni, Sow. Lias, Western Islands, Scotl., Murch. ; Lias, Yorks., Phil. 44. erugatus, Bean. Lias, Yorks., Phil. 45. fimbriatus, Sow.\ Lias, Lyme Regis, Buckl. ; Lias, Yorks., Phil. ; Lias, Mid. and S. Eng., Conyb. ; Lias, Norm., De C. ; Lias, Wurtemberg, Zieten ; Lias, Mende, Lozere, Banz ; Randen, Von Buck ; Inf. Oolite, Calvados, Desl. ; Lias, Gr. Gschaid ; Culmbach ; Rautenberg, Scheppenstadt, G. T. 46. nitidus, Y. fy B. Lias, Yorks., Phil. 47. anguliferus, Phil. Lias, Yorks., Phil. 48. crenularis, Phil. Lias, Yorks., Phil. 49. Clevelandicus, Y. $ B. Lias, Yorks., Phil. 50. Turneri, Sow. Lias, Watchet ; Wymondham Abbey, Sow. ; Lias, Yorks., Phil; Lias, South of France, Dufr. ; Lias, Wurtemberg, Zieten. 51.- geometricus, Phil. Lias, Yorks., Phil. * Am. Capricornus, Schlot. f According to Von Buch, this Ammonite is the same with A. lineatus and A. hircinus of Schlotheim. 2o 562 Organic Remains of the Oolitic Group. 52. Ammonites vittatus, Y. $ B. Lias, Yorks., Phil 53. sigmifer*, Phil Lias, Yorks., Phil ; Inf. Oolite, Haute Saone, Thir. ; Lias, Wurtemberg, Voltz. 54. Hawkserensis, F. $ B. Lias, Yorks., Phil. 55. Conybeari, Sow. Lias, Yorks., Phil. ; Lias, Mid. and S. Eng., Conyb.; Lias, Gundershofen and Buxweiller, Al. Brong.; Lias, Western Islands, Scotl., Murch. ; Lias, Southern Ger- many, G. T. 56. Bucklandi, Sow.^ Lias, Yorks., Phil ; Lias, Mid. and S. Eng., Conyb. ; Lias, Norm., De Cau. ; Lias, Malsch, near Heidel- berg, Bronn ; Inf. Oolite, Calvados, Desl 57. obtusus, Sow. Lias, Yorks., Phil; Lias, Mid. and S. Eng., Conyb. 58. Walcotii, Sow. Lias, Yorks., Phil. ; Inf. Oolite and Lias, Mid. and S. Eng., Conyb. ; Lias, S. of Fr., Dufr. ; Lias, Befort, HautRhin ; Lias, Boll, Voltz.; Achelberg, Ham.; Inf. Oolite, Calvados, G. T. 59. ovatus, Y- $ B. Lias, Yorks., Phil 60. Mulgravius, F. 8f B. Lias, Yorks., Phil ; Lias, Boll, G. T. 61. exaratus, F. $ B. Lias, Yorks., Phil 62. Lythensis, Y. 8f B. Lias, Yorks., Phil 63. concavus, Sow. Lias? Yorks., Phil; Inf. Oolite, Mid. and S. Eng., Conyb.; Coburg, Hott ; Inf. Oolite, Calvados, G. T. 64. elegansj, Sow. Lias? Yorks., Phil ; Inf. Oolite, Dundry, Co- nyb. ; Lias, Norm., De C. ; Inf. Oolite, Uzer, S. of Fr., Dufr. ; Lias, Wurtemberg, Zieten. ; Inf. Oolite, Calvados, G. T. 65. discus, Sow. Inf. Oolite, Dundry, Comb., Mid. and S. Eng., Conyb.; Inf. Oolite. Norm., De C.; Cornb., Wilts., Lons. ; Inf. Oolite, Wasseralfingen, Zieten. ; Comp. Great Oolite, Bernese Jura, Thur. ; Spaicbingen, G. T. 66. Banksii, Sow. Inf. Oolite, Dundry, Conyb. 67. Braikenridgii, Sow. Inferior Oolite, Dundry, Conyb. ; Inf. Oolite, Norm., De C. ; Gammelshausen, Zieten. 68. Brocchii, Sow. Inf. Oolite, Dundry, Conyb. ; Inf. Oolite, Haute Saone, Thir. 69. Sowerbii, Miller. Inf. Oolite, Dundry, Conyb. 70. falcifer, Sow. Inf. Oolite, Dundry, Conyb. ; Lias, Norm., De C. ; Lias, S. of Fr., Dufr. ; Lias, Wurtemberg, Zieten. ; Inf. Oolite, Barendorf ; Thurnau, Munst. ; Inf. Oolite, Bernese Jura, Thur. 71. Brownii, Sow. Inf. Oolite, Dundry, Conyb. 72. laeviusculus, Sow. Inf. Oolite, Dundry, Braikenridge ; Inf. Oolite, Norm., De C. 73. acutus, Sow. Oxford Clay, Inf. Oolite, Norm., De C. ; Lias, Western Islands, Scotl., Murch. ; Inf. Oolite, Haute Saone, Thir. ; Wasseralfingen, Zieten. 74. contractus, Sow. Inf. Oolite, Dundry, Sow.; Inf. Oolite, Norm., De C. 75. giganteus, Sow. Portland Stone, Coral Rag, and Lias, Mid. and S. Eng., Conyb. ; Portland Stone, Isle d'Aix, Brong. (var.) ; Inf. Oolite, Haute Saone, Thir. 76. Lamberti, Sow. Portl. Stone, Conyb.; RocbelleLimest.,Z)z//r. ; Coburg; Heinberg; Bamberg, Holl ; Oxf. Clay, Bernese Jura, Thur. * Am. costulatus. Rein. , f This Ammonite is, according to M. Deshayes, the A. bisulcatus of Bruguiere, and the A. Arietis of Schlotheim. J Am. radians, Rein. Organic Remains of the Oolitic Group. 563 77. Ammonites excavatus, Sow. Coral Rag, Mid. and S. Eng., Conyb. ; Oxford Clay, Norm., De la B. ; Lias, Norm., De C. ; Altdorf, G. T. 78. armatus, Sow. Oxford Clay and Lias, Mid. and S. Eng., Conyb.; Oxford Clay, Norm., De la B. ; Oxford Clay, Haute Saone, Thir.; Lias, Bath, Lons. ; Oxf. Clay, Bernese Jura, Thur. 79. modiolaris, Smith. Fuller's Earth ? Mid. and S. Eng., Conyb. 80. jugosus, Sow. Inf. Oolite, Mid. and S. Eng., Conyb. 81. Stokesii, Sow.* Inf. Oolite, Mid. and S. Eng., Conyb.; Lias, S. of Fr. Dufr. ; Inf. Oolite, Haute Saone, Thir. ; Lias, Wur- temberg, Zieten. ; Lias, Ubstadt, near Heidelberg, Bronn ; Oxf. Clay, Inf. Oolite, Bernese Jura, Thur. ; Inf. Oolite, Cal- vados, Desl. 82. Strangwaysii, Sow. Inf. Oolite, Mid. and S. Eng., Conyb. ; Lias, Norm., De C. 83. Brookii, Sow. Lias, Lyme Regis, Buckl. ; Lias, Gb'ppingen, G. T. 84. Bechii, Sow. Inf. Oolite and Lias, Mid. and S. Eng., Conyb. ; Was, Lyme Regis, De la B. ; Coburg, Holl. ; Lias, Rottweil ; Bahlingen, G. T. 85. stellaris, .0*0. Lias, Mid. and S. Eng., Conyb. ; Lias, Norm., De C. ; Lyme Regis, De la B. 86. Greenovii, Sow. Lias, Mid. and S. Engl., Conyb. ; Lias, Lyme Regis, De la B.; Halsbach; Diinkelsbiihl, G. T. 87. Loscombi, Sow. Lias, Mid. and S. Engl., Conyb. ; Lias, Lyme Regis, De la B. 88. - Birchii, Sow. Lias, Mid. and S. Engl., Conyb. ; Lias, Lyme Regis, De la B. ; Lias, Goppingen, G. T. 89. - omphaloides, Sow. Portland Stone, Sow. ; Oxford Clay, Norm., De la B. ; Gt. Arenaceous Formation, Western Islands, Scotl., Murch. ; Oxf. Clay, Bernese Jura, G. T. f90. quadratus, Sow. Inf. Oolite, Norm., De C. 91. - Gervillii, Sow. Inf. Oolite, Norm., De C. 92. Brongniartii, Sow. Inf. Oolite, Norm., De C. f 93. biplex, Sow. Inf. Oolite, Norm., De C. ; Lias, Ross and Cro- marty, Scotl., Murch.; Oxford Clay, Haute Saone, Thir.; Solenhofen, Keen.; Calc. Grit, Oxf. Clay, Bernese Jura, Thur.; Randen ; Rathhausen ; Streitberg ; Altdorf, G. T. 94. rotundus, Sow. Inf. Oolite, Norm., De C.; Kim. Clay, Pur- beck, Sow. f95. decipiens. Hohenstein, Saxony; Solenhofen, Munst. 96. Deslongchampi . Inf. Oolite, N. of Fr., Bobl. 97. vulgaris . Bradford Clay, N. of Fr., Bobl. 98. coronatus . Oxford Clay ? N. of Fr., Bobl. 99. Humphresianus J, Sow. Lias, S. of Fr., Dufr.; Inf. Oolite, Sherborne, Sow.; Lias, Boll, Zieten. 100. Pavkinsoni, Sow. Lias, Yeovil, Sow.; Inf. Oolite, Bayeux, Majendie ; Inf. Oolite, Hohenstein; Barendorf; Thurnau, Munst.; WaslSeralfingen ; Wisgoldingen ; Bopfingen, G. T. 101. Gulielmii, Sow. Oxford Clay, S. Engl., Sow. 102. Davsei, Sow. Lias, Lyme Regis, De la B. ; Lias, Wasseral- fingen, Wurtemberg, Zieten. 103. planorbis, Sow. Lias, Watchet, Somerset, Sow. * A. Amaltheus. f Found, according to Sowerby, in the Suffolk gravel. J Am. Bollensis, Zieten. 2 o 1 56 4- Organic Remains of the Oolitic Group. 104. Ammonites Johnstonii, Sow. Lias, Watchet, Somerset, Sow.; Lias, Bath, Lons. 105. comigatus, Sow. Inf. Oolite, Dundry, Braikenridge. 106. rotiformis, Sow. Lias, Yeovil, Sow. ; Lias, Bath, Lons. 107. multicostatus, Sow. Lias, Bath, Sow. 108. laevigatus, Sow. Lias, Lyme Regis, De la B. 109. lateecostata, Sow. Lias, Lyme Regis, Murch. 110. Murchisonae, -Sow. Micaceous Sandst., Holm Cliff, Western Islands, Scotl., Murch. ; Inf. Oolite, Allington near Bridport, Murch. ; Wasseralfingen ; Gundershofen ; Wisgoldingen ; Goslar, G. T. HI. serpentinus *, Rein. Inf. Oolite, Haute Saone, Thir. ; Lias, Gundershofen, Voltz. ; Lias, Ubstadt, near Heidelberg, Bronn ; Lias, Altdorf ; Boll, G. T. 112. cristatus, Defr. Weymouth, Bryer ; Oxford Clay, Haute Saone, Thir. ; Oxf. Clay, Bernese Jura, Thur. 113. interruptus, Schlot. Oxford Clay, Haute Saone, Thir.; Thir- nau, Roll. ; Oxf. Clay, Bernese Jura, Thur. 1 14. opalinus, Reinecke. Lias, Gundershofen, Voltz. 115. latina, Sow. Coral Rag, Wilts, Lons. 116. ammonius, Schlot. Lias, Gundershofen, Voltz ; Altdorf, Holl. 117. comptusf, Reinecke. Lias, Gundershofen, Voltz; Donzdorf, Zieten. 118. planulatus, De Haan. Baireuth, Holl. 119. Knorrianus, De Haan. Boll, Wurtemberg, Holl. 120. Reineckii, Holl. Coburg, Holl. 121. pustulatus, Rein. Coburg; Thurnau, Holl. 19.2. granulatus, Brug. Coburg, Holl. 123. bifurcatus, Brug. Coburg; Baireuth, Holl. ; Coral Rag, Ger- many, Von Buck. 124. trifurcatus, De Haan. Coburg, Holl. 125. macrocephalus J, Schlot. Aarau; Coburg, Holl.; Inf. Oolite, Southern Germany, Munst. ; Vaches Noires, Calvados, G. T. 126. Planula, Heyl. Donzdorf, Holl. 127. Fonticola, Mencke. Ferruginous Beds, Thurnau ; Langheim ; VonBuch; Gamelshausen, Zieten; Oxford Clay, Haute Saone, Thir. ; Oxf. Clay, Bernese Jura, Thur. 128. scutatus, Von Buck. Lias, Banz, near Bamberg, Von Buck. 129. canaliculatus, Munst. Woschnau, Aarau; Furstenberg; Lo- chenberg, Bahlingen, Von Buck. 130. flexuosus, Munst. Coral Rag, Streitberg, near Erlangen ; Donzdorf, Swabia ; Rathhausen, near Bahlingen ; summit of Mont Randen, near Schafhausen, Von Buck. 131. ' crenatus, Rein. CoraLRag, Germany, Von Buck. 132. sublaevis, Munst. Donzdorf, Zieten. 133. hecticus, Rein. Inf. Oolite, Gamelshausen, Zieten; Oxf. Clay, Bernese Jura, Thur. 134. Pollux, Rein. Inf. Oolite, Gamelshausen, Zieten; Vaches Noires, Calvados; Goslar; Thurnau, G. T. 135. aequistriatus, Munst. Lias, Wurtemberg, Zieten. 136. inaequalis, Merian. Bale, Merian. 137. tenuistriatus, Munst. Solenhofen, Hoen. 138. dnbius, Schlot. Gamelshausen, Zieten ; Oxf. Clay, Bernese Jura, Thur. * Am. Strangwaysii, Sow., according to G. T. f Am. gracilis, Munst. J Am. tumidus, Rein. According to Von Buch this Ammonite is figured as A. Lunula by M. Zieten. Organic Remains of the Oolitic Group. 565 139. Ammonites Kridion, Rein. Lias, Stutgard, Zieten. 140. Jason, Rein. Lias, Gamelshausen, Zieten. 141. alternans, Von Buck. Coral Rag, Muggendorf, Gailenreuth, &c. Von Buck. 142. Gigas, Zieten. Riedlingen on the Danube, Zieten. 143. denticulatus, Zieten. Lias, Boll, Zieten ; Oxf. Clay, Bernese Jura, Thur. 144. raricostatus, Zieten. Lias, Boll, Zieten. 145. decoratus, Zieten. Inf. Oolite, Guttenberg, Wurtemberg, Zieten. 146. . bipartitus, Zieten. Inf. Oolite, Guttenberg, Zieten. 147. toralosus, Schubler. Lias, Stuifenberg, Zieten. 148. oblique-costatus, Zieten. Lias, Kaltenthal, near Stuttgart, Zieten. 149. insignis, Schubler. Lias, Reichenbacb, Zieten. 150. oblique-in terruptus, Schubler. Lias, Wasseralfingen, Zieten. 151. . polygonius, Zieten. Lias, Zell, near Boll, Zieten. 1 52. discoides, Zieten. Lias, Reichenbach, Zieten. 153. bispinosus, Zieten. Wasseralfingen, Zieten. 154. biarmatus, *SW. Hohenstein, Saxony ; Bavaria, Wurtemberg, Switzerland; Munst. 155. laevis, Schlot. Inf. Oolite, Southern Germany, Munst. ; Lias, near Heidelberg, Bronn. 156. colubrinus, Rein. Oxf. Clay, Bernese Jura, Thur. 157. laevigatus, Schlot. Oxf. Clay, Bernese Jura, TAwr. 158. anceps, Rein. Oxf. Clay, Bernese Jura, Thur. 159. inflatus, Rein. Oxf. Clay, Bernese Jura, Thur. 160. Deluci, Al. Brong. Neuhausen, G. T. 161. Comensis, Von Buch. Neuhausen, G. T. 162. alternans, Von Buch. Coral Rag, Bamberg, G. T. 163. cristatus, Sow. Guttenberg, Streitberg, G. T. 164. polygyratus, Rein. Donzdorf; Randen, G. T. 165. tripartitus, Sow. Randen, G. T. 166. multiradiatus, Reng. Willibaldsburg, Eichstadt, G. T. 167. longidorsatus, Von Buch. Lias, Moutiers, Caen, G. T. 168. asper, Merian. Haute Rive, Neufchatel, G. T. 169. planorbiformis, Munst. Lias, Bavaria, G. T. 170. colubratus, Montf. Lias, Vaichingen; Dunkelsbiihl, G. T. 171. angulatus, Schlot. Lias, Neckar Thailfingen ; Wellersen, Scheppenstadt, G. T. 172. natrix, Schlot. Lias, Bahlingen ; Gr. Brunsrode ; Altdorf, G. T. 173. funicularis, Von Buch. Lias, near Strasburg, G. T. 1. Aptychus* laevis, Meyer. Solenhofen; Stafelstein, Bavaria, Meyer; Stufenberg; Banzberg, Amberg, G. T. 2. imbricatus, Meyer. Solenhofen, Meyer ; Lias, Banzberg, G.T. 3. bullatus, Meyer. Lias, Banz, Meyer ; Lias, Boll, G. T. 4. Elasma, Meyer. Lias, Banz, Meyer ; Lias, Boll, G. T. 1. Onychoteuthis angustaf, Munst. Solenhofen, Riippell. 1. Sepia antiquaj, Munst. Soleuhofen, Riippell. , remains of, with ink-bags preserved, Lias, Lyme Regis, Buckl. Rhyncolites, or Sepia beaks, Lias, Lyme Regis, De la B. ; Lias, near Bristol, Miller. * Trigonellites, Parkinson ; Tellinites, Schlotheim. f Loligo prisca, Riippell. t Septa hastaformis, Riippell. 566 Organic Remains of the Oolitic Group. CRUSTACEA. 1. Pagurus mysticus, Germar. Solenhofen, If oil. 1. Eryon Cuvieri, Desm. Solenhofen; Erchstadt, Pappenheim, Holl. 2. muticus, Germ. Solenhofen, G. T. 3. propinquus *, Germ. Solenhofen, Holl. 4. spinimanus, Germ. Solenhofen, G. T. ] . Scyllarus dubius, Holl. Solenhofen, Holl. 1. Palaemon spinipes, Desm. Pappenheim, Solenhofen, Holl. 2. longimanatus, Holl. Solenhofen, Holl. 3. Walchii, Holl. Pappenheim, Holl. 1. Astacus modestiformis, Holl. Solenhofen, Holl. 2. minutus, Holl. Solenhofen, Holl. 3. rostratus, Phil. Kelloway Rock and Coral. Oolite, Yorks., Phil. 4. leptodactylus, Germ. Solenhofen, G. T. 5. spinimanus, Germ. Solenhofen, G. T. 6. fusiformis, Holl. Solenhofen, G. T. , species not determined. Oxford Clay and Lias, Yorks., Phil. Crustacea, not yet determined. Lias, Midi, and S. Engl., Conyb. ; Lyme Regis, De la B. ; Forest Marble, Normandy, De Can. ; Stonesfield Slate, Conyb. ; Bradford Clay, North of France, Bobl. INSECTA. Insects of the families Libellula ; ^Eschna ; Agrion ; Myrmeleon ? Sirax ? Solpaga? Solenhofen, Munst., Murch,, G. T. Elytra of coleopterous insects. Stonesfield Slate, Leach, Buckl. PISCES. 1. Dapedium politum, De la B. Lias, Lyme Regis, De la B.; Lias, and Oxford Clay of Normandy, De Can. 1. Clupea sprattiformis, Blain. Solenhofen, Holl. 2. dubia, Blain. Solenhofen, G. T. 3. Knorrii, Blain. Solenhofen, G. T. 4. Salmonea, Blain. Solenhofen, G. T. 5. Davilei, Blain. Solenhofen, G. T. 1. Esox avirostris, Germ. Solenhofen, G. T. 2. acutirostris, Blain. Solenhofen, G. T. ] . Uraeus gracilis, Ag. Lias, Wurtemberg, Ag. 1. Sauropsis latus, Ag. Lias, Wurtemberg, Ag. 1. Ptycholepis Bollensis, Ag. Lias, Boll, Ag. 1. Semionotus leptocephalus, Ag. Lias, Zell; Boll, Ag. 1. Lepidotes Gigas, Ag. Lias, Ohmden, Boll, Ag. 2. frondosus, Ag. Lias, Zell ; Boll, Ag. 3. ornatus, Ag. Lias, Wurtemberg, Ag. 1. Leptolepis Bronnii, Ag. Lias, Neidingen, Ag. 2. Jaegeri, Ag. Lias, Zell ; Boll, Ag. 3. longus, Ag. Lias, Zell ; Boll, Ag. 1 . Tetragonolepis heteroderma, Ag. Lias, Zell ; Boll, Ag. 2. semicinctus, Bronn. Lias, G. T. 3. pholidotus, Ag. Lias, Zell ; Boll, Ag. 4. Traillii, Ag. Lias, England, G. T. Fish, species not yet determined. Several in the Lias, Lyme Regis, De la B.; Barrow, Leicestershire, Conyb.; Portland Beds, Tisbury, Wilts, Benett. Ichthyodorulites, Buckl. fy De la B. Different kinds. Lias, Lyme Regis, and elsewhere in Southern and Midland England, * Eryon Schlotheimii, Holl. Organic Remains of the Oolitic Group. 567 Conyb. 8f De la B. ; Kimmeridge Clay, near Oxford, Buckl.; Stonesfield Slate, Buckl. In the Great Oolite, Normandy, De Can. Fish palates and teeth. Lias, Lyme Regis, and Somersetshire, &c. Conyb. ; Stonesfield Slate, Buckl. ; Great Oolite, Normandy, De Can. ; Cornbrash and Forest Marble, North of France, Bobl. ; Coral. Oolite, Oxford Clay, Yorks., Phil. ; Portland Beds, Tisbury, Wilts, Benett. REPTILIA. 1. Pterodactylus macronyx, Buckl. Lias, Lyme Regis, Buckl.; Lias, Banz, Bavaria, Meyer. 2. longirostris, Cuv. Aichstadt, Collini. 3. brevirostris, Cuv. Aichstadt, Cuv. 4. grandis, Cuv. Solenhofen, Holl. 5. - crassirostris, Gold/. Solenhofen, Goldf. 6. medius, Munst. Monheim, Schnitzlein. 7. - Munsteri, Goldf. Monheim, Goldf. Pterodactylus, species not known. Stonesfield Slate, Buckl. 1. Macrospondylus Bollensis*, Von Meyer. Lias, Boll, Jag. 1. Crocodilus cylindrirostris, Cuv. Kim. Clay, Havre, Al. Brong. ; Alt- dorf, G. T. 2. brevirostris, Cuv. Kim. Clay, Havre, Al. Brong.; Altdorf, G. T. 3. Crocodile of Mans, Cuv. Great Oolite, Al. Brong. remains, species not determined. Lias, Yorks., Phil. ; Lias ? Lyme Regis, De la B. ; Cornbrash, Engl., Conyb. ; Stones- field Slate, Buckl.; Coral. Oolite, Yorks., Phil. ; Inf. Oolite, Calvados, Her. 1. Teleosaurus Cadomensis, Geoffroy St. Hilaire. Great Oolite, Caen, De Can. 1. Megalosaurus Bucklandi. Stonesfield Slate, Buckl. , species not known. Great Oolite, Normandy, De Can. 1. Geosaurus Bollensis, Jag. Lias, Boll, Jag. 2. Sommeringii, Cuv. Monheim, Sommering. \. Lacerta Neptunia, Goldf. Monheim, Goldf. 2. gigantea, Munst. Monheim, G. T. 1. Rhacheosaurus gracilis, Meyer. Daiting, Solenhofen, Meyer. 1. Jilodon priscusf, Von Meyer. Monheim, Sommering. 1. Pleurosaurus Goldfussii, Meyer. Daiting, Meyer. 1. Plesiosaurus dolichodeirus, Conyb. Lias, Lyme Regis, &c. 2. recentior, Conyb. Kim. Clay, Engl., Conyb. ; Kim. Clay, Hon- fleur, Al. Brong. 3. carinatus, Cuv. Great Oolite, Boulogne, AL Brong. 4. pentagonus, Cuv. Great Oolite, Ballon and Chaufour, Al. Brong. 5. ? trigonus, Cuv. Great Oolite, Calvados, .Al. Brong. 6. macrocephalus, Conyb. Lias, Lyme Regis, De la B. , species not determined. Oxford Clay, Stenay, Bobl. ; Oxford Clay, Calvados, De la B. ; Lias, N. of Ireland, Bryce ; Lias, Whitby, Dunn. 1. Ichthyosaurus communis, De la B.; Lias, Lyme Regis, &c. Engl., Conyb., &c. ; Lias, Boll, Wurtemberg, Jag.; Banz; Fried- richsgemund, G. T. 2. platyodon, De la B. Lias, Lyme Regis, &c. Engl., Conyb., &c. ; Lias, Boll, Jag. * Crocodilus Bollensis, Jag. f Crocodilus priscus, Sommering. 568 Organic Remains of the Red Sandstone Group. 3. Ichthyosaurus tenuirostris, De la B. Lias, Lyme Regis, &c. Conyb., &c. ; Lias, Boll, Jag. 4. intermedius, Conyb. Lias, Lyme Regis, &c. Conyb., &c. ; Lias, Boll, Jag. , species not determined. Lias and Inferior Oolite, Normandy, De Cau. ; Lias, Yorks., Phil.; Oxford Clay, England, Conyb.; Oxford Clay, Normandy, De la B. ; Great Oolite, Reugny, Brong. ; Coral. Oolite, Yorks., Phil. ; Calc. Grit, Midi. Eng., Conyb. ; Kim. Clay, Oxford, Buc/cl. ; Kim. Clay, Weymouth, De la B.; Kim. Clay, Honfleur, Brong. Saurian bones occur in the Kelloway Rock and Bath Oolite, Yorks., Phil. ; in the Portland Stone, Buc/cl. 8f De la B. ; Lias, Cal- vados, Her. Tortoise. Stonesfield Slate, Buckl ; Lias? Engl. Conyb.; Solenhofen, Munst. MAMMALIA. 1. Didelphis Bucklandi, Broderip. Stonesfield Slate, Buckl. Organic Remains of the Red Sandstone Group. Variegated Mark. PLANTS. 1. Equisetum Meriani, Ad. Brong. Neue Welt, Basle, Ad. Brong. 2. arenaceum*, Bronn. Near Heidelherg, Bronn; Wurtemberg; France, G. T. 3. _ column are, Ad. Brong. Lorraine ; Alsace ; Wurtemberg, Rozet. 1. Pecopteris Meriani, Ad. Brong. Neue Welt, Ad. Brong. 1. Tseniopteris vittata, Ad. Brong. Neue Welt, Ad. Brong.; Wurtem- berg, G. T. 1. Filicites Stuttgardiensis, Ad. Brong. Wurtemberg, Rozet, G. T. 2. lanceolata, Ad. Brong. Wurtemberg, Rozet. 1. Marantoidea arenacea, Jcsg. Stuttgart, G. T. 1. Pterophyllum longifolium, Ad. Brong. Neue Welt, Ad. Brong. 2. Meriani, Ad. Brong. Neue Welt, Ad. Brong. 3. Jaegeri, Ad. Brong. Wurtemberg ; France, Rozet. RADIARIA. Ophiura, species undetermined. Vosges, Rozet. CONCHIFERA. 1. Plagiostoma lineatum, Bronn. Wurtemberg, G. T. 1. Cardium pectinatum, Von Alb. Wurtemberg, G. T. 1. Trigonia vulgaris, Schlot. Louisburg, G. T. 2. - curvirostris, Schlot. Louisburg ; Schwenningen, G. T. 3. sulcata, Goldf. Villengen, G. T. 1. Mya musculoides, Schlot. Sulz on the Neckar, G. T. 2. elongata, Schlot. Sulz on the Neckar, G. T. 1. Avicula socialist, Desh. Sulz, G. T. 2. subcostata, Goldf. Sulz, G. T. 3. lineata, Goldf. Sulz, G. T. 1. Posodonia Keuperina, Voltz. Swabia; Hall, G. T. 2. minuta, Von Alb. Rottweil, G. T. 1. Modiola minuta, Goldf. Rottweil, G. T. * Calamites arenaceus, Jaeg. f Mytilus socialis, Schlot. Organic Remains of the Red Sandstone Group. 569 1. Venericardia Goldfussii, Von Alb. Rottweil, G. T. 1. Lingula tenuissima, Bronn. Rottweil, G. T. 1. Saxicava Blainvillii, Hoen. Ballbron, Hcen. MOLLUSC A. 1. Buccinum turbilinum *, . Sulz on the Neckar, G. T. PISCES. Genera not determined. Seidmannsdorf ; Neuses; Seidingstadt, Co- burg, G. T. REPTILIA. 1. Phytosaurus cylindricodon, Jceg. Boll, Jeeg. 2. cubicodon, Jceg. Boll, Jceg. 1. Mastodonsaurus Jaegeri, Hott. Gaildorf, G. T. 1. Ichthyosaurus Lunevillensis, . Wurtemberg, G. T. Plesiosaurus, species not determined. Diirrheim, Hcen. Muschelkalk. PLANTS. 1. Neuropteris Gailliardoti, Ad. Brong. Luneville, Ad. Brong. 1. Mantellia cylindrica, Ad. Brong. Luneville, Ad. Brong. ZOOPHYTE. 1. Astrea pediculata, Desk. Locality not stated. RADIARIA. 1. Cidaris grandaevis, Goldf. Wurtemberg, G. T. 1. Ophiura prisca, Munst. Baireuth, Goldf. 2. loricata, Goldf. Schwenningen, Wurtemberg, G. T. 1. Asterias obtusa, Goldf. Marbach, Villengen, G. T. 1. Encrinites moniliformis, Mill. Gottingen; Wurtemberg; Poland; France, &c., Far. Auih. 1. Pentacrinites dubius, Goldf. Rudersdorf, G. T. ANNULATA. 1. Serpula valvata, Goldf. Baireuth, Goldf. 2. colubrina, Goldf. Baireuth, Goldf. CONCHIFERA. 1. Terebratula communis f, Bosc. Gottingen, Hcen. ; Wurtemberg; Lu- neville ; Toulon, AL Brong. ; Richen, Basle, G. T. 2. perovalis, Schlot. Jena, Hoen. 3. sufflata, Schlot. Jena, Hoen. 4. orbiculata, Schlot. Dornberg, Jena, Hcen. 1. Delthyris semicircularis, Goldf. Villengen, G. T. 1. Lingula tenuissima, Bronn. Rottweil, G. T. 1. Ostrea placunoides, Munst. Baireuth, G. T. 2. subanomia, Munst. Baireuth, G. T. 3. ' reniformis, Munst. Baireuth, G. T. 4. difformis, Schlot. Baireuth ; Wurtemberg, G. T. 5. - multicostata, Munst. Wurzberg, G. T. 6. - complicata, Goldf. Baireuth ; Villengen, G. T. 7. - decemcostata, Munst. Baireuth, G. T. 8. - spondyloides, Schlot. Quedlinberg, //o?w. / Gottingen ; Luneville; Toulon, AL Brong. ; Baireuth ; Brombach, G. T. * Helix turbilinum, Schlot. f T. vulgaris, and T. subrotunda, Schlot. 570 Organic Remains of the Red Sandstone Group. 9. Ostrea comta, Goldf. Rottweil, G. T. 1. Gryphaea prisca, Goldf. Villengen, G. T. 1. Pecten reticulatus, Schlot. Gottingen, Hoen. ; Gotha, G. T. 2. Albertii, Goldf. Villengen, Rudersdorf, G. T. 3. laevigatus, Goldf. Wurtemberg j Hagen ; Bromberg ; Baireuth ; Gotha, G. T. 4. discites, Schlot. Wurtemberg ; Richen, Basle ; Rudersdorf; Po- land, G. T. 1. Plagiostoma lineatum*, Bronn. Michelstadt, Hoen.; Gottingen, AL Brong.; Mosbach, Heidelberg, Bronn; Wurtemberg; Bai- reuth; Weimar, G. T. 2. >- striatum, Schlot. Germany ; France ; Poland, G. T. 3. rigidum, Schlot. Rauhthal, Jena, Hoen. ; Gottingen, AL Brong. 4. laevigatum, Schlot. Mosbach, Hoen. 5. punctatum, Schlot. Gottingen ; Gotha ; Toulon, AL Brong. ; Weimar; Baireuth, G. T. 1. Avicula socialis, Desk. Gotha; Sachsenburg, Schlot. ; Weimar, Hosn. ; Gottingen; Mont Meisner; Wurtemberg; Luneville, AL Brong. ; Ibbenbiihren ; Rudersdorf; Nischwitz ; Wehrau ; Kalinowitz, G. T.; Heidelberg, Bronn. 2. costata, Bronn. Wurtemberg ; Baireuth, G. T. 3. crispata, Goldf. Friedrichshall, G. T. 4. Bronnii, Von Alberti. Villingen, G. T. 1. Mytilus vetustusf, Goldf. Gottingen; LuneVille, AL Brong. ; Wur- temberg; Hagen; Baireuth, G. T. 1. Trigonia vulgaris, Schlot. Weimar, Hoen.; Gottingen, AL Brong. ; Wurtemberg ; Riedern, Waldshut ; Baireuth, G. T. 2. Pes-anseris, Schlot. Luneville ; Mosbach, Hoen. ; Gottingen, AL Brong. 3. curvirostris, Schlot. Wurtemberg, G. T. 4. cardissoides, Goldf. Wurtemberg, G. T. 5. laevigata, Goldf. Marbach, G. T. 6. Goldfussii, Von Alberti. Marbach, G. T. 1. Area insequivalvis, Goldf. Freudenstadt, Schwarzwald, G. T. 1. Cardium striatum, Schlot. Wurtemberg; Gottingen, AL Brong. 2. pectinatum, Von Alb. Wurtemberg, G. T. 1. Mya musculoides, Schlot. Weimar; Wurtemberg; Upper Silesia; Po- land, G. T. 2. elongata, Schlot. Wurtemberg, AL Brong. ; Seewangen, Wald- shut ; Upper Silesia ; Poland, G. T. 3. ventricosa, Schlot. Luneville, AL Brong. ; Wurtemberg, G. T. 4. mactroides, Schlot. Marbach ; Upper Silesia ; Poland, G. T. 5. rugosa, Von Alberti. Rottweil, G. T. 1. Venus nuda, Goldf. Marbach, G. T. 1. Mactra? trigona, Goldf. Marbach, G. T. 1. Cuccullaea minuta, Goldf. Villengen, G. T. MOLLUSCA. 1. CalyptraeaJ discoides, Schlot. Villengen, G. T. 1. Capulus mitratus, Goldf. Villengen, G. T. 1 . Dentalium torquatum, Schlot. Gottingen, AL Brong. 2. laeve, Schlot. Gottingen, AL Brong. ; Alpirsbach ; Baireuth, G. T. 1. Trochus Albertinus, Goldf. Rottweil, G. T. 1. Turritella obsoleta, . Gottingen, Hoen. ; Weimar, G. T. * Chama lineata, Schlot. f Mytilus eduliformis, Schlot. J Patellites. Buccinum obsoleium, Schlot. Organic Remains of the Red Sandstone Group. 571 2. Turritella deperdita, Goldf. Weimar, G. T. 3. - detrita, Goldf. Culmbach, G. T. 4. scalata*, . Wiirtemberg; Riidersdorf, G. T. 5. ? terebralis, Schlot. Weimar, Hcen. 1. Buccinum gregarium, Schlot. Riidersdorf, G. T. 2. turbilinum f , . Wurtemberg ; Seewangen, Waldshut ; Rii- dersdorf, G. T. 1. Strombus denticulatus, Schlot. Riidersdorf, G. T. 1. Natica Gailliardoti, Lefroy. Wurtemberg, G. T. 2. Pulla, Goldf. Rottweil, G. T. 1. Turbo dubius, Munst. Hassel, Bronn ; Seewangen; Riedern, G. T. 2. giganteus, Schlot. Seewangen, G. T. 1. Nautilus bidorsatus, Schlot. Weimar, Hcen. ; Wurtemberg, Al. Brong. ; Gbttingen ; Riidersdorf; Luneville, G. T. 2. nodosus, Munst. Germany, Munst. 1. Ammonites nodosus, Schlot. Weimar, Hcen.; Gottingen; Wurtem- berg; Toulon, Al. Brong.; Lorraine, Beaum. ; Tarnowitz, G. T. 2. bipartitus, Gailliardot. Luneville, Al. Brong. Rhyncolites. Jena; Gottingen; Wurtemberg; Luneville; Rehainvil- lers, G. T. CRUSTACEA. 1 . Palinurus Sueurii, Desm. Durrheim, Villingen, Hcen. ; Blittersdorf, Saarbriick, G. T. PISCES. Fish, and Fish Teeth. Baireuth ; Wurtemberg ; Rudersdorf, Munst., G. T. REPTILIA. Plesiosaurus, species not determined. Wurtemberg, Jceg. ; Baireuth ; Riidersdorf, G. T. 1. Ichthyosaurus Lunevillensis, . Luneville ; Wurtemberg, G. T. Great Saurian, genus not determined. Luneville, Al. Brong. Crocodilus, species not determined. Riidersdorf, G. T. Chelolia, species not determined. Luneville ; Bindlocherberg ; Leinec- kerberg, G. T. Red or Variegated Sandstone. PLANTS. 1. Equisetum columnare, Ad. Brong. Sulz-les-Bains, G. T. 1. Calamitesarenaceus,^c7..Z?row#. Wasselonne; Marmoutier (Bas-Rhin), Ad. Brong. 2. remotus, Ad. Brong. Wasselonne, Ad. Brong. 1. Anomopteris Mougeotii, Ad. Brong. Wasselonne; Sulz-les-Bains, Ad. Brong. 1. Neuropteris Voltzii, Ad. Brong. Sulz-les-Bains, Ad. Brong. 2. elegans, Ad. Brong. Sulz-les-Bains, Ad. Brong. 1. Sphenopteris Myriophyllum, Ad. Brong. Sulz-les-Bains, Ad. Brong. 2. palmetta, Ad. Brong. Sulz-les-Bains, Ad. Brong. 1. Filicites scolopendrioides, Ad. Brong. Sulz-les-Bains, Ad. Brong. 1. Voltzia brevifolia, Ad. Brong. Sulz-les-Bains, Ad. Brong. 2. elegans, Ad. Brong. Sulz-les-Bains, Ad. Brong. 3. rigida, Ad. Brong. Sulz-les-Bains, Ad. Brong. * Strombus scalatus, Schlot. f Helix turbilinus, Schlot. 572 Organic Remains of the Red Sandstone Group. 4. Voltzia acutifolia, Ad. Brong. Sulz-les-Bains, Ad. Brong. 5. heterophylla, Ad. Brong. Sulz-les-Bains, Ad. Brong. 1. Convallarites erecta, Ad. Brong. Sulz-les-Bains, Ad. Brong. 2. nutans, Ad. Brong. Sulz-les-Bains, Ad. Brong. 1. Paleoxyris regularis, Ad. Brong. Sulz-les-Bains, Ad. Brong. 1. Echinostachys oblongus, Ad. Brong. Sulz-les-Bains, Ad. Brong. 1 . jEthophyllum stipulate, Ad. Brong. Sulz-les-Bains, Ad. Brong. CONCHIFERA. 1. Plagiostoma lineatum, Schlot. Sulz-les-Bains. 2. striatum, Schlot. Sulz-les-Bains. 1. Avicula socialis, Desk. Sulz-les-Bains; Domptail, Foltz. 2. costata, Bronn. Sulz-les-Bains, G. T. 1. Mytilus vetustus, Goldf. Domptail; Sulz-les-Bains. 1. Trigonia vulgaris, Schlot. Domptail. 1. Mya musculoides, Schlot. Sulz-les-Bains. 2. elongata, Schlot. Sulz-les-Bains. MOLLUSCA. 1. Natica Gailliardoti, Lefroy. Domptail. 1. Turritella scalata, . Domptail ; Sulz-les-Bains. 2. Schoteri, . Sulz-les-Bains. 1. Buccinum antiquum, Goldf. Sulz-les-Bains, G. T. Zechsteln. PLANTS. 1. Fucoides Brardii*, Ad. Brong. Cop. Slate, Frankenberg, Ad. Brong. 2. selaginoides, Ad. Brong. Cop. Slate, Mansfeld, Ad. Brong. 3. lycopodioides, Ad. Brong. Cop. Slate, Mansfeld, Ad. Brong. 4. frumentarius, Ad. Brong. Cop. Slate, Mansfeld, Ad. Brong. 5. pectinatus, Ad. Brong. Cop. Slate, Mansfeld, Ad. Brong. 6. digitatus, Ad. Brong. Cop. Slate, Mansfeld, Ad. Brong. 1. Pecopteris arborescens, Ad. Brong. Mont Muse, Autun, G. T. 2. abbreviata, Ad. Brong. Mont Muse, Autun, G. T. 1. Lycopodites Hceninghausii, Ad. Brong. Eisleben, G. T. 1. Asterophyllites ? bulbosa, . Thuringia, G. T. ZOOPHYTA. 1. Gorgonia anceps, Goldf. Gliicksbrunn, Thiiringerwald, G. T. 2. antiqua, Goldf. Gluksbrunn, G. T. 3. infundibuliformis, Goldf. Gliicksbrunn, G. T. 1. Calamopora spongites, (var.) Goldf. Gliicksbrunn, G. T. 1. Retepora flustracea, Phil. Shelly Mag. Limest., Durham, Sedg. 2. virgulacea, Phil. Shelly Mag. Limest., Durham, Sedg. Polypifera, genera not determined. Durham ; Northumberland, Sedg. RADIARIA. 1. Encrinus ramosus, Schlot. Gluksbrunn, Al. Brong. 1. Cyathocrinites planus, Mil. Mag. Limest., Durham; Northumberland, Sedg. Crinoidea, genera not determined. Durham ; Northumberland, Sedg. CONCHIFERA. 1 . Spirifer f trigonalis, Sow. Ropsen, Gera, Hoen. 2. undulatus, Sow. Midderidge ; Humbleton, Sedg. * Cupressus Ulmanni, Bronn. f Delthyris. Organic Remains of the Red Sandstone Group. 573 3. Spirifer multiplicatus, Sow. Humbleton, Sedg. 4. minutus, Sow. Humbleton, Sedg. 1. Terebratula cristata, Schlot. Ropsen, Hcen. 2. elongata, Schlot. Schmerbach, Al. Brong. 3. complanata, Schlot. Gera, G. T. 4. intermedia, Schlot. Ropsen, Hcen. 5. inflata, Schlot. Ropsen, Hcen. ; Schmerbach, Al. Brong. 6. lacunosa, Schlot. Cop. Slate, Schmerbach ; Zechstein, Ropsen, Hcen. 7. paradoxa, Schlot. Schmerbach, AL Brong. 8. pelargonata, Schlot. Schmerbach, Al. Brong. 9. pygmsea, Schlot. Leimstein near Schmalkalden, AL Brong. , species not determined. Durham, Sedg. 1. Producta* aculeata, Al. Brong. Bodingen, Hcen.; Gera; Thalitter; Goddelsheim, Logan on the Queiss, G. T. ; Durham ; Nor- thumberland, Sedg. 2. rugosa, Schlot. Ropsen, Gera, Hcen. 3. speluncaria, . Ropsen, Hcen. ; Gliickbrunn, Al. Brong. 4. antiquata, Sow. Midderidge, Sedg. 5. calva, Sow. Humbleton ; Midderidge, &c., Sedg. 6. spinosa, Sow. Humbleton, &c., Sedg. ? 7. - longispina, Sow. Cop. Slate, Schmerbach, Hcen. 1. Orbicula speluncaria, Schlot. Gliicksbrunn, G. T. 1. Axinus obscurus, Sow. Durham, Sedg. Ostrea, species not determined. Northumberland, Sedg. Pecten, species not determined. Humbleton, Sedg. Plagiostoma? species not determined. Humbleton, Sedg. 1. Avicula gryphaeoides, Sow. Humbleton (abundant), Sedg. 1. Mytilus keratophagus, Schlot. Gliicksbrunn, G. T. 2. striatus, Schlot. Gliicksbrunn, G. T. 3. squamosus, Sow. Ferrybridge, Sedg. ; Hasel near Goldberg, G. T. 1. Modiola acuminata, Sow. Black Rocks, Durham, Sedg. 1. Area tumida, Sow. Humbleton, Durham, Sedg. 1. Cucullaea sulcata, Sow. Humbleton, Sedg. Astarte? , . Whitley, Northumberland, Sedg. Venus? , . Humbleton, Sedg. MOLLUSCA. X ur bo ? , . Mag. Limest., Marr ; Hickleton, Sedg. ' Pleurotomaria ? , . Humbleton, Sedg. Melania? , . Hawthorn Hive, Phil. Ammonites, species not determined. Humbleton, Sedg. PISCES. 1 . Pakeothrissum macrocephalum, Blain. Cop. Slate, Mansfeld, Al. Brong. ; Marl Slate, Midderidge ; East Thickley, Sedg. 2. magnum, Blain. Mansfeld, AL Brong. ; Midderidge ; East Thickley, Sedg. 3. incequilobum, Blain. Cop. Slate, Rothenburg, G. T.; Bit. Slate, Mont Muse, Autun, Al. Brong. 4. macropterum, Bronn. Borschweiler, Thuringia, Hcen. 5. parvum, Blain. Bit. Slate, Autun, Al. Brong. 6. blennioides, Hott. Mansfeld, G. T. 7. - elegans, . Marl Slate, Midderidge ; East Thickley, Sedg. 8. Freieslebense, Blain. Mansfeld; Hessia, G. T. * Leptaena. 574? Organic Remains of the Coal Measures. Palaeothrissum, species not determined. Marl Slate, East Thickley, Sedg. ; Mag. Limest., Pallion, Winch. 1. Stromateus major, Blain. Cop. Slate, Hessia, G. T. 2. gibbosus, Blain. Cop. Slate, North Germany, G. T. 1. Clupea Lametherii, Blain. Cop. Slate, Eisleben, G. T. REPTILE. 1. Monitor antiquus, Cur. Cop. Slate, Mansfeld ; Rothenburg on the Saale ; Gliicksbrunn ; Memmingen, &c. Al. Brong. Organic Remains of the Coal Measures. PLANTS. [The following list of Plants, discovered fossil in the coal measures, is compiled from the labours of Adolphe Brongniart, Sternberg, Lindley, Hutton, Schlotheim, and other authorities. To have abridged it would have deprived the student not only of a valuable catalogue of localities, but also of an idea of the situations where plants of a similar general character probably existed. The names of the plants, when not otherwise noticed, are those assigned to them by M. Adolphe Brongniart.] VASCULARES. Subclass 1. DICOTYLEDONS. Euphorbiacea. STIGMARIA reticulata, England; S. Weltheimiana, Madgeburg; S. inter- media, St. Georges-Chatellaison ; Montrelais ; Wilkesbarre (N. America) ; S.faoides, Bristol; Dudley; Leeds; Newcastle; St. Georges-Chatellaison ; Montrelais ; St. Etienne ; Lie"ge ; Charleroi ; Valenciennes ; Muhlheim, near Dusseldorf; Bavaria; Silesia; S. tuberculosa, Montrelais; Wilkesbarre; S. rigida, Anzin, near Valenciennes ; S. minima, Anglesea, Charleroi ; S. Mo- sana (Sauv.), Liittich; S. gigantea (Sauv.), Liittich. Conifer CB. PINITES Brandlingi (L. & H.), Newcastle; P. Withami (L. & H.), Craig- leith, Edinburgh ; P. meduttaris (L. & H.), Craigleith, Edinburgh. PEUCE Withami (L. & H.), Durham. Doubtful Coniferee. SPHENOPHYLLUM Schlotheimii, Waldenburg, Silesia ; Somerset; Sph. emar- ginatum, Env. of Bath ; Wilkesbarre ; Sph. truncatum, Somerset ; Sph* den- tatum, Newcastle; Anzin; Geislautern; Sph. quadrifidum, Terrasson ; Sph. dissectum, Montrelais ; Sph. pusillum (Saur.), Liittich ; Sph. quadriphyllum (Sauv.), Liittich ; Sph. multijidum (Sauv.), Liittich ; Sph. erosum (L. & H.), Newcastle. Dicotyledonous Plants of Doubtful Affinity. ANNULARIA minuta, Terrasson ; A. brevifolia, Alais ; Geislautern ; A. fertilis, Env. of Bath ; St. Etienne ; Wilkesbarre ; A. floribunda, Saar- bruck (Sternb.) ; A. longifolia, Env. of Bath ; Geislautern; Silesia; Alais; Wilkesbarre; (Far.) Charleroi; Terrasson; A.spinulosa, Saxony (Sternb.); A. radiata, Saarbruck. ASTEROPHYLLITES equisetiformis, Mannebach ; Saxony; Rhode Island; As. rigida, Alais ; Valenciennes ; Charleroi ; Bohemia ; As. hippuroides, Alais ; As. longifolia, Eschweiler (Sternb.) ; Newcastle ; As. tenuifolia, Newcastle; Silesia; As. delicatula, Charleroi; Anzin; As. Brardii, Ter- rason ; As. diffnsa, Radnitz ; As. elegans (Sauv.), Belgium ; As. tuber- Organic Remains of the Coal Measures. 575 culata (L. & H.), Newcastle; As. grandis (L. & H.), Newcastle; As.ga- lioides (L. & H.), Newcastle. BECHERA grandis (Sternb.), Newcastle. Subclass 2. MONOCOTYLEDONS. Palmes. FLABELLARIA ? borassifolia, Swina. N Barometer on the mountain = 675 = 5206-1 / 975-9 Diff. of attached thermometers . . . = 5 = 7-4 Table C. Apparent height 968-5 Double .the sum of the detached thermometers mul- ) *,- - tiplied by the thousandth part of 968-5 . . . . j ' 1044- Correction for latitude 3-1 Table D. Height of the mountain 1047'] Table E. 612 Appendix. When the height of the barometer, graduated according to English inches and their parts, does not precisely correspond with a certain number of millimetres, and when great accuracy is required, it will be obvious, that instead of taking the next nearest number to it in the tables, as might otherwise be done, it will be necessary to calculate the difference. TABLE A. Inches. Mil. Inches. Mil Inches. Mil. Inches. Mil. Inches. Mil. Inches. Mil. 14-56 370 16-142 410 17-717 450 19-292 490 20-866 530 22-441 570 606 371 181 411 756 451 331 491 901 531 481 571 646 372 221 412 795 452 370 492 945 532 520 572 685 373 260 413 835 453 410 493 984 533 559 573 725 374 299 414 874 454 449 494 21-025 534 599 574 764 375 339 415 914 455 488 495 063 535 638 575 803 376 378 416 953 456 528 496 102 536 678 576 843 377 417 417 992 457 567 497 142 537 717 577 882 378 457 418 18-032 458 607 498 181 538 756 578 921 379 496 419 071 459 646 499 220 539 796 579 961 380 536 420 110 460 685 500 260 540 835 580 15-000 381 575 421 150 461 725 501 300 541 874 581 040 382 614 422 189 462 764 502 339 542 914 582 079 383 654 423 229 463 803 503 378 543 953 583 118 384 693 424 268 464 843 504 418 544 992 584 156 385 733 425 307 465 882 505 457 545 23-032 585 197 386 772 426 347 466 922 506 496 546 071 586 236 387 811 427 386 467 961 507 536 547 111 587 276 388 851 428 425 468 20-000 508 575 548 150 588 315 389 890 429 465 469 040 509 615 549 189 589 355 390 929 430 504 470 079 510 654 550 229 590 394 391 969 431 543 471 118 511 693 551 268 591 433 392 17-008 432 583472 158 512 733 552 307 592 473 393 047 433 622 473 197 513 772 553 347 593 512 394 087 434 662 474 237 514 812 554 386 594 551 395 126 435 701 475 276 515 851 555 426 595 591 396 166 436 740 476 315 516 890 556 465 596 630 397 205 437 780477 355 517 929 557 5.04 597 670 398 244 438 8191478 394 518 969 558 544 598 709 399 284 439 859 479 433 519 22-008 559 583 599 748 400 323 440 898 480 473 520 048 560 622 600 788 401 362 441 937 481 512 521 087 561 662 601 827 402 402 442 977 482 551 522 126 562 701 602 866 403 441 443 19-016 483 590 523 166 563 741 603 906 404 481 444 055 484 630 524 205 564 780 604 945 405 -520 445 095 485 670 525 244 565 819 605 985 406 559 446 134 486 709 526 284 566 859 606 16-024 407 599 447 174 487 748 527 323 567 898 607 063 408 638 448 213488 788 528 363 568 937 608 102 409 677 449 252489 827 529 402 569 977 609 1 Appendix. 613 TABLE A. (continued.} Inches. Mil. Inches. Mil. Inches. Mil. I nches. Mil. Inches. Mil. Inches. Mil. 24-016 610 25-197 640 26-378 670 27-560700 28-741 730 29-922 760 055 611 237 641 418 671 600 ! 701 780 731 961761 095 612 276642 -457 672 639702 819 732 30-000762 134 613 315 643 496 673 678703 859 733 040763 174 614 355 644 536 674 718704 898 734 079764 213 615 394 645 575 675 757705 937 735 119765 252 616 433 646 615 676 796706 977 736; 158766 292 617 473 647 654 677 836707 29-016 737 197(767 331 618 512 648 693 678 -875708 056 738 237768 370 619 551 649 733 679 915709 095 739 276 769 410 620 590 650 772 680 954710 134 740 316770 449 621 630 651 811 681 993711 174 741 355771 489 622 670 652 851 682 28-032712 213 742 394772 528 623 709 653 891 683 071713 252743 433773 567 624 749 654 931 684 111714 291 744 473774 607 625 788 655 970 685 150715 331 745 512775 646 626 827 656 27-010 686 189716 371 746 552 776 685 627 866 657 049 687 229717 410 747 591 777 725 628 906 658 089 688 268 718 449 748 630 778 764 629 945 659 128 689 308 719 489 749 670 779 804 630 985 660 167 690 347 720 528 750 709 780 843 631 26-024 661 206 691 386 721 567 751 749 781 882 632 063 662 246 692 426 722 607 752 788 782 922 633 103 663 285 693 465 723 646 753 827 783 961 634 142664 324 694 504 724 686 754 867 784 25-000 635 182665 363 695 544 725 725 755 906 785 040 636 221 666 403 696 583 726 764 756 945 786 079 637 260 667 442 697 622 727 804 757 985 787 119 638 300 668 482 698 662 728 843 758 31-024 788 158 639 339 669 521 699 701 729 882 759 064 789 TABLE B. Mil. Metr. I] Mil. Metr. Mil. Metr. Mil. Metr. Mil.! Metr. Mil. Metr. 370 418-5 ! 384 714-3 398 999-5 412 1274-8 4261540-8 440 1798-4 371 440-0 385 735-0 399 1019-5 413 1294-1 4271559-5 441 1816-5 372 461-5 386 755-6 400 1039-4 414 1313-3 4281578-2 442 1834-5 373 482-9 387, 776-2 401 1059-3 415 1332-5 4291596-8 443 1852-5 374 504-2388 796-8 402 1079-1 416 1351-7 4301615-3 444 1870-4 375 525-4 389 817-3 403 1098-9 417 1370-8 431J1633-8 445 1888-3 376' 546-6 390 837-8 404 1118-6 418 1389-9 4321652-2 446 1906-2 377 567-8 391 858-2 405 1138-3 4191408-9 4331670-6 447 1924-0 378 588-9 392 878-5 406 1157-9 420 1427-9 4341689-0 448 1941-8 379 609-9 380 630-9 393 394 898-8 919-0 407 408 1177-5 1197-1 421 1446-8 422:1465-7 4351707-3 4361725-6 449 450 1959-6 1977-3 381 651-8 395 939-2 409 1216-6 423 1484-6 4371743-8 451 1994-9 382 ! 672-7;396 959-3 410 1236-0 424 1503-4 438'l762-l J 452 2012-6 383 693-5 397 1 979-4 411 1255-4 425 11522-2 4391780-3 1 453 2030-2 614- Appendix. TABLE B. (continued.) Mil. Metr. Mil.l Metr. Mil Metr. Mil Metr. Mil. Metr. Mil. Metr. 454 2047-8 5022848-1 550 3575-3 598 4241-6 646 4856-4 694 5427-2 455 2065-3 503 ! 2864-0 551 3589-8 599 4254-9 647 4868-7 695 5438-7 456 457 2082-8 2100-2 5042879-8 505J2895-6 552 553 3604-2 3618-6 600 601 4268-2 4281-4 648 649 4881-0 4893-3 6965450-1 6975461-5 458 2117-6 506 2911-3 554 3633-0 602 4294-7 650 4905-6 6985472-9 459 2135-0 507 2927-0 555 3647-4 603 4307-9 651 4917-8 6995484-3 460 2152-3 508 2942-7 556 3661-7 604 4321-1 652 4930-0 700 5495-7 461 2169-6 509 2958-4 557 3676-0 605 4334-3 653 4942-2 701 5507-1 462 2186-9 510 2974-0 558 3690-3 606 4347-4 654 4954-4 702 5518-4 463 2204-1 511 2989-6 559 3704-6 607 4360-5 655 4966-6 703 5529-8 464 2221-3 512 3005-2 560 3718-8 608 4373-7 656 4978-7 704 5541-1 465 2238-4 5133020-7 561 3733-0 609 4386-7 657 4990-9 705 5552-4 466 2255-5 5143036-2 562 3747-2 610 4399-8 658 5003-0 706 5563-7 467 2272-6 5153051-7 563 3761-3 611 4412-8 659 5015-1 707 5575-0 468 2280-6 5163067-2 564 3775-4 612 4425-9 660 5027*2 708 5586-2 469 2306-6 5173082-6 565 3789-5 613 4438-9 661 5039-2 709 5597-5 470 2323-6 518 ( 3097'9 566 3803-6 614 4451-9 662 5051-2 710 5608-7 471 2340-5 5193113-3 567 3817"7 615 4464-8 6635063-3 711 5619-9 472 2357-4 5203128-6 568 3831-7 616 4477-7 6645075-3 712 5631-lj 473 2374-2 521 3143-9 569 3845-7 617 4490-7 665 5087-2 713 5642-2 474 2391-1 5223159-2 570 3859-7 618 4503-6 666 5099-2 714 5653-4 475 2407-9 5233174-4 571 3873-7 619 4516-4 667 5111-2 715 5664-6 476 2424-6 5243189-7 572 3887-6 620 4529-3 668 5123-1 716 5675-7 477 2441-3 5253204-9 573 3901-5 621 4542-1 669 5135-0 717 5686-8 478 2458-0 526 3220-0 574 3915-4 622 4554-9 670 5146-9 718 5697-9: 4792474-6 527 3235-1 575 3929-3 623 4567-7 671 5158-8 719 5709-0 480 2491-3 528 3250-2 576 3943-1 624 4580-5 672 5170-6 720 5720-1 481 2507-9 529 3265-3 577 3956-9 625 4593-2 673 5182-5 721 5731-1 482 2524-3 530 3280-3 5783970-7 626 4606-0 674 5194-3 722 5742-1 483 2540-8 531 3295-3 579 3984-5 627 4618-7 675 5206-1 723 5753-1 484 2557-3 532 3310-3 580 3998-2 628 4631-4 676 5217-9 724 5764-2 485 2573-7 533 3325-3 581 4011-9 629 4644-0 677 5229-7 725 5775-1 486 2590-2 534 3340-2 582 4025-6 630 4656-7 678 5241-4 726 5786-1 487 2506-6 535 3355-1 583 4039-3 631 4669-3 679 5253-2 727 5797-1 488 2622-9 536 3370-0 584 4052-9 632 4682-0 680 5264-9 728 5808-0! 489 2639-2 537 3384-8 585 4066-6 633 4694-5 681 5276-6 729 5819*0 490 2655-4 538 3399-6 586 4080-2 634 4707-1 682 5288-3 730 5829-9 491 2671-6 539 3414-4 587 4093-8 635 4719-7 683 5300-0 731 5840-8^ 492 2687-9 540 3429-2 588 4107-3 636 4732-2 684 5311-6 732 5851-7! 493 2704-1 541 3443-9 589 4120-8 637 4744-7 685 5323-2 733 5862-5 494 2720-2 542 3458-6 590 4134-3 638 4757-2 686 5334-8 734 5873-4 1 495 2736-3 543 3473-3 591 4147-8 639 4769-7 687 5346-4 735 5884-2 496 2752-3 544 3487*9 592 4161-3 640 4782-1 688 5358-0 736 5895-1 497 2768-3 545 3502-5 593 4174-7 641 4794-6 689 5369-6 737 5905-9, 498 2784-4 546 3517-2 594 4188-1 642 4807-0 690 5381-1 738 5916-7 499 2800-4 547 3531-8 595 4201-5 643 4819-4 691 5392-7 739 5927-5 500 2816-3 548 3546-3 596 4214-9 644 4831-7 692 5404-2 740 ->938-2 501 2832-2 549 3560-8 597 4228-2 645 4844-1 693 5415-7 741 5949-0 Appendix. 615 TABLE B. (continued.) Mil. Metr. Mil. Metr. Mil. Metr. Mil. Metr. Mil. Metr. Mil. Metr. 742 5959-7 751 6055-7 759 6140-1 767 6223-6 775 6306-2 783 6388-0 7435970-4 7526066-3 760 6150-6 768 6234-0 776 6316-5 784 6398-2 7445981*8 753 6076-9 761 6161-1 769 6244-4 777 6326-7 785 6408-3 745 5991-9 754 6087-5 762 6171*5 770 6254-7 778 6337-0 786 6418-5 746 6002-5 755 6098-0 763 6182-0 771 6265-0 779 6347-2 787 6428-6 747 6013-2 756 6108-6 764 6192-4 772 6275-4 780 6357-4 788 6438-7 748 6023-8 757 6119-1 765 6202-8 773 6285-7 781 6367-6 789 6448-8 749 6034-4 758 6129-6 766 6213-2 774 6296-0 782 6377-8 790 6458-9 750 6045-1 TABLE C. Deg. Metre. Deg. Metre. Deg. Metre. Deg. Metre. Deg. Metre. 0-2 0-3 4-2 6-2 8-2 12-1 12-2 17-9 36-2 23-8 0-4 0-6 4-4 6-5 8-4 12-4 12-4 18-2 16-4 24-1 0-6 0-9 4-6 6-8 8-6 12-6 12-6 18-5 16-6 24-4 0-8 1-2 4-8 7-1 8-8 12-9 12-8 18-8 16-8 24-7 1-0 1-5 5-0 7-4 9-0 13-2 13-0 19-1 17-0 25-0 1-2 1-8 5-2 7-6 9*2 13-5 13-2 19-4 17-2 25-3 1-4 2-1 5-4 7-9 9-4 13-8 13-4 19-7 17-4 25-6 1-6 2-3 i 5-6 8-2 9-6 14-1 13-6 20-0 17-6 25-9 1-8 2-6 ! 5-8 8-5 9-8 14-4 13-8 20-3 17-8 26-2 2-0 2-9 ; 6-0 8-8 10-0 14-7 14-0 20-6 18-0 26-5 2-2 3-2 6-2 9-1 10-2 15-0 14-2 20-9 18-2 26-8 2-4 3-5 6-4 9-4 10-4 15-3 14-4 21-2 18-4 27-1 2-6 3-8 6-6 9-7 10-6 15-6 14-6 21-5 18-6 27-4 2-8 4-1 6-8 10-0 10-8 15-9 14-8 21-8 18-8 27-7 3-0 4-4 7-0 10-3 11-0 16-2 15-0 22-1 19-0 28-0 3-2 4-7 7-2 10-6 11-2 16-5 15-2 22-4 19-2 28-2 3-4 5-0 7-4 10-9 11-4 16-8 15-4 22-7 19-4 28-5 3-6 5-3 7-6 11-2 ill-6 17-1 15-6 22-9 19-6 28-8 3-8 5-6 7-8 11-5 11-8 17-4 15-8 23-2 !| 19-8 29-1 4-0 5-9 8-0 11-8 12-0 17-6 16-0 23-5 1 616 Appendix. TABLE D. Appr. Ht. 5 10 15 20 25 Appr. Ht. 30 35 40 45 50 55 m. m. m. in. m. m. m. m. m. m. m. m. 200 1-2 1-2 1-2 1-0 1-0 1-0 200 0-8 0-8 0-6 0-6 0-6 0-4 400 2-4 2-4 2-4 2'2 2-0 2-0 400 1-8 1-7 1-4 1-2 1-0 0-8 600 3-4 3-4 3-4 3-2 3-0 2-8 600 2-6 2-4 2-0 1-8 1-6 1-2 800 4-5 4-5 4-5 4-3 4-1 3-8 800 3-5 3-1 2-8 2-4 2-0 1-7 1000 5-7 5-7 5-7 5-3 5-1 4-8 1000 4-3 3-8 3-4 3-1 2-6 2-2 1200 7-0 7-0 6-8 6-4 6-0 5-8 1200 5-1 4-6 4-2 3-6 3-1 2-6 1400 8-2 8-2 8-0 7-6 7-1 6-7 1400 6-1 5-4 4-8 4'2 3-6 3-0 1600 9-2 9-2 9-0 8-8 8-2 7-6 1600 7-0 6-2 5-6 4-8 4-1 3-4 1800 10-4 10-4 10-2 9-8 9*4 8-6 1800 8-0 7-0 6-3 5-4 4-6 3-8 2000 11-6 11-5 11-3 11-0 10-4 9-6 2000 8-8 7-8 7-0 6-0 5-1 4*2 2200 12-8 12-6 12-6 12-1 11-4 10-6 2200 9-7 8-6 7-6 6-6 5-6 4-6 2400 14-0 14-0 13*8 13-3 12-5 11-6 2400 10-6 9-4 8-4 7-2 6-1 5-1 2600 15-2 15-2 15-0 14-4 13-6 12-6 2600 11-6 10-5 9'2 8-0 6-8 5-6 2800 16-6 16-5 16-4 15-6 14-8 13-6 2800 12-6 11-4 10-0 8-8 7-4 6-2 3000 17-9 17-7 17-6 16-8 15-8 14-6 3000 13-6 12-2 10-8 9'4 8-0 6-6 3200 19-1 18-9 18-7 18-0 17-0 15-7 3200 14-6 13-1 11-5 10-1 8-6 7-0 3400 20-5 20-3 20-1 19'3 18-4 16-9 3400 15-7 14-1 12-4 10-9 9-2 7*7 3600 21-8 21-7 21-4 20-4 19'6 18-0 3600 16-7 15-0 13-4 11-6 9'8 8-2 3800 23-1 22-9 22-6 21-6 20-6 19-1 3800 17-7 15-9 14-3 12-4 10-5 8-7 4000 24-6 24-4 24-0 22-9 21-9 20-3 4000 18-7 17-0 15-1 13-1 11-2 9.4 4200 25-9 25-7 25-3 24-3 23-0 21-6 l 4200 19-9 18-0 15-9 14-0 12-0 10-1 4400 27-5 27*3 26-8 25-8 24-3 23-0 4400 21-1 19-1 16-9 15-0 12-9 10-8 4600 28-9 28-7 28*2 27-1 25-6 24-3 ! 4600 22-3 20-3 18-0 15-9 13-6 11-5 4800 30-4 30-2 29-6 28-4 27-0 25-5 4800 23-4 21-3 19-0 16-7 14-3 12-1 5000 31-8 31-6 30-9 29-8 28-4 26-7 5000 24-6 22-3 19'9 17-4 15-0 12-7 5200 33-0 ;32-8 32-1 31-0 29-7 28-0 j 5200 25-7 23-3 20-8 18-2 15-7 13-3 5400 34-3 34-1 33-5 32-4 30-8 29-2 5400 26-7 24-3 21-7 19-1 16-4 13-9 5600 35-7 35-5 34-8 33-7 32-1 30*2 5600 27'8 25-3 22-6 19-9 17-2 14-5 5800 37-1 36-9 36-1 35-0 33-2 31-3! 5800 28-9 26-3 23-6 20-7 17-8 15-1 6000 38-5 38-3 37-5 36-3 34-3 32-3 6000 30-0 27-3 24-6 21-5 18-5 15-7 Appendix. TABLE E. Reduction of Metres into English Feet and Inches. 617 Metr. Feet. Inches. Metr. Feet. Inches. Metr. Feet. Inches. 1 3 3-370 50 164 0-514 900 2952 9-261 2 6 6-740 60 196 10-217 1000 3280 10-290 3 9 10-111 70 229 7-920 2000 6561 8-58 4 13 1-481 80 262 5-623 3000 9842 6-87 5 16 4-851 90 295 3-326 4000 13123 5-16 6 19 8-222 100 328 1-029 5000 16404 3-45 7 22 11-592 200 656 2-058 6000 19685 1-74 8 26 2-963 300 984 3-087 7000 22966 0-03 9 29 6-333 400 1312 4-116 8000 26246 10-32 10 32 9702 500 1640 5-145 9000 29527 8-61 20 65 7-405 600 1968 6-174 10000 32808 6-90 30 98 5-108 700 2296 7-203 40 131 2-811 800 2624 8-232 ! Reduction of Decimetres, Centimetres, and Millimetres, to English Inches. Dec. Inches. || Cent. Inches. Milli. Inches. 1 3-937 1 0-393 i 1 i 0-039 2 7-874 | 2 0-787 2 0-078 3 11-811 3 1-181 3 0-118 4 15-748 4 1-574 4 0-157 5 19-685 5 1-968 5 0-196 6 23-622 6 2-362 6 0-236 7 27-559 7 2-755 7 0-275 8 31-496 8 3-149 8 0-314 9 35-433 9 3-543 9 0-354 10 39-370 10 3-937 ! 10 0-393 F. Comparison of English and French Measures. (From Daily's Astronomical Tables.) Fr. Metres. French Toise = 1-949036 Foot = 0-324839 Inch = 0-027070 English Foot = 0-304794 Inch . . . = 0-025399 Eng. Inches. French Metre =39-37079 Toise =76-739400 Foot = 12-789900 Inch = 1-065825 -Line . . . .= 0-088819 Constant logarithms (always additive) for converting French Toises into Met.0-2898200 French Ft. intoEng.Ft.0-0276860 Feet into Metr. 9-5116687 T. into Eng. Ft. 0-8058372 Met. into Eng. Ft.0-5 159929 Millimet. into Eng. In. 8-5951741 INDEX. ABEL, Dr. Clarke, on the bank thrown up at the Cape of Good Hope, 86. Aixin Provence, supracretaceous rocks of, 247. Alps, supracretaceous rocks of, 231, 237 ; cretaceous rocks of, 290, 292 ; oolitic rocks of, 329. Animal life, early, on the surface of the globe, remarks on the, 429. Arago, M., on the temperature of the earth's surface, 7. Arctic circle, numerous remains of ele- phants and rhinoceroses within the, 199. Argillaceous, or clay slate, 433. Arkose, 324. Artesian wells, temperature of, 12. Atchafalaya, raft of, 71 ; section of the alluvial banks of, ib. Attraction, local, of the magnetic needle, renders the determination of currents doubtful, 109. Auvergne, erosive action of rivers in, ,57 ; extinct volcanos and supracre- taceous rocks of, 271. Ava, organic remains found in the kingdom of, 268. Baculite limestone, of Normandy, 296. Bagshot sands, 263. Baku, inflammable gaseous exhala- tions of, 154; naphtha and petro- leum springs of, 163. Banda, Isle of, gradual rise of a pro- montory at, 127. Barometer, tables for calculating heights by, 610. Bars of rivers, 89. Basalt, 137 ; chemical composition of, 452 ; columnar structure of, 470. Basins, rock, 46. Basterot, M. de, on the supracretaceous rocks of Bordeaux, 250. Beach, raised, of Plymouth, 172; in Cornwall, 174; of Jura (Hebrides), 175; of Uddevalla, Sweden, 176; of St. Hospice, Nice, ib. ; in South America, 178. Beaches, shingle, 79 ; travel in the direction of the prevalent winds, ib. ; bar up the valleys, forming lakes, 81 ; sandy, 84. Beaufort, Capt., on the indurated beach, Selinty, Karamania, note, 85 ; on undercurrents in the Mediterranean, 1 14; on the gaseous exhalation of the Yanar, 152. Beaumont, M. Elie de, on the erratic blocks of the Alps, 194; on the gra- vels of the Lyonais, Dauphine, and Provence, 222 ; on the supracreta- ceous rocks of the Pertuis de Mira- beau, 246 ; on the oolitic rocks of Burgundy, 317; on the lias of the Alps, 327 ; on the Ores de Vosges, 356 ; on the elevation of mountains, 483. Beechey, Capt., on the temperature of the sea, 24. Belemnites, association of, with coal measure plants in the Alps, 328. Bertrand de Doue, M., on the volcanic rocks and ossiferous beds of the Ve- lay, 272. Bertrand-Geslin,M., on the Val d'Arno, 245. Beudant, M., on the volcanic rocks of Hungary, 278. Bigsby, Dr., on the erratic blocks of North America, 192. Black Head (Babbacombe Bay, Devon), relations of limestone and trappean rocks at, 464. Boase, Dr., on the submarine forest, Mount's Bay, Cornwall, 170. Boblaye, M., on the passage of the Meuse through the Ardennes, 58 ; on the shore-lines upon the lime- stones of Greece, 178; on the oolitic rocks of the North of France, 316. 620 Index. Bore, of the Ganges, 98 ; of the Ma- ranon, ib. ; of the Arouary, ib. Boue, Dr., on the salt of Wieliczka, 270; on the rocks of Gosau, 280; on the carboniferous rocks of Scot- land, 389. Bovey coal deposit, 223. Breakers, action of on coasts, 77, 84 ; large blocks moved by, during gales, 82. Breakwater, Plymouth, great blocks of rock forced over during a gale, 82. Brongniart, M. Adolphe, on the plants of the coal measures, 411. Brongniart, M. Alex., his classification of rocks, 39 ; on the raised mass of shells at Uddevalla, 176; on the er- ratic blocks of Sweden, 189 ; labours of, round Paris, 251 ; on the Vicen- tine, 276; on the Diablerets, 281; on the cretaceous rocks of the Isle d'Aix, 311 ; on the rocks of the Cotentin, 423. Brown coal, of Germany, 248. Buch, M. von, on the modern sand- stone of the Great Canary, 86 ; his theory of craters of elevation, 124 ; on the Caldera, Isle of Pal ma, 128; on the coral rag of Germany, 322 ; on the dolomite of the Lake of Lu- gano, 475. Buckland, Dr., on valleys of eleva- tion, 31 ; on the erratic blocks and gravel of Durham, 186 ; on the re- mains of elephants &c., within the Arctic Circle, 200 ; on ossiferous caverns, 201 ; on Kirkdale cavern, 204 ; on the German caves, 206 ; on the plastic clay of Woolwich, 259 ; on the magnesian conglome- rate of Somerset, 367. Buddie, Mr,, on the coal measures near Newcastle, 376 ; on the carbu- retted hydrogen of coal mines, 377. Cachin, M., on the action of the waves on the Digue, Cherbourg, 89. Calcaire grossier, of Paris, 252. Cantal, fresh-water rocks of the, 271. Carboniferous limestone, 381 ; sum- mary of the organic remains in the, ib. Carboniferous rocks, of Germany, 383; of Southern England, 386 ; of Cen- tral England, ib. ; of Northern En- gland, 387; of Scotland, 388; of Ireland, 393; of Northern France and Belgium, 394; of Saarbriick, 397 ; of Poland, ib. ; of Russia, ib. of Central France, 398; of the United States, ib. ; of India, 399 ; general remarks on the, 401. Carne, Mr., on the metalliferous veins of Cornwall, 492. Caspian, depression of land in the re- gion of the, 3 ; salts in the water of the, 4. Caverns, ossiferous, 201 ; onthemanner in which bones may occur in, ib. ; human remains in, 203 ; of Kirkdale, 204 ; of Germany, 205 ; of Kent's Hole, 206 ; of Echenoz, ib. ; of Fou- vent, 208 ; of Banwell, ib. ; of Choc- kier, 212. Cavities, funnel-shaped, produced du- ring earthquakes, 148. Chalk, converted into granular marble, 279 ; probable passage of, into the supracretaceous rocks, ib. analysis of that of Meudon, 283 ; of Wein- bohla, cut by granite, 294. Chesil Bank, Portland, 80 ; hypothesis respecting the, ib. China, natural exhalations of inflam- mable gas in, 152. Chlorite slate, 433 ; calculation respect- ing the chemical composition of, 441. Clefts of rocks, organic remains in, at Plymouth, 181. Cliffs, action of land springs on, 49 ; destruction of, 77 ; when partially protected, ib. Clifton, near Bristol, gorge of, 58. Climate, change of, in Europe, 215. Coal, in the cretaceous rocks of Ger- many, 290 ; in the cretaceous rocks of the Spanish Pyrenees, 293 ; in the oolitic rocks of Yorkshire, 314 ; in the oolitic rocks of Scotland, 319; in the oolitic rocks of Germany, 320, 321; of the coal measures, 376; chemical composition of, 377 ; re- marks on the accumulation of, 402 ; great thickness of, at St. Etienne, 405. Coal measures, 376; faults in the, 379; contortions of the, 380 ; summary of organic remains in the, ib. ; remarks on the, 404. Como, lake of, discharge of detritus into, 53 ; limestone of the vicinity of the, 330 ; dolomite on the shores of the, 477. Conybeare, Mr., on the superficial gra- vel and erratic blocks of Central En- Index. 621 gland, 186 ; his classification of the oolitic rocks of England, 314 ; on the magnesian conglomerate of Somerset, 367. Cooper, Mr., on Big Bone Lick, Ken- tucky, note, 198.' Coral islands, 163; defended to wind- ward by coral shingle beaches, 83. Corals, at what depth found, 164 ; re- attach themselves when freshly bro- ken off, 165 ; beds of, between lava currents, ib. Cordier, M., on the temperature of mines, 8. Crag, the English, 229. Craters of elevation, 124. Crawfurd, Mr., organic remains disco- vered in Ava by, 268. Cretaceous rocks, general characters of the, in Englandand Northern France, 285 ; of Sweden, 286 ; in Russia and Poland, ib.; of Germany, 288; of the Alps, 290, 293; variations in the mineralogical characters of the, 291 ; of Southern France, 292 ; of the Cotentin, 296; of Stevensklint (Seeland), 297 ; summary of or- ganic remains contained in the, 297 ; of the United States, 302. Cristie, Dr. Turnbull, on tabasheer, note, 215 ; on the osseous breccia near Palermo, 209. Croizet and Jobert, MM., on the ossi- ferous rocks of Auvergne, 271. Current, great Atlantic, 99 ; round Cape Lagullas, ib. ; Polar, 1 02 ; through Behring's Straits, 104; of the Straits of Gibraltar, 105; out of the Baltic, 106. Currents, in the West Indies, 100 ; of the Pacific, 104 ; in the Indian and Chinese Seas, 107 ; their general accuracy doubtful, 108; transport- ing power of, 112. Cuvier, Baron, on the dunes of the Landes, 84; labours of, round Paris, 251. Darby, Mr., on the Mississippi, 71 ; on the raft of the Atchafalaya, ib. Daubeny, Dr., on the gases evolved from volcanos, 117 ; on the heat and appearances of a lava current, ib. ; on the gases evolved from the Solfatara, near Naples, 135 ; on the lava of Auvergne, 136; on the Euganean Hills, 275; on thermal springs, 606. Daubuisson, M., on the decomposition of granite, 45 ; on the falls of the Rhine, 56. Oavy, Dr., on the volcanic island of Sciacca, 119. Dechen, M. von, on the passage of the Nahe to the Rhine, 58 ; on the er- ratic blocks of Germany, 190 ; on the brown coal of Germany, 248 ; on the cretaceous rocks of Germany, 288; on the oolitic rocks of Ger- many, 320, on the variegated or red marl of Germany, 354 ; on the roth- liegende of Gerrnany, 359 ; on the carboniferous rocks of Germany, 383. Delta, of the Nile, 69; of the Po, 70; of the Mississippi, 71; oftheGanges, 74. Deltas, relative importance and in- crease of, 76. Deposits, in the lake of Geneva, 52 ; in the lake of Como, 53 ; siliceous, from springs, 156; calcareous, from springs, 158. Deshayes, M., his classification of the supracretaceous or tertiary rocks of Europe, 221. Desnoyers, M., on the more modern supracretaceous rocks, 226 ; on the baculitelimestone of Normandy, 296. Detritus, superficial, in South Devon, 44 ; reconsolidated at Nice and in Jamaica, 45 ; talus of, at foot of cliffs, 49 ; delivery of, into the sea, 67. Devonshire, red sandstone and con- glomerate of, 362. Diablerets (Valais), on the rocks of the, 281. Diallage rock, 446 ; calculations re- specting the chemical composition of, 453. Digue, Cherbourg, action of the waves on, 88. Dirt bed, Isle of Portland, conclusions respecting the, 306. Dolomite, 474 ; association of with gypsum, 478. Dufrenoy, M., on the cretaceous rocks of Southern France, 292 ; on the cretaceous rocks of the Spanish Py- renees, 293 ; on the oolitic rocks of South-Western France, 318. Dunes, or Sand Hills, 84; advance of, ib. ; indurated, 85 ; of Lake Supe- rior, note, 88. Dykes, volcanic, 138 ; basaltic, con- verting chalk into granular marble, 622 Index. 279 ; trappean, 463 ; great length of certain trappean, note, 464; of serpentine, 466. Earth, figure of the, 1 ; density of the, ib. ; temperature of the, 6. Earthquake, of Lisbon, 141 j of Cutch, 144 ; of Jamaica, ib. Earthquakes, 140; felt at great di- stances, 141 ; shocks of, on different rocks, 142 ; relative importance of, 148. Electricity, effects of, on rocks, 46. Elephant, frozen, of Siberia, 199. Elevation of mountains, 481. Ellis, Mr., on the crater of Kirauea, Owhyhee, 121. Erman, M., on the maximum density of sea water, 24. Erratic blocks, of England, 186; of Scotland, 188; of Sweden, 189; of Russia, ib.-, of Poland, 190; of Ger- many, ib. ; in Belgium and Holland, 192 ; of North America, ib. ; of the Alps, 194; of the lake of Como, 195. Euganean Hills, volcanic and other rocks of the, 275. Eurite, 436 ; calculation respecting the chemical composition of, 442. Exhalations, gaseous, 151. Faults, at Dawlish, 183; act as Arte- sian wells, 379. Fingal's Cave, Staffa, how formed, 78. Fitton, Dr., on the Maestricht beds, 281 ; on the lower cretaceous rocks, 284; on the Wealden rocks, 304; on the dirt bed of the Boulonnois, 308. Fleming, Dr., on the submarine fo- rests of the Frith of Tay and the Frith of Forth, 167, 168. Fleuriau de Belle vue, M., on an Arte- sian well at Rochelle, 13. Floods, river, care required in estima- ting the real physical changes pro- duced by, 64. Forests, submarine, 166. Fossil plants, remarkable, in the oolitic rocks of the Alps, 327 ; vertical, in the Yorkshire oolite, 350 ; vertical in the carboniferous rocks, 406 ; re- marks on those contained in the coal measures, 409. Fourier, Baron, on the temperature of the earth and planetary spaces, 26. Fox, Mr., on the electro-magnetic pro- perties of mineral veins, 493. France, ossiferous caverns of, 203, 206; supracretaceous rocks of, 222, 227, 242, 246, 249, 251, 270; cretaceous rocks of, 283, 286, 290, 296 ; oolitic rocks of, 316, 318, 324, 327; varie- gated marl of, .353 ; muschelkalk of, 356 ; variegated or red sandstone of, 357; carboniferous rocks of, 398,404. Freshes, or freshets of rivers, 64. Fresh- water formation, upper, of Paris, 255 ; of the Isle of Wight and Hamp- shire, 264. Fuchsel, M., his geological researches, 213. Fundy, Bay of, great tides in, 93. Fusibility, relative, of rocks, 454. Ganges, delta of, 74 ; coarse gravel not transported by within 400 miles of the sea, 75 ; amount of detritus trans- ported by, ib. Gases, inflammable, natural jets of, 152 ; naturally produced, used for economical purposes, 152, 153, 154, 155. Gault, 285. Geneva, lake of, depth of water in, 22; discharge of detritus into, 52. Geological terms, explanation of, 598. Georges-gemiind, curious mixture of organic remains at, 228. Germany, erratic blocks in, 190; os- siferous caverns of, 205 ; brown coal of, 248 ; cretaceous rocks of, 288 ; oolitic rocks of, 320 ; variegated marls of, 354; muschelkalk of, 356; zech- stein of, 358 ; rothliegende of, 359 ; carboniferous rocks of, 383. Geysers, the, 20 ; siliceous deposits from the waters of the, 156. Giant's Causeway, basaltic rocks of the, 27_8. Glaciers, assist the degradation of land, 65 ; rate of advance of, 66. Globe, changes on the surface of the, 34. Gneiss, 436 ; calculations respecting the chemical composition of, 440. Gorges, cut by rivers, 54 ; otherwise produced, 57. Gosau, rocks of the valley of, 280. Gosse, Dr., on the baths of Vignone and San Filippo, 158. Granite, 445 ; calculations respecting the chemical composition of, 449 ; decomposition of, 54 ; covers chalk at Weinbbhla, 294 ; passage of, into basalt, note, 455 ; veins, 459. Index. 623 Granitic, rocks, position of, as regards the oolitic rocks of the Alps, 456 ; resting on fossiliferous rocks, 458. Grauwacke, 414 ; arrangement of the laminae in the slates of, 415 ; remarks on the limestones of the, 417; red, 420 ; rocks associated with the lower portion of the, 421 ; summary of organic remains contained in, 423 ; remarks on the organic character of the, 424 ; anthracite and coal de- tected in, 427; altered by granite, 479 ; upper beds of, in Shropshire and Wales, 609. Gravel, of Devonshire, 182; of central England, 186; rounded in caves, 206; of the Lyonais, Dauphine, and Pro- vence, 222. Green sand, upper, 284; analysis of green grains in, 285 ; lower, ib. Greenstone, 446 ; calculations respect- ing the chemical composition of, 450 ; porphyritic, ib. Gres de Vosges, 357. Group, modern, 43; erratic block, 181 ; supracretaceous, 213 ; cretaceous, 283 ; oolitic, 314 ; red sandstone, 353; carboniferous, 376; grauwacke, 414. Guevo Upas, or Valley of Poison, in Java, 155. Gulf Stream, 101. Gypsum, ossiferous, of Paris, 253. Hall, Sir James, on th e transported gra- vel near Edinburgh f 1 88; onthefusion of limestone beneath pressure, 465. Hall, Captain Basil, on the trees carried down by the Mississippi, 72 ; on the country fringing the delta of the Mississippi, 73 ; his plan for showing ships' tracks on charts, 109. Harris, Mr., on the large blocks of rock moved during heavy gales at Plymouth, 82. Hastings sands, 305. Heat, central, 8, 26 ; source of in thermal springs, 16. Hebrides, submarine forests of the, 168; oolite of the, 320. Hibbert, Dr., on the erratic blocks of Shetland, 188; on the rocks of the Velay, 274; on the passage of gra- nite into basalt, note, 445. Hitchcock, Mr., on the carboniferous deposits of Connecticut, 399. Hoffman, M., on the Valley of Pyr- mont, 32. Hooke, Dr., on inclined strata, and land raised by earthquakes, note, 21 4. Hornblende rock and slate, 434 ; cal- culations respecting the chemical composition of, 441. Hornei', Mr., on a submarine forest, Somersetshire, 169. Hugi, M., on the relations of the lias and granitic rocks at the Bbtzberg, 456. Human remains in caverns, 203. Humboldt, M. von, on the perpetual snow-line on various mountains, 28 ; on shocks of earthquakes in the Cor- dilleras, 142 ; on the red sandstone of Mexico and South America, 373. Hurricanes, 149 ; force of, 150. Hutton, Mr., on the lower (new) red sandstone of Durham, 361 ; on the state of carburetted hydrogen in coal, 378. Hypersthene rock, calculation respect- ing the chemical composition of, 450. Iceland, springs of, 20 ; volcanos of, 118; deposits from springs in, 156. Igneous rocks, remarks on, 468. Imatra, falls of, 63. Importance, geological, of tides and currents, 115. India, supracretaceous rocks of, 268 ; coal of, 399 ; hornblende rock of, 485. Inferior stratified, or non-fossiliferous rocks, 432; remarks on the, 437; calculations respecting the chemical composition of the, 439. Insects, fossil, at Aix in Provence, 248; fossil, at Solenhofen, 345. Iron ore, pisiform, of the Haute Saone, 309. Isle of Wight, supracretaceous rocks of, 260, 264. Ivory, Mr., on the heat disengaged by compressed air, 10. Jamaica, great earthquake at, 144; red sandstones of, 374. Jorullo, sudden elevation and formation of, 128. Kaepfnach, animal remains in the lig- nite of, 239. Kettle and Pans, rock basins so named, St. Mary's, Scilly, 46. Kirauea, crater of, Owhyhee, or Ha- waii, 121. Klipstein, M., on the relations of the cretaceous and granitic rocks at Weinbbhla. 294. 624 Index. Kotzebue, M., on the temperature of the sea, 24. Kovalevski, M., on the carboniferous rocks of Southern Russia, 398. Kupffer, Prof., on the temperature of springs, 14. La Spezia, dolomite and limestones of, 331. Lake Erie, supposed drainage of, 60 ; drainage of, could not produce a sudden deluge, ib, Lake Souvando, sudden considerable drainage of, 63. Lake waters, how only suddenly dis- charged, 61. Lakes, temperature of, 22 ; filled up by river detritus, 51 ; apparently drained, 54; Swiss, 238. Land, dry, superficial distribution of, 2 ; degradation of, 43 ; rise and de- pression of, by earthquakes, 143. Lava, flow of, beneath the sea, 1 25 ; current, heat and appearances of, 117. Lenz, M., on the saltness of the ocean, 5. Lias, remarks on the, 323 ; of the Alps, 327 ; organic character of the, at Lyme Regis, 346. Life, early animal, remarks on, 429. Limestone, siliceous, of Paris, 253 ; saccharine, 435. Lloyd, Mr., on the levels of the Pacific Ocean and Mexican Sea, 101. London clay, 260. Lonsdale, Mr., on the green sand of Wiltshire, 285 ; on the oolitic rocks near Bath, 315; on the oolitic rocks of Gloucestershire, ib. ; on the Stones- field slate, 316. Luidas Vale, Jamaica, remarkable drainage of, 54. Lyell, Mr., on the decrease of tempe- rature on the earth's surface, 8 ; on the gorge of the Simeto, 56 ; on a salt deposit in the Mediterranean, 105; on craters of elevation, 127; on surface changes produced by earthquakes, 144 ; on the Bakie Loch, 160; on the transport of er- ratic blocks, 193; on the evidences of a change of climate, 215 ; on the supracretaceous rocks of Aix in Pro- vence, 247 ; on the Hordwell beds, 265 ; on the fresh-water limestone of the Cantal, 271 ; on a serpentine dyke, Forfarshire, 466. Macculloch, Dr., on the tors of Devon and Cornwall, 43 ; on quartz rock, 434 ; on trappean rocks, 446 ; on the trap and serpentine at Clunie, Perthshire, 465 ; on the serpentine and diallage rocks of the Scottish Isles, 43.9. Macgillivray, Mr., on comminuted sea shells thrown up in the Hebrides, 87. Mackenzie, Sir G., on the Geysers, 20. Maestricht beds, 281. Mangrove trees, accumulation of land by means of, 90. Maps, geological, on the construction of, 600. Marcet. Dr., on the saltness and specific gravity of sea water, 4 ; on the max- imum density of sea water, 23. Marmora, M. de la, on subfossil shells with pottery in Sardinia, 177; on the supracretaceous rocks of Sardinia, 244. Measures, comparison of English and French, 617. Mediterranean, great saltness of the, 5; comparatively high temperature of the, 25 ; divided into basins, 106. Merian, M., on the ferriferous beds of Aarau, 310; on the oolitic rocks of the Jura, near Bale, 323. Metals, occurrence of, in rocks, 491. Meyer, M., on the mixture of organic remains at Friedrichsgemiind, 228. Mica slate, 436 ; calculations respecting the chemical composition of, 440. Michael, St., Azores, siliceous deposit from springs in, 156. Miner, quantity of heat disengaged by a, 9. Mines, temperature of, 8 : sources of error respecting the heat of, 9. Mississippi, course of, particularly in- structive, 72; changes produced in the, 73 ; delta of, ib. ; detritus carried by, does not enter the Gulf Stream, ib. Mitchel, Major, on the osseous breccia of Australia, 211. Modern deposits from springs, relative importance of, 160. Molasse and Nagelfluhe of the Alps, 231,237. Monsoons, effects of, on currents, 107. Monte San Primo, Lake of Como, mul- titude of erratic blocks on, 195. Morton, Dr., on the supracretaceous rocks of the United States, 269 ; on the cretaceous rocks of the United States, 302. Index. 625 Mountains, elevation of, 481. Murchison, Mr., on the lacustrine de- posit of (Eningen, note, 220; on the mixture of organic remains at Georgesgemiind, 228 ; on the supra- cretaceous rocks of the Austrian and Bavarian Alps, 240; on the supra- cretaceous rocks of Styria, 241 ; on the supracretaceous rocks of Aix in Provence, 247 ; on the fresh-water limestone of the Cantal, 271 ; on the rocks of Gosau, 280 ; on the oolitic rocks of Scotland, 319; on vertical stems of plants in the oolite of York- shire, 350 ; on the upper portion of the grauwacke in Shropshire and Wales, 609. Muschelkalk, 355. Naphtha and asphaltum springs, 162. Necker de Saussure, Prof., on the vol- canic dykes of Monte Somma, 138 ; on the rocks of the Buet and Vallor- sine, 329. Niagara, Falls of, 59 ; cutting back of the Falls of, ib. Nice, raised river-beds near, 67 ; sub- fossil shells near, 176; supracreta- ceous rocks of, 232. Nile, delta of, 69 ; cause of the erroneous opinion respecting the great advance of, 70. Nilsson, M., on the cretacous rocks of Sweden, 286. Old red sandstone, 382. Oltmanns, M., his tables for calculating heights by the barometer, 610. Oolitic group, general view of the, 349. Oolitic rocks, of England, 314; of Nor- mandy, 316 ; of the North of France, ib.; of Burgundy, 317; of the Haute Saone, ib. ; of the Bernese Jura, 318; of South-western France, ib. ; of Scotland, 319; of Germany, 320; of Poland, 326; of the Alps, &.; general remarks on the, 323 ; obser- vations on the organic contents of the, 338. Organic remains, of the modern group, 179; in superficial gravels, 197; in Kirkdale cave, 205 ; in the Mug- gendorf caverns, note, ib. ; in the osseous breccia of the Mediterranean, 210 ; in the osseous breccia of Aus- tralia, 211; of the crag, 229; of 2 the Swiss molasse, 240 ; of the su- pracretaceous rocks of Styria, 241 ; of the Val d'Arno, 245 ; of the supra- cretaceous rocks of Aix in Provence, 247 ; of the brown coal of Germany, 249; of the Parisian plastic clay, 252; of the calcaire grossier, ib. ; of the ossiferous gypsum of Paris, 253 ; of the upper marine sands of Paris, 255 ; of the upper fresh-water formation of Paris, ib. ; of the English plastic clay, 259; of the London clay, 261; of the fresh-water formations of the Isle of Wight and Hampshire, 264, 266 ; of supracretaceous rocks in India, 268; of supracretaceous rocks in the United States, 269 ; in the fresh- water limestone of the Cantal, 271 ; in the ossiferous beds of Auvergne, 272 ; in the Velay, 273 ; at Cussac, ib. ; of the Vicentine, 276 ; of the supracretaceous blue marls of the South of France, 495 ; in the supra- cretaceous rocks of Bordeaux and Dax, 501 ; of Gosau, 506 ; of the cretaceous group, 507 ; in the bacu- lite limestone of Normandy, 296; at Stevensklint (Seeland), 297 ; of the cretaceous rocks in the United States, 303 ; of the Wealden rocks of En- gland, 528 ; vegetable, in the Alps, 328 ; of the La Spezia limestones, 332 ; of the oolitic group, 529 ; of the red sandstone group, 568; of the variegated marls, ib. ; of the mus- chelkalk, 569 ; of the red or varie- gated sandstone, 571 ; of the zech- stein, 572 ; of the coal measures, 574; of the carboniferous limestone, 579; of the grauwacke group, 584. Osseous breccia, 208 ; of Nice, 209 ; of Palermo, ib. ; of Australia, 211. Ossiferous beds, of Auvergne, 272 ; of the Velay, ib. ; of Cussac, 273. Ossiferous caverns, 201 ; general ap- pearances in, ib. ; care necessary in the examination of, 202 ; containing human bones, 203. Palisades, Jamaica, remarks respecting the, 91. Papandayang, the sudden disappear- ance of, 134. Paris, remarks on the supracretaceous rocks around, 256. Paris, Dr., on the recent sandstone of Coi'nwall, 85. 626 Index. Passy, M., on the Wealden rocks of Normandy, 308. Pentland, Mr., on the remains dis- covered in the osseous breccia of Australia, 211 ; on fossil animals found in India, 268. Phillips, Mr. I., on the submarine forests and lacustrine deposits in Yorkshire, 166 ; on the erratic blocks and gravel of Yorkshire, 1 88 ; on the oolitic rocks of Yorkshire, 314. Pinhay, near Lyme Regis, cause of undercliffs at, 49. Pitchstone, chemical composition of, 452. Plants, remarkable fossil, in the oolitic rocks of the Alps, 327 ; vertical, in the carboniferous rocks, 406. Plastic clay, of Paris, 251; of England, 258. Po, raised bed of the, 67; delta of the, 70. Porphyry, chemical composition of, 451. Pratt, Mr., on the terrestrial animals in the fresh-water rocks, Binstead, Isle of Wight, 264. Products, mineral volcanic, 137. Prony, M., his calculations respecting the delta of the Po, 70. Protogine, 437; calculations respecting the chemical composition of, 442. Pterodactyles, occurrence of insects with, 346. Purbeck beds, 306. Pusch, Prof., on the erratic blocks of Poland and Prussia, 190 ; on the cretaceous rocks of Russia and Po- land, 286 ; on a ferriferous deposit of Poland, 311 ; on the white sand- stone of Poland, 355 ; on the carbo- niferous rocks of Poland, 397. Pyrmont, valley of, 32. Quartz rock, 416, 434. Quoy and Gaimard, MM., on coral reefs and islands, 163. Raffles, Sir Stamford, on the eruption from Tomboro, Sumbawa, 129. Raft of the Atchafalaya, the great ac- cumulation of transported wood so called, 71. Rasoumovski, Count, on the erratic blocks of Russia, 189. Red or variegated sandstone, 356. Red sandstone, of Devonshire, 362 ; of Mexico and South America, 373 ; of Jamaica, 374. Red sandstone group, general remarks on the, 366; organic character of the, 372. Rennell, Major, on the Ganges and its delta, 74. Rhine, falls of the, 56. Rhinoceros, frozen, of Siberia, 199. Riverbeds, when raised, 67; deepening of, 68. Rivers, 50 ; transporting power of, ib. ; fill up lakes with detritus, 51 ; cut- ting powers of, 55 ; action of, on their beds, 56; freshes or freshets of, 64 ; deflected from their courses by sea beaches, 89. Robert, M., on the ossiferous beds of Cussac and Solilhac, 273. Rocks, classification of, 35; tabular view of different classifications of, 38 ; decomposition of, 43; mineralo- gical differences in contemporaneous, 473. Rothliegendes, 359. Royle, Mr., on the fossil plants associ- ated with the coal deposit of India, 400. Ruffiberg, or Rossberg, fall of part of the, 48. Sabine, Capt, on waters, supposed to be those of the Maranon, flowing in the Atlantic, 96. Salses, or mud volcanos, 155. Salt of Wieliczka, 270. San Filippo, deposit from the thermal waters of, 158. Sands, transported, 84; Slapton, 81. Sands and Sandstones, upper marine, of Paris, 255. Sandstone, modern, of Cornwall, 85 ; of Guadaloupe, with human remains, ib. ; of the Great Canary, 86. Sardinia, sub-fossil shells with pottery in, 177; supracretaceous rocks of, 244. Saussure, M. de, ladder of, found in the Mer de Glace, 66. Schorl rock, 450 ; calculations respect- ing the chemical composition of, ib. Scott, Mr., on supracretaceous rocks in India, 268. Scrope, Mr. Poulett, on the classifica- tion of mineral volcanic products, 137. Index. 627 Sea, saltness and specific gravity of the, 3 ; temperature of, 21 ; action of the, on coasts, 77. Sections, geological, remarks on the construction of, 600. Sedgwick, Prof., on the erratic blocks of Northern England, 187; on the supracretaceous rocks of the Austrian and Bavarian Alps, 240 ; on the su- pracretaceous rocks of Styria, 241 ; on the rocks of Gosau, 280 ; on the (new) red sandstone of Northern England, 357, 360 ; on the magne- sian limestone, 359 ; on the carboni- ferous rocks of Central and Northern England, 386 ; on the old red sand- stone of Great Britain, 390. Serpentine, 446; chemical composition of, 453; passage of, into trappean rocks, 465 ; dyke of, in Forfarshire, 466 ; of Liguria, ib. Serres, M. Marcel de, on the supra- cretaceous rocks of the South of France, 242 ; on the fossil insects of Aix in Provence, 248. Shore lines on the limestones of Greece, 178. Slapton Sands, Devon, remarks re- specting, 81 ; fresh-water lake be- hind, ib. Smith, Mr. William, his identifica- tion of strata by organic remains, 213. Smith, Lieut-Col. Hamilton, on masses of coral thrown up during a hurricane at Curacoa, 83. Smith, Mr. C., on the submarine forest in the Hebrides, 168. Snow, line of perpetual, on different mountain chains, 28. Solfataras, 135. Soundings round the British Islands, note, 112. Sowerby, Mr. G. B., on the Isle of Wight, 264. Spezia, La, limestones of, 331. Springs, temperature of, 14; land, action of, on sea cliffs, 49 ; deposits from, 156; naphtha and asphaltum, 162; in faults, 379. Strangways, Mr., on the sudden drain- age of Lake Souvando, 63 ; on the falls of Imatra, ib. Stratified rocks, columnar structure of certain, the effect of heat, 471. Studer, M., on the molasse of Switzer- land, 237; on the relations of the 2 s oolitic rocks and gneiss of the Jung- frau, 456. Styria, Lower, on the supracretaceous rocks of, 241. Submarine forest, of Lincolnshire, 166; of Yorkshire, ib. ; of the Frith of Tay, 167; of the Frith of Forth, 168; of Orkney, ib. ; of the Hebrides, ib. ; of Cardiganshire, 169; of Somerset- shire, ib. ; of Mount's Bay, Cornwall, 170 ; of Morlaix, ib. ; of the Baltic, ib. ; probable cause of, 171. Submarine volcanic eruptions, 118. Substances, simple, entering into the composition of the inferior stratified rocks, 443 ; simple, constituting the unstratified rocks, 454. Supracretaceous rocks, 213; volcanic action during the deposit of the, 270. Surface of the globe, changes on the, 34. Svanberg, M., on the temperature of the planetary spaces, note, 27. Switzerland, supracretaceous rocks of, 237 ; remarks on the lakes of, 238. Talcose slate, 434; calculation respect- ing the chemical composition of, 441. Taylor, Mr. J., on ihe occurrence of metals in rocks, 492. Taylor, Mr. R., on the crag near Nor- wich, 230. Temperature of the earth, 6, 26 ; de- crease of, on the earth's surface, 8, 215; of mines, 8; of Artesian wells, 12; of springs, 14; of thermal springs, 18 ; of the sea, 21 ; of lake*, ib. ; of the planetary spaces, 27 ; of the at- mosphere, ib. Terni, falls of, 159. Tessier, M., on the human remains discovered in the caves of Southern France, 203. Thermal springs, in the Himalaya, 17; in the Alps, 18 ; in the Pyrenees, ib.; in North America, ib. ; in China, 19; in England, ib. Thirria, M., on the ossiferous caverns of the Haute Saone, 206 ; on the pisiform iron ore of the Jura, 210, 309 ; on the oolitic rocks of the Haute Saone, 317. Thomson, Dr., on the elementary sub- stances in coal, 377. 2 628 Index. Thurman, M., on the oolitic rocks of the Bernese Jura, 318. Tides, streams of, chiefly felt on coasts, 92 ; velocities of, ib. ; in the English Channel, ib. ; in the Bristol Channel, 93 ; in the Bay of Fundy, ib. ; dif- ference of, between the shore and the offing, 94 ; in the Straits of Gi- braltar, 95 ; in the Pentland Frith, ib. ; in rivers and estuaries, 96 ; trans- porting power of, 110. Tillard, Capt., on the Island of Sabrina, 118. Tomboro, in Sumbawa, great eruption of, 129. Tors in Devon and Cornwall, 43. Trappean rocks, 446, 461 ; dykes of, 463. Trachyte, 137 ; chemical composition of, 452. Travertin, at San Filippo, 158 ; at Ti- voli, 160. Trinidad, pitch lake of, 162. Tubes, lightning, near Drigg, Cumber- land, 46. Tuckey, Capt., on the mouth of the Zaire, or Congo, 90. Turner, Dr., analysis of springs in India, 157; analysis of organic remainsfrom the lias, 348 ; analysis of fish palates from the carboniferous limestone and chalk, 413. United States, North America, supra- cretaceous rocks of the, 269 ; creta- ceous rocks of the, 302 ; carboni- ferous deposits of the, 398. Unstratified rocks, 445 ; elementary substances entering into the compo- sition of the, 454. Val d'Arno, supracretaceous rocks of the, 245. Val de Bagnes, debacle from the, 61. Valleys, mountain, 29 ; lowland, ib. ; flat-bottomed, 30; of elevation, 31; of denudation, 32 ; dry, in Jamaica and Peru, 33. Variegated or red marl, 353. Vegetation, protection afforded to land by, 216. Velay, on the ossiferous and volcanic rocks of the, 272, 274. Vetch, Capt., on the raised beach, Isle of Jura, Hebrides, 175. Villeneuvc, M. de, on the carboniferous rocks of Belgium, 394. Volcanic cone, variable solidity of, 124. Volcanic explosions, sound of, trans- mitted through rocks, 143. Volcanic matter, injection of, amid stratified rocks, 140. Volcanic products, gaseous, 117; mi- neral, ib. Volcanos, active, 115 ; in Central Asia, 116; inland in America, ib. ; sub- marine, producing islands, 118; in and around the Pacific, 122 ; of the Atlantic, 123 ; extinct, 134. Walchner, Prof., on the pisiform iron ore of Candern, Brisgau, 310. Warburton, Mr., on the Bagshot Sands, 263. Water, superficial distribution of, 2 ; compressibility of, 6 ; temperature of, in Artesian wells and in mines, 12 ; maximum density of fresh, 22 ; maximum density of sea, 23 ; action of, in the destruction of rocks, 47 ; passage of, in faults, 379. Watt, Mr. Gregory, his experiments on fused basalt, 470. Watt, Mr., on a submarine forest in Orkney, 168. Waves, transporting action of, at diffe- rent depths, 88 ; action of, on shoals, ib.; produced by earthquakes, 141. Weald clay, 304. Wealden rocks, of England, 304 ; ge- neral observation on the, 307 ; of Normandy, 308 ; of other parts of Europe, 309. Weathering of rocks, 43. Weaver, Mr., his remarks on the car- boniferous rocks, 385 ; on the car- boniferous rocks of Ireland, 393 ; on the coal in grauwacke of Ireland, 428. Webster, Mr., on the plastic clay of the Isle of Wight, 260 ; on the fresh- water formation of the same island and Hampshire, 264 ; on the Pur- beck beds, 306. Webster, Dr., on the siliceous deposits from springs, St. Michael's, Azores, 156. Weiss, M., on granitic resting on cre- taceous rocks at Weinbohla, 294. Witham, Mr., on vertical plants in the carboniferous rocks, 406. Wollaston, Dr., on the currents of the Straits of Gibraltar, 105. Index. 629 Wood, Mr., on vertical stems in Kil- lingworth colliery, 407. Woodley, Mr., on rock basins, St. Mary's, Scilly, 46. Yates, Mr., on lakes produced by the fall of mountain masses, 62 ; on lakes formed by detritus discharged from a cross valley, 63 ; on a submarine forest, Cardiganshire, 169; on pris- matic clay ironstone near trap, duchy of Nassau, 472. Zechstein, or magnesian limestone, 358. THE END. LONDON: PRINTED BY RICHARD TAYLOR, KED LION COURT, FLEET STREET. ALERE FLAMMAM. By the same Author, SECTIONS AND VIEWS ILLUSTRATIVE OF GEOLOGICAL PHENOMENA. In one Volume 4to. With Forty Plates, mostly coloured. Price 2/. 2s. " We recommend to the particular attention of the numerous cultivators of geology throughout Britain these valuable sections and views. They are selected with judgment from many expensive works, which few have an opportunity of purchasing, and many of never seeing at all." Jameson's Edin. Phil. Journal, April, 1831. " This work tends, more than other publication that has yet appeared, to render easy and familiar many of the most difficult and complicated phse- nomena of the original structure of the earth's surface ; and also of the vio- lent changes and physical revolutions by which it has been disturbed." Philosophical Magazine and Annals, February, 1831. A TABULAR AND PROPORTIONAL VIEW of the Su- perior, Supermedial, and Medial (Tertiary and Secondary) Rocks. Second Edition. Coloured : Price 5$. GEOLOGICAL NOTES. 8vo, with Two Plates, Price 6*. RETURN EARTH SCIENCES LIBRARY TO +> 230 McCone Hall 642-2997 LOAN PERIOD 1 1 MONTH 2 3 4 5 6 ALL BOOKS MAY BE RECALLED AFTER 7 DAYS Books needed for class reserve are subject to immediate recall DUE AS STAMPED BELOW FORM NO. DD8 UNIVERSITY OF CALIFORNIA, BERKELEY BERKELEY, CA 94720