Celeste G Engel

Chemical Characteristics of Oceanic Basalts and the Upper Mantle

Tholeiitic basalts (oceanic tholeiites) that form most of the deeply submerged volcanic features in the oceans are characterized by extremely low amounts of Ba, K, P, Pb, Sr, Th, U, and Zr as well as Fe 2 O 3 /FeO 10 in unaltered samples. Oceanic tholeiites also have rare earth abundance-distribution patterns and ratios of K/Rb (1300) and Sr 87 /Sr 86 (0.702) similar to or overlapping those of calcium-rich (basaltic) achondritic meteorites. The close compositional similarities between the oceanic tholeiites and calcium-rich achondrites indicates the relatively primitive nature of the oceanic tholeiites. In contrast, the alkali-rich basalts that cap submarine and island volcanoes are relatively enriched in Ba, K, La, Nb, P, Pb, Pb 206 , Rb, Fe 2 O 3 , Sr, Sr 87 , Ti, Th, U, and Zr; i.e . in the same elements and isotopes that are concentrated in the sialic continental crusts by factors of 5 to 1000 more than the amounts readily inferred in the upper mantle. These analytical data coupled with the field relationships indicate that the alkali-rich basalts are derivative rocks, fractionated from the oceanic tholeiites by processes of magmatic differentiation, and that the oceanic tholeiites are the principal or only primary magma generated in the upper mantle under the oceans. Studies of the abundances and compositions of continental basalts show that essentially identical tholeiitic lavas, contaminated with Si, K, and the chemically coherent trace elements and radiogenic isotopes from the sial, also have been the predominant or only magma generated in the mantle under the continents. The chemical properties of oceanic tholeiites suggest that the upper mantle probably contains less than (in parts per million): Ba, 10; K, 1000; Pb, 0.4; Rb, 10; Th, 0.2; and U, 0.1. The Sr 87 /Sr 86 must be less than 0.7015; Th/U about 2; K/Rb about 1500–2000; and Fe 2 O 3 /FeO less than 0.1. The integration of field and petrochemical data with seismic, density, and shock-wave studies suggests that the oceanic tholeiites are either complete melts of the upper mantle or are generated from a mix of this tholeiite and a magnesium-rich peridotite or dunite in proportions up to perhaps 1:4. The Mohorovicic discontinuity under the oceans appears to mark the transition downward from a largely tholeiitic oceanic crust to either tholeiite reconstituted to blueschist or greenschist or to the ultramafic residue left after expulsion of oceanic tholeiite.

CHEMICAL DATA ON DESERT VARNISH

Desert varnish forms a dark coating up to 0.10 mm thick on the exposed surfaces of many stones and outcrops in southern California deserts. Wet chemical analyses were made of varnish, the underlying weathered rind, and fresh rock for a rhyolite and two andesites. The principal elements in varnish are O, H, Si, Al, Fe, and Mn, and the last two give the deposit its distinctive physical characteristics. H_2O, Fe_2O_3, and especially MnO show the greatest enrichment. Field observations and a number of partial analyses indicate that the best varnishes are on fine-grained rocks relatively rich in Fe and Mn. Spectrographic analyses were made of 22 varnishes, 14 rocks, 8 soils, and 5 samples of air-borne material. In the varnishes Ti, Ba, and Sr are by far the most abundant trace elements, followed by Cu, Ni, Zr, Pb, V, Co, La, Y, B, Cr, Sc, and Yb. Cd, W, Ag, Nb, Sn, Ga, Mo, Be, and Zn were recorded in some but not all varnishes. The trace-element content of all varnishes is similar, and the variations recorded are related to differences in the local geology. Most trace elements are considerably enriched in varnish—Cu and Co especially, and Ni, Pb, Ba, Cr, Yb, B, Y, Sr, and V. The chemical data suggest that (1) varnish on stones seated in soil or colluvium is derived largely from that material, (2) varnish on large bedrock exposures come from weathered parts of the rock, (3) air-borne material is probably a minor contributor. The formation of desert varnish is primarily a weathering process involving the solution, transportation, and deposition of Mn and Fe in particular and a host of trace elements. Most of these elements are derived from local sources, and the small amount of movement required can occur by transport in solution or possibly by ionic diffusion through moisture films. Dew may be as important a source of moisture as rain. Organic agents, such as bacteria, may cause deposition of varnish, but this has not yet been demonstrated. In the desert, evaporation and the catalytic action of MnO_2 should be capable of performing the task. The rate of varnish formation varies widely with local conditions. Hundreds and thousands of years may be required to form a dark coating in some situations, but at one locality in the Mojave Desert a good varnish formed on the surface stones of an alluvial deposit in 25 years. Although the widespread evidence of varnish deterioration may be due to climatological change, conditions in some parts of this desert area are currently favorable to varnish formation.

PROGRESSIVE METAMORPHISM AND GRANITIZATION OF THE MAJOR PARAGNEISS, NORTHWEST ADIRONDACK MOUNTAINS, NEW YORK PART II: MINERALOGY

The progressive metamorphism and partial granitization of a belt of quartz-mica feldspar-garnet paragneiss is considered in detail. This paragneiss is traced and sampled along a belt 35 miles long that extends across the Grenville Lowlands into the central massif of the Adirondack Mountains, New York. Geologic thermometers indicate temperatures of metamorphism of about 500° C. at the southwest end of the belt and about 600° C. near the perimeter of the massif. Minimum temperatures of metamorphism in the gneiss arc determined largely from solid solutions of magnesite in dolomite, FeS in sphalerite, paragonite in muscovitc, and TiO_2 in magnetite. Maximum temperatures of metamorphism are inferred principally from the absence of wollastonite in closely associated siliceous marbles. The gradient in T is checked by the δ O^(18) in quartz and coexisting magnetite in the gneiss. The composition of the paragneiss and its constituent minerals is determined from 75 new chemical analyses, 50 partial chemical analyses, 400 analyses of trace elements, and modal analyses of approximately 400 rocks. At the lower-temperature end of the belt the least altered gneiss is a quartz-biotile-oligoclase-muscovite gneiss averaging (weight per cent) 70.25 SiO_2, 0.67 TiO_2, 14.14 Al_2O_3, 0.55 Fe_2O_3, 3.83 FeO, 2.20 CaO, 1.76 MgO, 0.05 MnO, 3.43 Na_2O, 2.40 K_2O, and (in ppm) B, 10; Ba, 600; Co, 8; Cr, 35; Cu, 16; Ga, 11; Ni, 15; Pb, 12; Sc, 12; Sr, 300; V, 56; Y, 50; Yb, 3; Zr, 170. This is inferred to approximate the bulk composition of the parent sedimentary rock. With increasing temperature of metamorphism of the least altered gneiss, the mineral composition changes as follows: muscovite disappears, garnet appears, plagioclase increases in abundance, and average An content and quartz decrease. Complementary changes in chemical composition include an increase in Al, Fe^(++), total Fe, Mg, Ca, Cr, Ga, Ni, and V. Amounts of K, Si, Fe^(+++), H_2O, and Ba decrease. This “degranitization” or “basification” of the gneiss appears to be a metamorphic process that begins at about 550° C. and is well defined at 600° C. The mobilized Si, K, and H_2O appear to be partly liberated and partly trapped as a venitic migmatite. Granitization of parts of the gneiss is accompanied by an increase in K feldspar and Ab content of plagioclase and by a decrease in biotite, plagioclase, and quartz. Chemical changes in major elements include an increase in K and a decrease in Ti, Fe^(+++), Fe^(++), Mg, Ca, H_2O, and in Na–K ratio. Changes in the amounts of minor elements in granitized parts of the gneiss include increases in Ba and Pb and a decrease in Co, Cr, Ni, Sc, Sr, Ti, V, and Y. All granitizing substances in the gneiss in areas of lower-temperature metamorphism appear to be introduced either laterally or from below. Those in areas of highest-temperature metamorphism are partly introduced, partly derived locally from the gneiss. The implied basic front evolved during granitization of the gneiss may have been large, for the introduced granitizing substances replace one-third of the sedimentary rock throughout a zone over ½ mile thick, 75 miles wide, and more than 40 miles long. Calculations of the chemical composition of the more even-textured gneiss, from modal analyses and mineral analyses, show about the same deviation from the actual chemical analyses as exists in the 34 analyses of G-1 and W-1 as reported by Fairbairn et al, (1951).

Alga-like forms in onverwacht series, South Africa: Oldest recognized lifelike forms on earth

Spheroidal and cupshaped, carbonaceous alga-like bodies, as well as filamentous structures and amorphous carbonaceous matter occur in sedimentary rocks of the Onverwacht Series (Swaziland System) in South Africa. The Onverwacht sediments are older than 3.2 eons, and they are probably the oldest, littlealtered sedimentary rocks on Earth. The basal Onverwacht sediments lie approximately 10,000 meters stratigraphically below the Fig Tree sedimentary rocks, from which similar organic microstructures have been interpreted as alga-like microfossils. The Onverwacht spheroids and filaments are best preserved in black, carbonrich cherts and siliceous argillites interlayered with thick sequences of lavas. These lifelike forms and the associated carbonaceous substances are probably biological in origin. If so, the origins of unicellular life on Earth are buried in older rocks now obliterated by igneous and metamorphic events.

Ultramafic and Basaltic Rocks Dredged from the Nearshore Flank of the Tonga Trench

Deep dredging in the Tonga Trench (Southwest Pacific Ocean) at a depth of 9150 to 9400 m yielded fresh to granulated and serpentinized peridotite and dunite. Other rocks recovered there and at three stations deeper than 7000 m include basalts, tuffs, and tuffaceous agglomerates. Chemical analyses of the fresh peridotite, with combined H2O < 0.10 weight percent, indicate that the rock consists of Si, Mg, Fe (6 percent), and Cr + Ni about 0.7 percent. Mineralogically, the peridotite contains forsteritic olivine and enstatite with minor spinels. The ultramafic mass exposed at 9400 m probably is an accumulate exposed by faulting.

Lherzolite, anorthosite, gabbro, and basalt dredged from the mid-Indian ocean ridge.

The Central Indian Ridge is mantled with flows of low-potassium basalt of uniform composition. Gabbro, anorthosite, and garnet-bearing lherzolite are exposed in cross fractures, and lherzolite is the bedrock at the center of the ridge. The Iherzolites are upper-mantle rock exposed by faulting.

Igneous Rocks of the Indian Ocean Floor

Four dredge hauls from near the crest and from the eastern flank of the seismically active Mid-Indian Ocean Ridge at 23� to 24�S, at depths of 3700 to 4300 meters, produced only low-potassium tholeiitic basalt similar in chemical and mineralogic composition to basalts characteristic of ridges and rises in the Atlantic and Pacific oceans. A fifth haul, from a depth of 4000 meters on the lower flank of a seamount on the ocean side of the Indonesian Trench, recovered tholeiitic basalt with higher concentrations of K and Ti and slightly lower amounts of Si and Ca than the typical-oceanic tholeiite of the ridge. The last sample is vesicular, suggesting depression of the area since the basalt was emplaced. Many of the rocks dredged are variously decomposed and hydrated, but there is no evidence of important chemical modification toward conversion of the lava flows to spilite during extrusion or solidification.

Hornblendes Formed During Progressive Metamorphism Of Amphibolites, Northwest Adirondack Mountains, New York

Hornblendes in amphibolite interlayers in the paragneiss of the northwest Adirondack Mountains undergo systematic changes in color, composition, and density during progressive metamorphism from almandine-amphibolite to hornblende-granulite facies. In contrast, indices of refraction of the hornblendes remain about constant. In the almandine-amphibolite facies the amphibolite layers have the bulk composition of a saturated basalt and consist of bluish-green hornblende, andesine, and quartz. As these layers are traced into the hornblende-granulite facies, their composition undergoes a progressive change to that of an olivine basalt with brownish-green hornblende, clinopyroxene and orthopyroxene, and calcic andesine as major constituents. Compositional changes in the hornblendes with increasing grade of metamorphism include increases in Ti, Na, K, Cr, V, and Sc. Decreases occur in the amounts of Mn, Zn, OH + F + Cl, and in the ratios Fe2O3/FeO and Fe/Mg. Density of the hornblendes increases from 3.260 to 3.278 with the increasing grade of metamorphism. These changes in the hornblendes with increasing T and P, although well denned, are less pronounced than those measured in biotites and garnets of the enclosing paragneiss. Large variations in the physical and chemical properties of hornblendes in metamafic rocks reconstituted above the epidote-amphibolite facies appear to be induced principally by critical changes in the bulk composition of the total rock, and not by the regional gradients in T, P, or by changes in kind, or composition, of the coexisting minerals.

Basalts dredged from the Amirante ridge, western Indian ocean

Oceanic tholeiitic basalts were dredged from 2500 to 3000 m depth on each flank of the Amirante Ridge, 1200 km southeast of Somalia in the western Indian Ocean, by R.V. Argo in 1964. One sample, probably shed from a flow or dike in basement beneath the coralline cap, gave a wholerock KAr age of 82±16×106 years. The age is similar to those reported by others for agglomerate from Providence Reef, nearer Madagascar, and for gabbro from Chain Ridge, the southwest member of Owen Fracture Zone, nearer the Somali coast. The Amirante Cretaceous-Early Tertiary occurrence lies between the “continental” 650 × 106 years granites of Seychelles Archipelago and the large Precambrian “continental” block of Madagascar. Trends of major structures and distribution of the related topographic and magnetic-anomaly lineations in 7–8 × 106 km2of the surrounding Indian Ocean suggest that in addition to spreading of the seafloor from the seismically-active Mid-Indian Ocean Ridge-Carlsberg Ridge complex there has been, since mid-Mesozoic time, distributed left-lateral shear along 52°–54°E that has moved Madagascar at least 700 km south relative to Seychelles Bank. Measurements by other indicate the absolute movement of Madagascar has been southward as well. The emplacement of oceanic tholeiitic basalts at shallow depth, the development of volcanic topography between the sedimented Somali and Mascarene basins, and the existence of the faulted Amirante Trench and Ridge are consequences of the displacement.

GRENVILLE SERIES IN THE NORTHWEST ADIRONDACK MOUNTAINS, NEW YORK: PART II: ORIGIN AND METAMORPHISM OF THE MAJOR PARAGNEISS

The major gneissic metasediment within the Grenville series of the northwest Adirondacks extends northeastward across the Grenville Lowlands for about 35 miles, though areas of different degrees of metamorphism and igneous intrusion. Within this region, the gneiss is interpreted to be grossly monoclinal, but overturned to form the southeast flank of a regional anticlinorium. The opposite flank of this structure probably lies just northwest of the St. Lawrence River in the Brockville-Mallorytown-Kingston areas, Ontario. The gneiss is interstratified with thick zones of siliceous magnesian marble, thin quartzites, and schists. Two siliceous marble zones stratigraphically below the gneiss have a total thickness of about 8000 feet. A siliceous dolomite zone immediately above the gneiss (southeast of it) is at least 4000 feet thick and, with an overlying feldspathic quartzite, adjoins the central Adirondack igneous-rich massif. The gneiss itself is about 3000 feet thick and, except for thin interlayers of amphibolite and marble, the layers interpreted as relict beds are monotonous in texture and composition. These least-altered layers of gneiss are granoblastic, faintly foliated, quartz-greenish biotite-oligoclase gneiss in which the ratio of Na2O/K2O is approximately 1.3:1. Consequently the rock is far more sodic than typical shales. Layers of this composition are widely although sporadically distributed through the complex, and are injected lit-par-lit, replaced by and transitional into biotitic quartz-microcline-oligoclase granite, pegmatite, and alaskitic granite. Variants from these fades include garnetiferous and sillimanitic gneiss and migmatite, and tourmaline-bearing pegmatites. The simple assemblage quartz-greenish biotite-oligoclase, found principally in those areas farthest from large bodies of igneous granite, undoubtedly represents not only the least-altered metasediment but moreover the lowest-rank metamorphic facies into which the parent sediment is reconstituted. This rock type is abundant northwest of Hyatt, New York, in that segment of the gneiss farthest from the central Adirondack igneous complex. The alteration of the gneiss into various rock types reflecting increasing degree or rank of metamorphism as well as increasing interaction with magmatic fluids may be observed as the gneiss is traced northeastward toward Edwards, New York, which lies adjacent to the igneous massif. Alteration of the quartz-greenish biotite-oligoclase gneiss by granitic fluids commonly is accompanied by successive and marked decreases in biotite and in both the amount of plagioclase and in the per cent anorthite molecule of the plagioclase, as well as by a decrease in quartz content. Marked increases in potash feldspar content, particularly, serve as a useful index against which other mineralogical and textural changes may be plotted. As these changes appear in the gneiss, the greenish-brown biotite commonly gives way to a reddish-brown, probably more magnesian and titaniferous variety. A corollary to these successive mineralogical changes of gneiss toward granitic types are the chemical modifications involving, especially, a diminution in total iron, lime, and magnesia, and an increase in alkalis, particularly K2O, which becomes dominant over Na2O in many granitized facies. Intermediate types of this alkali-rich, potassic derivative of the gneiss most nearly approximate the composition of well-sorted shales and are possibly those considered by some earlier workers as representative of the sedimentary parent. Textural changes include the development not only of migmatite, pegmatite, and equigranular granite, but also of widespread augen gneisses and “porphyritic” granite, both of which contain large porphyroblasts and possibly some phenocrysts, chiefly microcline. The processes of injection and granitization seem to involve both magma and fluids from magma, which permeated the gneiss, interacting with and replacing it. Where the interaction of granitic fluids and gneisses occurred near, along, and in the central igneous-rich complex of the Adirondacks, in an environment of higher temperatures, the reddish-brown biotite appears almost or quite to the exclusion of greenish varieties. In addition, almandite garnet and sillimanite are formed, in that order, as products of the interaction. Almandite and some orthoclase also have formed as new phases presumably reflecting increasing rank of regional metamorphism without significant chemical modification of the parent sediment. Some of the earliest magmatic fluids appear to have included iron-rich derivatives, which by metasomatism formed scattered ferriferous biotite and hornblendic layers in and along the gneiss. These are largely granitized by the subsequent and dominant incursions of potassic, low-iron magma and associated fluids. Considerations of volume relations, compositional features, and paragenesis in the gneiss complex indicate that the potash in the “injecting” fluids had its source outside the gneiss; that is, it could not have been derived from the gneiss itself by secondary leaching, differential anatexis, or related processes. Nor seemingly could the relatively high soda content of the least-altered gneiss represent an earlier, but secondary, inoculation of the gneiss or parent sediment after diagenesis. Accordingly, the soda to potash ratio which approaches 1.3:1 in the quartz-biotite-oligoclase layers is regarded as either inherent in the sedimentary particles, or added essentially at the interface of sedimentation. If the composition of the quartz-biotite-oligoclase gneiss near Hyatt was inherent in the clastic particles deposited in Grenville seas, graywacke type sediments or tuffs are the most logical parent rocks. If soda was added to the clastic particles at the interface of deposition, or penecontemporaneously, the pre-existing sediment may have been one of the more common clay-weathering products. Unfortunately neither alternative is particularly compelling or more readily demonstrated than the others. The seemingly intimate and concordant intercalation of the gneiss with thick marbles and pure quartzites suggests that the gneiss was derived from shales or argillaceous sandstones which were products of marked residual weathering and good sorting. Graywackes, on the other hand, are presumed to imply a minimum of chemical weathering with rapid erosion, transportation, and deposition in an unstable crustal environment. Thick, monotonous tuffs are as anomalous within this Grenville association of sediments as are graywackes and hardly represent a logical parent to the gneiss. To assume soda-rich clays as the parent sediment poses problems equally great. The uniform addition of soda to great volumes of the commoner clayey products of residual weathering is almost unknown as a normal marine process. Commonly potash, rather than soda, is added to the accumulating clays, forming illitic shales. In addition to potash, both calcium and magnesium have greater replacing power than sodium and thus would replace it in requisite marine environments. The principal justification for assuming that soda could have been added to Grenville clays lies in indications that parts of the Grenville seas were, periodically at least, abnormally saline and soda rich. The great thickness of the carbonate sequence, its content of magnesia, silica, anhydrite, and halite indicate this, although most or all of these substances except the silica could be of secondary origin. If seas of abnormal salinity were involved, several possibilities suggest themselves. Some water-laid tuffs, altered tuffs, and clays are known to alter to zeolites and zeolitelike minerals. Thus, clinoptilolite, analcime, and apophyllite with more or less admixed clay are found as marine beds evolved through interaction of tuffaceous material and saline water. Some zeolites even form in open seas. The zeolitic rocks occur as both intermediate and end-stage products in the weathering of tuffs and probably of other rocks and form widespread uniform beds. As such they suggest a possible parent to the gneiss. Another alternative is that during sedimentation sodic shales are formed, by base exchange, from clay products of marked residual weathering. Base exchange of this type is credible if the seas were abnormally warm, rich in soda, and depleted in other bases which have greater energies of exchange than soda. In this environment, common clay minerals such as montmorillonite are known to take up soda in exchange for a previously held base, especially calcium, magnesium, or hydrogen. Lateral extensions of the Adirondack type gneiss should appear in the near-by parts of Canada where most gneisses in the Grenville series are described as “normal shales.” It is hardly reasonable to infer that the lateral transition from a sodic to potassic sediment coincides with the international boundary. The data, though fragmentary, suggest that gneiss of the Adirondack type occurs in the Brockville-Mallorytown-Kingston areas and possibly for many miles beyond. Along and northeast of the Ottawa River, however, most gneisses described to date appear to represent derivatives of more normal shales or argillaceous sandstones. Accordingly the Adirondack gneiss may be inferred to pinch out to the northeast or grade laterally into metasediment probably derived from much less sodic parent elastics. In the northern and northeastern parts of the subprovince, however, paragneisses which seem to approach the Adirondack type are reported in numerous areas. The relations of these rocks to those in the southwest part of the subprovince are, unfortunately, a major enigma.