USGS Logo Geological Survey 7th Annual Report
Obsidian Cliff, Yellowstone National Park


Having familiarized ourselves with the general outward appearance of the rock forming Obsidian Cliff, let us look deeper into the subject and by means of the microscope penetrate to the very heart of the matter. To this end thin sections of the rock were prepared by grinding one side of small fragments about the size of a 25-cent silver piece and polishing it smooth, cementing this to glass by means of heated Canada balsam, and then grinding away the other side of the rock until it became thin enough to be transparent, when by very skillful polishing the section was reduced to a thinness almost imaginary, 0.001 of an inch and thinner. With such sections magnifying powers as high as 1,500 diameters may be employed if necessary, permitting very little to escape notice. For most purposes, however, a power of 100 diameters is all sufficient.


The black obsidian in thin section becomes perfectly transparent and of a gray color. Under the microscope it is seen to be a colorless glass crowded with minute, transparent, pale-green crystals and short, black, hair-like bodies called trichites. The crystals are in thin rhombic tablets or irregular grains, 0.005mm (1/5000 inch) in diameter, either scattered about or strung on the short, opaque threads like conserved cherries on a straw (Pl. XV, Fig. 6). The larger of these transparent microlites, or microscopic crystals, prove to be a variety of angite, which only appears in microscopic forms in this rock. The trichites are about 0.0008mm (0.000032 inch) wide and may be traced through different gradations to grains of magnetite, which in larger form are recognized in the glass intimately associated with the angite, usually inclosed in a grain or crystal of the latter mineral. These trichites give the obsidian its black color. Such microlites and trichites are found in nearly all volcanic glasses and differ in shape and character according to the condition and composition of the lavas in which they occur. They are rudimentary crystals of minerals which develop in rapidly solidifying glasses, where a larger and more perfect crystallization is hindered by the viscosity of the glass.


The strings of microlites and trichites often have a distinctly parallel arrangement and are in layers of greater or less abundance, which mark the planes of flow within the glass.

Occasionally the trichites are in curved groups, radiating from a central grain and looking like the downy seed of a thistle. One variety of red obsidian derives its color from bright-red trichites and grains, due probably to the higher oxidation of the iron. Most of the brownish-red varieties are composed of colorless glass through which run orange and yellow, microscopic streaks and ribbons in the greatest complex of contortions and streams (Pl. XVI, Fig. 1), often producing very grotesque and curious shapes, among which one recognizes the forms of flowers, toad-stools, sea-anemones, jelly-fish, spearheads, and the like, a list only limited by the size of the thin section and the imagination of the observer.


These colored bands, when highly magnified, appear to be composed of the minutest yellow or orange granules. In places there is a shading through brown into black, which is accompanied by a shrinkage of the bands, and in their stead is clear glass streaked with clouds and lines of opaque grains and trichites, apparently magnetite. Some of the clouds of minute grains look blue in transmitted light.

The various shades of color in the obsidian arise from the different proportions in which these yellow, orange, brown, and black particles are mingled. In some the colored streaks are in broad, thin bands, either straight or twisted, according to the last movements of the viscous glass. In others they are in the most delicate threads, alternating with streaks of black grains running continuously through the rock, though sometimes interrupted by streaked patches of other character or appearing as though the rock had been broken into fragments and welded together again (Pl. XVI, Fig. 2). The transition of the yellow and orange bands into black grains, the larger of which are recognizable as magnetite, indicates that the former are made up of finely divided particles of iron more highly oxidized, which is confirmed by the chemical analyses of the red and black obsidian. The iron in the red variety is almost wholly sesquioxide, while in the black obsidian there is a slight excess of protoxide over that required to form magnetite in combination with the sesquioxide.

The only other minerals crystallized out of this obsidian are a few microscopic crystals, suggesting feldspar, and the various spherulitic bodies. We shall consider the different forms of crystallization in the order of their production or growth, having already started with the first, namely, the trichites and microlites of magnetite and augite.


Following these come the microscopic crystals of what at first appears to be feldspar. They are not, however, simple feldspar, but an intergrowth of this mineral with another in groups of small crystals with the nearly rectangular outlines of feldspar, averaging 0.2mm (0.008 inch) in diameter. Close inspection with a high magnifying power reveals the fact that each group is composed of several individuals of feldspar intersecting one another and that each individual has a fibrous structure in several directions through it and is granular in places. This is represented in Pl. XV, Fig. 3, where a section has been cut through two individuals, one end of one not having been developed. The fibers run nearly perpendicularly to the sides of each rectangle. The margin of what appeared to be a straight-edged crystal section is found to be serrated with the projecting ends of minute crystals. The component minerals are too finely divided to be determined optically; all that can be observed is the general structure, that they are colorless, and that between crossed nicols they extinguish light at various angles to the direction of the fibers. They inclose trichites and microlites, and have therefore crystallized after these.

Their structure recalls that of other groups of colorless minerals noticed in rocks of similar composition which occur in the neighboring regions, namely, the rhyolites of the Great Basin of Utah and Nevada, where we find similar groups of larger size, and in a rock from Eureka, Nev., there is one large enough to allow of the optical determination of the component minerals. This is represented in Pl. XV, Fig. 5, as it appears between crossed nicols when magnified 37 diameters. It is a section through three orthoclase feldspars that have crystallized about a grain of plagioclase, part of which has fallen out in grinding. The section has been so placed between crossed nicols that one orthoclase is dark and the others are light gray. Inclosed in the feldspar substance are strips of another mineral which appear white in two feldspars, but white, gray, and black in the third. This mineral is quartz, which has formed at the same time as the feldspar and has been inclosed in its mass. It is in shreds, which are long and thin or short and irregularly shaped. Many of them are bounded by crystallographic faces, which in cross-section give triangular and polygonal figures characteristic of pegmatite. All the quartz shreds in the largest two feldspars have the same crystallographic orientation, as though they belonged to one continuous crystal, but in the smallest feldspar the quartz shreds are in three sets, with different orientations. Thus, in one instance, quartz in one position is combined with feldspar in two positions, and, in another, quartz in three positions is combined with feldspar in one. In the largest individual, while the clear margin of feldspar extinguishes light at one angle and the quartz at another, in that portion where the thin shreds of quartz alternate with those of feldspar the maximum extinction of light takes place in various positions, according to the relative thickness of the two minerals through which the light passes. Hence we find in this fibrous portion different extinctions, none of which corresponds to that of either of the component minerals.

From some cause the quartz ceased to crystallize before the feldspar, as is seen by the clear border and sharp, straight outline of the latter mineral, pierced, however, in several places by shreds of quartz which protrude from the surface. If the feldspar had stopped forming a little before the quartz the outline would have been serrated like that of the groups formed in the obsidian. We have, then, strong evidence that these microscopic groups are composed of feldspar and quartz intergrown in the manner so frequently observed on a larger scale in many granites, porphyries, and rhyolites.

From the simple intersection of two feldspars the groups grow more complex with the increasing number of feldspars, the outline in cross-section becoming oval or circular, An extreme case is represented in Pl. XV, Fig. 4; in this the feldspars wedge out toward the center, their outer ends making an almost continuous outline and the crystal form being no longer recognizable. The fibration, however, is in wedge-shaped sets and does not radiate uniformly from the center. The extinction of light between crossed nicols is quite irregular, as shown in the drawing, and, as we observed in the rock from Eureka, Nev., the orientation of the quartz may vary greatly throughout this group, being quite independent of that of the feldspar. The inclosed trichites are crowded in a ring about the center.

These microscopic pegmatoid or granophyre groups, together with the trichites and microlites, crystallized before the lava came to rest and have been more or less twisted or turned about and arranged in layers along the planes of flow.


The next order of crystallization in the obsidian is the spherulitic. The simplest and smallest as well as the first formed spherulites appear as minute, colorless spheres about 0.2 to 0.05mm in diameter, scarcely noticeable in ordinary light. Highly converging light makes evident a finely fibrous structure, and between crossed nicols a more or less well defined, dark cross is observed. The arms of these crosses do not remain of constant width during the rotation of the section, but alternately contract and spread, and split into branches near their ends, as represented in Pl. XV, Fig. 2, and Pl. XVII, Fig. 1. The latter figure is from a photomicrograph taken by Mr. Clifford Richardson, of the chemical laboratory of the Agricultural Department, to whom the writer is greatly indebted for a number of beautiful photographs of rock sections taken with the aid of a solar microscope. Some of these photomicrographs are reproduced on Pls. XVII and XVIII.



A fibrous margin surrounds many of the granophyre groups; its character and length of fiber correspond to those of the smallest spherulites and from its optical behavior between crossed nicols it appears to be a continuation of the material of the inclosed kernel. The spherulites are often in rows and layers (Pl. XVII, Fig. 1) and sometimes their centers are along straight lines and so close together that they produce transparent, fibrous bands (Pl. XVIII, Fig. 1). These minute spherulites correspond to the dots observed in lines on the surface of the obsidian. After the microscopic spherulites those appearing blue in the hand specimen were formed. In thin section they are light gray in incident light, but brown by transmitted light. They range from less than one millimeter to five and rarely ten millimeters in diameter. Under the microscope they show a finely fibrous structure, radiating from a center at which there is frequently, though not always, a granophyre group or colorless spherulite. The fibers are so delicate that many are superimposed one on another within the thin section, preventing a determination of their optical characters. Most of these larger spherulites are traversed by the streams of microlites and minute colorless spherulites which pass through them and the surrounding matrix without change of direction; but in some cases the microlites and trichites have been pushed out or crowded into radial lines (Pl. XVIII, Figs. 1 and 2). Between crossed nicols the rays of shadow seldom form a perfect cross, but are scattered and broken into many arms, some branches lying at an angle of 45° to the principal plane of the nicols. One of these spherulites is represented in Pl. XV, Fig. 1, the scattered nature of the dark rays corresponding to that exhibited by the granophyre groups (Pl. XV, Fig. 4). The fibers are in sectors and do not radiate from a single point. The smallest colorless spherulites appear to be somewhat more regular forms.

The similarity in structure and optical behavior between these spherulites and the fibrous, granophyre groups indicates a correspondence in their mineral composition which would lead us to conclude that the spherulites are composed of feldspar and quartz that have crystallized from the molten glass at one and the same time and have intergrown with each other, the fibers not necessarily being individuals of these minerals elongated in the direction of their principal crystallographic axis. This view we shall see is confirmed by their chemical composition.

The striping and banding of these spherulites in different colors arises from the crowding together of minute brown or black particles into radial or circular bands. These particles, which appear opaque by transmitted light, are often white by incident light and are probably in part gas cavities, as these are recognized in large numbers in the coarser-grained spherulites to be described. The red surface of many spherulites is produced by the higher oxidation of the trichites and opaque grains which are inclosed in them; and it is quite noticeable that the black trichites and nearly colorless microlites of the surrounding glass as they pass into the spherulites become red, as though they had encountered an oxidizing agent not active in the surrounding glass.

The forms of these radially fibrous growths are not confined to spheres, but through unequal development in different directions take the shape of hemispheres, disks, and sectors, at times spreading out like plumes (Pl. XVIII, Fig. 3), others in section resembling a fox's tail.

Porous spherulites. —The large gray and red spherulites, which have an earthy and rather porous appearance in hand specimens, are seen in thin section to be quite coarsely fibrous, which permits their structure to be clearly made out. They frequently have a dense, dark spherulite at their center and appear under the microscope to be simply the continuation of its fibers under somewhat different conditions. The fibers consist of slender needles of feldspar, often twinned and generally showing low extinction angles. The needles are not straight-edged, are frequently jointed, and branch at low angles, in some cases having short, curved needles attached .ike those of a pine twig. Cross-sections of the needles show them to be polygonal, but the shapes are not uniform. Between the feldspar needles are rows of tridymite scales and scattered grains of magnetite, together with abundant gas pores. The tridymite is often aggregated in spherical masses surrounding a number of feldspar needles, leaving spaces between. The presence of tridymite instead of quartz (the latter being probably the form of free silica occurring in the smaller spherulites) suggests a difference in the conditions governing the formation of the two varieties of spherulites.

Similar feldspar growths are found starting in the clear glass from a stout stem, like the limb of a tree, from which branch smaller rods that continue to fork until a radiating bunch of thin needles results, forming a partial spherulite. This is illustrated in Pl. XVII, Fig. 2. The needles are all twinned, apparently according to the Manebacher law. Their form becomes still more arborescent when the feldspar crystals are somewhat tabular and assume foliate shapes, which strongly resemble oak leaves. All these structures arise from the combination of small crystals attached to one another in such a way as to produce apparently curved forms. When viewed between crossed nicols their effect is very striking and beautiful, the delicate feldspar crystals standing out in brilliant white against the black background of isotropic glass.

The light-gray bands or lithoidal layers of the rock unite two or more of the structures already described, together with coarser forms of crystallization of quartz and feldspar too numerous to describe, except to note, as another freak of mimicry, a structure resembling most excellent examples of microscopic eozoon. The microstructure of the lithoidal portion of this lava flow is the same as that of many other lithoidal rhyolites in various parts of the world.


In the porous portions of the larger spherulites, intimately crystallized with the feldspar and tridymite and generally surrounding the feldspar, as of later growth, are occasional crystals of fayalite, irregularly outlined, evidently one of the last minerals formed. They are sometimes noticed among the coarser crystals of feldspar, quartz, and tridymite, near the cavities of the more porous layers.

It is most unusual to see so basic and ferruginous a mineral as this iron olivine intimately associated with abundant quartz and acid feldspars in a highly siliceous, volcanic rock containing less than 2 per cent, of iron oxide. It is contrary to almost universal observation, and therefore to the laws which are supposed to govern the mineral composition of igneous rocks. It is formed especially within the lithophysæ, and an explanation of its occurrence may throw some light upon the origin of these interesting structures.

<<< Previous <<< Contents >>> Next >>>

Last Updated: 22-Jun-2009