Geological Survey 12th Annual Report (Part I) The Eruptive Rocks of Electric Peak and Sepulchre Mountain, Yellowstone National Park

GEOLOGICAL DESCRIPTION.

East of Electric Peak, across the deeply cut valley of Reese Creek, lies Sepulchre Mountain, so called from a mass of breccia on one of its high northwest spurs, which resembles a sarcophagus. The mountain rises to a height of 9,600 feet, and stands isolated from the surrounding peaks, from which it is separated by geological faults and also by deep drainage channels. It is composed of volcanic breccias and massive lavas that form a body of rock 3,000 feet thick, resting on Cretaceous and older strata.

The southern and southwestern slopes of the mountain are rounded from the action of the ice which has passed over the mountain from the Gallatin range. They are mostly covered with grass and sage brush, and present comparatively few rock exposures. This is also the character of the hills and ridges southwest of the mountain, which lie east of the fault at the base of Electric Peak, and form part of the geological body of Sepulchre Mountain.

The north and east faces of the mountain are precipitous and rocky, and afford excellent sections of the volcanic mass. The long northwestern spur is also rugged, and exposes the geological structure of this part of the mountain. This difference of surface character is shown by the illustration, Pl. LII, from a photograph of the north and west sides of the mountain.

Pl. LII. SEPULCHRE MOUNTAIN, FROM ITS NORTHWEST SPUR. (click on image for a PDF version)

The breccias exhibit little or no evidence of bedding, and are associated with flows of lava, the whole having a distinctly volcanic character. The western portion of the breccia is traversed by numerous dikes of andesite and dacite, which trend for the most part in a north and northeast direction from the vicinity of Cache Lake. A few trend east. The distribution and location of these later intrusions are shown on the map, Pl. LIII, in a general way. It is probable that there are a number of dikes cutting one another in the southwestern portion, rather than a few broad bodies, as represented on the map; the data were not sufficient to locate the different bodies, and the map has been drawn so as to represent what seems to have been the order and position of the eruptions.

Pl. LIII. GEOLOGICAL MAP OF ELECTRIC PEAK AND SEPULCHRE MOUNTAIN, YELLOWSTONE NATIONAL PARK. (click on image for a PDF version)

In the northwestern spur of Sepulchre Mountain the dikes are well marked, and stand out prominently from the surrounding breccia. They are from 5 to 25 feet wide, and are not perfectly straight, but maintain a generally uniform direction and can be traced by the eye for some distance. Long after the eruption of these dike rocks, when the region had been faulted and erosion had removed a great part of the rocks, and had cut the valley of Glen Creek, a flow of rhyolite flooded the country and filled this valley, covering the south and west base of Sepulchre Mountain to an altitude of about 8,100 feet. The rhyolite has been almost entirely removed, but remnants of the sheet are found in numerous places. This closed the series of volcanic events, as they are recorded in this vicinity, though in other parts of the region the rhyolite was followed by eruptions of basalt.

THE VOLCANIC ROCKS AT SEPULCHRE MOUNTAIN.

The volcanic rocks composing Sepulchre Mountain consist of andesitic breccias and tuffs with lava-flows of the same andesites, besides dikes of andesite and dacite. By far the greater part of the material is tuff breccia, which is easily separable into an older and a newer, or a lower and an upper breccia.

THE LOWER BRECCIA.

The lower breccia, which is about 500 feet thick, is light colored and is characterized by phenocrysts of biotite and hornblende. It carries a great amount of fragments of Archean schists which are not found in the overlying dark colored breccias. The lower breccia passes into tuff in places, containing fragments of carbonized wood; and at the extreme end of the northwest spur it is distinctly bedded with bowlders of foreign rocks scattered through layers of fine grained material. In places the upper portion of this bottom breccia is green and partly altered, as though it had been weathered before the upper breccia was deposited on it. In the northwestern spur of the mountain the upper breccia is distinctly seen to rest on an uneven surface of the lighter colored bottom breccia.

It is probable that the bottom breccia was thrown from some neighboring Archean area, and is considerably older than the overlying, basic breccia. This relation between a bottom breccia of hornblende-mica-andesite carrying Archean fragments, and overlying, basic breccias is found to exist in other places in this region.

An examination of the various specimens of this older breccia shows that it varies in mineral composition as well as in color and microscopical habit. It is mostly light colored, gray, white, and red. In places it is dark colored. Some varieties carry abundant large phenocrysts, others contain a multitude of small ones. Though the great bulk of it is characterized by porphyritical biotite, hornblende and plagioclase, some portions are poor in biotite, and are hornblende-andesite, while other parts approach dacite in composition, having biotite and quartz phenocrysts with those of plagioclase. The groundmass of the different fragments making up this breccia varies from glassy and microlitic to microcrystalline.

Associated with the bottom breccia at the northeast base of Sepulchre Mountain is a vesicular basalt with porphyritical augites and decomposed olivines. It is of small extent, is amygdaloidal, with quartz, agate, and calcite. Its exact relation to the breccia was not discovered but it appears to be an older basalt, more intimately connected with the bottom breccia than with the upper breccia. It does not resemble the younger basalts on the north side of the Yellowstone River opposite Sepulchre Mountain.

THE UPPER BRECCIA.

The upper breccia, which lies upon the one just described, is dark colored at its base, where it consists almost wholly of pyroxene-andesite with little or no hornblende. Many of the fragments are finely vesicular and basaltic in appearance, without macroscopic phenocrysts. At the south base of the mountain this breccia is accompanied by vesicular flows of pyroxene-andesite, with large porphyritical pyroxenes and feldspars. Intimately connected with this breccia is that of hornblende-pyroxene-andesite, which forms the uppermost portion of the mountain. They appear to grade into one another by an increase in the amount of hornblende. This later breccia is also accompanied by vesicular flows of the same kind of andesite. It is mostly lighter colored, though some of it is quite dark, with prominent hornblendes, and has an andesitic habit and not a basaltic one. There is no evidence of a geological break between the lower and upper portion of the upper breccia, which may be considered as a continuous geological body, made up of fragments and flows of andesite, which have been ejected from a common source during a prolonged series of eruptions.

The andesitic material composing it varies in mineral composition and outward appearance between certain limits. This variation will be described in detail.

TABLE XII.—Mineral variation in the upper breccias of Sepulchre Mountain.

Thirty-two thin sections have been made from the upper breccia, which upon investigation resolve themselves into a series of glassy andesites, some of which carry hypersthene, augite, and plagioclase phenocrysts, and others hypersthene, augite, hornblende, and plagioclase. They may be arranged, as in Table XII, according to the relative abundance of these minerals.

The first two specimens are very basaltic in appearance, with small phenocrysts. They carry a few decomposed crystals, which were probably olivine, but have the microstructure of the associated andesites.

The varieties without hornblende, B₁—that is, the pyroxene-andesites—have a groundmass of globulitic brown glass filled with microlites of feldspar, pyroxene, and magnetite grains. The shade of brown varies from very dark to light, and the size and abundance of the microlites also varies. The feldspar microlites are plagioclase with rather low extinction angles.

The pyroxene always includes hypersthene and augite. They have very much the same general appearance and habit, both occurring in some instances in large crystals, and in others in small ones. In the same rock section they differ as to the character of their crystal outline. Some individuals are bounded by crystal planes, especially in the prism zone, while others are rounded, particularly at the ends of the crystals. Some have rough surfaces, with multitudes of irregularly shaped tongues of glass penetrating the surface of the crystal. This does not appear to be the result of a corrosion of the crystal, but of a rapid crystallization of the mineral at the end of its growth, when the surrounding glassy magma was becoming filled with microlites, and was crystallizing from more numerous centers, for there are instances where larger depressions in the surface of the pyroxene crystals can be seen to have been occasioned by the presence of small crystals of feldspar, which must have hindered the growth of the pyroxene. They have numerous inclusions of glass, with gas bubbles, which are irregularly scattered through the minerals in most cases, but are occasionally arranged zonally; besides which are grains of magnetite and a few small crystals of apatite.

The hypersthene is pleochroic, and is green parallel to c, yellow parallel to a, and light red parallel to b. In most of the thin sections it is light colored, but in one instance there is a large individual with very strong colors and pleochroism, which carries brown inclusions in the shape of thin plates arranged in lines at right angles to the vertical axis of the crystal. These inclusions resemble those characteristic of many hypersthenes in coarse grained rocks. In this instance the hypersthene crystal occurs in a glassy, vesicular rock, and the inclusions do not appear to have resulted from an alteration of the mineral subsequent to the solidification of the rock, but to have been primary inclusions of some foreign substance. The lighter colored hypersthenes do not carry such inclusions. The color frequently varies in concentric zones, the center being light in some cases and dark in others.

In some forms of the rock the hypersthene and augite have narrow reddish brown borders which are in part opaque. This border, though not so strongly marked as the black margin to many hornblendes, appears to be of similar origin and to be due to an action of the magma on the crystals before the final consolidation of the rock. It affects the pyroxene microlites in the groundmass as well as the phenocrysts.

The color of the augite is light green without pleochroism in thin sections; and is easily confounded with the sections of hypersthene which exhibit little pleochroism. Its optical characteristics are the same as those of the augite in the diorites already described; in fact the pyroxenes of both rocks are alike optically, and have the same distinctions with respect to cleavage, which is more perfect in the augites than in the hypersthenes.

Instances of the complete inclosure of one of the pyroxenes by the other, or of their intergrowth, are rare. In the few cases observed small hypersthenes are surrounded by augite, indicating the earlier crystallization of the hypersthene. But the occasional intergrowths of the two, and the partial inclosures of adjacent individuals in groups, proves that the crystallization of most of the hypersthene and augite phenocrysts was contemporaneous. When decomposition has attacked the rock hypersthene yields before the augite, and is converted into a green fibrous mineral, probably bastite.

The feldspar phenocrysts are all plagioclase, which from their optical characters appear to be labradorite. They are small in most forms of the rock, but larger and more abundant in others. They are rectangular in long and short sections, a few are broad and polygonally outlined. The sections are mostly straight edged, some are rounded at the corners, and others are rough like the pyroxene crystals. The rough projections of the feldspars have crystal faces and appear to be due to an irregular checking of their crystallization. They exhibit the characteristic polysynthetic twinning of labradorite and are beautifully zonal. But the zones do not differ much in optical orientation, the extinction being quite uniform throughout each individual.

Glass inclusions are frequent in the feldspars; in some of the larger crystals the central portion is crowded with inclusions of the brown glass containing the same microlites as the surrounding groundmass These inclusions are usually in rectangular negative crystal cavities. Many of the smaller feldspars are almost free from them. There are occasionally grains of magnetite and pyroxene.

In most cases the feldspar and pyroxene phenocrysts are separated by the groundmass of the rock. But when they occur in juxtaposition it is evident that the feldspar is a younger crystallization which started after the pyroxene had commenced to crystallize, but before it had finished, for the feldspar interferes with the perfect development of the pyroxene.

Magnetite occurs in phenocrysts associated with the pyroxene and also isolated in the groundmass. It is in definite crystals and in irregular grains.

There are five representatives of the pyroxene-andesites which carry a few crystals of hornblende, B₂, and constitute transitional varieties between these rocks and the hornblende-pyroxene-andesites. They exhibit the same characters as the andesites just described.

The hornblende is in small irregular crystals, some being rounded and others in angular shapes. It is reddish brown and brownish green, with strong pleochroism. Many of the individuals, especially the rounded ones, have a narrow border of magnetite or one of small crystals of pyroxene, feldspar, and magnetite. There are all gradations, from rounded hornblendes with opaque borders to small angular pieces of hornblende surrounded by comparatively large crystals of pyroxene, feldspar, and some magnetite, which form a group of interlocked crystals in the glassy groundmass. The angular outline of the hornblende and its penetration between the crystals of feldspar and pyroxene would militate against the supposition that the hornblende is a remnant of a previous crystal that had been partially resorbed in the groundmass, were it not for the occurrence in one thin section of a group of different crystals with a hexagonal outline, corresponding to the cross section of the hornblende remnants contained in it which are properly oriented for such a section. The greater part of the group consists of feldspar and pyroxene with some magnetite. It is not to be supposed that these minerals crystallized out of the melted hornblende substance without interchange of material from the surrounding magma. The larger groups in the same rock section exhibit no definite outward form, but are bounded by the outlines of the outer crystals, so that we may conclude that the process of resorption of the hornblende phenocrysts was in some cases accompanied by the immediate formation of grains of magnetite and the absorption of the other chemical constituents by the magma; while in other cases the melted hornblende recrystallized in situ as pyroxene and magnetite. But in the instances just mentioned the partial resorption of the hornblende was followed by a greater tendency toward crystallization in the magma immediately surrounding the melted hornblende, which led to the development of a group of all the minerals then capable of forming. These minerals are the same in size and character as the small crystals scattered through the glassy groundmass.

In rock section No. 12 several small individuals of biotite occur with the same kinds of borders as those surrounding the hornblende. This thin section and one other, No. 32, are the only ones carrying biotite. It is in very small amounts in each case.

The remaining thin sections may be classed as hornblende-pyroxene-andesites, in which the proportions of hornblende and pyroxene vary. In the first four, B₃, the pyroxene is in excess of the hornblende. In the following ten, B₄, they are about equal and in the last five, B₅, the hornblende is in excess. The varieties thus form a series from those without hornblende to others with much hornblende and very little pyroxene.

In these andesites the microscopical character of the pyroxenes is the same as in those first described, except that they are in better shaped crystals, seldom rounded or with dark borders. The hypersthene is mostly light colored in thin section, but in several rock sections some of the individuals are strongly colored at the center, while others are more strongly colored at the margin.

The hornblende differs throughout these sections in color and in the extent to which it has been resorbed. In some cases it shows no sign of resorption. The form of the crystals when perfect is derived from the unit prism and clinopinacoid and the usual terminations. In many instances the crystal faces are poorly preserved and only the general characteristic form remains, especially in cross sections.

The color is intensely red in some varieties of the rock, in others it is reddish brown, chestnut brown, greenish brown, and also brownish green, with the corresponding pleochroism. This difference of color bears no relation to the presence or absence of opaque border nor to the amount of resorption exhibited by the hornblende. It does not appear to be due to secondary alteration of the hornblendes, since they all occur in perfectly fresh glassy rocks, and the color is generally uniform for all the hornblende in one rock section, when the rock is not a tuff.

The character of the border when present varies for different individuals of hornblende in one rock section. Around some it is a narrow margin of magnetic grains, while in a few instances it is a heavy opaque border. Other hornblendes in the same section are surrounded by crystals of pyroxene, plagioclase, and magnetite. In many sections, however, all the hornblendes have been affected to the same extent and have a narrow opaque border, while in others there are no borders at all.

It does not seem possible to connect the character or degree of the resorption with any definite degree of crystallization of the groundmass of the rocks. And, as just stated, different phases of resorption and of borders occur in one and the same rock section. It is often noticed that the center as well as the margin of the hornblende crystal has become an aggregate of pyroxenes and feldspars, and that very little of the hornblende substance remains. But it is also observed that many of the hornblendes which show no evidence of resorption have large and irregularly shaped inclusions and "bays" of the groundmass in them. So that it is probable that many of the cases of apparent extensive resorption or corrosion may be crystals which originally contained large bays of groundmass. Inclusions of glass are not very abundant, except in certain individuals.

There are numerous instances in which the hornblende incloses small pyroxenes and plagioclases, as well as magnetites, and others in which hornblendes and plagioclases have crystallized beside each other and have mutually interfered, proving that their growths was contemporaneous in part. As there are two or more generations of plagioclase and pyroxene, it is natural that the hornblende appears to be contemporaneous with the earlier feldspars and pyroxenes, and older than the later generations.

The feldspars are all plagioclase, but appear to belong to different species. They are in rather small crystals in most of the rock sections. The larger ones are generally labradorite, and many of the small ones are the same, but in a number of the sections the extinction angles indicate andesine or oligoclase. They are mostly rectangular with perfect crystallographic outline, some are tabular and polygonal, and in this position they exhibit the most striking zonal structure, which is almost universally present. The twinning is that characteristic of andesitic plagioclases. Glass inclusions are of frequent occurrence. In some cases the feldspar contains a great amount of glass which almost equals the bulk of the feldspar substance. Occasionally the feldspar has an irregular form and an indented outline, made by the projection of crystal points, the margin of the individual having a different optical orientation from the central portion, and appearing to be formed of more alkaline plagioclase. These are not very common.

The groundmass of these andesites is the same as that of the pyroxene andesites in some cases, and is composed of globulitic brown glass with microlites of pyroxene, feldspar and magnetite. But in most of the sections it consists of colorless glass crowded with small microlites of the same minerals. It carries microscopic crystals of these minerals which are porphyritical with respect to the groundmass when seen with a microscope, but which in turn form part of the groundmass which carries the macroscopic phenocrysts.

THE DIKE ROCKS.

The dike rocks of Sepulchre Mountain, as already mentioned, consist of a series of andesites and dacites, the earliest of which resemble the pyroxene-andesites and hornblende-andesites of the breccias. They vary in mineral composition as indicated by the porphyritical crystals of all sizes that are scattered through the groundmass, and range from rocks with phenocrysts of hypersthene, augite and plagioclase, to those with phenocrysts of quartz, biotite, hornblende and plagioclase. This variation is shown in the accompanying table (Table XIII), in which the 103 thin sections of these dike rocks are arranged according to the porphyritical minerals contained in them.

While the greater number of pyroxene-andesites and hornblende-pyroxene-andesites carry no biotite, there is a small amount of it in some of the latter varieties. In one instance biotite, hornblende, and pyroxene occur together in considerable amounts.

TABLE XIII.—Mineralogical variations in the dike rocks of Sepulchre Mountain.

There is a number of hornblende-andesites with neither pyroxene nor biotite, and others with a small amount of both. In most of the hornblende-mica-andesites there are 110 porphyritical pyroxenes; they occur in a few varieties only and in small amounts, and are equally rare in the dacites.

The greatest amount of porphyritical quartz is generally accompanied by considerable biotite and less hornblende.

Plagioclase feldspars are present in all the varieties of these rocks, but vary in composition from labradorite in the basic andesites to oligoclase or andesine in the dacites.

As to the microscopical characters of the essential minerals it may be said that they are like those already described for the essential minerals in the andesites which form the breccias.

The pyroxenes are the same, and consist of hypersthene and augite in all cases where they are fresh. In many instances a part of the pyroxene is entirely altered and part is fresh, and is augite, the hypersthene having been completely decomposed. They have the same color and pleochroism and crystal form as those in the andesites just described and need no further comment.

The hornblende in some of the pyroxene-andesites is represented simply by paramorphs, which consist of grains of magnetite and pyroxene with the outward form of hornblende crystals in others it is in small individuals, with a broad or narrow black border, occasionally with no border. In the hornblende-pyroxene-andesites the hornblende has a black border in some instances, but in the majority of cases it is entirely free from any border; the same is true of it in the hornblende-andesites. In the more acid andesites and dacites the hornblende exhibits no signs of black border.

In many instances where the crystal form is well developed both the orthopinacoid and clinopinacoid is present besides the unit prism faces, which is characteristic of the hornblende in the diorites of Electric Peak.

The color of the hornblende varies somewhat from brown and greenish brown to brownish green and green, with the usual pleochroism. It is brown and greenish brown in most of the pyroxene- and hornblende-andesites, but is very generally green and brownish green in the hornblende-mica-andesites and dacites. Its color is like that of the hornblendes in the porphyrites and diorites of Electric Peak.

Many of the hornblendes carry glass inclusions, and some have large bays and irregularly shaped inclusions of groundmass. They also inclose grains of magnetite and apatite, and occasionally are intergrown with pyroxene in such a manner that the two appear to have crystallized at the same time.

In some of the dacites the hornblende is entirely decomposed, while the biotite is still intact.

The biotite is chestnut brown, in thin section, with the ordinary absorption. The optic angle is very small and the mineral behaves like a uniaxial one. Its crystal form is simple and the individuals are generally quite thick. It is unaltered in almost all the rock sections, and carries a variable amount of inclusions of magnetite and apatite, with occasional zircon. In one instance it completely incloses a small crystal of plagioclase.

The feldspars are all plagioclase, and exhibit the characteristic polysynthetic twinning. In the more basic andesites they are mostly very small individuals, with rectangular sections and high extinction angles, indicating labradorite. They are usually very abundant. In some instances they are fewer in number, and do not exhibit high extinction angles or high double refraction, and appear to be oligoclase.

In the more acid andesites and dacites the plagioclases are larger and have more crystal faces, the sections being more polygonal and broader. The extinction angles are lower, and there seems to be several kinds of plagioclases among the phenocrysts; some are sharply rectangular with numerous twin lamellæ, and extinction angles indicating labradorite, while the majority of the individuals are not rectangular, have fewer lamellæ and lower extinction angles, and exhibit very marked zonal structure. They appear to be oligoclase; they all carry more or less glass inclusions, which are very abundant in some individuals and in some rock sections, and are quite scarce in others. The different specimens of the rocks vary greatly in the amount of inclusions in the phenocrysts. In one of the hornblende-andesites which has a brown, globulitic, glassy groundmass, many of the feldspars inclose patches and small bits of the brown glass, but one of the larger plagioclases also carries a great number of opaque needles and grains, arranged in several systems of parallel lines, which are identical with the inclusions in many of the labradorites in the diorites of Electric Peak. Besides this individual of feldspar there are several others which exhibit the same thing to a slight degree. There is an other fine example of it in a glassy hornblende-andesite; the feldspar in this case carries abundant inclusions of glass as well as the clouds of microscopic needles. This is important, as it proves the primary nature of these particular inclusions, and indicates that the phenocrysts containing them crystallized under conditions similar to those attending the crystallization of the labradorites in the diorites of Electric Peak.

The quartz phenocrysts occur in the biotite-hornblende rocks, and vary in amount from a few microscopic individuals to very abundant macroscopic ones. Their crystal form is well marked in many cases and corresponds to the double pyramid, but other individuals in the same rock section are rounded, and some have quite an irregular outline. It seldom, if ever, happens that all the individuals of quartz in one rock section exhibit the same degree of perfection of crystal form; rounded grains and idiomorphic crystals are scattered indiscriminately through the rock. The same is true in many instances of the hornblende individuals, as already described.

The quartzes occur singly in isolated crystals, and also in groups of two or more individuals with different orientations, grown together in the same manner as those of feldspar or of the ferromagnesian silicates. Glass inclusions are found in nearly all the quartzes, but in very different amounts, some being crowded with them, while others are almost free from them. They are usually in negative crystal cavities, occasionally in rounded ones. In some cases they are accompanied by the six-rayed cracks so common in the quartzes of rhyolites. The quartzes often inclose bays of groundmass, and occasionally small crystals of hornblende, biotite, and plagioclase. These latter inclosures show that the quartzes crystallized after part, at least, of the hornblende, biotite, and plagioclase had crystallized. The inclosing quartzes are rounded at the corners. In one instance a quartz contains small fluid inclusions besides those of glass.

Magnetite, which is very abundant in the more basic rocks, and is in much smaller amounts in the dacites, needs no special description.

The apatite occurs in short, stout, hexagonal prisms; it is colorless, and is rare in the basic andesites and more abundant in the more acid andesites and dacites. The same is true of the zircon which is seldom observed in the basic andesites.

Small individuals of allanite are found in three of the dacite sections. It is dark brown, with strong absorption.

The groundmasses of these rocks, which result from the processes of final solidification of the various magmas, differ in degree of crystallization, in mineral composition, and in structure.

In the pyroxene- and hornblende-pyroxene-andesites the groundmass in many cases is glassy, with multitudes of microlites of pyroxene and plagioclase and grains of magnetite. In many others it is completely crystallized and the outline of the microlites is no longer sharply defined. In one glassy hornblende-pyroxene-andesite there is a segregation of minerals, which is interesting from the fact that the mass is not holocrystalline, but contains in the interstices between the large crystals vesicular glass with skeleton feldspars, and much fewer microlites than the glassy groundmass contains. The segregation, at first glance, resembles those holocrystalline groups of hornblende and plagioclase so common in the andesites and porphyrites. It consists of large hornblende crystals, with a few small biotites and pyroxenes inclosed in them, besides some plagioclase. But the feldspars carry many fine glass inclusions, which are also found in the hornblendes. The interstitial glass is partly colorless, partly globulitic, carrying long, slender skeleton plagioclases, with square cross sections, and a few needles of pyroxene with grains of magnetite attached. This glass is quite vesicular, while the groundmass of the rock presents a wholly different appearance. The latter is compact, and crowded with small microlites of feldspar and pyroxene and magnetite, having a typical felt-like structure. The hornblende and plagioclase of the segregation have the same characters as those of the same minerals in the surrounding rock, but they carry more glass inclusions. The crystal form of the minerals on the outside of the segregation is perfect, and the large crystals project into the surrounding groundmass of the rock. The segregation can not be the broken fragment of some foreign rock mass, but must be a local crystallization which advanced more rapidly than that of the surrounding portion of the rock but did not result in complete crystallization. Within the interstitial glassy portion are numerous hollow cavities.

In the holocrystalline varieties of these rocks the groundmass has attained different degrees of crystallization, which may be compared within those exhibited by the intrusive rocks at Electric Peak. Separating the rocks into five groups to correspond to the preponderance of pyroxene with little hornblende; of pyroxene and hornblende of hornblende alone, or with little pyroxene; of hornblende and mica, and of mica, hornblende, and quartz (See Tables XII and XIII), and arranging them according to the size of grain of the groundmass, they fall into the order given in Table XIV. In this table the grades of crystallization correspond to those established for the intrusive rocks of Electric Peak, which are expressed in Table VIII, with the addition of five more divisions which embrace two finer grained degrees of holocrystalline structures and three degrees of glassiness.

TABLE XIV.—Grades of crystallization of the eruptive rocks of Sepulchre Mountain.

The microstructure of the acidic varieties is not the same as that of the basic, so that it is difficult to compare the grain of one directly with that of the other; but since the intermediate rocks possess microstructures intermediate between these extremes, it is possible to establish a kind of relationship between them, and it is admissible to place them in the same line across the table, it being understood that the correspondence is an approximation.

A glance at Table XIV shows that a great majority of the varieties are very fine grained forms that have only reached the crystallization of the few smallest grained forms of the Electric Peak rocks. A small number of them are more coarsely microcrystalline and correspond to the grain of the dike rocks at Electric Peak. A large number are finer grained than any of these rocks, or are glassy. The coarsest grained forms have been attained by the most basic varieties, but they do not represent bodies of any considerable extent. Specimen No. 42, grade 25, comes from a small exposure with no definite limits, surrounded by much finer grained rocks. It is properly a diorite-porphyrite, and carries much biotite of final consolidation, which has not been reckoned with the phenocrysts.

The coarsest grained forms of the acid varieties, however, represent larger bodies and are more abundant in the field.

In explanation of the degrees of crystallization indicated in the table it may be said that the first three are glassy groundmasses, the first one having fewer microlites than the second. In the third the microlites are closely crowded together. The next two represent microlitic structures in which no glass can be detected; they appear to be holocrystalline. In the sixth grade the form of the microlites is more indistinct, but the general structure is the same as before. Beyond this the different degrees indicate increasing grades of a structure which may be described in general as follows: Commencing with the lowest order, the groundmass is composed of a multitude of indistinct microlites of lath-shaped feldspars; between crossed nicols this aggregation extinguishes light in small patches, which bear no fixed relation to the position of the microlites within them. As the dimensions of the lath-shaped feldspars become larger it is observed that the patches of light and darkness arise from the cementing material between these feldspars. This cement possesses the same optical orientation for small spaces which in cross section produce the patches just alluded to. In still coarser grained forms it becomes apparent that the cementing material is quartz which has crystallized in irregularly shaped patches inclosing many smaller feldspars. The size of these feldspars and of the interstices between them is taken as the grain of the rock and not the size of the patches of quartz. For it is observed that as the rocks become more coarsely crystalline the feldspars, which are plagioclase, increase steadily in size and each quartz patch cements fewer of them, until in still coarser grades the quartz forms allotriomorphic individuals between the plagioclases and does not surround any, so that in these varieties of rock the size of grain is judged by the dimensions of the plagioclases and the interstices of quartz. The patchy structure just described is that already mentioned on page 589 and called micropoicilitic.

In the most siliceous varieties of the rocks the microstructure is different. The smallest grained forms appear to approach a granular structure in which, however, the feldspars exhibit a more or less rectangular shape and the quartz shows a tendency to appear in minute, poorly defined dihexahedrons. As the grain becomes larger the form of the quartz grains becomes more pronounced. They are rudely idiomorphic, with sections that are in many cases equilateral rhombs, extinguishing the light parallel to their diagonals. In the coarsest grained forms of the dacites these imperfectly idiomorphic quartzes are characteristic of the groundmass and reach a diameter of from 0.08mm to 0.10mm. Their surface is indented with the ends and corners of small plagioclases, the structure of the groundmass being hypidiomorphic. These quartzes often contain minute colorless inclusions in negative crystal cavities which have every appearance of being glass and correspond to the glass inclusions in the quartz phenocrysts of the same rocks. The partially idiomorphic quartzes in the groundmass are to a slight degree porphyritical with respect to the other constituents, but belong to the final consolidation of the magma.

GENERAL CONSIDERATION OF THE MINERAL AND CHEMICAL COMPOSITION OF THE ERUPTIVE ROCKS OF SEPULCHRE MOUNTAIN.

Mineral composition.—The mineral variations inn the group of rocks forming Sepulchre Mountain are much simpler and require much less discussion than those of the intrusive rocks of Electric Peak. They have already been expressed in the Tables XIII and XIV. From these tables it is evident that the so-called transitional forms of the rocks are as numerous and as important as those forms which would be considered type rocks. It is possible to describe those varieties of andesite with augite and hypersthene and no hornblende as typical pyroxene-andesite, those varieties with nearly equal amounts of pyroxene and hornblende as typical hornblende-pyroxene-andesites, those varieties with hornblende alone as typical hornblende-andesites, and so on for typical hornblende-mica-andesites and typical dacites. And for convenience of description this may be admissible. But in the occurrence at Sepulchre Mountain such a method of description would create a false impression and would lead one to expect definite bodies of such type rocks with facies which should present the transitional variations; whereas, there are definite bodies of the so-called type rocks and equally definite bodies of the intermediate varieties which are quite as numerous. There is no particular mineralogical modification of the rocks at this place, which from its greater abundance or its special mode of occurrence renders it a type rock. On the contrary, the whole accumulation of eruptive rocks which are subsequent to the bottom breccia within its admixture of Archean fragments, must be considered as a series of volcanic rocks that vary in mineral composition, through gradual changes from pyroxene-andesite to dacite.

Starting with those rocks which carry phenocrysts of pyroxene and plagioclase, it is observed that as the hornblende makes its appearance and increases in amount the pyroxene decreases. Biotite accompanies the hornblende in the more acidic varieties and increases in amount with the acidity of the rock. Quartz first appears in small quantities and increases within the acidity of the rock, the hornblende decreasing at the same time. To this rule there are exceptions which are indicated in the biotite is found to a slight extent in some of the hornblende-pyroxene-andesites and pyroxene occurs in small amounts in some of the hornblende-mica-andesites. It is, of course, understood that this relation between the essential minerals may be different for groups of andesites in other regions.

Chemical composition.—The chemical composition of the eruptive rocks of Sepulchre Mountain is shown in the accompanying table of chemical analyses:

TABLE XV.—Chemical analyses of rocks from Sepulchre Mountain.

Nos. 33, 80, 2, 95, and 129 were analyzed by Mr. J. E. Whitfield, Nos. 20, 21, and 102 were analyzed by Dr. T. M. Chatard, and No. 131 was analyzed by Mr. L. G. Eakins.

The first, No. 33, and fourth, No. 2, are analyses of pyroxene-andesites which carry no hornblende; the first is a dike near the summit of the mountain, the other is from a surface flow at its southwest base. Nos. 20 and 21 are of hornblende-pyroxene-andesites, occurring as breccia in the upper part of the mountain. No. 80 is of hornblende-andesite which is an intruded body in the small hill northeast of Cache Lake at the head of Reese Creek. No. 95 is a hornblende-mica-andesite from the same locality, also an intrusive rock. No. 102 is the same kind of andesite from an intrusive mass at the north base of Sepulchre Mountain, and Nos. 129 and 131 are dacites from the ridge south of Cache Lake.

The range of variation in the percentage of silica is about the same as that of the rocks at Electric Peak. The character of the variations of the other oxides in these rocks is shown by the accompanying diagram, which represents the variations in the molecular proportions of the essential oxides and has been plotted in the manner already described on page 628.

A glance at this diagram shows that it has the same form as that of the group of analyses of the rocks from Electric Peak. The variations in the oxides other than silica are quite irregular for a gradual change in the silica. The alumina varies rapidly in places and retains a high position in the diagram. The alkalies gradually increase with the silica the soda molecules being twice as numerous as those of potash and their variations being alike with one exception. Magnesia varies most widely and in striking contrast to the alumina; in each instance they vary in opposite directions. The lime is nearly as irregular as the magnesia, both decreasing rapidly from the less siliceous to the more siliceous end of the series. The two oxides of iron are strikingly reciprocal in their variations, the significance of which has been pointed out in discussing the diagram for Electric Peak. In the group of analyses from Sepulchre Mountain the oxidation of the iron bears a noticeable relation to the presence of hornblende, biotite, and magnetite in the rocks.

From a study of these analyses it is evident that the chemical variations in this group of rocks are the same in character and extent as those in the intrusive rocks of Electric Peak. Moreover, it appears that the variations between similar varieties of andesite—such as those between different pyroxene-andesites—are as great as and in some cases greater than the variations between varieties of andesites which are distinguished mineralogically from one another. Thus Nos. 33 and 2 are pyroxene-andesites without hornblende, Nos. 20 and 21 are hornblende-pyroxene-andesites, while No. 80 is a hornblende-andesite. It is not possible to point to any chemical character of these rocks which is distinctive of this mineral variation, with the exception of the oxidation of the iron, which, though slight, is an important one; for it undoubtedly relates to forces which did not alter the fundamental relation between the bases in the magma, but simply modified it by changing the oxidation of one of them. The last four analyses are of hornblende-mica-andesites and dacites. The chemical variations between them are as pronounced as those between the more basic members of the series, without there being the corresponding differences between the kinds of ferromagnesian silicates, so far as it can be detected microscopically. They all carry hornblende and biotite and no pyroxene, the relative proportions of these minerals varying. The character and amount of the feldspars differ in these rocks, and so do the abundance and mode of occurrence of the quartz. In Nos. 129 and 131 quartz appears as phenocrysts; in the other rock it is confined to the groundmass.

FIG. 81.—Molecular variation of the rocks of Sepulchre Mountain