Geological Survey Professional Paper 554—D Cenozoic Volcanic Rocks of the Devils Postpile Quadrangle, Eastern Sierra Nevada California

Andesite and basalt are exposed on Mammoth Pass, in the Mammoth Lakes basin, and in the valley of the Middle Fork of the San Joaquin River. This unit is here referred to as the andesite of the Devils Postpile. Two of the three analyzed samples (table 3) are trachybasalt, and the third is trachyandesite; but inasmuch as data from fused glass beads (fig. 5) give an average composition of approximately 53 percent silica, the map unit is called andesite.

Basalt predominates in the bottom of the Middle Fork valley and in the Mammoth Lakes basin; andesite is predominant elsewhere, except in the southern part of the outcrop area where more siliceous quartz latite occurs.

A potassium-argon age determination of 0.94±0.16 million years was reported by Dalrymple (1964b) for the andesite of the Devils Postpile. A redetermination on a second split of the same sample yielded an age of 0.63±0.35 million years.² Because the amount of potassium in the sample and the percentage of radiogenic argon relative to total argon are so low in both of these determinations, they indicate little more than that the age of the andesite probably is between a quarter of a million and a million years. It is thus possible to reconcile these data with an age of approximately 0.7 million years for the tuff of Reds Meadow, which the andesite unconformably overlies.

---

²The analytical data for this determination are as follows (G. B. Dalrymple, written commun., 1965): weight, 10.02 grams; K₂O, 0.268 percent; radiogenic Ar₄₀, 1.1 percent; radiogenic Ar₄₀, 2.50X10^-13 mole per g.

---

The andesite was erupted in large part in the Mammoth Pass-Mammoth Mountain area, as noted by Matthes (1930) and Erwin (1934), from whence it flowed eastward into the Mammoth Lakes basin and westward into the valley of the Middle Fork. There is some evidence for an additional local source in the northern part of Pumice Flat, where an apparent andesitic dike cuts a deposit of ash and volcanic rubble. The pattern of silica distribution (pl. 1) also suggests the possibility of separate source areas within the Middle Fork valley and in the Mammoth Lakes basin. Perhaps the samples anomalously high in silica from south of the Lower Falls also came from local vents. Three small outliers south of Lost Dog Lake are also derived from a local source; their correlation with the andesite of the Devils Postpile is uncertain. The Red Cones, considered to be all additional source by Erwin (1934, p. 49), have not been glaciated and are of Recent age rather than Pleistocene.

Prior to the eruption of the andesite of the Devils Postpile, the tuff of Reds Meadow had been almost completely removed from the central part of the Middle Fork valley (see above), and there the andesite rests directly upon granitic rocks. The andesite was subsequently largely removed by glaciation, leaving behind remnants on the valley bottom and slopes. Reconstruction from these remnants suggests an original thickness of at least 600 feet for the andesite in the vicinity of the Devils Postpile. The original extent of the andesite is unknown.

The andesite displays conspicuous jointing nearly everywhere, with three distinct types represented. Locally near the base of the flows, particularly along the Middle Fork in the vicinity of Rainbow Falls, nearly horizontal platy jointing has formed. Most commonly the jointing is of the orthogonal contraction-crack polygon type, giving outcrops a blocky appearance. Least common, but in places very well developed, are joints of the nonorthogonal contraction-crack polygon type, as for example, in the Devils Postpile itself (frontispiece and fig. 10). The origin of both orthogonal and nonorthogonal polygonal jointing has recently been studied by Lachenbruch (1962) and by Spry (1962), and we have discussed it briefly with reference to the Devils Postpile (Huber and Rinehart, 1965b). The columns of the Devils Postpile, some of which are as much as 60 feet long, are polygonal, with hexagonal columns slightly more abundant than pentagonal ones, and with other forms making up 10-20 percent of the total (Hartesveldt, 1952; Beard, 1959). The jointing which defines the columns formed near the base of the flow, and the curved and tilted forms of some of the columns are probably due to irregularities in the isotherms during cooling, which may have been caused, at least in part, by topographic irregularities on the bedrock surface over which the lava flowed.

FIGURE 10.—Polygonal joint pattern displayed by tops of columns in the Devils Postpile. Note glacial striations and remnants of glacial polish. Photograph by Gerhard Schumacher.

Mammoth Mountain has been described by Mayo (1941, p. 1068) as "the most impressive volcanic edifice in the region," an apt description, because it rises abruptly over 2,000 feet above the surrounding countryside. The eruptive history of Mammoth Mountain includes intrusion of a massive dome or domes, explosive activity, and the outpouring of extensive glassy flows, especially on the northeast side of the mountain.

Like the quartz latite of Two Teats, the rock which makes up this unit is extremely variable in color and texture. It ranges from shades of brown and pink through gray to nearly black in the glassiest flows. The rock is typically flow banded and porphyritic; plagioclase and biotite dominate and vary widely in crystal size and relative amounts. Despite the wide petrographic variation, the silica in 17 of 21 samples shows a range of only 4 percent and averages 68 percent, as inferred from the index of refraction of fused samples (fig. 5). The other four, with 73-74 percent silica, represent rhyolitic obsidian flows that are locally inter-layered with lithoidal quartz latite flows. Sampling of this unit and analytical work are inadequate to determine whether these obsidian flows are chemically distinct from the quartz latite, although the data available indicate such a possibility.

The quartz latite of Mammoth Mountain was considered by Erwin (1934, p. 45) to be of Miocene age and by Matthes (1930, 1960) to be "preglacial" in age. Rinehart and Ross (1964, p. 64) also considered the quartz latite of Mammoth Mountain to be of late Tertiary (Pliocene) age, because of its degree of dissection and its earlier correlation with the very similar quartz latite on Two Teats and San Joaquin Mountain, for which a late Tertiary age appears correct. Unlike Two Teats and San Joaquin Mountain, however, Mammoth Mountain does retain appreciable constructional form, and its degree of dissection is not incompatible with a Pleistocene—even a relatively late Pleistocene—age. Although he did not study the quartz latite of Mammoth Mountain in any detail, Gilbert (1941, p. 799 and fig. 2) implied that it is of Pleistocene age. Field relations between the quartz latite of Mammoth Mountain and the andesite of the Devils Postpile are equivocal as to their relative ages. However, two potassium-argon age determinations on the quartz latite yielded ages of 0.37±0.04 million years (Dalrymple, 1964a, table 1) and 0.18±0.09 million years.³ Mammoth Mountain is therefore considered to be of late Pleistocene age and younger than the andesite of the Devils Postpile.

---

³The analytical data for this determination are as follows (R. W. Kistler, written commun., 1961): biotite, 9.25 g; K, 6.8 percent; radiogenic Ar₄₀, 3.8 percent; radiogenic Ar₄₀, 2.21 X 10^-12 moles per g.

---

The construction of Mammoth Mountain appears to have been complex, involving extrusion of viscous flows of highly varied petrographic character that range from crystal-rich lithoidal types to crystal-poor obsidian. Some of the flows were sufficiently fluid to advance at least 2 miles beyond the north base of the mountain, but most appear to have moved only down the slopes over earlier flows, thus piling up around the vent to form a rather typical cumulodome. The high degree of dissection of the north side of the mountain is probably due to the removal of much material by violent explosive activity and to the subsequent modification by glaciation. In addition to the construction of Mammoth Mountain, this episode of volcanism produced a satellite dome 1 mile in diameter and 1,000 feet in height, about 1 mile northeast of the base of Mammoth Mountain. Glassy flows are discontinuously exposed over an extensive area north of Mammoth Mountain. Minor late-stage activity is manifested by small postglacial phreatic explosion pits at the north base of Mammoth Mountain and by present-day fumarolic activity at the south base and at the crest.

Mammoth Mountain lies on or immediately adjacent to the western perimeter fault that defines the Long Valley volcano-tectonic depression (Pakiser and others, 1964, p. 17 and pl. 1). Inasmuch as the location of a volcanic pile the size of Mammoth Mountain would probably be dictated by a major zone of crustal weakness, it is likely that this fault provided structural control for the vents from which the quartz latite erupted. In this respect Mammoth Mountain would lie in a structural position similar to the "late rhyolite domes" peripheral to the Valles caldera, New Mexico (Smith and others, 1961). When the structural depression began to form cannot be stated with any degree of certainty because not enough is yet known about the detailed relationships of the volcanic rocks in the Long Valley basin and adjacent areas, especially their relative and absolute ages. However, the vertical displacement along the western part of the fault, which resulted in the Sierran escarpment east of Deadman Pass, must have occurred later than approximately 3 million years ago, the age of the quartz latite of Two Teats. If this fault controlled the location of Mammoth Mountain, then it must have been active prior to approximately a quarter of a million years ago, and thus movement must have begun between approximately a quarter of a million and 3 million years ago.

In the Dry Creek drainage area, north of Mammoth Mountain, a number of scattered outcrops of andesite and basalt are exposed through the mantle of pumice and alluvium, chiefly along recent faults or in explosion pits or cinder cones. Because of the generally poor exposures of this unit, it is difficult to determine its stratigraphic position relative to the other volcanic rocks in the area and, indeed, whether all of the scattered outcrops are correlative.

Analysis of fused-bead data (fig. 5) indicates a distinct bimodal composition distribution with maximums at basalt (51 percent SiO₂) and andesite (56 percent SiO₂). The more silicic rocks are concentrated near the Inyo Craters and along the eastern edge of the quadrangle (pl. 1).

At three localities the andesite and basalt can be demonstrated to be younger than the quartz latite of Mammoth Mountain. These three exposures are (1) along a recent fault scarp about 1 mile north of Mammoth Mountain, (2) on the north flank of the quartz latite dome northeast of Mammoth Mountain, and (3) on the northeast flank of Mammoth Mountain just west of a recent fracture known locally as the "Earthquake Fault." The rock in exposures 1 and 3 has been glaciated, and therefore the unit is considered to be of late Pleistocene age rather than Recent. Although exact correlations are questionable, all the andesite and basalt in Dry Creek area appear to be approximately the same age and are arbitrarily assigned to the same map unit. In addition, some glaciated andesitic cinder cones near the base of the San Joaquin Mountain ridge are also assigned to this unit.

Pumice Butte is one of the two andesitic cinder cones in the southeast quadrant of the quadrangle. These cones, together with a vesicular andesite that appears to have originated as a flow from the base of Pumice Butte, were not overridden by glacial ice. They are above the elevation limit of the main valley glaciers of Wisconsin age, but a small tributary glacier of presumed Wisconsin age overrode a low shoulder on the northwesternmost of the two cones, and the margins of the flow were glaciated during latest Wisconsin (R. J. Janda, oral commun., 1965). This andesite is therefore thought to have been erupted during a pre-Wisconsin or intra-Wisconsin interglaciation. The rock is typically scoriaceous and appears fresh—quite unlike most of the andesitic flows and rubble south of Deer Creek and east of Pond Lily Lake, which are correlated with the andesite of Deadman Pass. Contacts shown between these units on the map are only approximate, however, and have been distinguished in part by aerial photo-interpretation. Included with the andesite of Pumice Butte is a domelike mass of dacite (at the north edge of the map unit), which appears to be more highly dissected and is probably older than the main mass of the andesite.

In the extreme northeast corner of the quadrangle, olivine-bearing quartz latite covers an area of about half a square mile. It is exposed somewhat more extensively in the Mount Morrison quadrangle, where it has been described by Rinehart and Ross (1964, p. 57). The three following lines of evidence suggest that the quartz latite is of late Pleistocene to Recent age: (1) There is no evidence that the quartz latite has been glaciated, although late Pleistocene glaciations probably did not extend this far from the range front, (2) only slight dissection is shown in the major area of outcrop, which consists of a relatively flat topped stubby steep faced flow that can be traced westward into steep-sided domes that retain much of their original form, (3) a north-trending fault scarp in the Mount Morrison quadrangle, along which old glacial till has been displaced, terminates abruptly at the edge of the flow and is probably buried by it.

The rock contains abundant white feldspar phenocrysts 2-5 millimeters long in a dark-gray aphanitic or glassy groundmass. Flow banding is visible locally, and in general appearance the rock is not unlike some of the darker varieties of the older quartz latites.