would preserve one suite of organisms in England, but a very different group at the foot of the Himalaya Mountains, yet the deposits at the two places might be absolutely coeval, even as to months and days. Hence it becomes apparent that while strict contemporaneity cannot be predicated of deposits containing the same organic remains, it may actually be true of deposits in which they are quite distinct. If, then, at the present time, community of organic forms obtains only in districts, regions, or provinces, it may have been more or less limited also in past time. Similarity or identity of fossils among formations geographically far apart, instead of proving contemporaneity, ought rather to be looked upon as indicative of great discrepancies in the relative epochs of deposit. For in any theory of the origin of species, the spread of any one species, still more of any group of species to a vast distance from the original centre of dispersion, must in most cases have been inconceivably slow. It must have occupied so prolonged a time as to allow of almost indefinite changes in physical geography. A species may have disappeared from its primeval birthplace while it continued to flourish in one or more directions in its outward circle of advance. The date of the first appearance and final extinction of that species would thus differ widely according to the locality at which we might examine its remains.

The grand march of life, in its progress from lower to higher forms, has unquestionably been broadly alike in all quarters of the globe. But nothing seems more certain than that its rate of advance has not everywhere been the same. It has moved unequally over the same region. A certain stage of progress may have been reached in one quarter of the globe thousands of years before it was reached in another; though the same general succession of organic forms might be found in each region.

The geological formations form the records of these ages of organic development. In every country where they are fully displayed, and where they have been properly examined, they can be separated out from each other according to their organic contents. Their relative age within a limited geographical area can be demonstrated by the mere law of superposition. When, however, the formations of distant countries are compared, all that we can safely affirm regarding them is that those containing the same or a representative assemblage of organic remains belong to the same epoch in the history of biological progress in each area. They are homotaxial; but we cannot assert that they are contemporaneous, unless we are prepared to include within that term a vague period of perhaps thousands of years. Doctrine of Colonies.-M Barrande, the distinguished author of the Système Silurien de la Bohême, drew attention more than a quarter of a century ago to certain remarkable intercalations of fossils in the series of Silurian strata of Bohemia. He showed that, while these strata presented a normal succession of organic remains, there were neverthe less exceptional bands, which, containing the fossils of a higher zone, were yet included on different horizons among inferior portions of the series. He termed these precursory bands "colonies," and defined the phenomena as consisting in the partial co-existence of two general faunas, which, considered as a whole, were nevertheless. successive. He supHe sup posed that during the later stages of his second Silurian fauna in Bohemia the first phases of the third fauna had already appeared, and attained some degree of development in some neighbouring but yet unknown region. At intervals, corresponding doubtless to geographical changes, such as movements of subsidence or elevation, volcanic eruptions, &c., communication was opened between that outer region and the basin of Bohemia. During these intervals a greater r less number of immigrants succeeded in making their way into the Bohemian area, but as the conditions for their

prolonged. continuance there were not yet favourable, they' soon died out, and the normal fauna of the region resumed its occupancy. The deposits formed during these partial interruptions, notably graptolitic schists, accompanied by igneous sheets, contain, besides the invading species, remains of some of the indigenous forms. Eventually, however, on the final extinction of the second fauna, and, we may sup pose, on the ultimate demolition of the physical barriers lutherto only occasionally and temporarily broken, the third fauna, which had already sent successive colonies into the Bohemian area, now swarmed into it, and peopled it till the close of the Silurian period.

This original and ingenious doctrine has met with much opposition on the part of geologists and paleontologists. Of the facts cited by M. Barrande there has been no question, but other explanations have been suggested for them. It has been said, for example, that the so-called colonies are merely bands of the Upper Silurian rocks or third fauna, which by great plications have been so folded with the older rocks as to seem regularly interstratified with them But the author of the Système Silurien very justly contends that of such foldings there is no evidence, but that, on the contrary, the sequence of the strata appears normal and undisturbed. Again it has been urged that the difference of organic contents in these so-called colonies is due merely to a difference in the conditions of water and sea bottom, particular species appearing with the conditions favourable to their spread, and disappearing when these ceased. But this contention is really included in M. Barrande's theory. The species which disappear and reappear in later stages must have existed in the meanwhile outside of the area of deposit, which is precisely what he has sought to establish. Much of the opposition, which his views have encountered has probably arisen from the feeling that if they are admitted they must weaken the value of paleontological evidence in defining geological horizons. A paleontologist, who has been accustomed to deal with certain fossils as unfailing indications of particular portions of the geological series, is naturally unwilling to see his generalizations upset by an attempt to show that the fossils may occur on a far earlier horizon.

If, however, we view this question from the broad natural history platform from which it was regarded by M. Barrande, it is impossible not to admit that such phenomena as he has sought to establish in Bohemia must have constantly occurred in all geological periods and in all parts of the world. No one now believes in the sudden extinction and creation of entire faunas Every great fauna in the earth's history must have gradually grown out of some preexisting one, and must have insensibly graduated into that which succeeded. The occurrence of two very distinct faunas in two closely consecutive series of strata does not prove that the one abruptly died out and the other suddenly appeared in its place. It only shows, as Darwin has so well enforced, the imperfection of the geological record. In the interval between the formation of two such contrasted groups of rocks the fauna of the lower strata must have continued to exist elsewhere, and gradually to change into the newer facies which appeared when sedimentation recommenced with the upper strata. Distinct zoological provinces have no doubt been separated by narrow barriers in former geological periods, as they still are to-day. There seems, therefore, every probability that such migrations as M. Barrande has supposed in the case of the Silurian fauna of Bohemia have again and again taken place. Two notable examples will be given in later pages, one in the Lower and one in the Upper Old Red Sandstone of Scotland.

Gaps in the Geological Record.-The history of life has been very imperfectly preserved in the stratified parts of the earth's crust. Apart from the fact that, even under the

most favourable conditions, only a small proportion of the total flora and fauna of any period could be preserved in the fossil state, enormous gaps occur where no record has been preserved at all. It is as if whole chapters and books were missing from an historical work. Some of these lacunæ are sufficiently obvious, Thus, in some cases, powerful dislocations have thrown considerable portions of the rocks out of sight. Sometimes extensive metamorphism has so affected them that their original characters, including their organic contents, have been destroyed. Oftenest of all, denudation has come into play, and vast masses of fossiliferous rock have been entirely worn away. That this cause has operated frequently is shown by the abundant unconformabilities in the structure of the earth's crust.

While the mere fact that one serics of rocks lies unconformably on another proves the lapse of a considerable interval between their respective dates, the relative length of this interval may sometimes be demonstrated by means of fossil evidence and by this alone. Let us suppose, for Example, that a certain group of formations has been disturbed, upraised, denuded, and covered unconformably by a second group. In lithological characters the two may closely resemble each other, and there may be nothing to show that the gap represented by their unconformability is not of a trifling character. In many cases, indeed, it would be quite impossible to pronounce any well-grounded judgment as to the amount of interval, even measured by the vagus relative standards of geological chronology. But if each group contains a well-preserved suite of organic re mains, it may not only be possible, but easy, to say exactly how much of the geological record has been left out between the two sets of formations. By comparing the fossils with those obtained from regions where the geological record is more complete, it may be ascertained perhaps that the lower rocks belong to a certain platform or stage in geological history which for our present purpose we may call D,. and that the upper rocks can in 'like manner be paralleled with stage H. It would be then apparent that at this locality the chronicles of three great geological periods E, F, and G were wanting, which are elsewhere found to be intercalated between D and H. The lapse of time represented by this unconformability would thus be equivalent to that required for the accumulation of the three missing formations in those regions where sedimentation went on undisturbed.

But fossil evidence may be made to prove the existence of gaps which are not otherwise apparent. As has been already remarked, changes in organic forms must, on the whole, have been extremely slow in the geological past. The whole species of a sea floor could not pass entirely away, and be replaced by other forms, without the lapse of long periods of time. If then among the conformable stratified formations of former ages we encounter sudden and abrupt changes in the facies of the fossils, we may be certain that these. must mark omissions in the record, which we may hope to fill in from a more perfect series elsewhere. The complete contrasts between unconformable strata are sufficiently explicable. It is not so easy to give a satisfactory account of those which occur where the beds are strictly conformable, and where no evidence can be observed of any considerable change of physical conditions at the time of deposit. A group of strata having the same general lithological characters throughout may be marked by a great discrepance between the fossils above and below a certain line. A few species may pass from the one into the other, or perhaps every species may be different. In cases of this kind, when proved to be not merely local but persistent over wide areas, we must admit, notwithstanding the apparently undisturbed and continuous character of the original deposition of the strata, that the abrupt transition

from the one facies of fossils to the other must represent u long interval of time which has not been recorded by the deposit of strata. Professor Ramsay, who called attention to these gaps, termed them "breaks in the succession of organic remains." He showed that they occur abundantly among the Paleozoic and Secondary rocks of England. It is obvious, of course, that these breaks, even though traceable over wide regions, were not general over the whole globe. There have never been any universal interruptions in the continuity of the chain of being, so far as geological evidence can show. But the physical changes which caused the breaks may have been general over a zoological district or minor region. They no doubt often caused the complete extinction of genera and species which had a small geographical range.

From all these facts it is clear that the geological record, as it now exists, is at the best but an imperfect chronicle of geological history. In no country is it complete. The lacunæ of one region must be supplied from another. Yet in proportion to the geographical distance between the localities where the gaps occur and those whence the missing intervals are supplied, the element of uncertainty in our reading of the record is increased. The most desirable method of research is to exhaust the evidence for each area or province, and to compare the general order of its succession as a whole with that which can be established for other provinces. It is, therefore, only after long and patient observation and comparison that the geological history of different quarters of the globe can be correlated.

Subdivisions of the Geological Record by means of Fossils.-: As fossil evidence furnishes a much more satisfactory and widely applicable means of subdividing the stratified rocks of the earth's crust than mere lithological characters, it is made the basis of the geological classification of these rocks Thus we may find a particular stratum marked by the occurrence in it of various fossils, one or more of which may be distinctive, either from occurring in no other bed above and below, or from special abundance in that stratum. These species might therefore be used as a guide to the occurrence of the bed in question, which might be called by the name of the most abundant species. In this way a geological horizon or zone would be marked off, and geologists would thereafter recognize its exact position in the series of formations. But before such a generalization can be safely made, we must be sure that the species in question really never does appear on any other platform. This evidently demands wide experience over an extended field of observation. The assertion that a particular species occurs only on one horizon manifestly rests on negative evidence as much as on positive. The paleontologist who makes it cannot mean more than that he knows the fossil to lie on that horizon, and that, so far as his own experience and that of others goes, it has never been met with anywhere else. But a single example of the occurrence of the fossil on a different zone would greatly damage the value of his generalization, and a few such cases would demolish it altogether. Hence all such statements ought at first to be made tentatively. To establish a geological horizon on limited fossil evidence, and then to assume the identity of all strata containing the same fossils, is to reason in a circle and to introduce utter confusion into our interpretation of the geological record. The first and fundamental point is to determine accurately the order of superposition of the strata. Until this is done detailed paleontological classification may prove to be worthless.

But when once the succession of the rocks has been fixed paleontological evidence may become paramount. From what has been above advanced it must be evident that, even if the several groups in a formation or system of rocks in any district or country have been minutely sub divided by means of their characteristic fossil- and if, after

the lapse of many years, no discovery has occurred to alter the established order of succession of these fossils, nevertheless the subdivisions can only be held good for the region in which they have been made. They must not be supposed to be strictly applicable everywhere. Advancing into another district or country where the petrographical characters of the same formation or system indicate that the original conditions of deposit must have been very different, we ought to be prepared to find a greater or less departure from the first observed or what might be regarded as the normal order of organic succession. There can be no doubt that the appearance of new organic forms in any locality has been in large measure connected with such physical changes as are indicated by diversities of sedimentary materials and arrangement. The Upper Silurian formations, for example, as studied by Murchison in Shropshire and the adjacent counties, present a elear sequence of strata well defined by characteristic fossils. But within a distance of 60 miles it becomes impossible to establish these subdivisions by fossil evidence. If we examine corresponding strata in Scotland, we find that they contain some fossils which never rise above the Lower Silurian formations in Wales and the west of England. Again, in Bohemia and in Russia we meet with still greater departures from the order of appearance in the original Silurian area, some of the most characteristic Upper Silurian organisms being there found far down beneath strata replete with records of Lower Silurian life. Nevertheless the general succession of life from Lower to Upper Silurian types remains distinctly traceable. Such facts warn us against the danger of being led astray by an artificial precision of paleontological detail. Even where the paleontological sequence is best established, it rests probably in most cases not merely upon the actual chronological succession of Organic forms, but also, far more than is usually imagined, upon original accidental differences of local physical conditions. As these conditions have constantly varied from region to region, it must hardly ever happen that the same minute paleontological subdivisions, so important and instructive in themselves, can be identified and paralleled, except over comparatively limited geographical areas.

It cannot be too frequently stated, nor too prominently kept in view, that, although gaps occur in the succession of organic remains as recorded in the rocks, there have been no such blank intervals in the progress of plant and animal life upon the globe. The march of life has been unbroken, onward and upward. Geological history, therefore, if its records in the stratified formations were perfect, ought to. show a blending and gradation of epoch with epoch, so that no sharp divisions of its events could be made. But the progress has been constantly interrupted; now by upheaval, now by volcanic outbursts, now by depression. These interruptions serve as natural divisions in the chronicle, and enable the geologist to arrange his history into periods. As the order of succession among stratified rocks was first made out in Europe, and as many of the gaps in that succession were found to be widespread over the European area, the divisions which experience established for that portion of the globe came to be regarded as typical, and the names adopted for them were applied to the rocks of other and far distant regions. This application has brought out the fact that some of the most marked breaks in the European series do not exist elsewhere, and, on the other hand, that some portions of that series are much more complete than in other regions. Hence, while the general similarity of succession may remain, different subdivisions aud nomenclature are required as we pass from continent to continent.

A bed, or limited number of beds, characterized by one cr more distinctive fossils, is termed a zone or horizon, and,

as already mentioned, is often known by the name of a typical fossil, as the different zones in the Lias are by their special species of ammonite. A series of such zones, united by the occurrence among them of a number of the same species or genera, is called a group. A series of groups similarly related constitute a formation, and a number of formations may be united into a system. The terminology employed in this classification will be discussed in the following part.

PART VI.-STRATIGRAPHICAL GEOLOGY.

This branch of the science arranges the rocks of the earth's crust in the order of their appearance, and interprets the sequence of events of which they form the records. Its province is to cull from all the other departments of geology the facts which may be needed to show what has been the progress of our planet, and of each continent and country. from the earliest times of which the rocks have preserved any memorial. Thus from mineralogy and petrography it obtains information regarding the origin and subsequent mutations of minerals and rocks. From dynamical geology it learns by what agencies the materials of the earth's crust have been formed, altered, broken, upheaved, and melted. From structural geology it understands how these materials were put together so as to build up the complicated crust of the earth. From palæontological geology it receives in welldetermined fossil remains a clue by which to discriminate the different stratified formations, and to trace the grand onward march of organized existence upon this planet. Stratigraphical geology thus gathers up the sum of all that is made known by the other departments of the science, and makes it subservient to the interpretation of the geological history of the earth.

The leading principles of stratigraphy may be summed up as follows:

1. In every stratigraphical research the fundamental requisite is to establish the order of superposition of the strata. Until this is accomplished it is impossible to arrange the dates and make out the sequence of geological history.

2. The stratified portion of the earth's crust, or geological record, as it has been termed, may be subdivided into natural groups or formations of strata, each marked throughout by some common genera or species, or by a general resemblance in the type or character of its organic remaius.

3. Many living species of plants and animals can be traced downward through the more recent geological formations; but they grow fewer in number as they are followed into more ancient deposits. With their disappearance we encounter other species and genera which are no longer living. These in turn may be traced backward into earlier formations, till they too cease, and their places are taken by yet older forms. It is thus shown that the stratified rocks contain the records of a gradual progression of organic formis. A species which has once died out does not seem ever to have reappeared. But as has been already pointed out in reference to Barrande's doctrine of colonies, a species may within a limited area appear in a formation older than that of which it is characteristic, having temporarily migrated into the district from some neighbouring region where it had already established itself.

4. When the order of succession of organic remains among the stratified rocks has been determined, they become an invaluable guide in the investigation of the relative age of rocks and the structure of the land. Each zone and formation, being characterized by its own species or genera, may be recognized by their means, and the true succession of strata may thus be confidently established even in a country which has been shattered by dislocation, or where the rocks have been folded and inverted.

But whatever be the name chosen to designate a particular group of strata, it soon comes to be used as a chronological or homotaxial term, apart altogether from the stratigraphical character of the strata to which it is applied. Thus we speak of the Chalk or Cretaceous system, and embrace under that term formations which may contain no chalk; and we may describe as Silurian a series of strata utterly unlike in lithological characters to the formations in the typical Silurian country. In using these terms we unconsciously allow the idea of relative date to arise prominently before us. Hence such a word as chalk or cretaceous does not suggest so much to us the group of strata so called, as the interval of geological history which these strata represent. We speak of the Cretaceous, Jurassic, and Cambrian periods, and of the Cretaceous fauna, the Jurassic flora, the Cambrian trilobites, as if these adjectives denoted simply epochs of geological time.

The geological record is classified into five main divisions: (1) the Archæan, Azoic (lifeless), or Eozoic (dawn of life) Periods; (2) the Primary or Paleozoic (ancient life) Periods; (3) the Secondary or Mesozoic (middle life) Periods; (4) the Tertiary or Cainozoic (recent life); and (5) the Quaternary or Post-Tertiary Periods. These divisions are further ranged into systems, each system into formations, each formation into groups, and each group or series into single zones or horizons. The subjoined generalized table exhibits the order in which the chief subdivisions appear.

Order of Succession of the Stratifica Formation of the Earth's Crust.

I. ARCHIEAN.

Underneath the oldest unaltered stratified and fossiliferous formations in Europe there occur masses of gneiss and other crystalline schistose rocks belonging perhaps to widely different geological periods, but, from want of satisfactory means of discrimination, necessarily united provisionally in one common series. That they are separated by a vast interval of time from the rocks which lie upon them is shown by the strong unconformability with which they are related to every formation of younger date than themselves. Everywhere thoroughly crystalline, they are disposed in rude, crumpled, often vertical beds, out of the ruins of which the overlying formations have been partly built.

BRITAIN.-In no part of the European area are these ancient rocks better seen than in the north-west of Scotland. Their position there, previously indicated by MacCulloch and Hay Cunningham, was first definitely established by Murchison, who showed that they possess a dominant strike to N.N.W., and are unconformably overlaid by all the other rocks of the Scottish Highlands. They consist of a tough massive gneiss usually hornblendic, with bands of hornblende-rock, hornblende-schist, quartz-felsite, granite, and other crystalline rocks. In two or three places they enclose bands of limestone, but neither in these nor in any other parts of their mass has the least trace of any organic structure been detected. It is impossible at present to offer any conjecture as to their probable thickness. It must be many thousand feet; but its approximate amount, if ever ascertainable, will only be made out after the region where they occur has been mapped in detail These gneisses and schists possess a massiveness and rudeness of bedding which strongly distinguishes them from all the other and younger metamorphic rocks of Britain. They form nearly the whole of the Outer Hebrides, and occupy a variable belt of the western parts of the counties of Sutherland and Ross. Murchison proposed to term them the Fundamental or Lewisian Gneiss from the isle of Lewis-the chief of the Hebrides. Afterwards he called them Laurentian, regarding them as the equivalent of some part of the great Laurentian system of Canada.

In recent years Mr Hicks and others have endeavoured to show that in Wales there exist here and there protrusions of an old crystalline group of rocks from beneath the Cambrian system, and they have described these "pre-Cambrian" masses as overlaid unconformably by younger formations, as in the north-west of Scotland. Professor Ramsay, however, who with his colleagues in the Geological Survey mapped the Welsh areas in detail, contends that the supposed older gneiss is merely a metamorphosed portion of the Cambrian rocks.

CONTINENTAL EUROPE-On the continent of Europe numerous areas of ancient gneiss rise from under the oldest fossiliferous formations. In Scandinavia the structure of part of the country resembles that of the north-west of Scotland: the fundamental-gneiss (Urgneiss), covering a large area, is overlaid unconformably by red sandstones which underlie the most ancient strata containing organic remains. The gneiss and its accompanying rocks range through Finland into the north-west of Russia, reappearing in the north-east of that vast empire in Petchora Land down to the White Sea, and rising in the nucleus of the chain of the Ural Mountains, and still further south in Podolia. In Central Europe they appear as islands in the midst of more recent formations. In the midst of the Carpathian Mountains they protrude at a number of points, but westwards in the Alpine chain they rise in a more continuous belt in the central portion of these crests, and show numerous mineralogical varieties, including protogine, mica-schist, and many other schists, as well as limestone