The Geology and Petrography of Crater Lake National Park, 1902
CLASSIFICATION OF ANDESITES.
HOLOCRYSTALLINE TYPE OF ANDESITE.
The following specimens are placed under this type: Nos. 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, and 74. These are mostly light-gray rocks with comparatively rough fracture and with relatively inconspicuous plagioclase phenocrysts. A few specimens of darker color are not thoroughly characteristic of the type.
Under the microscope these andesites do not differ very materially in appearance from those of the hypocrystalline type, when viewed in white light, except that they are much lighter in color and are entirely free from globulitic matter. Between crossed nicols, however, there is a marked difference between these andesites and all the preceding types. This difference is to be seen in the groundmass. The colorless constituents are of two distinct kinds, first, plagioclase laths not markedly different from the similar laths in the other andesites; and, second, a sort of residual product of crystallization occurring in irregular, allotriomorphic patches that have a poikilitic appearance, owing to crowded inclusions of the other groundmass ingredients. These patches vary considerably in size and in conspicuousness, but in general are from 0.2 to 0.5 millimeter in diameter. They do not show polysynthetic twinning, but very often possess a more or less undulous extinction. They are quite distinct from the plagioclase laths which lie embedded in them and appear to play the role of a glass base. The individual patches show usually simultaneous extinctions and are bounded only by meeting other similar patches.
This allotriomorphic material is in some cases undoubtedly quartz; in others it is probably feldspar. It was proved to be quartz in Nos. 63, 66, 71, and 73 by its slightly higher refractive power as compared with adjacent feldspar and by its giving a positive, uniaxial image in convergent polarized light. It is considered to be sometimes feldspar for the following reasons: First, failure to show uniaxial images; second, occasional undulous extinction; third, some of the plagioclase laths have allotriomorphic extensions that extinguish with the outer part of the lath and resemble these patches (63, 64); fourth, in a few of these andesites the lath-shaped plagioclase is nearly missing, so that the groundmass is almost entirely composed of these patches, except for the inclosed augite microlites and magnetite grains. If in this last case all this material were quartz, it would give a rock far too acidic in composition to be classed as an andesite, whereas the close association and resemblance between all of these rocks make it impossible to separate any one of them from this family.
The more characteristic of these holocrystalline andesites have a great abundance of well-defined plagioclase laths with fluidal arrangement, as well as innumerable augite microlites and octahedral crystals and grains of magnetite embedded in a base of these allotriomorphic patches of feldspar and of quartz (60, 61, 66, 71, 74). The augite microlites are, as a rule, somewhat larger and not quite so sharply developed as is the case in the other andesites. The phenocrysts of plagioclase, hypersthene, and augite differ in no essential respect from the descriptions given for all the Crater Lake andesites. Fig. C of Pl. XVI (p. 98) gives a fair idea of the allotriomorphic patches that characterize these holocrystalline andesites.
The presence of quartz in the groundmass of these rocks suggests a much more acid rock than the chemical analyses of Crater Lake andesites given elsewhere in these pages allow. As the only analysis made of this holocrystalline type is of No. 68, which is not thoroughly representative, inasmuch as the allotriomorphic patches are not very well developed, it will be of interest to compare these rocks with a similar rock collected by Mr. Diller and alluded to by Mr. H. W. Turner in his article on the Age and Succession of the Igneous Rocks of the Sierra Nevada. The rock in question was taken from Crater Peak, California (Lassen Peak quadrangle), and is No. 1829 of the Cascade Range collection. Mr. Turner does not describe the rock, but he refers to it as a pyroxene-andesite, and gives the chemical analysis made by W. F. Hillebrand.b This analysis will be found, repeated on page 94 of this paper. Through the kindness of Mr. Diller the writer was enabled to study a thin section of this Crater Peak andesite and to compare it with the thin sections of the rocks here under discussion. Without any noticeable difference it appears to be identical with the Crater Lake holocrystalline type of andesite. This is true not only of the groundmass, with its allotriomorphic patches inclosing plagioclase laths and augite microlites, etc., but also of the phenocrysts. It will he seen, in comparing the analysis with the analyses of other andesites from Crater Lake, that this contains between 6 and 10 per cent more silica. It is, in fact, closely analogous to the dacites, such as, for instance, those from Lassen Peak, analyzed by T. M. Chatard and W. F. Hillebrand.c It is highly probable, therefore, that these andesites are, in part at least, dacites, although no well-defined quartz is to be seen in them.
aJour. Geol., vol. III, 1895, p. 410.
bIdem, p. 407.
cBull. U. S. Geol. Survey No. 150, 1898, p. 218.
No. 198 probably belongs with the holocrystalline andesites, but owing to extensive alteration its characteristics are not readily made out. The groundmass consists of much homogeneous-looking chloritic matter, with rather faint lath-shaped plagioclase and not a little allotriomorphic, colorless material that is probably feldspar. The magnetite has been altered to leucoxene and the pyroxenes of both the phenocrysts and of the groundmass to chlorite. The phenocrystic plagioclases are partially altered to carbonates and probably to kaolin. A good deal of quartz is to be seen in distinct grains, two or three of which are quite large, while the rest occur in bunches of allotriomorphic grains. As this is the only andesite from Crater Lake that shows well-defined quartz, this mineral is in this case probably secondary in origin.
These holocrystalline andesites, as is the case with the other types, are widely distributed over the Crater Lake area. There may be some significance in the fact that six out of a total of fifteen are situated at the water’s edge; that is, a large proportion of them belong to the oldest lava flows exposed. No. 63, which comes from the basaltic cone called Crater Peak, south of the lake, is taken from an ejected fragment, and was doubtless torn loose from a deep-lying rock below.