origin of all kinds of rocks, and that without its assistance it would be and has been impossible to arrive at a natural classification of the igneous rocks.

For the purpose of examining a rock under the microscope, a thin chip or slice of it is taken, ground smooth, and polished on one side. The polished surface is then cemented with Canada balsam on to a piece of plate glass, and the other side is ground down until the section is of the required thinness and transparency. The Canada balsam is then melted, and the slice is carefully pushed off into a dish of turpentine, in which it is washed, so as to remove all traces of the emery powder and other substances used in the grinding and polishing process. The preparation is then mounted on a glass slide with Canada balsam and overlaid with a thin glass cover, care being taken to remove all air-bubbles from inside.

The advantages accruing from the microscopical examination of such slices are especially great in the case of rocks whose texture is so fine and close that their mineral composition could not be ascertained by examination with ordinary hand-lenses.

As Dr. Geikie observes: "A rock-section prepared in this way enables us to ascertain with precision the manner in which the different minerals are built into each other, and often throws a flood of light on the origin of a rock, and on the subsequent changes which the rock has undergone. It furnishes an opportunity of applying the delicate analysis of polarized light, and thus reveals points of structure in the composition of a rock which could not be ascertained in any other way."

For information regarding the methods of examining rock-slices under the microscope, the reader is referred to Geikie's "Textbook of Geology," p. 188, and Rutley's "Study of Rocks," p. 74.

Chemical Analysis.-Though in the case of crystalline and fragmental rocks, a microscopical examination will usually enable us to decide what minerals are present, yet it will not do so in the case of igneous rocks in a crypto-crystalline or

1 "Manual of Geology," Jukes and Geikie, 1872, p. 95.

vitreous state; neither will it reveal the actual proportions of the different minerals or elements present in stratified rocks. It is only by chemical analysis that we can ascertain the ultimate chemical composition of such rocks, and the relative proportion of the different elements they contain. Such information will enable us to determine what minerals can or cannot be present, and what compounds they would form if they had crystallized out, or we may compare the analysis with that of allied crystalline varieties, the mineral constituents of which are already known.

CHAPTER III.

ORIGINAL OR IGNEOUS ROCKS.1

A. Unaltered.

As these have solidified from a state of fusion by heat,

their present appearance will obviously be the result of two causes: 1, their chemical constitution; 2, the circumstances of cooling. To the former is due the presence of certain minerals and the absence of others; on the latter depends the structural condition of the rock, whether it is a glass, or has assumed a more or less crystalline structure.

For these rocks, then, which from the circumstances of their formation are called Igneous, the most natural basis of classification would seem to be that of their mineral composition. Each division thus formed will be either composed of certain minerals, or will consist of materials which are capable of forming these minerals. A further subdivision can be made of each group according to the crystalline condition in which the rocks occur, that is, according as the component minerals are more or less perfectly defined, or as the rock still remains in the state of a glass.

In order to carry these principles of classification into effect, it is necessary to make use of microscopical as well as of chemical analysis.

Mineral Composition.-All the minerals which ordinarily enter into the composition of igneous rocks are silicates, and free silica in the form of quartz is also frequently present. The total percentage of silica which they possess is sometimes as high as 80, sometimes as low as 40, or a little less; hence these rocks are often simply divided into acid and basic, according as they contain more or less than 60 per cent. of this mineral.

The chief mineral components of igneous rocks have

1 This chapter has been contributed by Prof. T. G. Bonney, F.R.S.

already been described, but it is seldom that many of these occur together in one rock, and certain laws or rules can be laid down regarding their paragenesis or association. We can easily understand that when the original molten mass contained a large percentage of silica, its crystallization would give rise to the formation of the more highly silicated felspars, and there might still be a surplus of silica which would remain as free quartz; while in a mass with a low percentage of silica, there might only be sufficient to form the less highly silicated felspars, without any surplus remaining. As a rule, therefore, free quartz is associated with the more highly silicated felspars, such Orthoclase, Albite, and Oligoclase; quartz may occur as an accessory mineral in company with Labradorite or Anorthite, but never as an important constituent of the rock.

Hornblende also is more commonly found in assocation with the highly silicated felspars, while Augite is generally associated with the more basic felspars, and with Leucite and Nepheline. Olivine, also, as might be expected, is always associated with the more basic felspars, and sometimes occurs alone with Augite and Enstatite. Hypersthene, a mineral less common than is usually supposed, has the same habit.1

The classification of igneous rocks adopted in the following pages is founded upon their mineralogical differences, or the association of different minerals in different rocks.

Structural Differences.-These afford the means of further subdivision, for we know, as a matter of experiment, that from the same materials, after fusion, may be obtained either a translucent glass or an opaque stone, built up more or less completely of crystalline particles, and by means of the microscope it is found that the matrix, or ground mass, of a rock may be in one of the following conditions:

A. It may be a true glass (when it is often termed the base). In this case, when a thin slice of it is placed between the crossed nicol prisms of a microscope with a

Recently, however, a mineral supposed to be hypersthene has been noted in association with some of the more highly silicated felspars.

polarizing apparatus, the field remains dark as the stage is rotated. Rocks with a glassy base are said to be vitreous. B. If small, faint, and rather ill-defined patches of light, like the ghosts of crystals, appear and disappear as the stage is rotated, so that no part remains dark throughout a whole revolution, the ground mass is said to be cryptocrystalline.

c. If the patches are more definite in shape and rather larger in size, sometimes showing bright tints, so that it is possible to identify the different component minerals, the mass is said to be micro-crystalline. In this state, however, the crystalline grains are often a little indefinite at the boundaries, and do not commonly exhibit the outline of their normal form.

D. If the slide is a mass of well-defined crystals, clearly distinguishable one from another, of which those first solidified give indications of their normal external form, the rock is then said to be crystalline.

The second of these conditions is in many cases the result of devitrification, the gradual change from a vitreous to a semi-crystalline state which a glass undergoes if it is kept for a considerable time at the point of incipient softening. In such case a crypto-crystalline rock might fairly be regarded as metamorphosed; but seeing that we are not yet able to distinguish satisfactorily between cases where the structure is original and where it is secondary, the separation is not generally attempted. Neither is it easy at present to separate rocks with this structure from those of the micro-crystalline group, so that we shall speak of both inclusively as semi-crystalline.

Glassy and semi-crystalline varieties are of frequent occurrence among the acid rocks, but are rare and more limited in mass among the basic rocks, and in some instances have not yet received distinctive names.

Conspicuous isolated crystals of any one or more of the component minerals may occur in glassy, semi-crystalline, or crystalline rocks, giving what is termed a porphyritic structure. The name Porphyry was first applied to the dark red or purple rock, speckled with light-coloured felspar, obtained from Jebel Dokhan in Egypt, and largely used by the Romans for decorative purposes. By some