It has been the current opinion that the earth was once in a molten state, and thence cooled to a solid condition, and hence that all the primitive rocks were igneous. Even those who think that the earth may never have passed through a molten state agree that the oldest known rocks are either true igneous rocks, or rocks of very similar nature. A molten magma may, therefore, be taken as the mother state of the rocks. Starting with this conception, the natural order of events suggests the inquiries (1) how rocks are formed from molten magmas, (2) what natures they assume, (3) how other rocks are derived from them, (4) how still other rocks are derived from these derivatives, and so on. To answer these inquiries is to trace out the generations of rocks and learn the general history of rock-formation.

(1) The process by which igneous rocks are formed from lavas is actually taking place in existing volcanoes. As these are widely scattered over the face of the earth, the material poured out by them represents different parts of the interior and varies in nature accordingly. This affords the means of studying the differences that arise from differences of material. This is a radical consideration, for variations in composition give rise to the most fundamental distinctions between rocks, though by no means the only ones. Rocks which have the same composition often differ greatly in texture or structure, owing to the varying conditions under which they were formed. In the solidification of rocks from the molten state, the rate of cooling causes many differences. A means of studying this is afforded by the various lava flows that are now being poured out on the surface under different conditions; but a more important means is afforded by extinct volcanoes, especially by those which have been deeply cut open by erosion. In certain very ancient volcanoes, not only have the solidified lava streams of the surface been cut across by erosion, but the lava that remained in the crater, or in the neck that led up from below, is laid bare for inspection. Exposures of even more profound nature have been made by the great disruptions which the outer part of the crust has suffered. In certain tracts there have been profound fractures, and the formations on one side of these have settled down and on the other side have been pushed up (faulted), so as to expose parts that were once much below the surface. Sometimes also the crust has been folded and crumpled, and the wrinkles thus formed have afterwards been worn away or cut open by deep valleys, and rocks that were once deeply buried have been laid bare. By the revelations made in these and other ways, it has been learned that at various times in the history of the earth molten matter has been thrust into fissures or intruded between layers of the crust and cooled there, without coming to the surface. Sometimes the lava appears to have forced its way into the rocks, and sometimes to have lifted the upper beds and formed great subterranean layers or tumor-like aggregates, called bathyliths and laccoliths (Fig. 334). Such intruded bodies of molten rock, solidifying under the varying conditions of such subterranean situations, are a fruitful source of instruction respecting the influence of varying rates and modes of cooling, as well as of other attendant conditions.

It will thus be readily seen that the rate of cooling of the various molten rocks must have differed very greatly. In the portions poured out upon the surface there were sometimes narrow streams and thin sheets, giving large exposure in proportion to the mass (Fig. 335), and sometimes thick flows and deep pondings in basins and choked valleys, giving massive bodies with relatively small surface exposure. There were explosions of the lava into minute particles with almost instantaneous cooling, and there were eruptions beneath the sea the peculiar effects of which are rather matters of inference than of positive knowledge. In the portions underground there were insinuations into thin fissures, on the one hand, and in-thrustings of thick bodies, on the other. Some intrusions entered the upper part of the crust where the rocks were cold and wet, and some were thrust into the deeper portions where the rocks were warmer and less penetrated by water. Sometimes the lava rose rapidly and was little cooled in passage, sometimes slowly with more cooling en route, and sometimes there were long halts between eruptions, with much opportunity to cool. An almost infinite variety of conditions is thus presented, and with it a rich field for the study of the modes of solidification.

In the underground intrusions the additional factor of high pressure was also present, and this is the third important condition in determining the nature of igneous rocks.

The three factors, composition, rate of cooling, and degree of pressure, require special consideration.

Composition of Igneous Rocks.

All or nearly all the chemical elements known on the earth are found in greater or less amounts in igneous rocks, and in a broad sense are constituents of them. If there are any exceptions, they are most likely to be found in the rarer elements in the atmosphere. Oxygen, nitrogen, hydrogen, aqueous vapor, and carbonic acid, which make up the mass of the present atmosphere, are all found in lavas and in their cooled products. Probably all the rarer elements also occur in igneous rocks. Helium is known to be given forth by springs.

Leading elements.—But although nearly or quite all the known chemical elements enter into the igneous rocks, only a few of them are abundant. These are regarded as normal or essential constituents, while the rarer substances are regarded as incidental. By combining a large number of the most trustworthy analyses of rocks of all sorts, F. W. Clarke[199] has estimated the relative amounts of the more abundant elements in the crust of the earth with the following result:

It will be seen that only eight of the elements hold a high rank in quantity. Many that are of the utmost importance in the history of the earth and the affairs of men are low in the list, or do not even appear in it at all, because their quantity is too small to be estimated in percentages. The precious metals, and even some of the more common metals, as lead, zinc, and copper, are too scarce to form an appreciable percentage.

Union of elements.—In a general study of the igneous rocks we may for the present neglect all but the first eight of these elements. Out of these elements spring various chemical combinations, and out of these combinations spring the various minerals, and out of the combinations of minerals come the various rocks. The union of oxygen with the other seven elements may be taken as a fundamental step in this series of combinations. The result is the following oxides: Silica (SiO₂), alumina (Al₂O₃), ferrous, ferric, and magnetic oxide (FeO, Fe₂O₃, and Fe₃O₄), magnesia (MgO), calcium oxide or lime (CaO), soda (Na₂O), and potash (K₂O). The oxygen sometimes unites in proportions different from those here given, but such exceptions may be neglected in a general study. We thus have nine leading oxides. Of these, silica acts as an acid, or more strictly according to the newer chemical view, as an acid anhydride. All the rest, except the magnetic oxide of iron, and sometimes the oxide of aluminum, act as basic oxides.

In the older chemical philosophy these oxides were supposed to combine by the simple union of an acid oxide with a basic oxide, and to remain as oxide joined to oxide; thus silica (SiO₂) and lime (CaO) formed silicate of lime (CaO,Si₂). The symbols express the idea better than the words. This method is used in the older geological works and in some of the later. But in the newer chemical doctrine, the oxides are not believed to remain so distinct after their union, and the symbols are written CaSiO₃, and the compound is named calcium silicate. According to the modern doctrine of solution, some of the calcium, silicon, and oxygen may exist as free ions in molten rock. The precise way in which the elements are related to each other in these compounds can scarcely be said to be known. For the general purposes of geology it is most convenient to think of these oxides as uniting in the simple fashion first named, and this involves no apparent geological error in general studies, since they are oxides when they enter the compound, and if the compound is decomposed they usually come forth again as oxides; but in closer studies more complex unions, attended by dissociations (ionization), must be recognized.

Formation of minerals.—As but one of the leading oxides that abound in an average magma plays the part of an acid, the silica, a very simple conception of the general nature of igneous rocks may be reached by noting that they are mostly silicates of the seven leading basic oxides—alumina, potash, soda, lime, magnesia, and the iron oxides. This general idea is a very useful one and represents a most important truth; but in its use we must not forget that there are many exceptions. Sulphur, phosphorus, chlorine, and other elements unite with the bases to form sulphates, sulphides, phosphates, phosphides, chlorides, etc. So also there are many minor bases that form silicates; and these minor bases unite with the minor acids to form many more or less rare minerals. Again, there are native metals in some igneous rocks. But altogether these hardly reach more than one or two percent. of the whole.

There are, however, two exceptions of more importance. In the molten magma the acid and basic elements are not always evenly matched. When there is an excess of silica, a portion remains free and takes the form of quartz (SiO₂). If there is an excess of the basic oxides, the weakest one is usually left out of the combination. This is commonly the iron oxide, which then usually takes the form of magnetite (Fe₃O₄). It is a singular fact that quartz often forms when there is no excess of silica, and magnetite when there is no excess of base. Quartz (free acid anhydride) and magnetite (free basic oxide) sometimes occur in the same rock. The explanation for this is yet to be found. These form rather important exceptions to the generalization that the igneous rocks are mostly made up of silicates, but, thus qualified, it expresses the essential truth, and has the merit of embodying the central chemical fact relative to these rocks.

Sources of complexity.—But here simplicity ends. As we pass on to the specific silicates that are formed, we encounter several sources of complexity. In the first place, the silica unites with the bases in different ratios and thus gives rise to unisilicates or orthosilicates (ratio of oxygen of bases to oxygen of silica, 1:1), subsilicates (ratio more than 1), bisilicates (ratio 1:2), trisilicates or polysilicates (ratio 1:3 or higher), and combinations of these. All the bases are not known to combine in all these ways, but many do in more than one of them. Still, if the silica were content to unite with each of the bases by itself alone, the results would remain comparatively simple; but instead of this it unites with two or more at the same time; and, more than that, it unites with them in varying amounts. The case would still remain measurably simple if these chemical compounds always crystallized out by themselves, each compound forming one mineral, and but one; but the different silicates have the confusing habit of crystallizing together in the same mineral. A crystal may thus sometimes be seen, under the microscope, to be made up of alternating layers of different silicates; e.g., a microscopic layer of an aluminum-calcium silicate may be overlain by a microscopic layer of an aluminum-sodium silicate, and the alternation may be repeated throughout the crystal, giving it a banded structure. There is reason to believe that this is true in many cases where the microscope fails to detect it, and that less symmetrical comminglings of silicates may take place. As such alternations or mixtures are not governed by any known mathematical law, as is the case in chemical compounds, there is no determinate limit to the number of combinations that may arise. As a matter of fact, new ones are still being discovered in the progress of research, and the total number that may ultimately be found can scarcely be prophesied.

As a result of all this fertility of combination, the total number of silicious minerals in igneous rocks is large. It is the function of the mineralogist to treat of these minerals as such. The geologist deals with them as constituents of the earth and as factors in its history. Only a few of them are so abundant as to require special individual notice in a general study of the earth. It may be remarked also that only a few of them can be identified by simple inspection as they occur in the rocks, partly because of the delicacy of the distinctions between many of them, and partly because of their minuteness and intricate intermixture. The resources of the polarizing microscope are necessary for safe determination in most cases. The student need not feel embarrassment or discouragement if he is often unable to recognize the constituents of the intimately crystalline rocks. Their determination has grown to be a profession by itself.

The leading minerals of igneous rocks.—Fortunately for the simplicity of geological study, a few minerals make up the great mass of the igneous rocks. These few are quartz, the feldspathic minerals, the ferromagnesian minerals, and the iron oxides. Quartz (silica, SiO₂) is the free acid already mentioned. The feldspathic and ferromagnesian minerals are the leading silicates of the earth’s crust, and vastly surpass all others in abundance. The feldspathic group embraces minerals formed by silica in union with alumina, together with either potash, soda, or lime, or two or more of these together. The ferromagnesian group embraces minerals formed by the union of silica with iron, magnesia, and lime, together with more or less of the other basic oxides. These statements are only true in a very general sense. Admixtures, replacements, and impurities are so frequent as to break down all sharp, simple definitions. The feldspathic minerals are normally light in color, ranging from white to red or gray. The ferromagnesian minerals are normally dark (commonly greenish) from the presence of iron, the great coloring element of rocks. But these color distinctions do not hold good in detail and cannot be much trusted as a means of identification.

The feldspathic minerals (p. 462) embrace the potash feldspars, orthoclase and microcline; the soda feldspar, albite; the lime feldspar, anorthite; and the mixed feldspars intermediate between albite and anorthite, viz., the soda-lime feldspar, oligoclase, the lime-soda feldspar, andesine, in which lime and soda are nearly equal, and the lime-soda feldspar, labradorite, in which the lime predominates; together with leucite, a potash silicate higher in alkali than orthoclase, and nephelite, a soda silicate higher in soda than albite. Leucite and nephelite are usually classified as feldspathoids, not as feldspars. It is to be understood that alumina is normally present in all these. Additional details respecting these minerals may be found in the reference list, p. 460.

Among the ferromagnesian minerals the most important are the pyroxenes, the amphiboles, and the biotite type of mica. Olivine is of subordinate importance. The pyroxenes (p. 465) and amphiboles (p. 460) have nearly the same chemical composition, but differ in crystallization and physical properties. Hornblende (an amphibole) has been melted, and on cooling under proper conditions found to take on the form of augite (a pyroxene). Pyroxene is sometimes altered into uralite, one of the amphiboles. The pyroxenes and amphiboles are the most abundant of the dark minerals in crystalline rocks. The leading members of the pyroxene group are augite, diallage, hypersthene, enstatite, and soda pyroxene. The chief members of the amphibole group are hornblende and the soda amphiboles. All are essentially silicates of magnesia and iron oxide, with or without the addition of lime, soda, and alumina. Details respecting these may be found in the reference list.

The two leading micas are the iron-magnesia mica, biotite, and the potash mica, muscovite, the familiar “isinglass” of the stove-door. Chemically, muscovite should go with the potash feldspars, but it is distinguished from them by its crystalline habit and physical properties. The biotite should go chemically with the pyroxenes and amphiboles, which it closely resembles except in its crystalline properties. Details respecting the micas may be found in the reference list, p. 464.

Two iron oxides, magnetite (Fe₃O₄) and hematite (Fe₂O₃) are widely disseminated in igneous rocks. They constitute the free bases already mentioned.

Summary of salient facts.—The salient facts are, therefore, (1) that out of the seventy-odd chemical elements in the earth, eight form the chief part of it; (2) that one of these elements uniting with the rest forms nine leading oxides; (3) that one of these oxides acts as an acid and the rest as bases; (4) that by their combination they form a series of silicates of which a few are easily chief; (5) that these silicates crystallize into a multitude of minerals of which again a few are chief; and (6) that these minerals are aggregated in various ways to form rocks. Possessed of these leading ideas, we are prepared to turn to the consideration of some of the conditions under which these combinations take place in the formation of rocks from molten magmas.

THE NATURE OF MOLTEN MAGMAS.

We easily fall into the habit of thinking of molten rock as we think of a molten metal, merely as a substance which has passed from the solid to the liquid condition because of high temperature. With the return of low temperature a molten metal returns to the solid state usually in the same molecular condition which it possessed before. The point of fusion and the point of solidification are the same and are rigidly fixed. If this were true of the constituents of a rock, a definite order for the solidification of the several minerals might be anticipated. As a matter of fact, the order is not the same under all conditions, and, what is especially significant, the order is far from being that in which the constituents would fuse or would solidify separately. For instance, in a granite composed chiefly of quartz, feldspar, and mica, the quartz is often the last to take form, although it is more infusible than the feldspar or the mica. This and other phenomena show that a molten magma is not to be viewed simply as a fused substance, but rather as a solution of one silicate in another, or as a solution of several silicates in one another mutually. The high temperature is to be regarded merely as a condition prerequisite to solution, or as the condition of fusion of some one constituent which then dissolves the others. If crystals of snow, sugar, and salt be mixed at a low temperature and compacted, the mass may be regarded as an artificial rock. On raising the temperature, all will pass into solution while the temperature is still somewhat below the melting-point of the snow, the most fusible, and while it is much below that of either the sugar or the salt. This particular case is instructive because the ice is not simply fused by temperature; the affinity of the salt plays a part. If the temperature were again lowered, the sugar and salt would not crystallize out at their fusing-points, but would remain in solution down to and even below the normal freezing-point of water; in other words, they would remain in solution until the water crystallized out and forced them to take the solid state. This holds good when the amounts of the sugar and salt are small relative to the water. If, on the contrary, their quantity is large relatively, crystallization will take place at higher temperatures and before the water crystallizes to ice. From this it appears that the salt and sugar might crystallize either before the water or after it, according to the degree of concentration. The behavior of mixtures of minerals in passing into and out of the molten condition appears to be quite analogous to this, and hence a great variety of results attend the process, dependent upon the number, the nature, and the relative quantities of the ingredients. The approved conception of the genesis of a rock from a molten magma (when ample time is given) is that one compound after another crystallizes out as the temperature falls and its point of saturation for each is reached, until the whole has been solidified. The modes of combination of the elements in the molten magma are not necessarily the same as those in the derivative crystals; indeed, the combinations doubtless change as the process proceeds; certain constituents being taken out, the remaining ones probably rearrange themselves.

Time required in crystallization.—The liquid magma of igneous rocks is essentially a fluid glass or slag. It is analogous to common glass, which is a silicate of potash, soda, or other base, except that usually common glass is relatively free from iron and other coloring substances, while these abound in the natural magmas and render them dark and more or less opaque; but the fundamental nature is the same, except that the natural lavas are usually mixtures of several silicates, while the artificial glasses consist of only one, or at most a few. Furnace slag is essentially an artificial lava.

When a lava is cooled quickly, the commingled silicates solidify in the diffused condition essentially as they were in the liquid; for there is no time for the silicate molecules of a like kind to come together, particle by particle, in regular systematic order, as required in crystallization. The essential feature of crystallization is this systematic arrangement of the molecules according to a definite plan, giving a specific crystal form, as a cube, a hexagonal prism, etc.

There are six (sometimes made seven) fundamental systems of crystallization, and a multitude of variations of special form in each system. The treatment of these forms belongs to mineralogy.

In a thick viscid liquid, this systematic arrangement of molecules into definite crystal forms takes place slowly, for the crystalline force in the silicates is far less energetic than that in water, which crystallizes into ice with much rapidity and with great force. Because of this slowness, the solidification of the lava may catch the process of crystallization at any stage. If the lava is cooled quickly, the result is a glass; if less quickly, part glass and part crystals; if slowly enough, all becomes crystalline. In general the slower the growth the larger the crystals. The solidification product may, therefore, range from a glass to a mass of crystals; i.e., it may be (1) wholly glass, (2) a glassy matrix with a few small crystals scattered through it, (3) a less abundant glassy matrix with more and larger crystals, (4) a mere remnant of glass in a mass of crystals, or (5) a mass of crystals with no glass.

Successive stages of crystallization.—Since eruptions take place intermittently, it is obvious that cooling of the lava may be in progress in its hidden reservoir during the quiescent intervals between eruptions. After a certain stage of partial crystallization has been reached during such time of quiet, a renewal of eruption may take place and the whole mass of lava be shifted into quite new conditions, and a second phase of solidification may be superposed on the one already started. The rock will then show two phases of crystallization: (1) large crystals of the kind or kinds most prone to develop in the given lava may have grown during the first long stage of slow subterranean cooling, while the greater part of the lava still remained liquid; and (2) small crystals or glass may have developed when the more rapid cooling under the new conditions took place. The result would be large crystals set in a matrix of small crystals or of glass, a combination styled porphyritic. In such cases the lava, in its later stages, carries the large crystals floating throughout its mass, and is not a simple liquid.

THE FRAGMENTAL PRODUCTS OF SUDDEN COOLING.

Pyroclastic rocks.—The extreme example of sudden cooling is presented when lavas are violently exploded into the air and solidify almost instantly. The resulting glassy particles or filaments, if small, constitute volcanic ash. The explosion appears to be due to steam and other gases which are held in the deeper lava under great pressure, but which, as they rise toward the surface of the lava where the pressure is relieved, expand with explosive violence. It is probably also due in part to progressive crystallization, which forces the gases out from the part that crystallizes and overcharges the rest. Sometimes the projected particles draw after themselves long filaments like the threads of spun glass, and sometimes while in the air they divide and draw apart, spinning a filament of viscid lava between them. A variety of this kind at the volcano of Kilauea in Hawaii is known as “Pele’s hair.” These light filaments drift with the wind and lodge on the lee side of the volcano, covering the surface “like mown grass” (Dana).

When the exploded fragments are coarser they fall about the volcanic vent and form the tuffs (tufa) of which most steep volcanic cones are chiefly built. In these larger fragments, crystals are not infrequently found, and the same is even true of the volcanic ash. These crystals are undoubtedly such as had already been formed in the lava before it exploded, and their formation, as suggested above, may have contributed to the explosion.

Fragments too large to be borne far away by the air, but still small, are known as lapilli, especially if they are somewhat rounded and gravel-like. A finer variety, of the nature of sand, much used in making Portland cement, is locally known as puzzolana.

The rougher, irregular fragments of a clinker-like nature ejected by volcanoes are known as scoriæ or cinders. They are more or less distended by gas-bubbles and are hence light and pumiceous.

The larger masses of lava ejected into the air are often caused to rotate by the unequal force of the projection, or by the unequal friction of the air, and to assume spheroidal forms, the internal gases at the same time often expanding and rendering the mass vesicular. These rounded projectiles are known as volcanic bombs (Figs. 336 and 337). Balls of lava that have originated in rolling movements of the seething mass, or in other ways, are also styled bombs. Usage is not altogether harmonious or consistent in the application of the term.

The larger masses that are projected into the air are more or less vesicular from the expansion of included gases, as already noted, and so the fragmental products of volcanic action grade into the vesicular. The type of this class is pumice, in which the gas cavities make up by far the larger part of the volume of the whole mass, and the whole is reduced to the condition of a solidified froth or foam. So thin are the dividing films of glassy material in some cases that the whole is pure white, though the same material in solid mass would be dark. This solidified glassy froth is often lighter than water and floats freely on the sea until it becomes “water-logged” and sinks. Dredgings of the deep sea show that much pumice has accumulated there, and being far from the land has escaped burial by the sediments borne in by the rivers.

All of these fragmental rocks produced by volcanic action are known as pyroclastic (fire-fragmented) rocks, a general term of much convenience in distinguishing them from lava-flows, on the one hand, and from the fragmental rocks produced by air and water (ordinary clastics), on the other.

THE GLASSY ROCKS.

The solid glasses.—The quick cooling of lava-flows into solid glasses is chiefly dependent on their exposure at the surface. Hence it is often the case that the exterior of a lava-flow is glassy in greater or lesser degree, while the interior is more or less crystalline. Quick cooling is sometimes also due to the intrusion of the lava in thin sheets into fissures in cold rocks. When massive bodies of lavas penetrate solid rocks, the lava does not usually cool so fast as to prevent some degree of crystallization, and the crystallization may even become complete; but if the intruded lava sheet be very thin, the lava is liable to be cooled to a nearly perfect glass. The glassy condition is, therefore, subject to indefinite gradations. As a rule, the acid lavas are stiffer at the same temperature than the basic ones, and crystallize more slowly, so that acid glasses are more common than basic ones. The basic rocks usually crystallize pretty thoroughly, except on the immediate surface of the flows.

The first stages of crystallization.—The microscopic study of the volcanic glasses reveals great numbers of minute forms known as crystallites, microlites, globulites, etc., that appear to be first steps in crystallization, though many of them do not take definite geometrical shapes and some do not show the optical characters of crystals. There are minute globules (globulites), needles, and hair-like bodies (trichites) of more or less indeterminate nature, together with other forms that can be seen to be certainly the initial forms of well-known minerals.

The obsidians.—Of the compact glassy rocks, obsidian is the best type. It is essentially a natural glass, formed usually of acid silicates. It has the close texture, conchoidal fracture, and other qualities of glass. It is usually black, but sometimes red, brown, purple, bluish, or gray. While chiefly of glass, it usually contains more or less of the incipient crystals above described, showing that even here the first step in the gradation to the next or the crystalline stage has been taken. These incipient crystals sometimes become so abundant as to change the texture from the vitreous to the stony order. In some cases, the stony texture seems to have been developed in the obsidian after it was formed, the change being a part of a subsequent process of devitrification, but in other cases the crystals seem to be original. Besides these, there are often small globular bodies known as spherulites.

Varieties of glassy rock in which the embryo crystals are more numerous and the glassy texture less perfect, are known as pitchstones. The fresh surfaces of these have rather the aspect of pitch or resin than that of true glass; hence their name. Like the obsidians, they are usually dark, but they take on greenish, brownish, yellowish, and light-colored hues as well. Sometimes glassy rock fractures in small spheroidal forms like pearls, and is known as perlite. Basic glasses are relatively rare, and while usually included under the term obsidian, are sometimes given special names.

SPECIAL STRUCTURES.

Flow structure.—Lavas that cool into glassy rocks frequently contain gas cavities, colored spots and variations of texture which, together with the hair-like embryo crystals, are drawn out into lines, streaks, and parallel belts by the flow of the viscous mass, giving rise to rhyolitic or flow structure (Figs. 338 and 339). Rocks in which this is the most pronounced feature were formerly known as rhyolites, though the term has drifted away from this original meaning and has been applied to a class of acidic rocks. The obsidians and pitchstones may be more or less rhyolitic under the microscope, though to the naked eye they may appear only as a glassy or resinous mass. The rhyolites generally have but an imperfect glassy texture, since the crystals and the cavities sometimes make up a notable part of the mass, the glassy portion being scarcely more than a matrix in which the crystals, spherulites, and cavities are carried. By an increase of the crystals in number and size, the rock passes by gradations into porphyry or phanerite.

Amygdaloids.—In lava-flows the included steam often collects in bubbles near the surface as the lava cools and forms a vesicular portion with a scoriaceous texture (Fig. 341). In its upper part, the vapor bubbles may be numerous, while below they become more and more scattered until they disappear. Similar bubbles are also often found near the bottom of a sheet of lava. This is perhaps due to the rolling under of the frontal surface of the lava-stream as it flows. Later, these cavities often become filled with minerals deposited from solution and the rock then becomes an amygdaloid, but this filling is a secondary action.

THE PORPHYRITIC ROCKS.

When the conditions are such that after a part of the magma has formed distinct crystals floating in the remaining liquid lava, there is a change which causes the rest to solidify as a glass or as a mass of small crystals, the structure is known as porphyritic, and the rocks possessing it are called porphyries. This differentiation into distinct crystals set in a ground-mass of minute crystals or of glass often gives a mottled or variegated aspect to the rock, especially if the matrix of glass or minute crystals differs in color from the distinct crystals. This structure is much oftener developed in acidic rocks than in basic ones, because the latter crystallize more readily. The most common porphyritic crystals are feldspar and quartz, though they are by no means the only ones. The matrix is also usually felsitic or quartzose, but not necessarily so. The character is a structural one, and is not dependent upon any special chemical or mineralogical constitution. The distinct crystals are known as phenocrysts, and the varieties of porphyries are named from the characteristic phenocryst, e.g., quartzophyre (quartz-porphyry) if the conspicuous crystals are quartz, orthophyre if orthoclase, augitophyre if augite, etc. A convenient classification has recently been proposed[200] into (1) leucophyre (white porphyries), which have a light-colored ground-mass set with phenocrysts of any kind, and (2) melaphyres (black porphyries), which have a dark-colored ground-mass, with phenocrysts of any kind. While it is to be hoped this usage will prevail, it is to be noted that these terms, especially the latter, have been used in a different sense. (See reference list of rocks, p. 445.)

In many cases the ground-mass itself becomes minutely crystalline and the porphyritic aspect is due simply to large distinct crystals set in a mass of minute obscure ones. The rock is then really holocrystalline, but the term porphyry is applied to it. In other rocks the crystals of the ground-mass become more and more distinct, the porphyritic aspect gradually disappears, and there is a graduation into the next class.

THE PHANEROCRYSTALLINE ROCKS.

The phanerites.—When time enough is given for the cooling process the molten magma becomes completely crystalline. The holocrystalline rocks hence include a large series, ranging from the most acid to the most basic. In this class the differentiation of the rock material and the formation of distinct minerals reach a high stage, and as a natural result the varieties of rock are numerous. Taken as a group they are phanerites. If they are to be more particularly characterized, it is usually done on the basis of the minerals of which they are composed. The following are the leading types, beginning with those which are rich in silica and poor in basic oxides, and ending with those which are rich in basic oxides and poor in silica.