Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Marine geology of the Cape Peninsula and False Bay wikipedia , lookup
The Marine Mammal Center wikipedia , lookup
Marine habitats wikipedia , lookup
Effects of global warming on oceans wikipedia , lookup
Marine pollution wikipedia , lookup
Abyssal plain wikipedia , lookup
Pacific Ocean wikipedia , lookup
Marine debris wikipedia , lookup
Ecosystem of the North Pacific Subtropical Gyre wikipedia , lookup
Clays and Clay Minerals, Vol. 27, No. 3, pp. 185-194, 1979. ORIGIN OF IRON-RICH MONTMORILLONITE FROM THE MANGANESE NODULE BELT OF THE NORTH EQUATORIAL PACIFIC JAMES R. H E I N , 1 H S U E H - W E N Y E H , z A N D E L A I N E A L E X A N D E R 1 U.S. Geological Survey, 345 Middlefield Road, Menlo Park, California 94025 2 Hawaii Institute of Geophysics, University of Hawaii, Honolulu, Hawaii 96822 Abstract--Clay minerals in the upper 50 cm of sediment that surround the Cu- and Ni-rich manganese nodules in the North Equatorial Pacific form two fractions: terrigenous (mostly eolian) illite, chlorite, and kaolinite, and authigenic smectite. Smectite increases with depth in box cores from 26 to 39% and from 53 to 66% in the easternmost and westernmost areas respectively, and with distance seaward from the Americas from 26 to 53% in surface deposits. The change in the amount of smectite relative to other clay minerals is due to dilution by terrigenous debris; smectite probably forms at a uniform rate over much of the North Pacific deep-sea floor. The 6018 value for the smectite is +29.6%0which suggests that it formed authigenically at a temperature characteristic of the deep-sea floor. The smectite is an Fe-rich montmorillonite that probably forms by the low-temperature chemical combination of Fe hydroxides and silica. Silica is derived from dissolution of biogenic debris, and the Fe hydroxide is from volcanic activity at the East Pacific Rise, 4000 to 5000 km to the east. AI in the authigenic montmorillonite may be derived from the dissolution of large amounts of biogenic silica or from river-derived A1 that is adsorbed on Fe-Mn hydroxides in the oceans. The Fe-montmorillonite contains relatively abundant Cu, Zn, and Mn and is of possible economic importance as a source of these and other metals. Key Words---Authigenic Formation, Deep-Sea Clay, Manganese Nodule Belt, Montmorillonite, Pacific Sediments, Smectite. INTRODUCTION GENERAL GEOLOGIC SETTING This investigation was undertaken as part of the DOMES (Deep Ocean Mining and Environmental Study) project designed to gain a more complete understanding of the formation of manganese nodules in the North Equatorial Pacific. The study concentrated on three geographic areas each measuring two degrees on a side, centered at 9~ 151~ (Area A), 12~ 138.5~ (Area B), and 15~ 126~ (Area C) (Figure 1). Areas A, B, and C are within a larger area (bounded by 5~ to 20~ latitude and 110~ to 180~ longitude) that is of great commercial interest for the mining of deep-sea nodules. Area A and C are about 2500 km apart. Preliminary results of the DOMES geologic and geochemical studies have been reported in Open-File Reports of the U.S. Geological Survey compiled by Bischoff (1976) and Piper (1977). Within each 2~area several box cores measuring 50 • 50 cm were taken at each of several stations. Between 20 and 50 cm of sediment was penetrated and recovered at each station. Stations 46-50 are in Area A, 51-55 are in Area B, and 11-27 are in Area C. Every 2-cm interval of the box core collected at Stations 47, 52, 16, 18B, 20, and 25 was sampled; other cores were sampled at less frequent intervals. This report discusses the distribution of smectite, its chemical and oxygen isotopic composition, and its mode of formation. Conclusions are also drawn on the possible significance of authigenic smectite to the geochemical balance of the oceans. Areas A, B, and C are located between the Clarion and Clipperton fracture zones, in abyssal-hill topography (Figure 1). Sediment overlying basaltic basement is 150 to 200 m thick. The upper 50 cm of sediment sampled by DOMES consists of detrital and authigenic clay minerals, quartz, feldspar, siliceous microfossils, locally abundant silicic volcanic glass, Fe and Mn hydroxides and oxides, microscopic Mn nodules, and minor apatite (fish debris), zeolites, and opaque minerals (Hein et al., 1979; Piper et al., 1979). Based on radiolarian biostratigraphy, the sediment is Quaternary, but contains minor to abundant reworked Miocene and Eocene microfossils (Quinterno and Theyer, 1979). Classified according to Deep-Sea Drilling Project (DSDP) standards, DOMES sediments range from siliceous ooze to red clay. This is consistent with their location within a narrow belt of siliceous biogenic deposits that occur between deposits of thick calcareous ooze at the equator and thin, but extensive red clay in the North Pacific. Twenty to 100% of the seafloor in the DOMES areas is covered with Ni- and Cu-rich Mn nodules. Copyright 9 1979, The Clay Minerals Society METHODS Clay mineralogy was determined on the <2-/zm size fraction with a Norelco diffractometer using C u K a radiation, a curved crystal carbon monochrometer, and a theta-compensating slit (Hein et al., 1976). Weighting factors used in calculating percentages are those of Bis- 185 186 160 ~ 150 o 140 ~ r * 150"o r 120 o i t II 0 o i00 o ~X~:.,I.. c 90 ~ .~ 30 ~ Hawaiian Islands 20 ~ ............... - Clays and Clay Minerals Hein, Yeh, and Alexander ;[ i - ON-E- - "D ': cD 10~ A O _ .CL~~ IF "_R .TONFRACTUREzONE ...... ~I 0o ~// I000 kilometers alt the equator I ] 1 ] 7 L I I0~ Figure 1. Location of DOMES Areas A, B, and C between the Clarion and Clipperton fracture zones. caye (1965), that is, two times the sum of the area of the chlorite and kaolinite peaks and four times the area of the illite peak. In general, authigenic smectite was isolated for chemical and isotopic analysis by obtaining finer and finer size fractions, but due to the limited amount of sample available, not all samples were monomineralic (see Table 4). All samples were treated with Morgan's solution (sodium acetate + glacial acetic acid) and HzO2 to remove calcite, organic matter, and Mn oxyhydroxides. These chemical treatments may also have some influence on the exchangeable cations in the smectite. The < 1-/zm size fraction was used for chemical analysis. Forty elements were scanned by the semiquantitative emission spectroscopic method of the U.S. Geological Survey, the reliability of which can be estimated by comparing the spectroscopic values with the values determined by atomic absorption (AA) spectroscopy on the <0.1-/zm size fraction of sample 47-10 (Table 2). Si was not determined by the emission spectroscopic technique, which is primarily used to determine minor and trace elements. Oxygen isotopic compositions of the <l-/xm size fractions (<0.1 ~m for sample 47-10) were determined by the techniques of Yeh and Savin (1977). Isotope values are reported in the 8-notation and are presented relative to the Standard Mean Ocean Water (SMOW) standard (Table 4). RESULTS Distribution of clay minerals In Areas A and B clay minerals in the <2-/zm size fraction are dominated by smectite; illite is of secondary importance, and kaolinite + chlorite is minor (Table 1). At stations 52, 53, and 54 in Area B, illite is equal to smectite in abundance. In contrast, illite is dominant (40-50%) in Area C except locally in the lower part of most cores where smectite is the dominant clay min- Table I. Lateral and temporal distribution of clay minerals in DOMES cores. Depth (cm) Surface 0-202 >20 Area C % Area B % Area A % Clay mineraP 26 48 26 25 44 31 19 37 44 19 40 42 16 31 53 16 29 56 K+ C I S K + C I S 23 38 39 18 36 46 13 22 66 K + C I S Percentages are averaged for the cores in each area. K = Kaolinite; I = Illite; S = Smectite; C = Chlorite. =0-15 cm used for Area C. Vol. 27, No. 3, 1979 Iron-rich montmorillonite from the Pacific manganese nodule belt Table 2. Sample ~ no. 187 Chemistry of DOMES smectites. Si % AI % Ti % Fe % K % Ca % Na % Mg % P % Cu ppm Ni ppm Co ppm Mn ppm 47-10 47-10 z 48-19 52-42 55-53 16B-49 18A-36 18B-37 nd 21.32 nd nd nd nd nd nd 7.5 7.45 6.6 7.1 5.9 7.6 4.0 6.2 0.5 0.39 0.5 0.5 0.5 0.5 0.5 0.7 7 6.47 7 7 7 7 10 7 nd 1.86 nd nd nd nd nd nd 0.2 0.63 0.3 0.2 0.2 0.2 2.0 0.2 3.0 2.06 3.0 3.0 2.0 2.0 1.5 1.5 2.0 3.14 3.0 3.0 3.0 2.0 2.0 2.0 ------0.3 -- 1500 1100 1500 1000 1500 1000 1000 700 150 500 150 70 150 100 100 100 100 200 70 50 70 70 30 50 2000 3300 2000 1000 2000 1500 3000 2000 Sample no. Zn ppm Ba ppm Zr ppm Cr ppm Pb ppm Sr ppm V ppm Li ppm B ppm Be ppm La ppm Nb ppm Sc ppm 47-10 47-102 48-19 52-42 55-53 16B-49 18A-36 18B-37 300 1300 300 300 300 300 500 300 1500 500 1000 1000 1000 700 2000 1000 300 nd 500 500 500 500 700 300 30 nd 30 50 20 50 15 50 50 nd 50 70 50 70 100 70 70 nd 100 70 70 70 300 100 150 nd 150 150 100 200 200 150 500 nd 300 200 300 200 --- 200 nd 200 300 200 300 100 300 5 nd 7 7 7 7 5 7 50 nd 70 30 50 30 70 30 -nd 15 15 15 15 20 15 50 nd 50 30 50 30 20 30 Sample no, Sn ppm Y ppm Ga ppm Yb ppm Pr ppm 47-10 48-19 52-42 55-53 16B-49 18A-36 18B-37 15 20 15 15 15 --- 150 t50 150 100 200 200 150 50 50 50 50 50 30 50 10 15 10 10 10 15 10 50 50 50 50 50 50 50 1 See Table 5 for purity of smectite analyzed. 2 Ignition loss is 18.43% at 1000~ <0,1-p.m size fraction analyzed by atomic absorption spectroscopy, all other samples are the < 1.0-/zm fraction analyzed by emission spectroscopy. nd, means not determined; - - , means below limit of detection. Other elements scanned that were below the limits of detection in all samples are: Cd (15), Mo (5), TI (7), Ce (100), Nd (50), Ag (1.5), Au (15), As (200), Bi (15), Pt (10), Sb (30), W (20), Te (700), U (300), Hf (100), Sm (100), Eu (2), Re (15), Ta (50), Th (300), Ge (15), Pd (2), In (3); detection limits in brackets in ppm. Detection limits of Li and Sn are 200 and 5 ppm, respectively. eral. Similarly, at Stations 47 and 48 in A r e a A and at Station 55 in A r e a B, smectite increases with depth in the cores (Table 1); it makes up 90% of the clay minerals in the l o w e r m o s t deposits at Station 48. At other stations in Areas A and B, the clay mineralogy does not change with depth. Smectite not only increases with depth below the seafloor but also increases in abundance with greater distance from the Americas, that is, from A r e a C to Area A (Table 1). Smectite can, however, be locally abundant in any of the three areas. Smectite chemistry and mineralogy E x c e p t for sample 18A, little variation in the chemistry of the < 1-/zm size fractions was o b s e r v e d (Table 2). Sample 18A contains less AI and more Fe and Ca than the other smectites. In all smectites, Cu, Ni, Mn, and possibly Zn are relatively abundant. The Zn and Ba concentrations for sample 47-10 obtained by A A and emission spectroscopy are too different to be due simply to the difference in grain size of the samples and suggest an analytical error for these two elements in the emission spectroscopy technique (Table 2). The structural formula for smectite (sample 47-10; Table 3) was derived by the technique of Ross and Hendricks (1945). Si makes up most of the tetrahedral layer (Si/A1 - 6.8), and AI, Fe, and Mg are primary constituents of the octahedral layer. All Fe was arbitrarily considered as Fe a§ Other trace elements (Table 2), such as Cu, Zn, Co, Mn, Pb, Ni, Cd, and Cr, can be structurally bound in octahedral coordination or can be present as exchangeable cations in clays (Counts et al., 1973; Pronina and Varentsov, 1973; McBride, 1976; K o p p e l m a n and Dillard, 1977). The exchangeable cations, h o w e v e r , are mainly Na, with some K and Ca (Table 3). Because sample 47-10 may contain up to 10% illite interlayers, or may contain minor illite contami- 188 Table 3. Atomic properties calculated for DOMES sample 47-10. No. of ions Tetrahedral Octahedral Interlayer Clays and Clay Minerals Hein, Yeh, and Alexander Si AI Total A1 Mg Fe 3+ Ti Mn Zn Cu Ni Total 3.484 0.516 4.000 0.7506 0.5921 0.5324 0.0367 0.0275 0.0090 0.0078 0.0037 1.9598 Na K Ca Total 0.4085 0.2203 0.1441 0.7729 Octahedral + tetrahedral excess charge is -1.24. (Ca.14K.22Na.41) (Ni.004Cu.01Zn.olMn.03 Ti.o4Mg.5~Fe3+53AI.Ts)(A1.52Si3.4s) (O~o(OH)z) Table 4. Oxygen isotopic composition of DOMES smectite, split of same samples analyzed for chemistry. Size Sample no. Depth in core (cm) 47-10 47-10 48-19 52-42 55-53 16B-49 16B-49 18A-36 18B-37 39 to 42 39 to 42 27 to 31 43 to 47 26 to 30 16 to 20 16 to 20 17 to 20 13 to 17 Smectite ~ s (#.m) 80 TM Smectite (%) d(001) (.~) <1.0 <0.1 <1.0 <1.0 <1.0 <1.0 >1.0 <1.0 <1.0 +27.6 +29.6 +26.9 +24.7 +28.3 +24.9 +20.4 +26.4 +22.7 95 to 100 98 to 100 98 to 100 65 to 75 94 to 100 65 to 75 -10 85 to 95 -46 16.9 S 16.7 MS 16.9 S 16.8 M 16.9 S 16.7 B -17.7 VB 16.9 M 1 Percentage of the respective size fractions estimated qualitatively from X-ray powder diffractograms. 2 Glycolatedsamples;S =sharp, M = moderate, B = broad. 16B which is a mixture of terrigenous, biogenic and authigenic material. In general, the variation in isotopic compositions can be attributed to changes in the relative proportion of authigenic smectite to the detrital minerals such as illite, kaolinite, quartz, and so on (Table 4); detrital minerals lower the ~O is values. DISCUSSION nation, some of the K may be present in illite rather than as an exchangeable cation in the smectite. The excess tetrahedral and octahedral charge (Table 3) is large but is in the range for smectites (Weaver and Pollard, 1975). Smectites from the deep sea commonly have larger layer charges (>0.8) than do those formed on continents (e.g., Bischoff, 1972; Weaver and Pollard, 1975; Scheidegger and Stakes, 1977). The reason for this difference is unknown; marine smectites may be interlayered with amorphous oxyhydroxides, a condition that would change the structural formula and layer charge. Alternatively, a large number of minor elements (Table 2) that are not included in the layer charge calculation may occupy the octahedral layer. The Si, AI, Mg, and Fe atomic proportions show that sample 47-10 is an Fe-rich montomrillonite. The (060) X-ray spacing of most samples is very weak and broad (from about 1.508 to 1.528/~) and cannot be used to support the identification of a dioctahedral smectite structure. The (001) peaks are, in general, sharp, of high amplitude, and occur near 16.9 /~ when glycolated, whereas the peak for sample 18A is very broad, weak, and centered near 17.7/~ (Table 4). Higher order X-ray spacings are very weak or absent. Oxygen isotopic composition of smectite 8018 values range from +22.7 to +28.3 for the < l wm size fractions (Table 4). The isotopically heaviest measurement (+29.6), however, is from the <0.1-/xm fraction (pure smectite) of sample 47-10, and the lightest value (+ 20.4) is from the > 1-/~m fraction of sample Abyssal sediment may be composed of terrigenous, biogenic, volcanic, and authigenic debris. Also found in abyssal sediment but not discussed much in the literature are amorphous Fe, AI, Mn, and Si hydroxides. Amorphous debris enters the oceans from rivers (Si, AI, Fe), from hydrothermal activity at oceanic spreading centers (Fe, Mn, Si), and from low-temperature weathering of oceanic basaltic basement (Fe, Mn, Si, A1). Most of these components are found in DOMES cores, and the distribution and origin of the terrigenous, volcanic, and biogenic debris is discussed by Hein et al. (1979). The authigenic fraction consists of smectite (discussed here), Mn nodules and micronodules, zeolites (phillipsite and clinoptilolite), and possibly apatite. Source of clay minerals DOMES illite, chlorite, and kaolinite are terrigenous, probably mostly eolian in origin, whereas nearly all the smectite is authigenic. K-Ar dating of illite isolated from Holocene abyssal deposits shows that the illite is continental and formed several hundred million years ago (Hurley et al., 1959). It has also been shown that illite abundance in the Pacific correlates with that of quartz and that both correlate with the atmospheric circulation patterns. This implies that illite and quartz are probably of mostly eolian origin (Heath, 1969; Windora, 1969). Dust collected from the atmosphere is rich in quartz and illite (Windom, 1969). Oxygen and hydrogen isotopic compositions of deep-sea illite also support a continental origin (Savin and Epstein, 1970b). Illite probably will not form from smectite on the sea- Vol. 27, No. 3, 1979 Iron-rich montmorillonite from the Pacific manganese nodule belt floor as it does in thick sedimentary sequences on land, even if the reaction is favored thermodynamically, because the reaction rate is too slow (Eberl and Hower, 1976). Griffin and Goldberg (1963) concluded that chlorite and kaolinite distributions in the Pacific indicated their derivation from the continents. Therefore, the illite, chlorite, and kaolinite from DOMES cores are likely derived from the continents and are transported primarily by winds. In contrast, abyssal smectite is formed in the marine environment. Smectite is not an important part of atmospheric dust (Windom, 1969), nor is there a major continental source for the Fe-rich smectite found in the deep sea (Aoki et al., 1974). Griffin and Goldberg (1963) suggested that smectite in the South Pacific formed from the in situ alteration of volcanic debris. Apparently, only half of the smectite in South Atlantic deposits and a third of that in Indian Ocean sediment is derived from the continents (Chester and Stoner, 1974). Recently, Heath and Dymond (1977) found that smectite in the Bauer Basin (west of South America) formed in the marine environment. By plotting the percentage of smectite in DOMES samples versus the 8018 values, it can be shown that the 8018 values of the <0. 1-/zm fraction for each sample are about the same (+29.6%0 as that measured in sample 47-10 (Table 4). Thus, DOMES smectite must have also formed in a deep-sea environment (see below). Detrital smectites have 80 TM values less than +22%o (Savin and Epstein, 1970a, 1970b; Yeh and Eslinger, 1979), whereas, the 8018 endmember value for authigenic smectite formed at 0~ in the deep sea is +31%o (Yeh and Savin, 1977). Hence, in contrast to the speculation of Hein and Jones (1977), the isotopic data indicate that essentially all the smectite in the < 1-~m size fraction is authigenic, with terrigenous smectite making up less than about 5%. Smectite in the > 1-/zm fraction is likely to have had, but is not limited to, an authigenic origin. Formation o f D O M E S smectite Four general mechanisms have been postulated for the formation of authigenic smectite in the deep sea. First, smectite may precipitate directly from hydrothermal fluids; for example, nontronite in the Red Sea (Bischoff, 1972). Such smectite is rich in Fe and low in AI and Mg (Table 5). Second, authigenic smectite may precipitate from solutions at low temperatures into vesicles and fractures of basalt at oceanic spreading centers (Seyfried et al., 1976). Elements for this smectite were derived from seawater or were leached from basalt by seawater. These smectites are rich in Fe and Mg and have moderate AI contents (Table 5). Third, authigenic smectite may form by the alteration of volcanic rock fragments and glass in the marine environment. This process produces moderate to high Fe and Mg contents and higher Al-smectite than the first two mechanisms and is the most commonly reported mech- 189 anism for the formation of smectite in basal marine deposits, in deep-sea ash beds, and in altered oceanic basaltic crust (Scheidegger and Stakes, 1977; VaUier and Kidd, 1977; Hein and SchoU, 1978). Fourth, authigenic smectite may form from the low-temperature combination of Fe oxyhydroxide and silica (Heath and Dymond, 1977). This reaction produces high-Fe, low-Al, and probably low-Mg smectite when the clay forms near where Fe oxyhydroxide is generated, and moderate- to high-Fe, moderate-A1, and low- to moderateMg clays at great distances from the Fe oxyhydroxide source (Table 5; see discussion below). A hydrothermal origin for DOMES smectite can be eliminated because of the oxygen isotopic compositions. Using the smectite-water equilibrium equation of Yeh and Savin (1977) and assuming 8018 of 0.0 for seawater, a maximum temperature of formation for DOMES smectite is 7.0~ If as little as 2% terrigenous quartz, chlorite, and illite contaminate the sample, the temperature of formation would be near 0.0~ essentially the same as bottom water temperatures in DOMES areas (0.8~176 (Gordon and Gerard, 1970). Bischoff and Rosenbauer (1977) suggested that there is a large component of metalliferous sediment at stations 18A and 18B that was produced by hydrothermal activity. The smectite examined at station 18B is from a layer that occurs above the metalliferous sediment; however, the smectite from station 18A included metalliferous deposits, and the oxygen isotope data indicate a low temperature of formation. A low-temperature, vesicle-filling origin for the smectite can be eliminated because DOMES sites are not at an oceanic spreading center. The bulk chemistry of DOMES smectites does not preclude their formation from the alteration of basaltic volcanic debris, however, analyses of all 2-cm intervals of many cores show little volcanic material. Also, the relatively high percentage of some trace elements (Cu, Zn, Mn, Co) is inconsistent with altered volcanic debris as a source of the smectite. Petrographically, light-colored andesitic volcanic glass shards in DOMES cores are very fresh (Hein et al., 1979) and possibly were transported from Central and South America by winds. Palagonite occurs in trace amounts in some cores and is a significant component at only two DOMES stations-10 and 15--where it has been reworked from nearby basement outcrops (Cook et al., 1977). Thus, the formation of DOMES smectite by alteration of volcanic debris in the near-surface sediment is not probable (see further discussion on this point below). The oxygen-isotope ratios suggest that the smectite formed in the marine environment at temperatures typical of the deep-sea floor. Therefore, it is possible that smectite formed by alteration of volcanic debris where such debris is abundant and was then transported to the DOMES sites by bottom currents. Basaltic volcanic sediment is abundant near mid-Pacific volcanic island Hein, Yeh, and Alexander 190 Table 5. Reference Clays and Clay Minerals Elemental ratios for smectites that formed in the deep sea. Ti/AI Si/AI Mg/AI Ca/AI Origin ~ 0.87 2.50 0.05 0.13 2.86 -- 0.42 0.50 0.08 0.50 4 ~0.45 1.07 0.13 0.29 --- --- --- 3 -20.00 -- 15.33 0.64 0.32 1 16.10 -- 18.93 -- -- 4 0.76 2.05 2.61 0.56 0.03 0.03 0.30 0.38 3.13 3.46 3.55 2.34 0.64 1.84 0.42 1.40 0.04 0.07 0.07 0. l0 3 Drever (1976) 0.73 0.76 to 1.50 --- 2.79 2.96 to 3.76 0.24 0.26 to 0.41 0.09 0.15 to 0.50 5 3 Scheidegger and Stakes (1977) 2.95 2.72 1.96 to 3.12 0.06 0.08 0.08 5.93 6.67 4.62 to 5.84 2.56 3.59 1.41 to 2.10 0.42 0.41 0.28 to 0.42 3 Banks (1972) 3.22 0.07 7.35 6.91 0.16 3 Seyfried et al. (1976) 4.23 0.05 7.77 3.80 0.06 This work, sample 47-10 sample 18A Hein and Scholl (1978) Bischoff (1972) Heath and Dymond (1977) Melson and Thompson (1973) Kastner (1976) Fe/AI -0.50 1.70 Aoki et al. (1974) 1.82 to 2.97 . . . . 0.08 to 0.25 . . 5.61 to 6.93 . . 0.59 to 0.68 2 3 0.02 to 0.18 4 1 Origin of smectite as described in text: (1) hydrothermal precipitation; (2) low temperature precipitation at spreading centers; (3) alteration of volcanic debris; (4) chemical combination of Fe oxyhydroxides and biogenic silica; (5) detrital smectite for comparison. chains; however, the major deep-sea circulation is from south to north, with water-flow velocities of about 0.050.1 cm/sec (Knauss, 1962). Antarctic bottom water may also enter the Equatorial East Pacific through passes in the Line Island Ridge (Reid, 1969). These current directions suggest that deep-sea sediment surrounding the north-south-trending volcanic island chains north and southwest of the D O M E S sites are not likely sources for DOMES smectite. If the extensive DOMES clay deposits were formed by low-temperature reaction of Fe oxyhydroxide and silica in a deep ocean-floor environment, the source of the Si, Al, and Fe must be discussed. Heath and Dymond (1977) proposed that for smectite in the Bauer Basin, Fe in the form of Fe hydroxide or oxyhydroxide was derived from hydrothermal activity at the East Pacific Rise (EPR) about 650 km to the west. Si was derived by the dissolution of siliceous microfossils. We propose a similar origin for the Fe and Si in DOMES samples. Because there are no local sources for Fe, colloidal Fe oxyhydroxides must have been transported 4000-5000 km from the EPR source. Volcanic cones are scarce in this part of the Pacific. The Line and Hawaiian Islands and the Clarion fracture zone, 500-800 km from D O M E S Area A, may have provided some Fe, but only very minor amounts compared to that supplied by the EPR. The EPR is a welldocumented source for Fe and Mn hydroxides that are apparently widely distributed adjacent to the ridge (Bostrom, 1970; Lyle, 1976). The present study requires that volcanic Fe and Mn hydroxides produced at the EPR are distributed for many thousands of kilometers. It is possible that oceanic spreading centers (especially where spreading rates are high) provide abundant colloidal Fe-Mn compounds that are distributed throughout the deep sea and become major sediment builders in areas where terrigenous and biogenic sedimentation is minor. Biogenic silica (5-50%) in DOMES cores decreases in abundance with depth (Hein et al., 1979). Spicules become progressively more abundant relative to other siliceous microfossils with depth, suggesting that wholesale dissolution of the more soluble, delicate forms has taken place. Thus, biogenic silica was likely available to combine with the Fe hydroxide derived from the spreading centers to produce smectite. These deposits probably cover extensive areas of the seafloor. It is not clear at this time in what proportions the extensive smectite deposits in the South Pacific (Peterson and Griffin, 1964) are due to the alteration of locally derived basaltic volcanic debris versus the combination of Fe hydroxide and silica. The problem is under investigation. The origin of Al is not as easily understood, although there are three possible sources. First, Al bound in or adsorbed on biogenic silica will be released upon dis- Voi. 27, No. 3, 1979 Iron-rich montmorillonite from the Pacific manganese nodule belt solution. Second, A1 entering the oceans from rivers may be deposited on the seafloor in the form of amorphous A1 hydroxides or it may be adsorbed on other particles. Third, Al may be released during the alteration of minute volcanic dust particles in the sediment. Hydes (1977) and MacKenzie et al. (1978) suggested that Al is removed from seawater by living diatoms, and Hurd (1973) believed that Al is adsorbed on frustules of dead diatoms. In fact, it has been proposed that smectite forms in the frustuies of living diatoms (Van Bennekom and Van der Gaast, 1976). Sayles and Bischoff (1973) reported the occurrence of smectite pseudomorphs after radiolaria, and Johnson (1976) proposed that authigenic smectite formed in siliceous deposits from the silica released by dissolution. F. T. MacKenzie and co-workers (personal communication, 1978, Northwestern University) found that A1 covaries with Si in pore waters of sediment containing biogenic silica. Their data suggest a release of Al and Si by dissolution of diatoms and later uptake of these elements by formation of an aluminosilicate. It is likely that dissolution of siliceous biogenic debris provided some or perhaps most of the A1 for formation of smectite, although other sources are also required. Some of the Al may have been derived from Al hydroxides supplied by rivers. A1 may either have been adsorbed on ferric hydroxide in the water column or, less likely, it may have been deposited uncombined as amorphous A1 hydroxide. Fine volcanic dust in the sediment cannot be completely ruled out as a source for Al or for the smectite in general. However, if volcanic dust were present, it would be andesitic or more silicic (there is no source for abundant basaltic dust) in composition and would probably not alter within the short time period represented by the sediment. The coarser grained andesitic glass is petrographically very fresh. Hein and Scholl (1978) showed that deep-sea andesitic ash layers in the Bering Sea take 3-5 million years to alter to smectite, but fine dust should not take so long. Further, the chemistry of smectite commonly reflects the chemistry of the parent ash (Kastner, 1976; Hein and Scholl, 1978), and DOMES smectite more closely resembles basaltic rather than andesitic ash. From analysis of the bulk chemical composition of DOMES deposits, Bischoff et al. (1979) concluded that Areas A and B contain virtually no volcanic debris and Area C contains perhaps up to 10%. The minor and trace elements (Table 2) are derived primarily from seawater and biogenic debris. High Cu and Zn values relative to Ni in the smectite is similar to that found in nodules, where Cu and Zn are associated more with Fe phases and Ni with Mn phases. Smectites, well known for their adsorptive properties, probably scavenge trace elements from seawater. Fe and Mn hydroxide also scavenge metals from seawater 191 (Hem, 1977; James and MacNaughton, 1977) and may account for some trace metals in nodules and in clays. SUMMARY AND CONCLUSIONS DOMES clay minerals form two fractions: detrital illite, chlorite, and kaolinite; and authigenic smectite. The increase in smectite relative to the other clays with depth in some cores is due to dilution of the authigenic smectite by terrigenous debris in the upper deposits (Hein et al., 1979). Authigenic smectite probably forms at a relatively uniform rate over much of the North Pacific from the combination of Fe hydroxide produced at the East Pacific Rise, Si and A1 released from biogenic silica, and perhpas from A1 that enters the ocean from rivers. Harder (1978) synthesized iron-rich smectite at low temperature from hydroxides and silica. He found that reducing conditions and the presence of Mg in solution aid in the synthesis. DOMES smectite formed near the sediment-water interface, where Mg is abundant. The bottom waters, however, are highly oxygenated, and, consequently, smectite may not form until after some burial, perhaps to 10-20 cm where conditions are more reducing. Thus, silica released from dissolution of biogenic debris in the upper few decimeters of sediment is being used very rapidly after production, probably by chemisorption on hydroxides (production of smectite), or is being released into the water column. Volcanic debris is an unlikely source for DOMES smectite because: (1) Basaltic volcanic debris has not been found in DOMES deposits. Even if extensive alteration of the volcanic debris occurred within the short time period represented by DOMES cores, one should find at least traces of the parent basalt. (2) There is no source of abundant basaltic glass or rock fragments in or adjacent to the area of this study. (3) If volcanic dust, too fine grained to be detected, were present it would be andesitic or more silicic in composition and would alter to smectite with a different chemical composition than that determined for DOMES smectite. (4) More volcanic debris would have to have been present in DOMES deposits in order: to produce the large amount of smectite found than is typically found in deep-sea biogenic deposits of Quaternary age. (5) The trace element content of DOMES smectite is too high for volcanic debris to have been the parent material. (6) Phillipsite, commonly produced with smectite during alteration of basaltic debris is rare in DOMES deposits. Clinoptilolite, also uncommon in DOMES cores, can form from siliceous biogenic debris rather than from alteration of volcanic debris (Hein et al., 1978). Authigenic smectite forms over an area at least 2700 km (E-W) by 700 km (N-S). Based on a conservative mean clay content of 30% (Hein et al., 1979) of which 30% is smectite (Table l), authigenic smectite comprises about 10% of the sediment. In some areas it 192 Hein, Yeh, and Alexander makes up more than 50%. Thus, there may be about 20 x 10a cubic meters of Cu-, Zn-, and Mn-containing smectite in the upper 10 cm of sediment that connects the three DOM ES areas. These deposits may offer a long-range resource if areas where smectite, rich in some of the uncommon and trace metals, can be delineated. Authigenic smectite must be considered in any evaluation of the geochemical balance of the oceans, especially in regard to Si, Fe, AI, Cu, Zn, Mn, and other trace metals. ACKNOWLEDGMENTS Our gratitude to G. Ross Heath, University of Rhode Island, J. I. Drever, University of Wyoming, and Tracy L. Vallier, U.S. Geological Survey, for reviewing this manuscript. C. Robin Ross and Jeanne Henning gave technical assistance. Kam Leong provided the A A resuits, and Robert Mays and Chris Heropoulos the emission spectroscopic data. Support for the isotopic work is from N S F Grant DES71-00558-A03 to S. Epstein. REFERENCES Aoki, S., Kohyama, N., and Sudo, T. (1974) An iron-rich montmorillonite in a sediment core from the northeastern Pacific: Deep-Sea Res. 21,865-875. Banks, H. H. (1972) Iron-rich saponite: Additional data on samples dredged from the Mid-Atlantic Ridge, 22~ latitude: Smithson. Contrib. Earth Sci. 9, 39-42. Biscaye, P. E. (1965) Mineralogy and sedimentation of recent deep sea clay in the Atlantic Ocean and adjacent seas and oceans: Geol. Soc. Am. Bull. 76, 803-832. Bischoff, J. L. (1972) A ferroan nontronite from the Red Sea geothermal system: Clays & Clay Minerals 20, 217-223. Bischoff, J. L. (Compiler) (1976) Deep Ocean Mining Environmental Study, NE Pacific Nodule Province, Site C, Geology and Geochemistry: U.S. Geol. Surv. Open-file Rep. 76-548, 275 pp. Bischoff, J. L., Heath, G. R., and Leinen, M. (1979) Geochemistry of deep-sea sediments from the Pacific Manganese Nodule Province: DOMES sites A, B, and C: in Marine Geology and Oceanography of the Pacific Manganese Nodule Province, J. L. Bischoff and D. Z. Piper, eds., Plenum, New York, 397-436. Bischoff, J. L. and Rosenbauer, R. J. (1977) Recent metalliferous sediment in the North Pacific Manganese Nodule Area: Earth Planet. Sci. Lett. 33, 379-354. Bostrom, K. (1970) Submarine volcanism as a source for iron: Earth Planet. Sci. Lett. 9, 348-354. Chester, R. and Stoner, J. H. (1974) Montmorillonite in surface detritus: Nature 249, 335-336. Cook, H. E., Piper, D. Z., and Gardner, J. V. (1977) Geologic and oceanographic framework of DOMES Sites A, B, and C: Central Equatorial Pacific: in Deep Ocean Environmental Study: Geology and Geochemistry of DOMES Sites A, B, and C, Equatorial North Pacific, D. Z. Piper, Compiler, U.S. Geol. Surv. Open-file Rep. 77-778, 15-81. Counts, M. E., Jen, J. S. C., and Wightman, J. P. (1973) An electron spectroscopy for chemical analysis study of lead adsorbed on montmorillonite: J. Phys. Chem. 77, 19241926. Drever, J. I. (1976) Chemical and mineralogical studies, Site 323: Initial Reports of the Deep-Sea Drilling Project 35, 471-477. Clays and Clay Minerals Eberl, D. B. and Hower, J. (1976) Kinetics ofillite formation: Geol. Soc. Am. Bull. 87, 1326-1330. Gordon, A. L. and Gerard, R. D. (1970) North Pacific bottom potential temperatures: in Geological Investigations of the North Pacific, J. D. Hayes, ed., Geol. Soc. Am. Mem. 126, 23-39. Griffin, J. J. and Goldberg, E. D. (1963) Clay mineral distributions in the Pacific Ocean: in The Sea Vol. 3, M. N. Hill, ed., John Wiley and Sons, New York, 728-741. Harder, H. (1978) Synthesis of iron layer silicate minerals under natural conditions: Clays & Clay Minerals 26, 65-72. Heath, G. R. (1969) Mineralogy of Cenozoic deep-sea sediments from the equatorial Pacific Ocean: Geol. Soc. Am. Bull. 80, 1997-2018. Heath, G. R. and Dymond, J. (1977) Genesis and transformation of metalliferous sediments from the East Pacific Rise, Bauer Deep, and Central Basin, northwest Nazca Plate: Geol. Soc. Am. Bull. 88, 723-733. Hein, J. R. and Jones, M. G. (1977) Mineralogy and diagenesis of DOMES Sites A and B: in Deep Ocean Environmental Study: Geology and Geochemistry of DOMES Sites A, B, and C, Equatorial North Pacific, D. Z. Piper, Compiler, U.S. Geol. Surv. Open-file Rep. 77-778, 146-178. Hein, J. R., Ross, C. R., Alexander, Elaine, and Yeh, H-W (1979) Mineralogy and diagenesis of surface sediments from DOMES sites A, B, and C: in Marine Geology and Oceanography of the Pacific Manganese Nodule Province, J. Bischoff and D. Piper, eds., Plenum, New York, 365396. Hein, J. R. and Scholl, D. W. (1978) Diagenesis and distribution of late Cenozoic volcanic sediment in the southern Bering Sea: Geol. Soc. Am. Bull. 89, 197-210. Hein, J. R., Scholl, D. W., Barron, J. A., Jones, M. G., and Miller, J. (1978) Diagenesis of late Cenozoic diatomaceous deposits and formation of the bottom simulating reflector in the southern Bering Sea: Sedimentology 25, 155-181. Hein, J. R., Scholl, D. W., and Gutmacher, C. E. (1976) Neogene clay minerals of the far NW Pacific and southern Bering Sea: Sedimentation and diagenesis: Proc. Int. Clay Conf., Mexico City, 1975, 71-80. Hem, J. D. (1977) Reactions of metal ions at surfaces of hydrous iron oxide: Geochim. Cosmochim. Acta 41,527-538. Hurd, D. C. (1973) Interactions of biogenic opal, sediment and seawater in the central equatorial Pacific: Geochim. Cosmochim. Acta 37, 2257-2282. Hurely, P. M., Hart, S. R., Pinson, W. H., and Fairbairn, H. W. (1959) Authigenic versus detrital illite in sediments: Geol. Soc. Am. Bull. 70, 1622, (abstract). Hydes, D. J. (1977) Dissolved aluminum concentration in seawater: Nature 268, 136-137. James, R. O. and MacNaughton, M. G. (1977) The adsorption of aqueous heavy metals on inorganic minerals: Geochim. Cosmochim. Acta 41, 1549-1555. Johnson, T. C. (1976) Biogenic opal preservation in pelagic sediments of a small area in the eastern tropical Pacific: Geol. Soc. Am. Bull. 87, 1273-1282. Kastner, M. (1976) Diagenesis of basal sediments and basalts of Sites 322 and 323, Leg 35, Bellingshausen Abyssal Plain: Initial Reports of the Deep-Sea Drilling Project 35, 513527. Knauss, J. A. (1962) On some aspects of the deep circulation of the Pacific: J. Geophys. Res. 67, 3943-3953. Koppelman, M. H. and Dillard, J. G. (1977) A study in the adsorption of Ni (II) and Cu (II) by clay minerals: Clays & Clay Minerals 25, 457-462. Lyle, Mitchell (1976) Estimation ofhydrothermal manganese input to the oceans: Geology 4, 733-736. MacKenzie, F. T., Stoffyn, M., and Wollast, R. (1978) Aluminum in seawater: control by biological activity: Science 199, 680-682. Vol. 27, No. 3, 1979 Iron-rich montmorillonite from the Pacific manganese nodule belt McBride, M. B. (1976) Exchange and hydration properties of Cu 2§ on mixed ion Na§ z+ smectites: Soil Sci. Soc. Am. J. 40, 452-456. Melson, W. G. and Thompson, G. (1973) Glassy abyssal basalts, Atlantic sea floor near St. Paul's Rocks: Petrography and composition of secondary clay minerals: Geol. Soc. Am. Bull. 84, 703-716. Peterson, M. N. A. and Griffin, J. J. (1964) Volcanism and clay minerals in the southeastern Pacific: J. Mar. Res. 22, 13-21. Piper, D. Z. (Compiler) (1977) Deep Ocean Environmental Study: Geology and Geochemistry of DOMES Sites A, B, and C, Equatorial North Pacific, U.S. Geol. Surv. Open-file Rep. 77-778, 527 pp. Piper, D. Z. and Cook, H. (I979) Litho- and acousticstratigraphy of Equatorial North Pacific DOMES sites A, B, and C: in Marine Geology and Oceanography of the Pacific Manganese Nodule Province, J. L. Bischoff and D. Z. Piper, eds., Plenum, New York, 309-348. Pronina, N. V. and Varentsov, I. M. (1973) Adsorption of nickel and cobalt from seawater by natural iron and manganese hydroxides: Dokl. Akad. Nauk SSSR 210, 241-244. Quinterno, Paula and Theyer, F. (1979) Biostratigraphy of the Equatorial North Pacific DOMES sites A, B, and C: in Marine Geology and Oceanography of the Central Pacific Manganese Nodule Province, J. L. Bischoff and D. Z. Piper, eds., Plenum, New York, 349-364. Reid, J. L. (1969) Preliminary results of measurements of deep currents in the Pacific Ocean: Nature 226, 848. Ross, C. S. and Hendricks, S. B. (1945) Minerals of the montmorillonite group: U.S. Geol. Surv. Prof. Pap. 205-B, 79 ppSavin, S. M. and Epstein, S. (1970a) The oxygen and hydrogen isotope geochemistry of clay minerals: Geochim. Cosmochim. Acta 34, 25-42. 193 Savin, S. M. and Epstein, S. (1970b) The oxygen and hydrogen isotope geochemistry of ocean sediments and shales: Geochim. Cosmochim. Acta 34, 43-63. Sayles, F. L. and Bischoff, J. L. (1973) Ferromanganoan sediments in the Equatorial East Pacific: Earth Planet. Sci. Lett. 19, 330-336. Scheidegger, K. F. and Stakes, D. S. (1977) Mineralogy, chemistry and crystallization sequence of clay minerals in altered tholeiitic basalts from the Peru Trench: Earth Planet. Sci. Lett. 36, 413-422. Seyfried, W. E., Shanks, W. C., and Bischoff, J. L. (1976) Alteration and vein formation in Site 321 basalts: Initial Reports of the Deep-Sea Drilling Project 34, 385-392. Van Bennekom, A. J. and Van der Gaast, S. J. (1976) Possible clay structures in frustules of living diatoms: Geochim. Cosmochim. Acta 40, 1149-1152. Vallier, T. L. and Kidd, R. B. (1977) Volcanic sediments in the Indian Ocean: in Indian Ocean Geology and Biostratigraphy, J. R. Heirtzler, H. M. Bolli, T. A. Davies, J. B. Saunders, and J. G. Sclater, eds., Am. Geophys. Union, Washington, D.C., 87-118. Weaver, C. E. and Pollard, L. D. (1975) The Chemistry of Clay Minerals: Elsevier Scientific Pub. Co., New York, 213 PP. Windom, H. L. (1969) Atmospheric dust records in permanent snowfields: Implications to marine sedimentation: Geol. Soc. Am. Bull. 80, 761-782. Yeh, H. W. and Eslinger, E. V. (1979) Hydrogen and oxygen isotopic and mineralogic studies of the Mississippi River sediments (in preparation). Yeh, H. W. and Savin, S. M. (1977) Mechanism of burial metamorphism of argillaceous sediments: 3. O-isotope evidence: Geol. Soc. Am. Bull. 88, 1321-1330. (Received 19 June 1978; accepted 12 January 1979) Pe3mMe---F.rIHHHCTbIe MHHepa21bl a BepXHefi 50 CM HaCTH oca~IKOB, BMeI~amIRHX 6oraTb~e Cu H Ni MapraHueahle KOHKpeUHH B ceBepnofi 3KBaTOpHa2IbHOfi qaCTH THxoro oKeana no~Ipa3~le3Lq}OTCa Ha ~se qbparRnu: TeppnrenHb~e (a ocuoanoM ~O.rlOBble) HJI21HT, XJIOpHT H KaO~HHnT ~ ayTHreHHblfi CMeKTHT. Co/Iepx~anHe CMeKTHTa ynena~naaeTcn c rny6enofi a ropo6qaTblX repHax OT 26 ~Io 39% n OT 53 ~O 66% COOTBeTCTBeHHO B CaMOf[ 8OCTOqHOfi H CaMOg 3anaJIHOg 3onax a C y~a~enHeM OT AMeparancrnx KOHTHHeHTOBB CTOpOUyoKeana OT 26 nO 53% a HOBepXHOCTHblXOTJ10>KeHltflX.FIaMenenne conepxam~a CMeKTnTa no OTHOmeRu~O K ~lpyrHM MuHepa.qaM o6n3aHo pa3y6ox~uaauu~ TeppHrenHbtMU oca~lKaMtt. CMeKTHT BO3MO)KHOorpa3yeTcn paanoMepHo na 6Oabtuefi cesepao~ qacTn TnxooKeaHcKoro rny6ogoBo:Iuoro :1Ha. 3naqen}te 80 ~a ~n~ CMeKTHTa +29,6%~, YTO npeano~araeT ero o6pa3oaanne ayTnrennTnqecKn npn TeMnepaType, xaparTepnofi ~ n rny6oKoao~moro aria. CMeKTHT npe~IcTaa~LaeT co6ofi 6oraTblfi Fe MOHTMOpH2121OI-IHT,KOTOpbII4BO3MO)KHOo6paayeTcn H3 UH31~OTeMuepaTypnhlX XHMHqeEKHX coeaaHeuafi r~Ipoorncaoa F e u gpeMae3eMa. KpeMueaeM abl~ennexcn B pe3y.rlhTaTe pacTaopenHa 6HoreueTnqecKHx oca~lKoa, a FrUIpOOKHCJlhlFe o6paay~oTca a pe3y3IbTaTe ay,gannqecgofi aKTHBHOCTH Ha BOCTOqHO-TuxooKeaHCKOMxpe6Te, 4000-5000 rM K BOCTOKy. A1 a ayTnreHnoM MOHTMOpH~OHHTe MOr o6pa3oaaTbCa s pe3y~bTaTe pacTaopen~t~ 60~bU~HX ~O~HqeCTB 6aoreanoro rpeMne3eMa H~H 8 pe3y~lhTaTe a~lcop6u:aH AI peqnoro npoHcxo;~cj~ennn,a~Icop6apoaaHHoro a ogeanax rn~lpOOKHC~aMa F e - Mn. Fe-MOHTMOpH2uIOHHT CO~lepTgnT OTnOCHTe.qhnO MnOrO Ca, Zn n Mn a BO3MO)KHO HMeeT 3KOHOMHqecKoe 3naqenne KaK HCTOqHHKS*THXH :lpyrnx MeTa~.rlOa. Resiim~ Tonmineralien in den oberen 50 cm yon Sediment, welches die Cu- und Ni-reichen ManganNeste im nrrdlichen/iquatorialen Pazifik umgibt, formen zwei Fraktionen: terrigenes (meist fiolisches) Illit, Chlorit und Kaolinit und authigenisches Smektit. Mit zunehmender Tiefe nimmt der Anteil des Smektit in Kernen yon 26 bis 39% und von 53 bis 66% in den am weitesten 6stlich gelegenen beziehungsweise den am weitesten westlich gelegenen Gegenden zu und nimmt mit seew/irtigem Abstand von Amerika von 26 bis 53% in Obertl~ichenablagerungen zu. Der Unterschied in der Menge des Smektiten im Vergleich zu anderen Tonmineralien ist auf Verdiinnung mit terrigenem Schutt zur/ickzufiihren; wahrscheinlich forint sich Smektit mit einheitlicher Geschwindigkeit fiber dem grrBten Tell der Tiefseesohle des Nord-Pazifik. Der 80 TM Weft f'fir die Smektite ist +29,6%o, was vorschl/igt, dab es authigen geformt wurde, bei einer Temperatur, welche charakteristisch fiir die Tiefseesohle ist. Das Smektit ist ein Fe-reiches Montmorillonit, dab sich wahrscheinlich dutch die chemische Reaktion yon 194 Hein, Yeh, and Alexander Clays and Clay Minerals Fe-Hydroxyden und Silika formt. Silika wird von der Aufl6sung von biogenischem Schutt abgeleitet und das Eisenhydroxyd kommt von vulkanischer Aktivit/it in der Ost-Pazifik-H6he, 4000 bis 5000 km 6stlich. A1 im authigenischen Montmorillonit k6nnte von der Aufl6sung von grol3en Mengen von biogenischem Silika herstammen oder von vom Flu8 abgeleitetem AI, welches auf Fe-Mn Hydroxyden im Ozean adsorbiert ist. Das Fe-Montmorillonit enth~ilt verh~iltnismS.13ig viel Cu, Zn und Mn und k6nnte m6glicherweise wirtschaftliche Bedeutung erhalten als eine Quelle for diese Metalle. R6sum6---Les min6raux argileux des 50 cm du dessus du s6diment entourant les nodules de manganese riches en Cu- et en Ni dans l'Oc6an Pacifique 6quatorial Nord forment 2 fractions: l'illite, la chlorite, et la kaolinite terrigineuses (surtout 6oliennes) et ia smectite authig6nique. La smectite dans des carottes augmente proportionellement h la profondeur de 26 ~t 39% et de 53 h 66% dans les r6gions le plus /~ l'est et les plus h l'ouest, respectivement, et elle augmente de 26 h 53% proportionellement /t la distance des Am6riques darts les d6p6ts de surface. Le changement dans la quantit6 de smectite relative aux autres min6raux argileux est dfi /~ la dilution par des d6bris terrigineux; la smectite est probablemerit form6e /~ une allure unfforme sur une grande partie du sol profond de l'Oc6an Pacifique Nord. La valeur 8 0 t8 pour la smectite est +29,6 per rail ce qui sugg~re qu'elle est form6e authig6niquement /~ une temp6rature caract6ristique du sol profond de l'oc6an. La smectite est une montmorillonite fiche en Fe qui est probablement form6e par la combinaison chimique /i basse temperature d'hydroxides F e e t de silice. La silice est d6riv6e de ia dissolution de d6bris biog6niques, et l'hydroxide Fe provient de l'activit6 volcanique ~ l'East Pacific Rise, de 4000 ~ 5000 km A 1'est. AI dans la montmorillonite authig6nique peut ~tre d6riv6 de la dissolution de grandes quantit6s de silice biog6nique ou d'A1 d6rive de rivi6res, adsorbe sur les hydroxides Fe-Mn dans les oc6ans. La montmorillonite-Fe contient assez bien de Cu, Zn, et Mn et est possiblement d'importance 6conomique en tant que source de ces m6taux et d'autres.