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

KEECHELUS ANDESITIC SERIES

The Keechelus andesitic series is a large mass of rocks covering hundreds, and possibly thousands, of square miles of the Cascades and averaging more than one half mile in thickness within the Park. The dominant types are a series of massive tuffs, breccias and porphyries of andesitic composition which have been altered and indurated. The subordinate types include andesite flows, felsites, basalts, hornfels and sediments.

The name, Keechelus andesitic series, was applied originally to similar rocks occurring in the Snoqualmie quadrangle which adjoins the Mount Rainier National Park sheet immediately to the north and east. The Keechelus rocks of the Snoqualmie region can be traced with continuity directly across into the Mount Rainier region. Smith and Calkins (34) predicted that: "The Keechelus series is probably the most voluminous assemblage of rocks in the Cascade Range for some distance to the south and west," and, later work has proved their prediction. Fuller (11) found Keechelus in the northern part of the Cedar Lake quadrangle, which lies west of the Snoqualmie and immediately north of Mount Rainier. (4) Observations made by the writer during brief visits to the southern half of the Cedar Lake region indicate that Keechelus rocks occupy a major portion of that quadrangle. This same material has been traced into the Mount Aix quadrangle, lying to the east of Mount Rainier. To the south, the extent is unknown. To the west, the sediments of Puget Sound come almost up to the Park boundary and thus limit the extension of the Keechelus in that direction. Within the Mount Rainier sheet this series outcrops almost continuously along all four of the Park boundaries and extends inward, finally disappearing under the lavas of Mount Rainier.

Regarding the distribution of rock types, the Keechelus formation may be divided into areas which are relatively homogeneous and those which present the extreme in variability. In practically every case, the former are far removed from intrusive bodies; while the latter are adjacent to known exposures of granodiorite. Homogeneity prevails in the Ohanepecosh district in the southeastern corner, where the only rocks exposed are altered and indurated porphyries and breccias of intermediate composition. The same is true in the southwestern corner, in the neighborhood of Mount Wow. The 3,000-foot cliff on the east side of this mountain is an excellent place to observe the massive character of these porphyries and breccias.

Elsewhere, considerable more variety is shown. This is notably true in the northeast corner, along the White River, where granodiorite invades the Keechelus with many interesting effects. The rocks vary from massive porphyries 30 meters thick, to thinly-bedded pyroclastics displaying marked changes almost every centimeter. The colors may be chocolate-brown, purple, tan, and black, but the inevitable greens and grays, so typical of the Keechelus, prevail. The rocks in this vicinity are about as fresh as any to be found in the formation and all have partaken in a gentle dip to the northward. The Mather Memorial Parkway has provided almost continuous exposures for miles in a long, diagonal section across this region, extending from the granodiorite below, across the xenolith swarms of the contact, and up into the Keechelus. Numerous intrusives of small dikes and sills, as well as irregular tabular masses, have contributed several new types to this already hetrogeneous group of rocks.

FIG. 3. Mother Mountain from Seattle Park, looking northward. (This entire mountain is composed of Keechelus andesite. The upper portion of Mother Mountain is a series of almost flat-lying flows; the lower portion is largely massive and altered breccias and porphyries.)

Another locality noteworthy for its variety is the Tatoosh Range, paralleling the southern border of the Park. Here the Keechelus is the capping, while the granodiorite forms the base of the range. The intrusive contact is well exposed on the northern face. This area is worthy of more detailed study than the writer could possibly allow at present and will receive further attention.

Before going deeper into the discussion of the Keechelus, a few of the rock types will be described. It is impossible to obtain an adequate representation of this formation from a suite of specimens. There is not only great diversity in the rocks, but infinite gradations that exist between the end types. However, a number of specimens have been selected for petrographic description in the hope that they will convey a general impression of the material in this formation.

MINERAL MOUNTAIN ANDESITE PORPHYRY

At Mineral Mountain, near Mystic Lake, a porphyry occurs as an irregularly shaped mass, several hundred meters in width. Although definitely invading the lower portion of the Keechelus in this vicinity, it continues upward and forms an integral part of this series above. The specimens from Mineral Mountain represent a type of porphyry commonly found in the Keechelus and are good examples of the younger portions of this formation.

The rock is medium-gray in color with numerous stringers of hornblende 1 mm. in width traversing the mass in every direction. Phenocrysts of feldspar occur frequently and attain an average length of 4 mm. The groundmass is fine grained, with the individual crystals averaging about 1 mm. or less in diameter.

Microscopical Petrography. The porphyry consists of euhedral plagioclase crystals imbedded in felt of plagioclase, hornblende, biotite, and quartz. The phenocrysts exhibit a very intense oscillatory zoning and have lines of inclusions consisting of flakes of biotite, hornblende and magnetite dust parallel to the euhedral borders. Intermediate andesine, with a composition of Ab₆₂An₃₈, is most prevalent. However, a few of the phenocrysts are rather ragged and these contain clear patches with the composition of labradorite Ab₄₄An₅₆.

The ragged and rather moth-eaten appearance is characteristic of the smaller phenocrysts but it reaches its greatest development in the feldspars of the fine grained groundmass. It is due to innumerable particles of included mafic minerals. With higher magnification, the particles are easily distinguished as being identical to those found outside the feldspars and which make up so large a portion of this rock.

The hornblende is usually deeply-colored with X, a pale yellow; Y, a dark, brownish-green; and Z, a dark green. The refractive index reaches a maximum of 1.660. Some of the crystals contain abundant subrectangular magnetite grains and, in such cases, the hornblende is noticeably paler in color. The magnetite is always intimately associated with the hornblende and occurs in grains up to 1 mm. in length.

The biotite is of the siderophyllite variety with X a clear yellow and Y and Z both a deep reddish-brown. Quartz occurs both interstitially and as a mosaic associated with the more idiomorphic hornblende crystals. The percentage of each is approximately as follows: feldspar phenocrysts, 15 per cent; feldspar in the fine grained groundmass, 30 per cent; hornblende, 25 per cent; magnetite, 10 per cent; quartz, 12 per cent; and, biotite, 8 per cent.

FIG. 4. Mineral Mountain andesite porphyry. A (top). x25, plane light. (Large L-shaped plagioclase phenocryst in a matrix of hornblende, plagioclase, biotite and quartz.) B (bottom). Same view as above, under crossed nicols.

SHEEPSKULL GAP TUFFS

A rather typical Keechelus tuff outcrops in the vicinity of Sheepskull Gap. Here the fragmental rocks are well-bedded to massive; all are essentially flat-lying, and have associated with them a small amount of igneous material in the form of minor injected bodies.

In the specimen chosen from Sheepskull Gap, the color is predominantly gray, but, on closer inspection, the individual fragments may be seen to be light gray, dark gray, green or even purple. Because of this variation in color, the fragments range in size from 50 mm. or more down to particles too small to be seen with the unaided eye. The average is close to 3 or 4 mm. in diameter. As a result of induration, the rock is now so hard and compact that it can scarcely be scratched with a knife.

Microscopical Petrography. Under the microscope, the fragments can be identified as chips of andesite and angular pieces of feldspar crystals. Both are embedded in a dusty, almost opaque matrix of extremely fine ash.

The andesitic chips usually exhibit a distinct felted texture but they may be porphyritic or even cryptocrystalline. In any case, cloudy feldspars are the dominant mineral. A few clearer crystals indicate the composition to be andesine (Ab₆₃-An₃₇). The only other distinguishable minerals in these chips are quartz, in amygdules, chlorite, with properties very similar to clinochlore, and magnetite, occurring as small specks disseminated throughout the rock. The percentage of matrix between the fragments is relatively small. It contains numerous patches of clear, green clinochlore, stringers and angular pieces of calcite, gray dusty ash, and small flakes of andesite and feldspar.

The approximate content of the tuff is: andesite chips, 35 per cent; feldspar chips, 30 per cent; ash matrix, 18 per cent; chlorite, 10 per cent; quartz, 3 per cent; calcite and magnetite, 2 per cent each.

FIG. 5. Sheepskull Gap tuffs, x25, plane light. (The rock is composed of angular chips of several andesitic types. A few cloudy plagioclase fragments are also embedded in the fine ash matrix.)

SOURDOUGH MOUNTAIN BRECCIAS

The breccias exposed along the northern face of the Sourdough Mountains near Yakima Park are quite different from the distinctly fragmental tuffs of the Sheepskull Gap area. At first sight, the breccias appear to be black porphyritic lavas but, on closer examination, the lighter phenocrysts are abundant in some patches and wholly lacking in others. In the field, the so-called patches have no distinct boundaries but grade from a lighter to darker colored, or from a porphyritic to a non-porphyritic type. Some of the larger patches average over 1 meter in diameter but the more ordinary ones range from 30 to 40 mm. in length. This patchy texture is one of the most characteristic features of the Keechelus formation and usually is very obvious because of the differences in color of the various components. In the case of the Sourdough Mountain breccias, the colors are limited to a very dark grey and black and, hence, the general texture is more obscure. The lighter grey color is caused by swarms of plagioclase phenocrysts, approximately 3 mm. in length; while in the darker patches, the phenocrysts are either lacking or are very sparse in the dull, dense groundmass.

Microscopical Petrography. Several kinds of phenocrysts are visible. The larger ones so readily recognized in the hand specimen are, in reality, merely glomerophyritic groups of plagioclase phenocrysts which average less than .4 mm. in length. These range in composition from acidic to basic andesine but, generally, are close to Ab₅₅-An₄₅. Some show pronounced zoning, while others are twinned according to the Carlsbad law. All are charged with quantities of magnetite dust, or antigorite flakes, or both. The arrangement of the inclusions is either in zones, parallel to the periphery of the crystal, or confined almost exclusively to the center and surrounded by a clear, and slightly more acidic, rim.

The pyroxene (?) phenocrysts were of the same size as the feldspars, but since have been uralitized and completely replaced by a fibrous antigorite with smaller amounts of biotite, magnetite and quartz.

The groundmass is so heavily charged with a dense magnetite dust that it is almost opaque. However, minute flakes of feldspar and antigorite are scattered throughout the groundmass, and under high magnification, the alternating transparent and opaque minerals give it a salt and pepper effect.

Cutting across the rock are thin stringers of clear, green antigorite and titanite. These clearly indicate the type of solutions which permeated the mass and contributed so much to its present condition. The composition is as follows: plagioclase, 35 per cent; antigorite, 20 per cent; groundmass (including magnetite), 35 per cent, biotite, 7 per cent; and, quartz, 5 per cent.

FIG. 6. Sourdough Mountain breccias, x25, plane light. A (top). Section view. (The larger and more irregular phenocrysts of plagioclase are charged with magnetite dust and flakes of antigorite. A small veinlet of antigorite and titanite may be seen cutting across the rock.) B (bottom). Section view. (The fragmental nature of the rock is not always apparent. This section shows the darker fragments—heavily charged with magnetite dust—in a much lighter colored matrix.)

CHINOOK PASS DIORITE PORPHYRY

A diorite porphyry outcrops 1 mile below the summit of Chinook pass as a tabular mass approximately 25 meters in thickness. It is separated from the country rock, an andesitic breccia, by a narrow border noticeably finer in grain than the central portion of the porphyry. Following the general color scheme of the Keechelus, this rock is dark greenish-gray and contains swarms of lighter gray feldspar phenocrysts.

Microscopical Petrography. The plagioclases form a definitely seriated fabric with the size of the euhedral crystals ranging from .5 mm. up to 4 mm. in length. The larger crystals represent the "phenocrysts" identified in the hand specimens. The remaining minerals occur interstitially as roughly triangular patches filling up what little space is left by the abundant feldspar. All the minerals have participated in a widespread alteration in the form of saussitierization and uralitization. This may be attributed to paulopost action or, as an alternative, to mild regional met amorphism.

The plagioclase is traversed by numerous stringers of phrenite, chlorite, epidote, and more acid plagioclase. Labradorite, with a composition of Ab₄₀An₆₀ is general, but more acid rims reach Ab₅₀An₅₀. Water-clear plagioclase with a markedly lower refractive index occurs irregularly throughout the feldspars. The twinning is not entirely obscured by the alteration and follows the Carlsbad and Albite laws, the latter displaying relatively wide lamellae. About 50 per cent of the feldspars contain included particles of chlorite that are independent of the stringers and which choose to align themselves either parallel or at right angles to the twinning lamellae.

The pyroxenes have not fared so well as the feldspars. Few remnants of the original augite remain and these are enclosed by wide borders of chlorite and actinolite. Large octahedra and cubes of magnetite are always associated with the altered pyroxenes. Quartz, showing undulatory extinction, and most of the other minerals listed below, occur interstitially. The mineral percentage is roughly as follows: plagioclase, 50 per cent; quartz, 10 per cent; actinolite, 7 per cent; phrenite, 5 per cent; augite, 4 per cent; epidote, 4 per cent; and, clinozoisite, 3 per cent.

FIG. 7. Chinook Pass diorite porphyry. A (top). x25, plane light. (The plagioclase forms a definitely seriated fabric of roughly euhedral crystals. Altered pyroxenes and chlorite occur interstitially between the blocky feldspars.) B (bottom). Under crossed nicols. (A portion of one of the larger plagioclase crystals can he seen in the upper right hand side.)

LONGMIRE ACID BRECCIAS

At the settlement of Longmire, on the Nisqually River, are a series of almost white breccias. These are exposed to the best advantage in the low, rounded knobs behind the government houses and along the road leading across the suspension bridge to the auto camp grounds. Being so massive and light-colored, it is difficult, from a distance, to distinguish these breccias from the granodiorite exposed in the Paradise River valley nearby. In the hand specimen, the true texture of the rock is revealed as the fragments display a wide assortment of colors, ranging from a purplish-gray to white. The larger ones average 30 mm. or more in diameter, but the smaller ones, ranging from 5 to 10 mm., are much more plentiful. Little regularity exists in regard to the size; shape, and color of the rock particles. A noticeable feature in the hand specimen is the clear lustrous quartz grains embedded in a dense, dull matrix.

FIG. 8. Longmire acid breccias, x25, plane light. (The clear quartz crystals show resorbed boundaries and embayments. The andesitic fragments are so altered and turbid that it is often difficult to determine their constituents and outlines.)

Microscopical Petrography. In thin section, the fragmental texture is not so pronounced as the megascopic examination would lead one to expect. All the fragments are turbid and altered and none differ markedly from the others in texture, color, and amount of alteration. The only relief from the cloudiness are the clear, fresh grains of quartz. These have a general subangular shape with beautifully resorbed and sinuous borders. Occasionally small embayments will protrude one-third the diameter of the crystal. Some of the fragments show a dense pilotaxitic texture wherein the small feldspar laths may be determined only under the highest magnification as intermediate andesine. The fragments may display either a microcrystalline mosaic of quartz and feldspar, or may consist of actinolite, chlorite, clinozoisite, and feldspar, with accessory magnetite. A few display an amygdaloidal structure; the vesicles being filled with quartz, chlorite, and titanite. Rare patches of intensely pleochroic brown biotite and clear quartz occur as a mosaic within these fragments.

The remaining portion of the rock is made up of cryptocrystalline fragments and a microfelsitic base containing scattered crystals of plagioclase. The feldspars are subhedral in shape, cloudy with kaolin and filled with inclusions, and the determinable ones have a composition ranging from andesine (Ab₅₃-An₄₇) to oligoclase (Ab₇₈-An₂₂). It is impossible to estimate with any degree of accuracy the percentage composition of a rock of this kind.

STARBO ALTERED TUFFS

In Glacier Basin, between the Emmons and Winthrop glaciers, the Keechelus tuffs and porphyries have been invaded by granodiorite. The old Starbo mining camp has its tunnels immediately above the contact where an attempt was made to mine the sulphide seams in the narrow joints of the tuffs and porphyries. This massive, indurated, and altered series continues upward to the base of Mount Ruth where it is overlain unconformably by the lavas of Mount Rainier. The texture varies from porphyritic to a porphyry with inclusions and, finally, results in a mass so charged with chips of rock and minerals, averaging between 1 and 4 mm. in length, as to make it a tuff. Yet the physical properties of the rock, such as its extreme hardness, the even, dark gray to black color, and general compactness, certainly do not suggest a fragmental origin. The true texture is well brought out under the microscope.

FIG. 9. Starbo altered tuff, x25, plane light. (The plagioclase is represented by the lighter, partially-rounded crystals. The groundmass is crypto- to microcrystalline material and contains angular andesite chips identified by their felted texture.)

Microscopical Petrography. The rock may be divided into three roughly equal parts; the chips of rock, the phenocrysts and fragments of feldspar, and the micro- to cryptocrystalline matrix. Most of the rock chips have a fine pilotaxitic felt of slender plagioclase laths and abundant interstitial magnetite with a little actinolite. The feldspars are very cloudy but both the low refractive index and the small extinction angle suggest a composition close to intermediate oligoclase. Certain chips are charged so completely with magnetite dust that they are almost opaque. Other of the rock fragments display a felsitic texture of quartz and feldspar with abundant magnetite and occasionally a cloudy feldspar phenocryst.

The widely distributed chips and angular fragments of feldspar present many features. They may be either clear or very cloudy, euhedral or anhedral, twinned or untwinned, or range in size from 2 to .4 mm. or less. The composition lies between Ab₁₃ and Ab₃₀. The untwinned ones are often remarkably clear and show wavy extinction, due to zoning. These strongly resemble the quartz found in other Keechelus tuffs but the negative biaxial sign discloses their character. The majority of the feldspars are twinned in some fashion or else show oscillatory zoning.

The micro- to cryptocrystalline matrix contains magnetite specks, and minute flakes of actinolite and feldspar, in addition to much unidentifiable material, and serves, more or less, as a base for the rock and mineral fragments.

CAYUSE PASS ACID HORNFELS

At Cayuse Pass on the Mather Memorial Parkway, the road is blasted out of an acid hornfels overlain, conformably by andesitic tuffs, all partaking in a gentle northward dip.

The rocks are generally buff colored with all gradations from a pale cream color to a deep rusty brown. Individual beds vary from a centimeter to many meters in width with each layer distinguished by a noticeable change in color and grain size. The examination of the hand specimen shows small spots of buff colored material, alternating with dense particles of a white substance—all having a dull appearance. Nevertheless, the rock is extremely hard and cannot be scratched with a knife blade.

Microscopical Petrography. The rock exhibits a palimsest structure indicative of psammitic sediments, now partially hornfelsed with the production of feldspar, sericite and biotite. Added to this is an abundance of quartz with a smaller amount of limonite, which is partially responsible for the buff color. Lesser quantities of calcite occur interstitially.

Haphazardly distributed throughout the rock are groups of feldspar porphyroblasts with as many as ten individual crystals occurring in one of these clusters. The turbid appearance of the plagioclase is not due to alteration but rather to the presence of many inclusions, of the same type and size, as are found elsewhere in the groundmass. So perfect is the continuation of the groundmass materials through the porphyroblasts that it is practically impossible to locate the larger crystals in plane light. The porphyroblasts attain lengths up to one mm. and vary in shape from idioblastic to hypidioblastic. They may be twinned according to either the Carlsbad or Albite laws or both, or may show a very complex twinning. The composition is albite (Ab₂₂-An₈).

The groundmass contains quartz and feldspar as the essential minerals. The quartz is typically xenoblastic and averages .2 mm. in length. The feldspars contain inclusions of a type and number comparable to the porphyroblasts but they do not display the prominent twinning of the larger feldspars. Judging from the low refractive index the smaller laths of plagioclase have the composition of albite and are, undoubtedly, similar to the porphyroblasts in the percentage of Ab to An.

The biotite flakes range around .02 mm. in size, but they are usually in clusters that attain an overall length of .1 mm. or more. The sericite and limonite occur around the quartz and feldspar grains and have dimensions similar to the biotite specks. The calcite, on the other hand, has entered the rock in small veinlets, replacing other minerals, notably the feldspar phenocrysts, and often has clear quartz as an associate.

The percentage composition is approximately as follows: phenocrysts (plagioclase), 10 per cent; plagioclase (groundmass), 40 per cent; quartz, 30 per cent; biotite, 8 per cent; sericite, 5 per cent; limonite, 4 per cent; and, calcite, 3 per cent.

FIG. 10. Cayuse Pass acid hornfels. A (top). x25, plane light. (A microcrystalline matrix of quartz and feldspar containing plagioclase porphyroblasts. Note that it is almost impossible to detect the presence of the larger feldspars in this photomicrograph.) B (bottom). Under crossed nicols. (This view is the same as above; note the cluster of feldspar porphyroblasts.)

MOWICH HYPERSTHENE BASALT

Breccias and porphyries make up by far the greater portion of the Keechelus, but to emphasize the wide divergence of rock types encountered in this one formation, a felsite and a basalt are also included in this description. Some of the freshest basaltic flows in the entire formation are exposed between Mowich and Eunice lakes in the northwestern corner of the Park. Vertical cliffs in these thick, horizontal sheets often display well-defined columns, reaching a height of 30 meters or more. The rock is a coarsely porphyritic, black basalt. The feldspar phenocrysts are but slightly lighter than the groundmass in color but their presence is betrayed by their clear, glassy lustre and the light reflected from the cleavage planes. The phenocrysts average 3 mm. in length and are set in a rich, black, aphanitic groundmass.

Microscopical Petrography. The phenocrysts of feldspar, pyroxenes and olivine are crowded in a holocrystalline groundmass, displaying a granulitic texture. The larger feldspars invariably show strong oscillatory zoning and they may be filled with concentric or scattered inclusions or be water-clear. Albite twinning, with wide lamellae, is common; Carlsbad, less so. The composition varies from very basic andesine through labradorite with the percentage ranging from Ab₄₇-An₅₃ to Ab₃₅-An₆₅.

The mafic phenocrysts include augite, olivine and hypersthene. The fresh augite crystals occur as euhedral stubby prisms exhibiting lamellar twinning on 100. The faint pleochrosim and reddish tinge suggest a titaniferous variety. The rounded olivine grains are traversed with widened cracks filled with lamellar, greenish-yellow antigorite. The centers are remarkably clear and fresh. Hypersthene is especially abundant, both as euhedral phenocrysts and as smaller grains in the groundmass. In each case the mineral is noticeably pleochroic and has suffered a slight amount of serpentinization but to a lesser extent than the olivine.

The groundmass contains small laths of feldspar, grains of hypersthene, augite, and olivine, all averaging .06 mm. or less in diameter. Magnetite cubes and octahedra, twice the size of the other grains, are scattered through the groundmass. A striking feature of the feldspathic base is the presence of fine acicular apatite, often penetrating through several plagioclase crystals and attaining a length of .3 mm. or more. These are so numerous that even under the limited area of high magnification no portion of the groundmass has failed to show crowds of these fine needles. The base also contains considerable alteration products in the form of antigorite, chlorite, and cloudy material. The composition is approximately as follows: feldspar, 48 per cent; hypersthene 19 per cent; augite, 13 per cent; antigorite and other alteration products, 12 per cent; olivine, 4 per cent; and, magnetite, 3 per cent.

FIG. 11. Mowich hypersthene basalt. A (top). x25, plane light. (Glomeroporphyritic plagioclase with concentric lines of inclusions near the periphery. Note alteration of the hypersthene along the cleavage cracks and in the center; magnetite often fills these cracks, producing a Schiller effect.) B (bottom). Under crossed nicols.

SUMMARY

After examining several hundred thin sections of Keechelus rocks, the writer was impressed by three characteristics that were almost unfailingly present.

One of the most constant characters is a porphyritic texture. This is seen is best advantage in the numerous bodies of andesite and diorite porphyries, but it to also prevalent in the various flows and serves as a modifying texture in the fragmental rocks. The average size of the phenocrysts for all the specimens examined is 1.5 mm. Exceptions to this texture are confined to a very few dense pyroclastic layers and to small dikes and sills (less than .5 meter in width) intruding the younger portions of this formation.

Another distinguishing trait is the presence of abundant opaque minerals. In the vast majority of cases, magnetite may be seen as dust, small granules, or moderately sized grains—the size being roughly proportional to the dimensions of the other minerals in the groundmass. In the dense flows, the magnetite is usually in the form of dust but, in the porphyries, the octahedral grains often attain a size sufficiently large to be distinguished megascopically. The only exception noted in regard to the opaques was one specimen of felsite and it contained limonite in place of the magnetite. Less plentiful opaques include ilmenite, leucoxene, and pyrite.

The green color of so many of the Keechelus rocks is undoubtedly related to the presence of chlorites, serpentines, actinolite, and, to a smaller extent, to the pyroxenes. Such minerals as antigorite clinochlore, actinolite, and bowlingite are especially plentiful and one or more may be found in most any of the Keechelus rocks, with the exception of the felsites.

Other traits are not so constant. However, certain strong tendencies are thought to be worthy of mention. The feldspars are zoned more often than not and usually contain inclusions of material identical to that in the groundmass. There is a distinct inclination toward a glomeroporphyritic texture in many of the rocks. The pyroxenes are almost totally lacking in the lower, and older, part of the formation and are but slightly more abundant in the younger. The chlorites and serpentines are far more prevalent and continue down to the actual contact of the granodiorite.

The jointing of the formation is rather distinctive in that the blocks are huge and roughly rectangular—simulating those of granite. The joint planes are remarkably flat and continuous. This type of jointing is not confined to the massive porphyries but is seen in the younger tuffs as well. The columnar structure of the topmost flows is a notable exception.

RELATIONS AND AGE

The Keechelus series can scarcely be regarded as a unit. From a lithologic standpoint, the formation is largely composed of andesitic porphyries and breccias in various stages of alteration but a great wealth of other types are likewise included in sufficient amount to make it impossible to consider as a formation. Reference has been made in the previous pages to the older and younger portions of this series. Such a division was recognized by Smith and Calkins (34) while mapping the Snoqualmie quadrangle. This grouping is evident to field workers in the Keechelus, yet innumerable difficulties arise when an attempt is made to represent the component parts on a map. The reason is well summed up by Smith and Calkins in the following paragraph:

In short, the criteria, while sufficient to establish the presence of two distinct groups of these volcanics, fail, except locally, to serve as a basis for the determination of the boundaries between them. On the average, the later andesite is much fresher, less tilted and less dissected than the earlier, but, in contiguous areas, certain phases of the two are so similar that they cannot be discriminated with certainty, and the endeavor to map them separately was, therefore, abandoned.

The lower and altered portion is extremely massive and indurated and, where the attitude may be distinguished, displays a gentle folding. Rocks of this character average 2,000 feet in thickness within the Park.

The age relationships of both the older and younger parts of the Keechelus leave much to be desired. In the northern half of the Snoqualmie quadrangle, the lower part of the Keechelus was found to overlie, unconformably, the Swauk, Teanaway, and Guye formations. The youngest of these Tertiary formations is the Guye; its age being determined as Miocene on the basis of fossil leaves. In the southern half of the quadrangle, the Keechelus lavas are overlain by beds of the Ellensburg formation, which has been assigned to the late Miocene in age on paleo-botanical evidence, Hence, the Miocene age of the Keechelus formation, seemingly, is well established.

The exposures of the Guye are known, only, in the northwest corner of the Snoqualmie quadrangle. Concerning it, Smith states:

The Guye consists of detrital rocks with some chert and limestone and interbedded basalts and rhyolites. The base of the formation is nowhere exposed; the top has been removed by erosion so that its limits and thickness are unknown. The formation is much-folded; its structure cannot be worked out in detail, nor can any general section of it be compiled.

Obviously, great care should be exercised in using this formation as a basis for correlation for so large and important a series as the Keechelus. Before wholly accepting the evidence limiting the Keechelus between two Miocene formations, caution should be taken for the following reasons:

1. It is doubtful if the Keechelus overlies that portion of the Guye wherein the leaves have been found. If that be true, it is not known whether the younger or older Keechelus covers the leaves.

2. On a lithologic basis, the limestone strata in the Guye have no counterpart in any known Tertiary formations, either to the east or west of the Cascades. Limestone is, however, found in the older rocks.

3. A considerable time interval must have elapsed between the two Miocene formations, the Guye and the Ellensburg, for the Guye is overlain unconformably by thousands of feet of Keechelus material and this, in turn, is overlain by the Ellensburg.

4. Smith and Calkins remark:

The stratigraphic relations to the overlying rocks, added to the lithologic resemblance of the Guye to the Eocene formations, would have lead to its reference to the Eocene were it not for the paleobotanical evidence.

5. The entire age determination was based on two fossil leaves referred to the Miocene and one to the Fort Union (Eocene or Cret.).

The younger Keechelus may be separated from the older on lithologic structural, and possibly topographic evidence.

Lithologically, the younger fragmental rocks are characterized by a well-developed bedding. The flows have a freshness that makes them exceedingly difficult to distinguish from the lavas of Mount Rainier. The older rocks, on the other hand, are more massive, dull and altered looking.

Structurally, the younger flows and pyroclastics still retain a horizontal position, or, locally, they may be slightly warped. The older rocks have been tilted and folded and suffered minor fracturing.

Topographically, the attitude of the younger lavas is still expressed in some of the flat-topped park areas. The older part is deeply dissected, and, being so massive, exerts little influence on the drainage pattern. A possible exception would be a tendency toward a subsequent drainage following the general northwest-southeast structural trend of these older rocks.

The exact upper age limit of the Keechelus is unknown. The only materials definitely overlying the younger capping flows in the vicinity of the park are the Rainier volcanics. Elsewhere, volcanic activity in the Keechelus may have extended well into the Pliocene, and judging from the freshness of the flows, they may be equivalent to the Rainier lavas in age.