Geological Survey Professional Paper 3 The Geology and Petrography of Crater Lake National Park

HYPERSTHENE-ANDESITES.
(continued)

ANDESITE DIKES.

With the exception of two dikes under Llao Rock all the dikes that cut the walls surrounding the lake belong, as far as may be judged from the specimens collected, to the andesites.

Two specimens were collected from the rather large dike that runs diagonally down from The Watchman to near the lake level. These two specimens (92 and 93) may be classed with the hypocrystalline andesites, subtype B. The phenocrysts do not present any specially distinctive features, but the groundmass of the two specimens shows two extremes. No. 92 is relatively coarse grained and presents a transition to the holocrystalline type, in which, however, the allotriomorphic patches are suppressed. It is further characterized by the presence of unusually abundant plagioclase laths of a very uniform and fairly large size. No. 93, on the other hand, may be considered as transitional to subtype A in that the plagioclase laths sink to minute size. But this groundmass also contains a great many squarish or short rectangular feldspars that suggest an affinity with the dacites.

From the upper part of the dike that lies between Glacier Peak and The Watchman No. 94 was collected. In the hand specimen this presents a strikingly uniform appearance. It has a rather light-gray color, is very porous, and at a casual glance appears to be free from phenocrysts. On the contrary, phenocrysts are very abundant, as in thin section they appear to form about one-half of the entire mass. This is, in fact, a thoroughly characteristic andesite and is very strongly porphyritic, with great contrast between the phenocrysts and the groundmass. Plagioclase, hypersthene, and augite are all abundant. The pyroxenes occur both in sharp crystals and in nests of granular individuals. The feldspar occurs in two distinct types; in the larger many-faced forms having the customary glass inclusions, and in the smaller rectangular forms. From the character of the groundmass this may be considered a typical illustration of the hypocrystalline type, subtype A, characterized by the abundance of minute plagioclase microlites

The analysis of this rock is given on page 94. It indicates that this is thoroughly representative of the Crater Lake andesites.

Under the north end of Llao Rock occurs a dike of which No. 95 is a sample. It very closely resembles the rock just described. Under the microscope it appears that the phenocrysts are not quite so abundant and not so uniform in size. The groundmass contains considerable allotriomorphic feldspar, which to a certain extent hides the microlitic plagioclases. It belongs, with the above-described rock, to subtype A of the hypocrystalline andesites.

At the head of Steel Bay, a little to the east of Llao Rock, occurs a dike from which No. 96 was collected, which shows under the microscope a very pretty development of the hyalopilitic structure. Very abundant phenocrysts of the customary characters lie embedded in a groundmass consisting of a deep-brown glass that appears to impregnate a loosely felted aggregate of slender augite and plagioclase microlites. The glass base is free, or nearly free, from globulitic matter, but contains, in addition to the above-mentioned microlites, octahedral crystals and grains of magnetite.

On the east side of the lake are to be seen two dikes near Sentinel Rock. One of these (33) occurs exactly below Sentinel Rock and the other (21) about a third of a mile to the northeast. Both of these rocks have already been referred to in connection with the description of the andesites. No. 33 belongs to the subtype A of the hypocrystalline andesites and No. 21 to the hyalopilitic type. As these types have been fully described above, a repetition here is hardly necessary.

In concluding the description of the andesites in dike form, we may note that many of the various types found in the andesite flows are repeated in the dikes, while no markedly different types have been met with in the dike andesites.

CHEMICAL ANALYSES OF ANDESITES.

The analyses given in the following table are all of Crater Lake rocks, with the exception of the last, which is from Crater Peak, California, and is given for comparison. The analyses were made by the chemists of the United States Geological Survey. The numbers at the heads of the columns are the specimen numbers.

Analyses of Crater Lake andesites.
[Analyses by H. N. Stokes, except of 1829.]

It is a little surprising and also unfortunate that out of five specimens of andesite selected by Mr. Diller for chemical analysis four should turn out to belong to one type, and this not the most common one. It is surprising, because these specimens were selected with a view to representing typical occurrences. As the selection was made before an opportunity was given for microscopic study, it was impossible to distinguish the different types from the hand specimen alone.

SECRETIONS IN ANDESITES.

In a number of the andesite rocks at different localities are found inclosures or nodules that appear to be secretions of the more basic minerals. Sometimes these inclosures have more or less roundish form and are evidently inclosed in the sense that they are not literally formed in place. This is particularly evident in a specimen (97) collected in 1883 from the southern rim of the crater. The rock itself, of which no thin section has been made, has a nearly black groundmass—which is probably hyalopilitic—inclosing numerous and very conspicuous plagioclase phenocrysts. Embedded in this rock is a roundish mass 2 or 3 inches in diameter from which the andesite appears to have shrunk away in cooling as though from a foreign inclosure. The inclosure is blackish gray, compact, without phenocrysts, and slightly porous. It has a very basaltic look.

In thin section this specimen is fairly characteristic of most of the inclosures. It consists of a deep clear-brown glass in which lie in loosely felted arrangement slender and very sharply defined laths of plagioclase and equally slender but not always so sharply cut prisms of hypersthene and augite; also a little magnetite in the customary octahedral form. Hypersthene is more abundant than augite, but both are much more abundant than is usual in the andesites. The two pyroxenes appear to form about as large a percentage of the rock as does the plagioclase. Parallelism of growth between the augite and hypersthene is common. The length of these pyroxenes is often ten times as much as the width. The plagioclase is rather simply twinned with but two or three bands and shows large extinction angles. They have square cross section with diagonal extinction. Phenocrysts are entirely wanting.

Fig. A of Pl. XVI presents a photograph of this thin section in polarized light and Fig. D of Pl. XVI a photograph of thin section of No. 9 in white light, showing both the secretion and the inclosing hyalopilitic andesite.

Analogous to the above are No. 9, from the path that leads from the camp ground down to the water's edge; No. 16, from Kerr Notch, at the head of Sand Creek; and No. 75 from near the road on Anna Creek. These three specimens consist of small inclusions in rather glassy andesites, the first two of the hyalopilitic and the last of the dacitic type. The inclusions form only a small part of the thin section in each case. They do not, however, appear to be foreign to the rock. On the contrary, the junction of the secretion with the surrounding andesite is such as apparently to preclude the idea of an entirely separate origin. The clear glass of the section is exactly of the same color as that of the main rock and the junction is not sharp, but the ingredients seem to have grown from the secretion out into the surrounding rock mass. The first two contain phenocrysts of plagioclase, and to a less extent of hypersthene similar to the corresponding phenocrysts in the inclosing andesite. In some cases one of these plagioclase phenocrysts appears to belong about equally to the secretion and to the main mass. In No. 16 the inclusion greatly resembles an enlarged area like these one frequently meets with in the more glassy andesites of Crater Lake. where a deep and clear brown glass, free from the customary microlites, occurs, filling the space between two or three adjacent phenocrysts.

No. 98 is one of these inclusions from a large bowlder between Sun Creek and Sand Creek. The rock in which this is inclosed (76) belongs to the dacitic type of andesite, and has already been described. This inclusion contains large, characteristic plagioclase phenocrysts that are crowded with the unusual inclusions, and that present clear margins; it contains also phenocrysts of hypersthene. The glass is stained a very deep red with ferritic matter.

Nos. 99 and 100 are sections in the above-described hypocrystalline andesite (40) found at the spring near the camp ground. They both resemble No. 97 in being entirely without phenocrysts and in having the same felted structure and the same composition as that rock, but they differ materially in that crystallization appears to have progressed until but little glass remains, and this little is nearly colorless and not readily discernible. In fact, these two inclusions were at first taken for basalt inclosed in the andesite, and were it not for numerous intermediate stages between these and the more evident secretions the determination as basalt might hold. Other cases of secretions resembling basalts are mentioned in connection with the description of the dacites of Crater Lake.

Plate XVI.—THIN SECTIONS OF ANDESITE.

FIG. A.—Secretion in andesite from the southern rim of the crater. Magnified 48 diameters. Specimen No. 97. A photomicrograph in polarized light with crossed nicols. The secretion consists of a deep-brown glass in which lie well-defined laths of plagioclase and equally slender prisms of hypersthene and augite; also a little magnetite. The pyroxenes can not well be distinguished from the plagioclase in the photomicrograph. See page 96.

FIG. B.—Andesite of the hypocrystalline type, from Cathedral Rock. Magnified 20 diameters. Specimen No. 45. A photomicrograph in white light. The ragged black crystal is a brown hornblende almost completely altered through resorption to a black aggregate of magnetite and augite. The section also shows two generations of plagioclase. The large phenocryst on the right represents the oldest generation with glass inclusions. For description see page 83.

FIG. C.—Andesite from the water's edge at the head of Sun Creek. Magnified 48 diameters. Specimen No. 73. A photomicrograph in polarized light with crossed nicols. Illustrates the third or holocrystalline type of andesite and is characterized by the presence of allotriomorphic patches of feldspar, which appear as light spots hint the photomicrograph See page 89.

FIG. D.—Andesite from the path leading from the camp ground to the water at Eagle Cove. Magnified 48 diameters. Specimen No. 9. Shows the junction between a hyalopilitic andesite on the right and a secretion in the same on the left. The secretion consists of a deep-brown glass and slender laths of plagioclase and prisms of hypersthense and augite; also a little magnetite. See page 96.

In No. 25, a hypocrystalline andesite from Wizard Island, is to be seen an inclosure that may with more probability be taken for basalt. This consists of small, irregular, and apparently isolated grains of augite, or possibly of hypersthene—the distinction is not certain in this case—that extinguish simultaneously over considerable areas. The spaces between these grains are filled with short and not very sharp plagioclase laths.

A rock collected by the writer from the bottom of the crater on Wizard Island (2017, 4 of the private collection of the writer) seems to throw some light on these secretions. In the hand specimen it consists of a rather compact and nonporous gray andesite, with irregular, lighter gray patches scattered promiscuously through it. In thin section these lighter patches are seen to be holocrystalline, with a groundmass composed of plagioclase, augite, and magnetite. The plagioclase is largely in the form of short laths, but between these laths the same mineral appears in allotriomorphic form, but not in poikillitic patches. The darker portions of the rock do not appear to differ from the lighter part, except that the allotriomorphic feldspathic part of the groundmass is replaced by a deep clear-brown glass. That several such alternations in the groundmass may occur within the area of a thin section makes it more likely that the apparent inclosures, in which brown glass is so very conspicuous, are only local differentiations in a common magma.