browncoal into common coal; common coal into anthracite; and anthracite into graphite. Gabbro or granite passes over into serpentine.

These last-mentioned transitions or transmutations are such in the strictest sense of the word, having been occasioned by changes of the rocks' substance in the course of time; whereas the term transition, as applied to the two former classes is only a conventional term for a progressive series of rocks, all of which were from the first different from each other, and remain so.

It need hardly be said that these several transitions and transmutations multiply not a little the difficulties of nomenclature and classification, and frequently render the desired precision and accuracy impossible. We are always driven back to this--that every name applied to a rock can only be considered as establishing an especially characteristic form of its development as a kind of centre point, which, however, in nature is surrounded by numerous varieties and derivative forms of more or less doubtful character.

PART II.

THE ROCKS.

INTRODUCTORY CHAPTER.

CLASSIFICATION.

A SCIENTIFIC classification of rocks is a task of more difficulty than might at first sight appear; as yet, no one has succeeded in producing a perfectly consistent and comprehensive system. Not only do the nature of the subject and our own imperfect knowledge present many serious obstacles to consistent arrangement; but in many cases established usage and nomenclature, too firmly rooted to be lightly disturbed, prevent our changing an old classification even when based on error.

Even were our knowledge far more certain than it is, and were we free to overthrow all previous errors and misconceptions, we could not lay down a logically complete system of classification to embrace all rocks, on any principle, whether of ORIGIN, TEXTURE, or COMPOSITION (chemical or mineralogical). We do not find the mineralogical differences between rocks coincide with those of their chemical composition, nor are either of those dependent on geological position or stratification. There are no rigidly defined classes in nature.

The student must not, therefore, expect too much from any system; but, as we are driven to choose some basis for arrangement of our subject, we consider, on the whole, that the best scheme for our purpose will be one in its general features coinciding, as far as possible, with what we know of the origin of the various rocks, making use, however, of the distinctions arising from differences of

texture, composition, or otherwise, for the subdivision of our subject, as the nature of each case may seem to render advisable.

A great number of distinctions have been established by custom between many rock formations, which in truth do not differ from each other very materially. These we shall as far as possible drop, endeavouring to make uniform connected groups, and treating many rocks, which have hitherto been known by different names, as varieties only of one and the same rock. On the same principle, we avoid as far as possible the introduction of new names for rocks. It is impossible, and would be unprofitable, to dignify every slight modification of texture or structure (perhaps only of local occurrence) by a separate name. Even in treating the most important and prevalent rocks, we should seek to confine our nomenclature to their most characteristic forms of development, establishing these as central points of departure, from which manifold transitions are found leading towards other central points in the next group of rocks. One observer may pronounce a doubtful rock to be granite, which another will call a gneiss, without our being able to say that one is right and the other wrong. In cases of this kind there constantly arises the temptation to give new names, but in the interest of science this temptation should be resisted as far as possible.

The following are the general heads under which we have grouped the rocks in this work.

I. IGNEOUS ROCKS* (Eruptive Rocks), all of which are most probably products of igneous fusion.

* The term 'IGNEOUS ROCKS' is used throughout this book as the equivalent of the German ERUPTIV-GESTEINE. The Germans object to the term 'igneous,' as conveying the idea of fire or burning (which could not take place in the absence of air), and also because the metamorphic rocks may have been subjected to heat as well as those we call igneous. Most of our rocks have, however, been named in an imperfect state of knowledge of their origin, and with reference to erroneous ideas; and if we are agreed on the signification of a term, we need not always go back to its derivation. Mr. Jukes objects to the German term eruptive,' as applied indiscriminately to these rocks. He thinks that in speaking of the plutonic rocks, we should use the terms, irruptive' or 'intrusive,' &c., as they did not, or are not supposed by us to have reached the surface at the time of their upheaval.

A. Rocks poor in silica, or basic rocks.

(a) Volcanic.-Of which the BASALTS are the principal representatives.

(b) Plutonic. Of these the principal representatives are the so-called GREENSTONES (diabase, diorite, &c.).

B. Rocks rich in silica, or acidic rocks.

(a) Volcanic, e. g. the TRACHYTES. (b) Plutonic, e. g. the GRANITES.

II. METAMORPHIC CRYSTALLINE SCHISTS.- Most probably the product of the transmutation of sedimentary rocks, but in respect of their mineralogical composition closely allied to the igneous, e. g. GNEISS, MICA-SCHIST, CHLORITE-SCHIST, &c.

III. SEDIMENTARY ROCKS.-The products of deposit. 1. Argillaceous rocks, such as CLAY and ARGILLACEOUS

SHALE.

2. Limestone rocks, such as LIMESTONE and DOLOMITE (including gypsum and anhydrite).

3. Siliceous rocks, e. g. SANDSTONES and CONGLOME

RATES.

4. Tufa formations.

The above are the groups of principal rocks which occur in masses of great extent.

IV. We shall next range those rocks of less frequent occurrence, or which only form subordinate strata or separate beds, and whose origin is in part still doubtful, without attempting in their case a logical classification. To this series belong, for instance, many silicates, the CARBONACEOUS ROCKS, the IRONSTONES, SERPENTINE, &c., and some other rocks of problematical character.

V. Finally we shall instance those rocks which are essentially composed of one mineral, such as QUARTZ, OPAL, &c.

The first book on rocks, at that time a most masterly treatise, was von Leonhard's Charakteristik der Felsar

We have, however, kept to the term ' eruptive' as a general term for describing the action of all igneous rocks; and any other course would have compelled us to put a construction of our own on the origin of each rock, although not in the mind of our author, and unnecessary for his immediate purpose.-TRANSLATOR.

ten' (1823). In it is to be found a reprint of Alexander Brongniart's Classification minéralogique des roches mélangées,' which had appeared in the 34th vol. of the Journal des Mines.

The following are the most important among the more recent works on this subject:

Naumann's Geognosie, vol. i., a second edition of which appeared in 1858.

Senft's Classification der Felsarten, 1857, in which the rocks are arranged with special reference to one or more characteristic ingredients.

Durocher, Essai de Pétrologie comparée in the Ann. des Mines, 1857; ii. pp. 217 and 676. He separates the igneous rocks, in the same way that Bunsen did before him, into acidic and basic rocks. He subdivides these again according to the degrees of their acidity or basic composition; these subdivisions nearly correspond with Bunsen's 'Mittelgesteine.' G. Bischof has examined and pronounced upon a large number of rocks from a chemical point of view. The arrangement of the separate treatises in his Lehrbuch der Geologie appears to be entirely accidental, and the geological relations of the rocks are hardly regarded.*

Rammelsberg's Handwörterbuch der Mineralogie (with supplements) contains numerous analyses.

Roth has recently attempted to collect all the known analyses of rocks and to arrange them according to fixed principles, accompanying them with critical remarks.

Having referred the reader to the above-named comprehensive works, we shall abstain from again quoting them in detail at the mention of each different rock. In dealing with the particular views of their several authors we shall only give the name of the author in question.

On the other hand, we shall have occasion to cite at the proper places the valuable treatises of Abich, Bäntsch, Bergemann, Blum, Breithaupt, Bunsen, Delesse, Deville, Ehrenberg, Fischer, Girard, von Hochstetter, Hochmuth, Jentsch, Knop, List, Naumann, Oppermann, vom Rath, G. Rose, Freiherr von Richthofen, Scheerer, Söchting, Stache, Streng, von Walterhausen, &c.

* In the second edition (translated into English) there is much provement in this respect.