University of Washington Publications in Geology The Geology of Mount Rainier National Park

THE MOUNT RAINIER VOLCANICS

Mount Rainier is a typical strato-volcano comparable, in many respects, to the other andesitic cones scattered along the Cascade Range from Lassen in California to Baker near the Canadian border. Spread out in a very irregular fashion, the Rainier lavas occupy approximately 100 square miles in areal extent. Of this amount, 45 square miles, or nearly half, are covered by perennial snow and ice. Vertically, the lavas range from the upland surface of the Cascades, at an elevation of 6,000 feet, to the crater, which towers 14,408 feet above sea level. Thus the actual volcano is well over 8,000 feet high.

Although exhibiting much less diversity than Mount Shasta and Mount Lassen, in general form and composition, Mount Rainier has suffered intense glaciation and the original symmetry of the cone is now destroyed. An erosional remnant worthy of mention is Little Tahoma. Because of its 11,117-foot elevation, shape, and position, it has been mistaken countless times for a parasitic cone. When viewed from Seattle or any of the neighboring towns, this lesser peak rather closely resembles Shastina, a parasitic cone on Mount Shasta. However, on closer inspection, the alternating flows and beds of pyroclastics all partake in a common dip to the southeast, away from the crater of Rainier and pass under the summit of Little Tahoma without interruption. The peak is only one of the many wedge and cleaver remnants of the original cone.

The Rainier volcanics may be roughly divided into two groups; the loose and crumbly pyroclastics, and the compact flows. On the higher slopes of the mountain, the pyroclastics are abundant. The material presents a wide assortment of sizes and includes dust, ash, tuff, tuft-breccias, breccias volcanic conglomerates, and small mud flows. The volcanic conglomerates and mud flows extend down toward the base of the mountain, while the greater portion of the true ejectamenta (with the exception of the widely-scattered small pumicious lapilli and ash) are confined above the 9,000-foot elevation.* Vertical sections, exposed on the sides of the various wedges and cleavers, show the unconsolidated nature of the tuffs and ash to be in strong contrast with the intercalated flows. The pyroclastics are especially annoying to climbers who attempt to scale the higher peaks. These rocks allow no secure hand or foot holds and they present the added difficulty of continuously falling from above. In color, the fragmental rocks vary from a light to dirty-brown, through all shades of red and pink, to maroon and black. The red tuff-breccia and pumice beds along Gibraltar are easily seen from Paradise. Others, even more completely exposed, outcrop along the sides of Little Tahoma and the Cathedral rocks. These probably have a counterpart in the Red banks near the summit of Mount Shasta. (43)

---

*Since there has been such a wide diversity of opinion regarding the terminology of the pyroclastic rocks, the writer has used the definitions set forth by the National Research council in describing these pyroclastics. Cf. Wentworth and Williams. (39)

---

The observed fragmental beds generally dip away from the central vent with angles between 7° and 30°, the steeper dips being nearer the summit. Dips up to 25° have been observed as low as 9,000 feet in the coarse, loose breccia of Steamboat Prow.

Not all of the pyroclastics have been dropped on the slopes of Mount Rainier. Ash and pumice fields, undoubtedly derived from this cone, are scattered over much of the contiguous area. These vary from a few centimeters to a few meters in thickness and contain fragments ranging from dust to lapilli size. Similar fields, but of much larger extent, have been described by Williams (41) in the vicinity of Crater Lake and Mount Theilson.

The lava flows of Mount Rainier attain their greatest development in the basal portions of the mountain but, as mentioned previously, this material may also occur on the higher slopes intercalated with the pyroclastics. The earlier flows were comparatively fluid and spread out as tongues, partially filling previously formed valleys. A high degree of fluidity is not to be inferred, however, for lavas extending more than 6 miles from the central vent are exceptional. Individual flows up to 40 meters in thickness are not uncommon, but the average would be closer to 25 meters.

Quantitatively, a very small amount of flow material has been added through fissures located at St. Elmos Pass on the Interglacier side of the divide, and also on the west side of the mountain below Emerald Ridge adjoining the South Tahoma Glacier, and possibly near Yakima Creek, although here the relations are obscured by talus and slope wash.

The great majority of flows are fresh in appearance and compact. Scoriaceous and vesicular facies are of minor importance. A platy jointing is very typical with examples exposed at Ricksecker Point, below the inn at Paradise, and on Burroughs Mountain. At the higher elevations, vigorous frost action, together with wide diurnal temperature changes, have been effective in wedging the plates apart and, as a result, the peaks are covered with loose slabs of andesite.

Columnar structure is well shown at a number of localities. At the terminus of the Nisqually Glacier, the flows resting on the granodiorite have vertical columns 25 meters or more in height. At Basaltic Falls, on the east side of the Cowlitz Glacier, and at Pearl Falls, on Pyramid Creek, the columns are larger and even more perfectly developed. On the Yakima Park road, near Yakima Creek, perfect horizontal columns are piled one on the other like cord-wood. These are modified by a secondary parting at right angles to the long axis of the columns and on the north end of the outcrop a platy parting is developed parallel to the long axis.

In color, the lavas are chiefly shades of gray. The lighter shades are restricted to the earlier flows, while the darker grays to blacks may be encountered anywhere on the mountain from the first flows overlying the granodiorite to the crater rim. Other colors include red, pink, purple, bluish-gray, and brown. The red color, as in the case of the fragmental rocks, is prevalent near the summit. As will be shown later, the color is largely dependent on the condition and amount of glass in the groundmass.

In the literature there is a slight disagreement concerning the proportion of pyroclastics as compared to flow material. Russell (27) states:

The main mass of Mount Rainier is composed of andesite and basalt, which were ejected to a considerable extent in a fragmental condition as scoria, pumice, lapilli bombs, etc. Lava flows were not abundant during the latter stages of eruption. The mountain ranks as a composite cone but so far as its structure is revealed in the canyons. . . it was built largely by the material thrown out by explosions from a summit crater.

On the other hand, Smith (30) declares: "The breccias, agglomerates, and tuffs, although of striking appearance, are, perhaps, less important elements in the construction of the composite cone."

This difference in opinion may be explained by the districts visited by each man, or, as an alternative, to their interpretation of the areal extent of the Rainier lavas. Russell evidently spent much of his time on the glaciers and higher slopes where the fragmental rocks are dominant. Smith, however, visited many of the lower, as well as the higher reaches, and encountered large quantities of flows. The writer concurs with Smith in estimating the bulk of the mountain to be composed of flows.

It is interesting to note that on the limited excursions taken by the average Park visitor, no pyroclastics are encountered; with the possible exception of the small and surficial beds of ash. The majority of trails lead over the marginal tongues of the Rainier lavas and the Keechelus andesites.

The exact time of issuance of the Rainier lavas is unknown, but it is thought that the greater portion of the volcano was formed during Pleistocene time. Paleobotanical evidence (46) indicates post-Pliocene eruptions, as leaves have been found intercalated with some of the Rainier pyroclastics. Witnesses (18) have observed eruptions in the form of a series of brown, billowy clouds in 1879 and again in 1882.

COMPOSITION

After examining the first specimens ever gathered from the mountain, Hague and Iddings (15) came to the conclusion that "Mount Rainier is formed almost wholly of hypersthene andesite." Later work on the mountain has not only failed to alter, but has forcibly emphasized this statement. In comparing the volcanoes of the United States Pacific coast, these men stated further:

While the rocks from the volcanoes, in general, present the closest resemblances, there is a wider range and a greater variety of structure in the more acid types from Lassens peak and Mount Shasta. On the other hand, judging from the collections, the range in the character of the extrusions is most restricted at Mount Rainier.

Smith attempted a more detailed classification of types and mentioned:

Four rock types are represented; hypersthene andesite, pyroxene andesite, augite andesite, and basalt—any of which may carry small amounts of hornblende. A rigid separation of these rock types, however, is impossible since insensible gradations connect the most acid with the most basic. In the same flow, hypersthene andesite may occur in one portion while in close proximity the lava is an augite andesite.

Even in the classification suggested by Smith, the majority of rocks from Mount Rainier are hypersthene andesites. Pyroxene andesites, in which both hypersthene and augite are essential, would be a close second. The augite andesites and basalts are of minor importance.

A wide diversion of opinion exists among petrographers as to just what constitutes a basalt and how it may be separated consistently from similar rocks, as, for example, andesites. If the distinction between basalts and andesites is based on the nature of the feldspars, andesine characterizing andesites and the more basic varieties the basalts then the lavas from Rainier contain approximately 7 per cent basalts. Following the other school which regards the abundance of olivine and the preponderance of mafic over felsic minerals as criteria for basalts, then less than 1 per cent of the Rainier rocks are basalts.

In the chemical analyses which follow, No. 1 is from the crater. This, undoubtedly, is one of the darker and more glassy types so abundant along the crater rim. The analysis shows it to be a rather acid andesite.

Unfortunately the locality for No. 2 is unknown. Smith (30) mentions that it was collected from the "northern slope of the mountain." Because of the close similarity between the Rainier and the Keechelus lavas, and this is especially true on the northern slope, it is not improbable that this analysis might be of a Keechelus rock. The low alumina and magnesia and the high soda, potash and lime are unusual for the andesites of the Cascade volcanoes.

MICROSCOPICAL PETROGRAPHY

The rocks of Mount Rainier are extremely monotonous in their mineral content, but, happily more varied in their textural characteristics. In the pages to follow, the minerals occurring as phenocrysts will be described in the order of their importance. Added to this is a brief description of some of the textures encountered. The data is compiled from the examination of several hundred thin sections, and, while lacking in detail, should provide a general acquaintance with the Rainier volcanics.

The dominant phenocrysts are plagioclase, hypersthene, and augite. Such minerals as hornblende and olivine may be present but are quantitatively so subordinate as to require little attention.

Plagioclase. It occurs in two, and possibly three, more or less distinct generations. Generally the difference in size will serve as a criteria to distinguish the various sets of feldspar. However, this characteristic is not always dependable as in several of the flows all gradations in size are to be found from the smallest to the largest. In such cases, other features, to be described below, will serve to separate the various generations.

FIG. 14. Andesite from north side of the South Puyallup Glacier. A (top). x25, plane light. (A blotchy, plagioclase crystal.) B (bottom). Under crossed nicols. (The component parts of the plagioclase phenocryst are, for the most part, arranged in a parallel fashion; however, one side is formed by individual crystals oriented in different directions, indicating a glomeroporphyritic mechanism.)

FIG. 15. Hypersthene andesite from Saint Elmo Pass. A (top). x62, plane light. (The light-colored area in the center of the photomicrograph is a cluster of plagioclase crystals.) B (bottom). Under crossed nicols. (The orientation of the various crystals in the glomeroporphyry can be seen in this photomicrograph.)

The largest phenocrysts are unique in several ways. In plane polarized light they display a chunky, blocky outline, seemingly made up of several smaller crystals having a common orientation. Under crossed nicols many of these blocky phenocrysts are composed of smaller crystals, each twinned on the Carlsbad plan with the composition planes usually parallel to each other. A zoning is invariably present and is superimposed on the Carlsbad twins. Pericline twinning is rare. The feldspars are normally crowded with inclusions, the chief constituent being a brownish-glass, subordinately, mafic minerals or portions of the groundmass may be included. Frequently, smaller feldspar crystals are attached to the larger ones with a marked difference in orientation. The effect of the attached crystals is obviously a glomeroporphyritic tendency and this, together with a cumulophyric texture, is very typical of the Rainier lavas. In some of the larger clusters, the individuals are welded together so perfectly that it is impossible to distinguish the component parts in ordinary light. Under polarized light, the sets of Carlsbad twins can often be distinguished, but sometimes even this is lacking and the resulting phenocrysts are strongly zoned, complexly twinned, and full of cuneiform-shaped inclusions. The fact that all steps exist, even in the same section, between the larger phenocrysts and the glomeroporphyritic clusters, strongly suggests the phenocrysts were formed by the accretion and welding together of smaller feldspars of an earlier generation. This would also account for their oversize dimensions (averaging close to 2 mm. in length) when compared to the second generation feldspars, which average only .3 mm. in length. This explanation would offer a plausible reason for the complex twinning; the abrupt and angular lamellae and composition planes being inherited from the arrangement of the former individuals and not obliterated by later cohesion effects. The chunky shape, with blocks protruding from the margin and also outlined within the phenocryst by continuous lines of brown glass, fits well into this explanation. The tendency toward a glomeroporphyritic texture is not only abundantly displayed in the Mount Rainier lavas but also in many other of the Cascade andesitic volcanoes. (42)

Patton (24) has figured a feldspar phenocryst from Crater Lake which fits the above description perfectly. He attributes the formation of the crystal to "secondary enlargement," and his figure indicates a thin rim of clear material added to the inclusion-filled core. Thin rims are also found on the plagioclase in the Rainier lavas and they may be either clear, as compared to the cores, or full of inclusions with relatively clear cores. In either case, while the rim has added to the dimension of the crystal, it is, nevertheless, quite thin and not of sufficient magnitude to account for the great difference in size between the first and second generation feldspars. The secondary enlargement idea fails to account for the blocky shape of the feldspars as the rims would be merely added, in a constant thickness, to the previously formed crystal. Then, too, by secondary enlargement the only difference in the twinning should be encountered in the rims as compared to the cores. This, however, is not the case. The complex twinning extends through rims and core alike and roughly divides the crystal into wedges and blocks.

FIG. 16. Andesite from near the snout of the Nisqually Glacier. A (top). x25, plane light. (Note the blocky character of the plagioclase; perhaps indicating a glomeroporphyritic mechanism in the formation of these crystals.) B (bottom). Under crossed nicols.

The large phenocrysts have other features worthy of mention. In the majority of rocks with a holocrystalline or hypocrystalline groundmass, the feldspars are whole and show little evidence of cleavage. In the more glassy flows and, especially, in the pyroclastics, as for example, the Muir pumice, the feldspars are a maze of cracks. Quick chilling and violence of ejection probably are responsible for much of the fracturing. Glass inclusions are also more numerous in rocks with a glassy groundmass. So many of the plagioclases have interesting peripheries. In a few of the flows, near the terminus of the Nisqually Glacier, the phenocrysts have ill-defined margins which seem to grade insensibly into the groundmass. On further examination, most all of the feldspars showed this effect; an amount out of all proportion to the percentage expected by tapering wedges. Williams (42) has figured a similar effect from the Mount Harkness lavas. Many of the feldspars of both the first and second generation have moderately-rounded corners, due to resorption.

In composition, the largest phenocrysts range from acidic to basic andesine and into acidic labradorite (Ab₆₃ to Ab₄₈). The majority are basic andesine.

The second generation of plagioclases must be considered as phenocrysts for they are distinctly larger than any minerals in the groundmass. Although they range considerably in size, most of them fall fairly close to the average of 0.3 mm. in length. The crystals are, for the most part, clear and fresh and present crisp, euhedral outlines to the groundmass. The shape is characteristically tabular to stubby rectangles with square cross-sections. Polysynthetic twinning is not always distinguishable, but a mild, oscillatory zoning is common. Carlsbad twinning is almost universally present, dividing the crystal into two equal halves. The composition varies from acidic andesine to intermediate labradorite (Ab₃₀ to Ab₆₁). The second generation plagioclases differ from the first in many respects. In size they are from one-tenth to one-fifth as large as the first. The outlines are euhedral, regular, and sharp; while the older phenocrysts are blocky and jagged. The twinning is most often confined to simple Carlsbad halves which are combined with a mild zoning; in the larger crystals, the twinning is very complex and the zoning is more pronounced. The smaller feldspars are relatively free from inclusions as compared to the larger ones.

The third generation of plagioclase must be considered as a part of the groundmass. They can often be detected in the glassy rocks as microlites but reach their greatest perfection in the more holocrystalline types. In size, the crystals of the third generation vary between microlites and 0.07 mm. in length, with the average falling close to 0.04 mm. The shape is typically elongated, microlithic laths often displaying castellated terminations. In spite of their incomplete terminations, they strongly resemble the feldspars of the second generation and, in addition, correspond almost identically to them in composition, in so far as the composition can be determined.

FIG. 17. Hypersthene andesite from Spray Park. (x62, plane light. This shows a hypersthene crystal surrounded by a jacket of augite. The ground mass is a deep red glass.)

FIG. 18. Hypersthene andesite from Faraway Rock. (x62, plane light. This association of hypersthene and magnetite is exceedingly common. The small, lighter colored crystals also in the hypersthene cluster are apatite. This type of pilitic groundmass is quite common.)

Hypersthene. As early as 1883, Oebbeke was fascinated by the hypersthene in a rock brought back to Germany from Mount Rainier by Professor Zittel. Due to his failure to isolate enough of the mineral for chemical analysis, Oebbeke (23) gave a rather complete description of its optical properties. A portion of the description follows:

Leider gelang es nicht, den pleochroitischen Pyroxen zu isoliren, um ihn einer chemischen Prüfung zu unterziehen. Es blieb daher nichts übrig, als sich auf die microscopische Untersuchung zu beschränken.

Die Schnitte senkrecht zur Längsrichtung ziegen ausser der prismatischen noch eine pinakoidal Spaltbarkeit; in den Längsschnitten, besonders in denjenigen der kleineren Krystalle, ist die Spaltbarkeit nicht immer deutlich, häufig beobachtet man in ihnen enie zur Längsrichtung senkrecht verlaufende Querabsonderung. An Einschlüsssen sind die erwähnten Krystalle arm. Ausser glas, Magnetit und Apatit wurden keine Einschlüsse gefunden.

Der pleochroismus der Längsschnitten ist der Richtung der C Axe grünlich (hellgrünlich-blaulichgrün), senkrecht dazu gelblichgrün, hellbrüunlich odor rötlichbraun.

Die Schnitte senkrecht zur Längrichtung ziegen parallel der pinakoidalen Spaltbarkeit hellbraune bis rötlichbraune, senkrecht dazu grünlich gelbe bis hellbraunliche Farben.

Wurden diese Schnitte im convergenten polarisitiren Licht untersucht, so sah man eine optische Axe austreten; die Ebene der optischen Axen geht der pinakoidalen Spaltbarkeit parallel.

Die Längschnitte in gleicher Weise untersucht liessen bald den Austritt einer optischen Axe zeimlich am Rande des Gesichtsfeldes erkennen, bald konnte deutlich wahrgenommen werden, dass eine Mittellinie senkrecht zu ihnen stehen und dass der Axenwinkel ein ziemlich grösser sein müsse.

Rarely can a rock be found on the slopes of Rainier which does not contain at least a few crystals of hypersthene. It normally presents euhedral to subhedral outlines and forms short to long rectangular crystals with rather abrupt terminations. When pyramids and domes occur, they are remarkably flat and quite often have slightly-rounded corners. The prismatic cleavage is well shown and the 010 parting, while not so regular, is usually present. The phenocrysts average 0.5 mm. in length, but may be either much larger or smaller than this mean. In the more holocrystalline varieties, hypersthene may occur in the groundmass as small stubby crystals or displaying a lath like habit and averaging .05 mm. in length.

The pleochrosim is one of the most outstanding characteristics of the mineral and, while varying in intensity among the different flows, it is always pronounced. Generally X is an orange color, Y a yellowish brown, and Z a green. The optic angle changes considerably depending, in part, on the amount of magnetite included within the different crystals. In those heavily charged with magnetite the 2 V drops as low as (-) 60 degrees, but in the clearer crystals, and these are far more common, the 2 V averages (-) 70 degrees. In a few cases an optic angle of 90 degrees was obtained. The determination of refringence based on immersion oils was Np=1.765 to 1.680 and Ng=1.700 to 1.705.

Following the example set by the feldspars the hypersthene has a tendency to form glomerophyritic clusters or to associate with the augite, plagioclase and magnetite in cumulophyric groups. Among the inclusions found in hypersthene, large grains of magnetite and rounded blebs of glass are equally important with clear stubby apatite prisms being slightly less common.

An interesting association is that of hypersthene and augite. Occasionally the augite will crystallize in jackets around the partially resorbed hypersthene so that both pyroxenes will have their C axes and their prismatic cleavages parallel to each other. A similar structure was observed by Williams (42) in the Red Mountain basalts. In one instance, in a brilliant red flow from Spray Park the augite jacket showed a lamellar twinning (001) which occurred in a direct line on each side of the jacket but was interrupted by hypersthene in the center. It is difficult to determine whether the jackets continue over the end of the hypersthene or not. In most of the sections containing the jackets the hypersthene ends are free. At times the augite completely encircles the rhombic pyroxenes but, in these, the section is not cut exactly parallel to the C axis.

FIG 19. Andesite from McClure Rock. (x25, plane light. A cumulophyric group of plagioclase, augite. and hypersthene in a glassy groundmass charged with magnetite.)

FIG 20. Andesite from Register Rock at the summit of the mountain. (x25, plane light. It is quite usual for the rocks near the summit to display a well-developed fluxion structure.)

Monoclinic Hypersthene

A noteworthy feature of the rocks of Mt. Rainier is the clino-hypersthene encountered in the lavas. Scarcely a section examined failed to show the presence of at least one crystal of hypersthene with a definitely inclined extinction. These angles range from 1° up to a maximum of 15° although by far the greater percent age of this particular type of hypersthene usually had maximum angles of 8°. Occasionally zoned crystals were observed in which the extinction angle varied from center to periphery by as much as 3° The remaining optical properties for the clino-hypersthene are very similar for those given for the rhombic varieties. The sign was universally negative, the 2 V and pleochroism were approximately identical and the refringence was only slightly higher in the inclined types.

Clino-hypersthene has been described by Winchell (47) as having indices of Ng=1.73, Nm=1.715 and Np=1.713 and by graphic solution he found the (+) 2 V to be 30° and ZAC of 46° From this description it is readily seen that Winchell's clino-hypersthene has little in common with that from Mount Rainier.

Recently J. Verhoogen (48) found clino-hypersthene in the lavas from Mount St. Helens and has since been studying pyroxenes in lavas from Lassen Peak, Mount Shasta and Mount Theilson. The results of his work will be published in a forthcoming paper. An exchange of material and information on these pyroxenes indicates that the clino-hypersthenes from the various Cascade volcanoes are practically identical although similar material has never been described elsewhere.

Augite. Hypersthene and augite go hand-in-hand as the typical mafic minerals of the Rainier volcanics. Seldom is one present without the other and, to have both absent is, indeed, a rarity. The monoclinic pyroxenes closely approach the hypersthene from a quantitative standpoint but only in a limited number of cases can they be considered the dominant mineral. Usually in the Cascade volcanoes the augite is not so persistently idiomorphic as the hypersthene and the Rainier pyroxenes are no exception to this generalization.

Subhedral crystals commonly assume either an elongated tabular form or stubby prisms modified by pyramidal terminations. Most prominently displayed are the phenocrysts, similar in dimensions to the hypersthene (0.5 mm.), but small grains of augite as a groundmass constituent are very abundant. These rounded grains or microlithic laths average 0.02 mm. in length. A good many crystals show a greenish color and some exhibit a weak pleochroism with X and Z greenish and Y brownish. Ng=1.700±.001. The maximum extinction angle (ZAC) reaches 49° Inclusions in the augite are identical to those in the hypersthene with brownish glass and magnetite being prevalent.

The augite does not cluster with its own kind as do the feldspars and the hypersthene but it is included in the cumulophyric groups. Both of the pyroxenes are exceedingly fresh. In a few instances, slight leaching by iron-rich solutions has resulted in the deposition of hematite, limonite and possibly gothite along the cleavage cracks. Although the pyroxenes resemble each other, they may be differentiated by the stronger pleochroism and idiomorphism, lower birefringence and parallel extinction of the hypersthene.

Olivine. It can scarcely be classed as an essential mineral of the Mount Rainier rocks. In rare cases, it becomes almost as important as the pyroxenes but, out of 200 sections picked at random, olivine was found to be plentiful in only 4 cases. Out of this same number, a few small grains of olivine were detected in 21 cases.

The mineral is present both as a phenocryst, averaging 0.2 mm. in greatest dimension, and as a constituent of the groundmass, in which the granules average .05 mm. in diameter. The shape of each is typically subhedral. In this section, olivine is outstanding because of its clear, clean appearance, the scarcity of cleavage cracks as compared with the pyroxenes, and the lustrous sheen of the interference colors. Of all the phenocrysts, olivine is the least contaminated with inclusions.

As a rule, the mineral is little altered. However, certain interesting alteration and reaction effects have taken place along the peripheries of some of the crystals. In a flow from Spray Park, the olivine is surrounded by a margin of golden-brown bowlingite with a very small 2 V (about 10°). A few grains have wide marginal rims of hornblende, which are, in turn, studded with magnetite dust. Iddings (17) first mentioned this effect in one of the Rainier rocks from the Survey collection. In the lavas, wherein the groundmass is heavily charged with hematite dust, there is a concentration of hematite encircling the olivine, forming deep red rims. At times the iron-rich solutions have seeped along widened cleavage cracks, coloring them a brownish-red.

Hornblende. Hornblende occurs sporadically throughout the Rainier rocks as an accessory, or, in a few limited cases, as an essential constituent. Examples of the latter are found at Edith Creek in Paradise Valley or at St. Elmos Pass between Interglacier and the Winthrop Glacier. Without exception the hornblende crystals are edged with magnetite or are entirely replaced by that mineral. This effect is common to all the described Cascade volcanoes. When the groundmass is hematitic, the surrounding rims or pseudomorphs are also of hematite or limonite.

The hornblende is of the basaltic or oxyhornblende variety and occurs as long, euhedral prisms, or as stubby basal pinacoids with the characteristic cleavage. The length varies from 0.3 mm. to 0.6 mm. with the average being closer to the smaller figure. The pleochroism changes from a greenish-yellow (X), to a deep red dish-brown (Y and Z), and the extinction (ZAC) is from O° to 1° In addition to the large quantity of magnetite, both plagioclase and apatite are present as inclusions. This is quite comparable to the hornblende described by Ransome (25) from Goldfield, Nevada.

FIG 21. Hornblende andesite from St. Elmos Pass. (x25, plane light. The black, elongated crystals are hornblende now largely replaced by magnetite. Note the fluxion structure.)

FIG. 22. Hornblende andesite from Edith Creek, Paradise Valley. (x25, plane light. Both the basal and prismatic sections of hornblende are rimmed by a border of magnetite.)

After a petrographical examination of a number of rocks from Mount Rainier, one is strongly impressed by the monotonous regularity and repetition of the few constituent minerals. They not only vary little in kind, but maintain a constant size and shape and relation to their associates. The only relief is to be sought in the diversity of the groundmass. Smith (30) well appreciated this condition when he stated:

The megascopic differences are mostly referable to the groundmass characters; the color of the rock being dependent on the color and proportion of glassy base present. Therefore, the degree of crystallization of the groundmass constituents is of more importance in determining the megascopic appearance than is the mineralogical composition.

In the pages to follow, the types of groundmass will be mentioned according to their percentage of glass; the holohyaline coming first. Succeeding this will be the description of some unusual textural features.

Holohyaline Groundmass. Approximately 16 per cent of the Rainier rocks have a holohyaline groundmass. These vary in color from shiny black, brilliant red, brown, to almost white. Of all the colors, black is the most characteristic, and the converse is generally true that all the black rocks have a glassy groundmass. A number of black specimens from the crater rim at the summit show, in thin section, a dark brown, glassy base, crowded with magnetite dust and innumerable crystallites, probably of feldspar. In another black specimen from Old Desolate, the glass is also dark brown and contains microliths of feldspar 0.03 mm. in length and, in addition, granules of magnetite and prisms of apatite one-third the size of the feld spars.

In the brilliant red varieties, the glass is so full of hematite dust as to be almost opaque. Specimens of this type are numerous on the north side of the mountain in the vicinity of Seattle and Spray parks and also above the 11,000-foot contour line.

Less brilliantly colored than the red andesites and agglomerates, the pumice fragments so liberally scattered over the Park are shades of dirty brown to black. In thin section, the glass varies from a deep black through all shades of brown to colorless.

A milky-white lava was encountered on the west side of the Winthrop Glacier at an elevation of 6,000 feet. The microscope revealed the base to be a clear, colorless glass sporadically studded with a few grains of magnetite.

With the exception of certain pumice fragments, all the above mentioned rocks contain the usual phenocrysts. The glassy groundmass imparts to the andesites a brilliance in color and a lustrous freshness that makes them easily separable from the more holocrystalline varieties.

Hypo- and Holocrystalline Groundmass. The hypocrystalline base is found in approximately 80 per cent of the Rainier rocks, while the holocrystalline type is limited to but 4 per cent. Because of the small percentage of the holocrystalline material, and its insensible gradation into the hypocrystalline, it is convenient to group the two types together.

Contrasted with the dark holohyaline bases, the rocks of this group are normally medium to light gray in color. Abundant and easily accessible exposures of the light gray rocks may be seen along the Longmire-Paradise road at the Ramparts, Miller cut-off, Mazama Ridge and at Paradise. On the other side of the mountain, at Yakima Park, the nearest Rainier lavas with the hypocrystalline base outcrop on top of Burroughs Mountain.

The groundmass of these rocks, as seen under the microscope, presents a wealth of textures ranging from predominantly glassy to holocrystalline. The more glassy bases contain innumerable thin microlites and crystallites of feldspar, and less frequently augite and hypersthene. Magnetite dust, or granules, are never lacking. With a few exceptions, the glass is not at all conspicuous because of its lack of color. The flows outcropping at Frog Heaven and near the highest peak on Burroughs Mountain are exceptional in that their base is a coffee-brown glass and, were it not for the myriad of incipient minerals, these would belong to the holohyaline group.

As the percentage of glass diminishes, the narrow microlites and crystallites tend to widen, enlarge and to assert a little more sharply their crystallographic habits. The resulting laths of plagioclase, 0.02 mm. in length, are normally felted in a hyalopilitic texture; less commonly they are aligned in a sub-parallel fashion indicative of flowage. Hypersthene assumes a size and tabular shape very similar to the plagioclase, while the augite forms sub-rounded granules 0.01 mm. or less in diameter. When clear, the colorless, glassy residuum has an index of refraction of 1.511. However, magnetite dust or granules are so universally present that a dull, dusty-gray color is typical of more than 60 per cent of the Rainier lavas. It is with difficulty that the microlithic laths can be distinguished in so turbid a base; only the clear phenocrysts are really outstanding. The dull, platy andesites of the Cowlitz rocks are a beautiful example of the dusty, gray, magnetite-charged base.

As the matrix becomes more and more feldspathic, a pilotaxitic texture is universal. In this instance, the ubiquitous magnetite no longer is in a dusty form but rather is present as small grains, averaging 0.03 mm. in diameter. With the concentration of the magnetite into grains, the cloudiness so characteristic of the hyalopilitic types disappears and the pilitic base is crisp and clear. The feldspars of the groundmass have also grown in size until they are scarcely distinguishable from the smaller, or second generation, phenocrysts, and, indeed, they may be one and the same thing. The augite and hypersthene have increased their dimensions in proportion to the feldspars and attain lengths of 0.1 mm.

Miscellaneous Features of the Groundmass

Blotchy Groundmass. At least 20 per cent of the rocks from Mount Rainier have a peculiar patchy effect in the groundmass. The blotches are irregular in shape, typically with rounded or sub-rounded margins and averaging slightly larger than the phenocrysts in size. These are caused by the concentration of the magnetite dust in patches, and noticeably darkening the pale or colorless glassy matrix. Less frequently, as in a specimen taken near Sluiskin Falls, the blotches are dark, glassy areas in a lighter and more holocrystalline portion of the groundmass. Williams (42) mentions similar blotches from Mount Diller.

FIG. 23. Andesite from Panorama Point. (x25, plane light. Note the peculiar blotchy effect in the groundmass. The darker areas represent a concentration of magnetite and kaolinitic material.)

FIG 24. Andesite from Gibraltar. (x25, plane light. Diktylaxitic texture in a porphyritic andesite in which there is a minimum amount of glassy matrix.)

Open Textured Andesites. Along the base of Gibraltar, on the summit route, is an interesting occurrence of a highly porphyritic material which contains a minimum amount of glassy residuum. The phenocrysts are plagioclase, augite, and hypersthene—all presenting idiomorphic outlines. These average 0.7 mm. in length and have a more rectangular and stubby shape than is usually characteristic of the Rainier phenocrysts. The base is a colorless to pale-brown glass, present in sufficient quantity to serve only as a binding agent to hold the phenocrysts together. In the interstitial areas between the large crystals, the glass is wanting. The resulting effect is very similar to the diktytaxitic texture as described by Fuller. (12) The feldspars, in this case, are more stubby than the "delicate laths of light-gray labradorite" mentioned in the original description. However, similarities are to be found in the net-like arrangement of the feldspars, and the ends of the minerals protruding into the cavities. In both cases the residuum must have possessed sufficient fluidity to permit its easy escape from the crystal mesh.