The Geology of Crater Lake National Park, Oregon With a reconnaissance of the Cascade Range southward to Mount Shasta by Howell Williams
Welded Dacite Tuffs (Ignimbrites)
From the standpoint of the petrographer, perhaps the most interesting rocks on the caldera wall are the welded tuffs. A younger series forms the topmost cliffs between Pumice Point and Skell Head; an older, interandesitic series is exposed in the Cottage Rock section, on the cliffs above Cloudcap Bay. They are all characterized by fine, streaky banding and by the presence of wisps, clots, and lenses of glistening, black obsidian in a buff, brown, or pinkish matrix. In one place, the obsidian predominates by far; in another, the varicolored matrix carries only a few small bands of obsidian. No matter which type is subordinate, the rocks show such beautiful fluidal banding that they might readily be mistaken for lavas. Indeed, Diller was so doubtful about their origin that he compromised by referring to them as tuffaceous dacitic lava. Patton, in his description of the younger series, spoke of them as “the Wineglass flow.”
In passing from lower to higher ground, the finely laminated, lava-like t&s merge into more or less incoherent varicolored pumice tuffs and tuff breccias devoid of obsidian streaks. Similarly, they grade downward into a basal layer composed of pulverulent lump pumice. The inference seems justified that the finely banded rocks represent pyroclastic ejecta firmly welded by reason of high temperature, high gas content, and the weight of overlying material. Microscopic study serves to confirm this interpretation.
In thin sections of typical samples from the brim of the Wineglass, the streaks of obsidian, which appear black in the hand specimen, are seen to be composed of clear, colorless glass finely interlaminated with gray glass choked with trichites, globulites, and irresolvable dust. There is no sign of devitrification. The stream lines may be gently curved or bent into the most complex swirls and eddies. Commonly they are tightly molded around phenocrysts. The latter constitute approximately a quarter of the total volume. Plagioclase is four or five times as plentiful as all the other phenocrysts combined. Medium andesine is the dominant variety, but all the crystals are strongly zoned in an oscillatory fashion, and the full range in composition is from acid labradorite to acid andesine. All are much corroded and charged with inclusions of glass. Next in importance among the porphyritic minerals is pale-green augite. Hypersthene is quite subordinate; reddish-brown oxyhornblende and granular ore are minor accessories.
These minerals recur in about the same proportions in the buff and pink matrix between the streaks of obsidian. Viewed by reflected light, the varicolored matrix contrasts vividly with the obsidian, for whereas the latter appears gray, the former appears a bright rusty red, on account of dusty hematite. Here and there the obsidian streaks have frayed ends and merge indefinitely into the matrix, but usually the two are sharply separated. The outstanding feature of the matrix, however, is its unmistakably pyroclastic character, for it is composed of densely packed shards of glass and interstitial glass dust. In normal vitric tuffs, the glass shards lie at random; here, on the contrary, they show a distinct alignment. Many were obviously flattened while still plastic, and some show distortion from having been squeezed between neighboring crystals. In other words, this is a pyroclastic rock showing the results of deformation prior to solidification. The obsidian streaks represent deformed clots of magma, and the matrix represents fine magmatic spray and pulverized glass dust. Together the constituents descended the slopes after the manner of a glowing avalanche, and suffered plastic flow.
The welded tuff bands of the Cottage Rock section are generally more porphyritic and contain more hypersthene. Other minor differences are the absence of oxyhornblende, and the development locally of an autobrecciated appearance, caused by the break-up of glass lapilli and bombs into angular fragments.
No analysis of the Cottage Rock tuff is available, but the one reproduced from Diller’s monograph (no. 19) shows that the tuff of the Wineglass is a typical dacite.