an increase of pressure against and from the retaining walls, which pressure we have assumed tends to decrease the mobility of the molecules. And of course the actual mobility or liquidity of the magma will depend on the relation between these two tendencies, which, if balanced, will result in the storing up of potential mobility, which would show itself should the pressure be relieved.

Heated, hydrated magma under pressure; unstable rigidity. -We may then imagine a heated, hydrated magma within the earth under such a pressure that it has become solid, the rigidity being an unstable one; and we may further assume that this solidification is either to an amorphous, glassy mass, or to a more or less crystallized one, that is, to one which is wholly crystallized to an aggregation of minerals, or to one which is made up of crystallized minerals and rigid amorphous material; a glassy, porphyritic mass. Under these conditions the solid magma will possess a potential mobility or liquidity, which will exert itself to melt the solid mass, if the pressure is relieved or lessened, as may happen when the fracturing of the earth's crust ruptures the retaining walls and permits the escape of this pent-up body, stored with gigantic expansive energy.

Refusion. The melting up of an unstable, solidified mass by the heat inherent in it would undoubtedly proceed differently in the different minerals or the glass composing it, some fusing more rapidly than others. The difference would be the greatest between anhydrous refractory minerals, like quartz, and hydrated glass, through which they might be scattered porphyritically. If the refusion is complete, nothing will remain to indicate a former state of solidification. If the process of fusion is checked by the cooling of the magma in consequence of its eruption through cooler rock masses, there might remain in the cooling magma reinnants of the minerals previously existing in it.

Final consolidation.—Moreover it is evident that the nature of the minerals of final consolidation will be affected by the nature and amount of the minerals formed at the time of unstable consolidation which remain unmelted. If these are highly acid the minerals of final consolidation will be proportionately basic, and vice versa.

Application to quartz-bearing basalt.-Applying the foregoing general considerations to the occurrence of porphyritic quartz grains in basalts, it seems reasonable to suggest that the production of extremely acid and basic silicate ninerals in deep-seated magmas may have been brought about, like their production in certain magmas after they have reached the surface, by the influence of absorbed water acting under favorable conditions of pressure and temperature, which combined to

solidify the magma more or less completely for the time being, but which, as the quartz grains themselves show, was an unstable solidification, which subsequently yielded to the potential liquidity of the magma, resulting in the partial resorption of the quartz crystals before the final consolidation of the rock to its present form.

Reyer, (1. c. p. 166), suggests that pressure and different degrees of saturation with absorbed water may lead to metameric processes. And also states that the development of quartz in rock magmas requires a considerable saturation of the magma with water.

Prof. Lagorio (1. c.) refers the concurrence of quartz and olivine in the same rock to the super saturation of-the magma with silica and magnesia, but this idea of itself is not sufficient to account for the occurrence of quartz in magmas with the normal basaltic composition, where it generally does not occur, as it is not sufficient to explain the occurrence of fayalite in rhyolitic obsidian, having 75 per cent of silica and less than 2 per cent of iron oxide.

Confirmatory observations.-If the foregoing explanation which refers the production of the quartz in these basalts to physical conditions apart from chemical ones is correct, we should expect to find such anomalous associations of minerals in other varieties of volcanic rocks, and should not expect to find a necessary correspondence in the chemical composition of all basalts which carry porphyritic quartzes. Nor should we expect to find the quartz-bearing varieties, which are exceptional, necessarily holding a definite relation in point of age to the other volcanic rocks with which they are associated. These expectations, I think, are realized by the following observations.

Porphyritic quartz in other volcanic rocks.-The first point is beautifully illustrated in the suite of rocks in the collection from New Mexico, for it shows that similar quartz grains occur in almost all of the varieties of volcanic rocks from this region, and that their occurrence is not uniform throughout the series. Thus in most of the rhyolites rounded grains of quartz are very abundant, but in some specimens they are absent (obsidian and lithoidite). In the single specimen of mica-andesite they are wanting. In the hornblende-mica-andesites, some specimens show a considerable number of quartzes; one, a few, and others, none. In the hornblende-pyroxene-andesites, one specimen shows many quartz grains; others, considerable; and some, none. Of the five specimens of pyroxene-andesite, one shows a few grains of quartz; the rest, none. And of the twelve specimens of basalt, seven show much quartz; the others, none.

Another group of volcanic rocks, specially characterized by abundant rounded grains of quartz, occurs in the vicinity of Crescent Peak, Colorado. The group embraces basalt, andesite and possibly trachyte. These all bear rounded grains of quartz, and some of the olivine-bearing varieties also carry hornblende paramorphs, which furnish additional evidence of a change of physical condition from one which induced the crystallization of certain minerals, to a later one, in which they were partially resorbed.

Chemical similarity of basalts with and without quartz.That the chemical composition of quartz-bearing basalts is not characteristic of a particular modification of rock magma will be seen from the accompanying analyses:

I. Quartz-bearing basalt, Rio Grande Cañon, N. M. (L. G. Eakins.)

II.

IV. Basalt without quartz,

V. Quartz-bearing basalt, Cinder Cone, Lassen's Peak, Cal.

brand.)

(W. F. Hille

VI. Quartzose diorite. Electric Pk., Yellowstone Park. (J. E. Whitfield.)

The first three are of three forms of quartz-bearing_basalt from Rio Grande Cañon; the first is a light gray dense basalt; the second, a greenish black, dense basalt; and the third, a dark red vesicular basalt. They have practically the same composition with slight variations. The higher oxidation of the iron in the red rock is indicated by the high percentage of ferric oxide in analysis III.

The fourth analysis is of a gray, dense basalt from Rio Grande Cañon, which resembles the basalt from which the first analysis was made, except that it exhibits no quartz, either in macroscopic or microscopic grains. These four are normal basalt analyses resembling one another as closely as analyses of

similar rocks usually do. There are no greater differences between the analysis of the variety without quartz and those of the quartz-bearing varieties than there are between the analyses of the three quartz-bearing varieties. So that the occurrence of the quartz in this instance cannot be ascribed to anything exceptional in the chemical composition of the magma.

Chemical differences between basalts with quartz.-The fifth analysis is that published by Mr. Diller in the paper already cited. It shows a different chemical composition for this form of quartz-bearing basalt, which is more acid than typical basalt, and corresponds more closely to some andesitic forms of volcanic rocks.

It is probable that additional analyses of other quartz-bearing basalts will show as great a variation in their chemical composition as exists between that of basalts free from quartz grains.

Different mineral development of chemically similar magmas.-The sixth analysis is presented for comparison with analysis V. It is that of a magma of very nearly the same chemical composition, slightly more basic, which has consolidated under different conditions. It may serve to illustrate two points: first, the mineralogical extremes to which chemically similar magmas may be developed. Second, the possi bility of a basaltic magma having existed at some previous period in a condition of unstable consolidation, in which quartz might have been crystallized out. Analysis VI is of a coarse grained, quartzose diorite, perfectly fresh and unaltered, of quite recent geological age, and which is composed of plagioclase feldspar, quartz, hornblende, biotite and pyroxene, with accessory magnetite, apatite and zircon. The quartz is in considerable quantity, very much more than the amount of quartz observed in quartz-bearing basalts. The discussion of this diorite is reserved for another paper, which is in process of preparation.

Summary.

The principal points brought out in this paper may be briefly stated as follows:

The quartz-bearing basalt from Rio Grande Cañon belongs to a series of volcanic rocks, characterized by a variable amount of porphyritic quartz in rounded grains.

These quartzes are primary crystallizations from the molten magma, and exhibit no definite relation to its chemical composition.

Their production is to be referred to certain physical conditions attending some earlier period of the magma's existence. From analogy with the occurrence of iron olivine in rhyolitic obsidian, it seems probable that the formation of primary quartz in basalt took place under the influence of water-vapor at a great pressure.

ART. XXI.-Mineralogical Notes; by GEO. F. KUNZ.

1. Phenacite from Maine.

IN May, 1888, some crystals of phenacite were found near Stoneham, Me., in a vein of coarse albitic granite,* associated with crystals of smoky quartz, topaz and muscovite. Some of the crystals were implanted on smoky quartz. A few of them were attached to the matrix by one of the rhombohedral faces so loosely that they could be removed without being broken. They were about thirty in number, lenticular in shape and measured from 3 to 12mm across, and from 1 to 3mm in thickness. They were all white or colorless, and had polished faces; the form being for the most part very simple. Figure 1 of the series from Pike's Peak, Colorado, described by Penfield,† is an exact counterpart of the more highly modified form.

Of the topaz there were found about a dozen crystals which had unfortunately been broken from the gangue. They were colorless, light-green or cherry colored on the outer sides and colorless in the center. The largest crystal measured 13 inches in height and thickness. Almost all the crystals contained irregular hollow spaces from 1 to 10mm across. In habit the crystals closely resemble those from Cheyenne Mountain, Colorado.

It is through the courtesy of Messrs. E. D. Andrews, T. F. Lamb and N. H. Perry, that I have obtained the crystals and the facts in regard to their occurrence.

2. Quartz Pseudomorphs after Spodumene.

In 1887, at the spodumene locality at Peru, Maine, which has furnished tons of that mineral for commercial purposes, there were found some crystals in which the original spodumene had been almost entirely replaced by white quartz, with the exception of a white core of crystallized albite. These crystals are remarkable for the sharpness of the striated prismatic faces; the terminations are not so distinct. The surface of the crystals is covered with a coating of damourite. The alterations of quartz after spodumene are fully described by Julien, from the granite veins of Hampshire County.

3. A remarkable variety of transparent Oligoclase.

In December, 1887, some specimens of feldspar were sent to me for examination by Mr. Daniel A. Bowman, who ob*This Journal, III, xxvii, 212. + This Journal, III, xxxiii, 131.

Annals New York Acad. Sciences, vol. i, No. 10.