Mottl, M.J., Davis, E.E., Fisher, A.T., and Slack, J.F. (Eds.), 1994Proceedings of the Ocean Drilling Program, Scientific Results, Vol. 139

17. CHEMICAL COMPOSITION AND FORMATION OF A MASSIVE SULFIDE DEPOSIT,MIDDLE VALLEY, NORTHERN JUAN DE FUCA RIDGE (SITE 856)1

S. Krasnov,2 T. Stepanova,2 and M. Stepanov2

ABSTRACT

Studies of sulfide cores from Site 856, Middle Valley, northern Juan de Fuca Ridge, established the vertical zonation of a largeinactive oceanic massive sulfide deposit. Zone 1 consists mainly of pyritic massive sulfide with minor sphalerite and magnetite (0to 28 mbsf in Hole 856H and 0 to 40 mbsf in Hole 856G). Compared with the rest of the deposit, this zone is strongly enriched inZn and a large number of trace metals (in particular Cd, Mn, Sn, and Sb), and in elements contained in gangue minerals (Si, Al,Mg, and Ba). Most of these elements reach their maximum concentrations in the upper part of the zone; the lower part is enrichedin Cu, Se, and Ca. Zone 2 consists dominantly of pyrrhotite-rich massive sulfide with minor pyrite and magnetite (28 to 48 mbsfin Hole 856H) and is depleted in minor elements. Zone 3 is almost pure pyritic massive sulfide (48 to 75 mbsf in Hole 856H and40 to 65 mbsf in Hole 856G) and is relatively enriched in Ge, As, Sb, Pb, and Tl. Zone 4 consists of dominant pyrrhotite with minorchalcopyrite (75 to 95 mbsf in Hole 856H) and shows a maximum enrichment in Cu, Co, and Bi.

Correlation analysis shows that elements enriched in zone 1 are associated with different ore-forming and gangue minerals:sphalerite (Cd and Pb), talc and/or chlorite (Sc, Th, and REE), silica (Hg and Ta), dolomite (V), and barite (Sr, Zr, and Hf). Sb, As,Sn, Au, Ag, Mn, Mo, and Tl form a separate association, presumably related to sulfosalt minerals.

Enrichment of Zn and minor elements in the upper part of the deposit, which is typical of ancient massive sulfide deposits, isattributed to the zone-refining process of progressive upward replacement of lower temperature minerals by higher temperatureminerals during growth of the sulfide body, with the concomitant hydrothermal leaching of trace metals.

Later, the deposit was hydrothermally altered, probably due to lateral flow of evolved seawater along turbidite layers. Thealteration produced replacement of primary pyrrhotite by pyrite in zone 3 and by pyrite + magnetite assemblages in the upperzones. In the upper levels of the deposit, sulfate-rich water acted as a strong oxidant. At lower levels a hotter and more evolvedsulfur-enriched pyritizing solution also introduced additional Pb, Sb, As, and other sediment-derived elements into zone 3. Theabsence of light rare earth element enrichment and the existence of a positive Eu anomaly in the massive sulfides of this zone isconnected with the alteration of plagioclase in the associated turbidites.

The secondary convection system that caused alteration of the massive sulfides could have been driven by one of the sillsemplaced into Middle Valley sediments subsequent to sulfide formation.

INTRODUCTION

During Leg 139 of the Ocean Drilling Program (ODP) a large bodyof massive sulfides was drilled in the rift valley (Middle Valley) of theEndeavour Segment, northern Juan de Fuca Ridge. With the exceptionof several relatively shallow holes drilled during Leg 106 wi'Mn theSnake Pit Hydrothermal Field in the Atlantic (Honnorez et al., 1990),this was the first opportunity to core an oceanic sulfide deposit andstudy its inner structure.

Similar massive sulfides within sedimentary-volcanic sequences,genetically related to mafic volcanic rocks, are commonly referred toas Besshi-type deposits (Franklin et al., 1981). These deposits com-prise an important group of ancient volcanogenic massive sulfides,which have been actively studied and mined on land. Therefore,information on the structure of these modern equivalents is of primaryimportance for understanding the genesis of economically significantbase-metal deposits.

The aim of this paper is to describe the chemical compositions ofthe recovered massive sulfides, compare them to those of other oce-anic sulfide deposits, and discuss possible genetic implications.

GEOLOGIC SETTING

Hydrothermal vents and massive sulfide deposits of the northernJuan de Fuca Ridge are situated in Middle Valley, a 15-km-wide,sediment-filled abandoned rift valley. Heat-flow anomalies in excess

Mottl, M.J., Davis, E.E., Fisher, A.T., and Slack, J.F. (Eds.), 1994. Proc. ODP, Sci.Results, 139: College Station, TX (Ocean Drilling Program).

2 Institute for Geology and Mineral Resources of the Ocean, 1 Maklin Ave., 190121,St. Petersburg, Russia.

of 1 W/m2 were discovered in the eastern part of the valley between48°20'N and 48°30'N (Davis et al., 1987). Ahydrothermal origin hasbeen suggested for a number of mounds in Middle Valley, eachseveral hundred meters across and up to 60 m high, which were foundby SeaMARC-I acoustic imagery in this area (Davis et al., 1987). Ashallow core, taken at the base of one of the mounds during the 1985cruise of C.H.S. Parizeau, contained 2.4 m of sediments rich indisseminated sulfides and massive sulfide clasts (Goodfellow andBlaise, 1988).

Later photographic and video studies and subsequent samplingdocumented the presence of extensive massive sulfide outcrops alongthe margins of the mounds, which confirmed the large scale of thedeposits. Little hydrothermal venting was discovered in the area, andthe sulfide deposits are mostly relict. However, another area withnumerous high-temperature vents of the black and white smoker typewas found about 3 km to the northwest, closer to the axis of the valley(Kappel and Franklin, 1989).

During ODP Leg 139, eight holes were drilled at Site 856 in themound area, near where the Middle Valley sulfides were first sampledin 1985 (Shipboard Scientific Party, 1992). Two holes (856A and856B) were drilled into Bent Hill, a large mound typical of several inthe area. This mound appeared to be covered by more than 100 m ofundisturbed sediments that were uplifted by sill intrusion. In Hole856B, at the southern edge of Bent Hill, massive sulfide clasts wererecovered in the sediment column at a depth of about 30 m. Six moreholes were drilled at Site 856 (856C to 856H) within a sulfide moundthat is up to 100 m in diameter and 20 m high, lying immediately tothe south of Bent Hill. These holes are distributed over a distance of60 m across the southern flank of the sulfide mound, but none reachedthe base of the deposit. Drilling of the deepest hole (856H) wasterminated for technical reasons at 94 m below seafloor (mbsf), or

353

S. KRASNOV, T. STEPANOVA, M. STEPANOV

Table 1. Comparison between certified and measured element concentrations (%) in the standardsamples, used in atomic absorption analyses at VNIIoceangeologia.

Standard

SNK-1

#3031

#681

#710

RUS-3

#2891

ST-1

Fe

18.67 ±0.1319.09

4.01 ±0.004.07

10.15 ±0.2410.24

16.26 + 0.5416.38

Element concentrations, measured (top) and certified (bottom)Mn

0.173 ±0.0070.16

0.21 ±0.000.21

Cu

0.40 ± 0.000.39

3.38 ±0.083.37

12.62 ±0.1212.50

Zn

0.038 + 0.0010.039

10.40 ±0.1010.39

1.18 ±0.021.23

6.43 ± 0.076.55

Cd

0.0037 ± 0.00030.0035

0.0028 ± 0.00010.0029

Ca

20.0020.04

7.297.29

Mg

0.86 ± 0.0020.80

3.43 ± 0.073.44

Note: Certified concentrations are from Chemyakhovskaya and Osyko (1983). Measured concentrations are average valuesof two to three individual determinations.

2518 m below sea level (mbsl). This is much deeper than the level ofthe seafloor near the mound.

ANALYTICAL METHODS

All chemical analyses were carried out in St. Petersburg. SiO2 andA12O3 were analyzed by photometry at the Institute for Geology andMineral Resources of the Ocean (VNIIoceangeologia) by N. Luneva.The samples were decomposed by melting with sodium carbonateand borax. Silica was determined by the reduction product of yellowsilica molybdenum heteropolyacid complex, with ascorbic acid usedas a reducer. Aluminum oxide was determined using alumocreson, anammonia salt of dioxytricarbonic acid of triphenylmethane row. Sul-fur was determined gravimetrically, by means of precipitation withbarium chloride.

Fe, Mn, Cu, Cd, Ca, and Mg were also analyzed at VNIIocean-geologia by N. Luneva, using atomic absorption spectrometry (C-115spectrophotometer with flame atomization), with preliminary acidictreatment. Certified standards were used for control (Table 1).

Zn, Pb, Mo, and Ti were determined by the same method, by S.Shishkova and Ye. Chikhacheva at the Russian Geological Institute(VSEGEI). A Perkin Elmer M 305 spectrophotometer with flameatomization was used for the determination of Zn and Pb. A PerkinElmer M 603 with electrothermic atomization was used for Mo analy-ses, and an AAS-3 spectrophotometer with electrothermic atomiza-tion was used for Ti determination. Comparison between certified andmeasured concentrations in the standards, analyzed by atomic absorp-tion at VSEGEI, is given in Table 2.

Sr, Sn, Bi, V, Ge, and TI were analyzed by Ye. Khitric (VNII-oceangeologia) by quantitative optical emission spectrometry, usinga DFS-13 spectrograph with a flat grating (600 lines/mm). The work-ing range was 200-1000 nm, relative aperture 1:40, and linear disper-sion 0.4 nm/mm. Certified and measured element concentrations inthe standards used are given in Table 3.

A wide range of minor elements was determined by I. Shtangeeva(St. Petersburg State University) by instrumental neutron activationanalysis (INAA): Ag, Au, As, Sb, Co, Ni, Sc, Cr, Zr, Cs, Rb, Ba, Hf, Ta,Th, U, Se, Hg, La, Ce, Sm, Eu, Tb, and Yb. The samples and stand-

Table 2. Comparison between certified and measured ele-ment concentrations in the standard samples, used inatomic absorption analyses at VSEGEI.

Element concentrations, measured (top) and certified(bottom)

Standard Zn (%) Pb (ppm) Mo (ppm) Ti (ppm)

RUS-1

ST-2

SDPS-3

BM

SKR-3

SG-IA

3.39 ± 0.043.49

0.0112 ±0.00120.0112

253 ± 23250

12.3 ±0.713

10.7 ±0.713

436 ± 26430

Note: Certified concentrations are from Chemyakhovskaya and Osyko(1983). Measured concentrations are average values of two to fourindividual determinations.

ards were irradiated with epithermal neutrons at a flux density of 5 ×10 n/cm2s for two days. Measurements were carried out by means of aGe(Li) detector (resolution 2.1 keV for line 13332 keV). Results ofINAA for the standard samples are given in Table 4. The method ofinner standard (Vaganov, 1975) was used for Au, Ag, As, Hg, Ni, andSe determinations.

X-ray diffraction was conducted by M. Ostroumov (St. PetersburgMining Institute), using DRON-2 (V = 30 kV, I = 10 ma) and URS-50-IM (V = 15-20 kV, I = 3-4 ma) diffractometers. Electron micro-probe analyses were conducted by O. Yakovleva (VSEGEI), using aCameca MS-46 instrument. The detection limit was 0.01%, with therelative analytical error of 2% for major elements. The ZAF correc-tion method was used in data processing.

354

MASSIVE SULFIDE DEPOSIT, SITE 856

Table 3. Comparison between certified and measured element concen-trations (ppm) in the standard samples, used in quantitiative opticalemission spectrometry.

Element concentrations, measured (top) and certified (bottom)

Standard Bi Ge Sr Sn Tl V

SG-1A

ST-1A

SGD-1

1.7 ±0.4 3.5 ±0.1 <IOO 8.6 ±2.5 6.7+ 0.7 5± 0

2.0 3.3 7.0

268 ± 18

270

232+18

3.8 ±0.5

3.5

3.8 + 0.6

3.7

322 ±12

320

240 ±10

240

Note: Certified element concentrations are from Chernyakhovskaya and Osyko (1983).Measured concentrations are average values of three to four individual determina-tions.

MINERAL COMPOSITION AND STRUCTURE OFTHE MASSIVE SULFIDE DEPOSIT

Results of detailed ore petrology studies of the massive sulfidedeposits are given by Duckworth et al. (this volume). We studied andconsider here mainly the distribution of the major minerals, whichdirectly determine the chemical composition of the ores, across thedrilled section of the massive sulfide deposit.

Vertical changes in the mineralogy of the core from the deepest(95 mbsf) and most extensively sampled hole (856H; Fig. 1) aredescribed herein. The percentage of core recovery and quantity ofmassive sulfide samples do not allow us to describe these changes indetail. However, certain trends in mineral distribution can be tracedthat make it possible to divide the deposit into several zones.

Zonel

Massive sulfides from the upper part of the deposit, to about 28mbsf, are the most complex in composition. They consist of pyrite asthe dominant mineral, together with pyrrhotite, sphalerite, and magnet-ite in different proportions, and with minor quantities of marcasite,chalcopyrite, and isocubanite. Chlorite, talc, dolomite, and amorphoussilica are the main gangue minerals; barite occurs in minor amounts.The uppermost part of the deposit (Core 139-856H-1R) was destroyedin the course of drilling. The resultant disintegrated mass was sorted bygrain size during drilling, from fine-grained clayey sand in the upperpart to coarse-grained sand with gravel and separate clay lumps in thelower part. The sands contain clasts of massive, cavernous, partlyoxidized sulfides up to 4 cm in size. The 15-cm-long lowermost inter-val of Core 139-856H-1R consists almost totally of this massive sul-fide debris, which resembles clasts recovered from conventional blacksmoker chimneys (e.g., Krasnov et al., 1992).

The pyrite from Core 139-856H-1R typically forms round, rose-like aggregates, intergrown with gangue minerals. Traces of primarypyrrhotite, most of which is replaced by pyrite, occur locally. Subhe-dral sphalerite is interstitial and dispersed in the matrix of gangue min-erals. The sphalerite contains tiny chalcopyrite inclusions oriented inone direction. Chalcopyrite (locally with isocubanite) and pyrrhotitealso occur as isolated grains. Marcasite is intergrown with pyrite, replac-ing pyrrhotite. Rhombic barite crystals and crystal aggregates, anddark-brown grains of hydroxides, are also present. Talc and silicalocally replace the clayey sediment matrix.

The largest sulfide clasts are composed of tabular pyrrhotite crys-tals up to 0.1 mm in size, forming a framework with interstitialsphalerite. In some of the clasts pyrrhotite is totally replaced by pyrite(Fig. 2). Clasts are commonly semioxidized.

Table 4. Comparison between certified and meas-ured element concentrations (ppm) in the standardsamples AGV-1, used in neutron-activation analy-ses.

Element

BaCoCrCsHfRbSbSeTaThUZrLaCeSmEuTbYb

Certifiedconcentrations

1221.015.212.01.265.1

67.04.4

12.10.916.51.89

225.035.964.96.11.710.791.73

Measuredconcentrations

1240 ±6015.0 ±0.89.5 ± 0.5

1.50 ±0.085.4 ±0.365 + 35.0 ±0.2

13.4 ±0.70.90 ± 0.046.4 ± 0.3

1.75 ±0.09240 ± 1237.0 ±1.865.0 ±3.06.1 ±0.3

1.76 ±0.090.80 ± 0.041.75 ±0.09

Note: Certified concentrations are from Baedecker andMcKown (1970). Measured concentrations are averages offive individual determinations, using the BCR-1 standard.

The sulfides of zone 1 in Core 139-856H-2R and the followingcores are massive. Major minerals are pyrite, magnetite, and sphaleritewith a variable pyrrhotite content, and small amounts of chalcopyrite.Abundant gangue minerals (dolomite, local silica, and traces of talc)form veins and are characteristic of the zone. Two generations of pyritecan be recognized: anhedral pyrite that is intergrown with tiny, orientedmagnetite grains, and a subhedral variety that lacks magnetite inter-growths. Another form of magnetite occurrence is thin spongy inter-growths with gangue minerals. Sphalerite forms subhedral crystals thatvary in size from 0.05 to 1 mm and are partly dispersed in massive sul-fide and partly grouped in blebs. The largest grains occur in veins,together with pyrite and dolomite. Traces of pyrrhotite replacement bypyrite are evident in zone 1. The two pyrite generations reflect succes-sive stages of the replacement process. The paragenetically earlier gen-eration involved the pseudomorphic replacement of tabular crystals ofpyrrhotite by pyrite, intergrown with magnetite. Pyrite of the youngergeneration, associated with sphalerite, replaces both pyrrhotite relictsand early pyrite-magnetite aggregates. Only general outlines of thereplaced pyrrhotite crystals are seen in the resultant structure (Fig. 3).

The upper meter of Core 139-856H-4R (at about 27 mbsf) is asedimentary layer heavily veined with sulfide minerals and magnetite.The altered sediment consists of chlorite with minor talc and traces ofsiderite and dolomite rimming pyrite veins. Crystals of hexagonalpyrrhotite form sheaflike aggregates between pyrite crystals in veinsand are dispersed in the sediment, along with minor sphalerite andchalcopyrite. Sporadic large sphalerite crystals (up to 0.2 mm) occurin veins. The spongy magnetite is mostly intergrown with dolomite.

Zone 2

Massive sulfides of zone 2 (to about 48 mbsf) are characterized bythe dominance of pyrrhotite, which forms a framework of elongatedplates up to 3 mm in length. Based on X-ray diffraction data, pyrrhotiteis mostly hexagonal, and partly monoclinic. The crystals of pyrrhotiteare overgrown and partly replaced by pyrite (Fig. 4), the second mostabundant mineral. Pyrite also fills interstitial spaces, forming large (upto 1.5 mm) anhedral grains, and occurs in thin ramifying veins. Mar-casite is locally present together with pyrite, forming rims and over-growths on pyrrhotite. Anhedral sphalerite grains, up to 1 mm in size,fill the interstitial spaces. Small, mostly anhedral chalcopyrite grainsfill the interstices and form tiny (up to 0.1 mm) inclusions in pyrrhotiteand sphalerite.

355

S. KRASNOV, T. STEPANOVA, M. STEPANOV

Hole 139-856H

20 40 60 80Sample

Hole 139-856G

20 40 60 βo % Sample

1R-1, 146-147

2R-1, 67-69

6R-2, 101-103

• 6R-3, 95-97

• 7R-1, 38-40

• 7R-4, 67-69

2490

• 2495

• 2500

• 2505

h 17R-1, 35-36L 2515

Depth(mbsl)

Figure 1. Mineral composition (percentage of total ore minerals) and zonation of sulfide cores from Holes 856G and 856H. Py = pyrite, Po = pyrrhotite, Mt =magnetite, Sp = sphalerite, Cp = chalcopyrite, Ic = isocubanite, Me = marcasite, and Hm = hematite.

2425

2430

2435

2440

2445

2455

2460

2465

2470

2475

Magnetite occurrences in zone 2 are different from those in zone 1.The intergrowth of magnetite with pyrite is atypical, and more com-monly it forms subhedral crystals in pyrrhotite and spongy, mossyaggregates of skeletal crystals in cavities (Fig. 5). Gangue minerals,represented mostly by chlorite, with minor carbonates and local silicaand barite, are less abundant than in zone 1. The lower part of zone 2(Core 139-856H-7R) forms a transition to zone 3. Typical of this tran-sition is Sample 139-856H-7R-1,80-82 cm, where fine-grained pyriteforms boxwork textures including overgrowths and partial replace-ments of pyrrhotite. Magnetite and hematite are associated with thenewly formed pyrite, and occur in thin veinlets. Abundant fine anhe-dral grains of marcasite, intergrown with pyrite in its finest aggregates,are also characteristic of this core. Sphalerite and chalcopyrite are rare.Carbonate content is higher than that in the upper part of zone 2.

Zone 3

To about 75 mbsf, the core from Hole 856H consists of almostpure pyrite. Different types of pyrite aggregates in this zone form athick megascopic framework (on the scale of several millimeters),and cells are either void or filled with gangue minerals, mostly amor-phous silica and talc. Colloform, banded, and embayed pyrite texturesare common. Structures that display local pyrrhotite replacement canalso be seen, although pyrite is mainly recrystallized (Fig. 6). Smallquantities of sphalerite and chalcopyrite are dispersed in pyrite.

Strong enrichment in barite is seen in Sample 139-856H-8R-1,124—126 cm. Barite crystals are overgrown by euhedral pyrite andradial aggregates of hematite (Fig. 7). A talc aggregate, partly re-placed by pyrite and magnetite, occurs in the lower part of zone 3

356

MASSIVE SULFIDE DEPOSIT, SITE 856

0.2 mm

Figure 2. Boxwork texture of pyrite, growing over dissolved pyrrhotite crystals.Photomicrograph in reflected light, Sample 139-856H-16R-1, 108-110 cm.

0.1 mm

Figure 3. Outlines of tabular crystals of pyrrhotite, totally replaced by pyrite(Py), sphalerite (Sp), and magnetite (Mt). Photomicrograph in reflected light,Sample 139-856H-3R-2, 67-70 cm.

Figure 4. A large tabular crystal of pyrrhotite (Po), replaced and overgrown bypyrite (Py), with interstitial sphalerite (Sp). Photomicrograph in reflected light,Sample 139-856H-5R-1, 52-53 cm.

30 µm

Figure 5. SEM photomicrographs of magnetite aggregates from cavities inmassive sulfides. A. Mossy magnetite with a crystal of chalcopyrite (Cp),Sample 139-856H-5R-1, 52-53 cm; area in box is shown in (B). B. Skeletalmagnetite crystal, shown enlarged from (A). C. Skeletal magnetite, Sample139-856H-4R-1, 144-146 cm.

(Sample 139-856H-13R-1, 48-49 cm). Pyrite and magnetite cementthe sediment or form network structures within it.

Zone 4

Below 75 mbsf, homogeneous fine-grained pyrrhotite defines zone4 (Fig. 8). The pyrrhotite is mainly monoclinic, with traces of thehexagonal variety. Aggregates of anhedral grains and interpenetrating

357

S. KRASNOV, T. STEPANOVA, M. STEPANOV

* • r

%

0.1 mmFigure 6. Recrystallized pyrite (Pyl), with chalcopyrite (Cp) and sphalerite(Sp), overgrown by euhedral Pyrite (Py2). Photomicrograph in reflected light,Sample 139-856H-10R-1, 13-15 cm.

300 µmFigure 7. Overgrowth of a rhombic crystal of barite by cubic pyrite withhematite spherulites. SEM photomicrographs, Sample 139-856H-8R-1, 124-126 cm. The scale bar refers to the left, general view; box shows area of detailenlarged in right photomicrograph. Close-up (right) is enlarged 5.5x).

plates are most common. Pyrite forms rare euhedral crystals, up to 4mm in size. Small (0.1 to 0.3 mm) anhedral chalcopyrite grains are dis-persed in the massive sulfide. Magnetite is scarce in the ore matrix, al-though it occurs locally, together with pyrite, in thin (0.5 mm) veinlets.

At the bottom of Hole 856H, at about 90 mbsf, two pieces of brec-ciated sulfide were recovered. This sulfide differs in composition fromthat typical of zone 4, in that pyrrhotite is partly recrystallized andreplaced by pyrite containing fine-grained dispersed magnetite. Pyriteand minor marcasite also overgrow pyrrhotite. Chalcopyrite emulsiontextures occur in coarse-grained, recrystallized pyrrhotite. Aragonite,the main gangue mineral, is dispersed in the sulfide and forms, alongwith calcite, dolomite, and chlorite, thin ramifying veinlets.

Hole 856G

The massive sulfides recovered from the second deepest hole(Hole 856G, to 64 mbsf), differ mainly from those of Hole 856H bythe almost complete absence of pyrrhotite. Evidence of the formerexistence of pyrrhotite is clearest in the upper part of the core, wherethin boxwork textures of pyrite, overgrowing and partly replacingdissolved pyrrhotite crystals, can be seen. Two generations of pyrite

0.1 mmFigure 8. Fine-grained pyrrhotite aggregate. Photomicrograph in reflectedlight, Sample 139-856H-16R-1, 79-81 cm, under crossed nicols.

also are present, similar to those observed in Hole 856H (Fig. 9). Sul-fides from the upper part of Hole 856G differ from the correspondingdeposits of zone 1 in Hole 856H by containing skeletal magnetite crys-tals and small quantities (up to 5%) of hematite. Morphologically sim-ilar magnetite is present in zone 2 of Hole 856H, which is absent inHole 856G. Voids and pores between pyrite grains in the lower part ofHole 856G locally contain aggregates of spherulitic talc (Fig. 10).Sphalerite and magnetite are less evenly distributed in the upper partof Hole 856G than in zone 1 of Hole 856H.

In spite of these differences, massive sulfides in the upper part ofHole 856G to about 40 mbsf share the most characteristic feature ofzone 1 in Hole 856H, the presence of pyrite with tiny oriented magnet-ite intergrowths. This distinctive feature makes correlation between thetwo holes possible. Massive sulfides from the lower part of Hole 856Gtherefore clearly correlate with those of zone 3 in Hole 856H.

Holes 856C through 856F

Massive sulfides from the shallower Holes 856C, 856D, 856E,and 856F were completely destroyed during drilling. On the wholethey correspond to the upper part of zone 1 in Hole 856H. Two sulfidevarieties can be distinguished: silica-rich (more than 20% SiO2), andrelatively silica-poor. The silica-rich sulfides contain round, rose-shaped pyrite aggregates that are generally similar to those in thesands of Hole 856H, and are evenly dispersed in the siliceous matrix.The relatively silica-poor ores are similar to those in large clasts fromthe upper part of Hole 856H, where pyrite replaces tabular crystals ofpyrrhotite, forming boxwork textures.

Sulfide clasts, discovered in sediments from Hole 856B between18.4 and 24.2 mbsf, consist almost entirely of pyrite. Colloform,banded, and network structures are typical.

ANALYTICAL RESULTS

Chemical Composition of the Cores

Table 5 gives the results of bulk chemical analyses of sulfides fromSite 856, and Table 6 gives summary statistics, along with comparativedata from other locations. The two elements that dominate the compo-sition of the sulfide deposit, Fe and S, are distributed normally; otherelements are distributed log-normally. In plotting element distributionwith depth (Fig. 11), we have to consider drilling disturbances in theuppermost parts of the holes. The evident sorting of sulfides accordingto grain size along the length of cores may have influenced the distri-bution of elements. Therefore, in Figure 11 we plot average elementconcentrations between pairs of samples representing the bottom and

358

MASSIVE SULFIDE DEPOSIT, SITE 856

0.1 mmFigure 9. Two generations of pyrite (Py): early anhedral pyrite, totally replacingpyrrhotite, with tiny magnetite (Mt) intergrowths (gray) on the right, and latersubhedral pyrite, without magnetite, on the left. Photomicrograph in reflectedlight, Sample 139-856G-4R-1, 75-77 cm.

top parts of the disturbed cores, instead of data from each of thesesamples. Analytical values corresponding to the middle parts of thedisturbed cores are plotted individually for each sample. We presumethat this method of data presentation gives a more realistic trend ofelement distribution in massive sulfides that were disintegrated andsorted during the course of drilling.

The vertical profile of the deepest hole (856H) is characterized bya nonuniform distribution of most elements. Only several of the ele-ments show general trends throughout the cored section. Concentra-tions of Fe, Co, and Bi tend to increase downhole; Zn, Mn, Cd, Sn,Ta, and Mg increase uphole. Zn, Sn, and Ta concentrations increaseabruptly in the uppermost part of the deposit. All of the other elementsreveal a more complex behavior with depth in the sulfide deposit (Fig.11, Table 5).

Changes in elemental concentrations from zone to zone, com-pared to average values, are described as follows.

Zone 1 is distinguished, along with relatively low Fe values in thesulfides, by variably high concentrations of many elements. In mostsamples throughout the zone, Zn, Cd, Mn, Se, Sn, and Al contents arehigh. Some samples from the zone are rich in Si, Cr, Cs, and Ni, andseveral are significantly enriched in V and Sc.

On the basis of the geochemical data, we divided zone 1 into twoparts: the upper 1A and the lower IB (Fig. 1). Many elements havedistinctly different concentrations in the upper and lower parts of zone1. In the upper part, to approximately 18 mbsf, Fe values are the low-est of the entire hole. Cu content is also low. Concentrations of Zn, Al,Mg, Si, Au, Ba, Sb, Sn, Sr, Ta, Tl, La, Sm, and Tb are at a maximumhere. Ag, As, Mn, Mo, and Ni concentrations are highly variable, butin individual samples they also reach maximum values for the entirehole. The lower part of zone 1 is marked by increased concentrationsof Cu and Se and low Au, As, and Sb. The peak in Cu concentrationin the lower part of zone 1 is based on only three analyses (Fig. 11),but its existence is confirmed by the results of study aboard ship(Shipboard Scientific Party, 1992). Ca and Mg are enriched, due tothe abundance of dolomite.

Zone 2 is characterized by relatively high Fe concentrations. Cu islow, and Zn is enriched in one sample from the upper part of the zone.Also relatively high in most samples are Mn and Se, whereas Pb, Bi,Co, Ge, Sn, and U contents are low compared to the average valuesfor the deposit. Of the main gangue mineral elements, Al and Mgconcentrations are higher than those in lower zone 3 due to thepresence of chlorite.

Zone 3, which is dominated by pyrite, contains more S than anyother zone. Its most characteristic features are very high concentra-

30 µm

Figure 10. Talc spherulites from a cavity in massive sulfides. SEM photomi-crograph, Sample 139-856G-6R-3, 95-97 cm.

tions of Pb, As, Sb, Ge, and Tl relative to those in the adjacent zone 2and especially in zone 4. Concentrations of Pb and Ge are maxima forthe entire hole. Co content increases downward within zone 3. Aushows relatively high, though variable, concentrations compared toadjacent zones. All gangue mineral elements are low, except forenrichment of Si and Mg in the lowermost sample, which containssignificant amounts of talc, and of Ba in the uppermost sample. At thelower boundary of zone 3, Sb, As, Ge, and Tl concentrations de-crease abruptly.

Zone 4 is characterized by relatively low S and high Fe contents,owing to the dominance of pyrrhotite. Of the other elements, only Cu,Co, and Bi show high concentrations. As, Au, Sb, Ag, Hg, and Pbvalues are low. The concentrations of all gangue mineral elementsexcept Al are also low.

Not all features of the element distribution in Hole 856G arerepeated in Hole 856H. Although enrichment of the upper cores in Ba,As, Au, Mn, Sb, Se, Sn, and Ta relative to the rest of the cores isevident (Fig. 11), no enrichment in Zn, Cd, Si, and Al can be seen.This may be attributed to the aforementioned uneven distribution ofsphalerite, as well as that of the gangue minerals, in the upper part ofHole 856G. Element concentrations in the shallower Holes 856C,856D, 856E, and 856F also display maximum enrichment of theuppermost part of the deposit (corresponding to subzone 1A) in Znand most of the minor elements.

Based on the average data of Davis et al. (1987) and Goodfellowand Blaise (1988), the surficial part of the same deposit that wassampled before drilling by conventional methods (Table 6) corre-sponds in chemical composition to that shown by the uppermost partof the cores (subzone 1A).

REE Distribution

Figure 12 shows chondrite-normalized patterns for rare earth ele-ments (REE), averaged for each of the four compositional zonesrecognized in Hole 85 6H. For comparison, similar patterns for mas-sive sulfides from of the East Pacific Rise (EPR) and Mid-AtlanticRidge (MAR), based on previous data of the authors (Krasnov et al.,1992), are also shown. In the upper part of the sulfide deposit, in zones1 and 2, light rare earth elements (LREE) and intermediate REEs areenriched slightly over heavy rare earth elements (HREE). There areno Eu anomalies. The average Eu/Sm ratio is 0.95. Lower in Hole856H, in pyrite ores of zone 3, abundances of LREEs do not exceedthose of HREEs, and there is a well-pronounced positive Eu anomaly.The average Eu/Sm ratio is 1.72. Pyrrhotite-rich massive sulfide inthe lowermost part of Hole 856H (zone 4) also lacks a Eu anomaly.The Eu/Sm ratio is 0.78.

359

S. KRASNOV, T. STEPANOVA, M. STEPANOV

Table 5. Element concentrations in bulk sulfide samples from Site 856.

Core, section,interval (cm)

139-856B-3H-5, 110-113

139-856C-1H-1,48-50

139-856D-1H-1, 99-1021H-2, 64-671H-4, 62-641H-5, 113-116

139-856E-1H-1, 106-1081H-2, 84-86

139-856F-2X-CC, 1-32X-CC, 32-34

139-856G-1R-1,145-1471R-4,42-442R-1,67-692R-3, 29-312R-5,64-663R-1,61-6441^-1,49-514R-1,75-775R-1,47^96R-2, 101-1036R-3, 95-977R-1, 38-407R-4, 67-69

139-856H-1R-1, 108-110lR-2,4-71R-3,59-621R-CC, 3-52R-1,8-103R-1,34-373R-2, 67-703R-3, 14-164R-1, 18-204R-2, 144-1465R-1,52-536R-1,24-267R-1,80-828R-1, 124-1269R-1,80-8210R-1, 13-1511R-1, 49-5113R-1,48^914R-1,4_615R-1,60-6216R-1,79-8117R-1,35-36

Depth(mbsf)

18.40

0.48

0.992.145.127.13

1.062.34

11.7112.02

1.454.929.27

11.8915.2418.2127.4927.7537.1748.8150.2556.1860.77

1.081.543.593.96

13.5822.4423.9324.8526.7829.5432.9237.6443.9049.2453.3057.1361.7971.3875.7481.1085.9990.45

Description

Sulfide clasts from sediments

Sulfide clasts

Sulfide-silicate clastsSulfide-silicate clastsSulfide-silica clastsSulfide-silicate clasts

Sulfide clastsSulfide clasts

Sulfide-silica clastsSulfide clasts

Sulfide clastsSulfide sandSulfide-silicate clastsSulfide clastsSulfide sandMassive sulfideMassive sulfideMassive sulfideMassive sulfidePorous massive sulfidePorous massive sulfidePorous massive sulfidePorous massive sulfide

Sulfide-silicate clastsSulfide sandSulfide sandSulfide clastsMassive sulfideMassive sulfideMassive sulfideMassive sulfideSulfide in altered sedimentMassive sulfideMassive sulfideMassive sulfideMassive sulfidePorous massive sulfidePorous massive sulfidePorous massive sulfidePorous massive sulfideSulfide-talcose aggregateMassive sulfideMassive sulfideMassive sulfideSulfide breccia

Fe(%)

45.55

41.48

23.9823.6316.2824.32

32.2044.80

17.5034.30

43.4041.9319.7841.5023.8045.5542.7046.2036.4043.4044.8042.7043.40

31.7028.6029.5735.5229.4039.9039.2043.7538.5058.1049.3546.5538.8541.3045.5044.4544.8033.6052.1552.8555.3029.40

Zn(ppm)

0.76

5.60

4.053.533.113.32

17.401.11

1.091.15

0.100.300.92

0.501.530.290.061.080.04

0.08

0.652.312.04

10.004.622.911.513.210.770.544.280.270.120.420.060.250.080.040.410.230.090.04

Cu(ppm)

0.12

0.50

0.210.270.160.25

0.910.18

0.100.22

0.480.400.230.350.230.710.740.290.220.060.020.900.41

0.410.180.160.400.230.661.480.650.280.340.210.130.100.310.740.580.400.910.910.431.740.21

CaO(ppm)

0.13

0.10

0.090.110.100.11

0.150.15

0.201.15

0.090.090.220.900.250.350.250.350.200.200.200.200.10

0.660.640.410.104.204.504.251.253.600.200.500.255.550.250.100.100.100.200.100.750.135.00

MgO(ppm)

0.53

0.13

10.055.914.077.01

0.170.21

0.727.35

0.240.55

11.980.229.882.902.273.259.201.121.272.501.97

1.047.186.410.739.404.206.202.236.200.130.622.923.771.410.540.171.52

10.000.831.950.292.87

Mn(ppm)

70

460

330220120180

1000730

30330

680630530

1600

130

901705060

<20

2000350330

1400610150210150

30

23050

50n.a.1402090

<20190

S(%)

50.94

39.68

28.1327.8718.7027.69

37.6150.21

21.2028.20

50.6150.3523.8352.6328.9035.1140.5735.4035.0750.4951.8450.0949.11

28.8334.3037.2245.9024.5231.3628.1535.6628.6835.8538.5533.1439.8447.6852.5651.1452.9730.8135.1235.1837.4131.44

SiO2

(%)

1.12

3.80

24.0031.2047.8030.50

0.060.36

55.169.44

1.001.80

32.400.404.603.325.261.807.702.380.783.102.64

14.4018.9017.50

1.608.621.684.261.687.240.100.628.121.802.700.701.501.68

20.060.842.140.304.24

A12O3

(%)

0.16

0.62

0.360.74

<O.Ol0.26

0.04n.a.

0.100.14

0.420.44

<O.Ol0.380.260.260.060.140.040.120.040.160.16

20.00<O.Ol

0.260.580.360.161.080.641.880.160.16

0.120.020.050.080.040.020.08

0.121.24

Cd(ppm)

27

67

64543044

95434

3037

<IO1315

<IO19

<IO37

<IO<IO

30<IO

13<IO

46322771

441082417

12920

<IO17

<IO<IO<IO<IO<IO<IO<IO<IO

Au(ppb)

170

145

1608615

175

n.a.82

110108

44120140290260

2360351250564464

110156130250

81342

822

8113252

2436993622

71412

Ag(ppm)

5.0

2.0

2.03.0

<l.O<l.O

n.a.5.0

2.04.0

7.08.04.0

18.03.03.94.31.51.02.63.05.54.7

3.04.03.0

10.0<l.O

1.73.05.51.72.63.21.02.83.92.83.93.42.52.72.02.71.6

As(ppm)

180

33

4030

528

n.a.52

34<l

110

14035

1657

489

9355

60140

280507466<l

<l74

<l6

7412

160536794

s:2

<l<l

2

Sb(ppm)

13.0

4.0

17.013.02.03.2

n.a.6.8

3.60.6

13.016.012.011.013.0

1.73.20.80.30.50.71.41.6

13.07.88.6

13.00.70.60.70.40.70.10.55.50.55.53.93.63.74.10.20.20.20.3

Co(ppm)

6.0

2.1

4.07.53.68.0

n.a.1.0

3.29.0

14.022.010.03.45.89.28.6

24.06.62.13.8

83.023.0

20.05.46.55.73.78.9

54.024.014.084.07.5

14.013.088.040.056.0

130.0380.0210.0

60.0190.0100.0

Element Associations in Sulfides

Linear correlation analysis was used to study the relations amongelements in massive sulfide samples. The results are given in Table 7.Figure 13A shows the main positive correlations. For comparison, asimilar scheme is also given, based on geochemical data (Krasnov etal., 1992) for samples of massive sulfides from several sites at sedi-ment-free mid-oceanic ridges (Fig. 13B).

Massive sulfides from the studied cores show rather weak corre-lations among the elements. The main ore-forming metals have fewsignificant correlations. Fe correlates only with S (group 1, Fig. 13 A)and slightly with Cu. Cu forms a separate group (2) with two stronglycorrelated metals that enrich the pyrrhotite ores of the lowermost partof Hole 856H (zone 4): Co and Bi.

Zn shows the strongest correlations with Cd and Pb (group 3),typical of all oceanic massive sulfides. Zn, however, differs in itsbehavior in the sulfides at Site 856 relative to those of sediment-freeridges because it is not correlated with Ag nor with elements such asBa and As that are characteristic of low-temperature surficial faciesof oceanic sulfide bodies.

Ba and Sr tend to form their own small element group (4). Thisgroup is associated through the usual Ba-Sb correlation with the most

dense association of elements (group 5), including Sb, Sn, Mn, Ag,Au, As, Mo, and Tl. All of these elements are enriched in the surficialpart of the deposit. They are, however, neither directly related to zincsulfides nor to the main gangue minerals.

Three additional geochemical groups (6, 7, and 8) are associatedwith different gangue minerals that are also enriched in the upper partof the ore body. The Al-related group (6), including most of thelithophile elements, is the largest and most tightly associated throughthe Mn-Al correlation with group 5. Si (with Ta and Hg) and Ca (withV) form separate groups (7 and 8, respectively).

MINERAL CHEMISTRY

Microprobe Analyses

Results of microprobe analyses of minerals are shown in Tables 8through 10. Iron concentrations in sphalerites (Table 8) are mostly inthe range of 11 to 16 wt% Fe. The only sphalerite grain analyzed fromalmost pure pyritic massive sulfide of zone 3 (Sample 139-856G-6R-2, 101-103 cm) has a much lower Fe content (4.8 wt%) thansphalerites from the other zones. Among the analyzed pyrites (Table9), varieties from magnetite-bearing zones are mostly Fe-deficient incomparison to their theoretical compositions. Iron concentrations are

360

MASSIVE SULFIDE DEPOSIT, SITE 856

Table 5 (continued).

Core, section,interval (cm)

139-856B-3H-5, 110-113

139-856C-1H-1,48-50

139-856D-1H-1, 99-1021H-2, 64-671H-4, 62-641H-5, 113-116

139-856-1H-1, 106-1081H-2, 84-86

139-856F-2X-CC, 1-32X-CC, 32-34

139-856G-1R-1, 145-1471R-4,42-^42R-1,67-692R-3,29-312R-5, 64-663R-1,61-644R-1,49-514R-1,75-775R-1,47-496R-2, 101-1036R-3,95-977R-1,38-407R-4, 67-69

139-856H-1R-1, 108-1101R-2,4-71R-3, 59-621R-CC, 3-52R-l,8-103R-1,34-373R-2, 67-703R-3,14-164R-1, 18-204R-2, 144-1465R-1, 52-536R-1,24-267R-1, 80-828R-1, 124-1269R-1, 80-8210R-1, 13-1511R-1,49-5113R-1,48-^914R-l,4-615R-1, 60-6216R-1, 79-8117R-1,35-36

Ni(ppm)

27

10

66601020

n.a.10

40<6

302466

<610<613

24284317

110<632

3013

831737143317513317242532225112

Sc(ppm)

n.a.

0.82

0.540.190.170.32

n.a.0.17

0.370.52

0.20<O.IO0.270.16

<O.IO0.80

<O.IO0.300.500.501.000.200.20

3.100.180.130.530.800.301.901.101.900.900.700.60

<O.IO0.600.20

<O.IO0.400.300.400.600.601.20

Cr(ppm)

85.0

12.0

20.019.010.036.0

n.a.56.0

27.018.0

28.037.036.035.014.030.015.015.0<5.015.030.010.013.0

63.036.080.030.020.028.0<5.057.020.036.0

<s.o30.017.019.040.017.020.0<5.010.020.028.026.0

Zr(ppm)

<IOO

130

460

150330

n.a.150

150100

250430n.a.1002201904490210803409095

120260260150210320n.a.40013069120140

120270n.a.14014013050

100

Cs(ppm)

0.55

0.30

0.830.560.120.20

n.a.0.27

0.230.27

0.200.350.650.340.220.400.600.410.230.300.600.600.30

0.940.800.390.781.100.200.281.10

<O.IO0.300.200.500.250.301.000.580.200.56

<O.IO0.301.200.18

Ba(ppm)

290

250

100021000<50510

50260

120<50

220062006700180

770029026018026020512057270

29701000078005100550n.a.<50610210220300

2000270

16000120500350400140210210370

Hf(ppm)

0.23

0.48

1.602.600.520.35

n.a.0.37

1.400.83

1.100.86n.a.0.60n.a.0.900.700.501.500.802.200.601.30

0.900.902.901.101.402.200.921.401.50

<O.IO2.40

<O.IO0.900.90n.a.0.800.801.201.101.900.401.80

Ta(ppm)

<0.05

0.32

1.102.003.600.92

n.a.<0.05

0.070.16

0.621.500.971.000.74

<0.050.06

<0.050.12

<0.050.500.050.07

0.661.201.500.820.050.050.120.140.110.060.070.080.080.100.06

<0.050.120.070.070.08

<0.050.08

Th(ppm)

0.3

0.2

0.30.20.10.2

n.a.0.1

0.2<O. 1

0.20.2

<O.l0.20.20.40.20.30.30.70.90.5

<O.l

1.10.40.5

<O.l0.40.30.20.51.10.50.20.50.20.4

<O.l0.40.40.30.2

<O.l0.50.4

U(ppm)

2.6

3.0

10.019.01.35.0

n.a.23.0

2.74.5

2.320.030.07.0

26.08.23.52.03.81.21.63.56.5

4.38.25.06.04.02.02.05.52.04.02.59.02.53.82.43.010.04.82.18.09.06.0

Rb(ppm)

<IO.O

13.0

31.040.0

<IO.O<IO.O

n.a.25.0

30.014.0

20.0<IO.O<IO.O<IO.O22.016.010.0<5.022.0n.a.43.043.030.0

18.018.0

<IO.On.a.24.0

<IO.O28.025.047.020.033.018.025.020.044.013.0

22.045.038.023.0

Sr(ppm)

100

300

100

300200

100200

100200

<IOO300100200200200200100200200200

100

300500400300200200300200200100200200200440

<IOO200100100200

100300

La(ppm)

<0.5

3.1

2.91.73.62.4

n.a.0.5

3.01.5

1.33.43.01.22.72.0

<0.51.30.70.92.11.21.5

7.42.10.53.70.82.0

<0.51.60.9n.a.1.71.25.41.30.70.91.5

<0.50.9n.a.2.00.8

Ce(ppm)

n.a.

n.a.

3.70.80.63.0

n.a.1.7

5.8n.a.

5.9n.a.n.a.n.a.n.a.n.a.3.03.04.014.020.017.0n.a.

5.05.93.99.014.08.06.513.014.05.09.0

44.04.09.01.56.02.03.03.08.05.0n.a.

Sm(ppm)

0.62

0.25

3.402.700.451.30

n.a.n.a.

0.481.50

2.30n.a.n.a.0.82n.a.1.700.500.300.90n.a.n.a.0.901.70

3.600.511.302.500.900.500.791.700.700.200.70n.a.1.401.401.200.700.900.900.502.201.001.40

Eu(ppm)

0.23

0.14

0.970.680.450.15

n.a.0.13

0.570.22

0.370.210.870.060.440.400.400.200.900.600.701.200.60

0.860.240.100.440.200.300.200.600.300.300.300.800.300.600.800.800.700.400.500.300.400.30

Tb(ppm)

n.a.

0.26

0.130.900.200.10

n.a.<O.IO

0.23n.a.

n.a.0.200.360.100.120.45n.a.0.210.360.400.30n.a.0.25

0.670.140.580.300.18n.a.n.a.0.490.21n.a.0.12n.a.0.30n.a.n.a.0.200.150.400.25n.a.0.200.18

Yb(ppm)

0.27

1.00

0.551.100.430.38

n.a.0.10

0.17n.a.

0.13n.a.0.400.230.141.20n.a.0.500.702.403.40n.a.0.90

1.500.431.500.700.800.700.341.300.80n.a.n.a.1.000.401.700.500.900.400.300.501.600.801.00

Se(ppm)

5.0

160.0

27.038.0115.058.0

n.a.8.0

41.068.0

94.0n.a.38.018.042.039.030.083.017.03.03.017.02.0

58.050.048.019.030.091.0100.053.036.032.049.067.017*011.014.013.028.046.031.022.030.096.0

Hg(ppm)

<l.O

<l.O

12.0<l.O270.020.0

n.a.7.0

4.0<l.O

<l.On.a.13.0n.a.2.0n.a.6.014.049.0n.a.20.09.08.0

n.a.7.0n.a.18.0n.a.39.037.0n.a.14.028.026.0n.a.48.0n.a.29.0n.a.37.0n.a.6.010.0n.a.n.a.

Mo(ppm)

17.0

30.0

70.058.09.01.0

84.013.6

26.013.0

62.088.062.04.6

42.027.013.019.07.617.048.08.812.0

7.028.030.013.012.06.810.012.014.017.06.8

46.012.09.0

46.020.012.039.017.019.010.042.0

higher in the cores of pyrite grains than in the rims of the same grains.Cu-Fe sulfides are deficient in Cu (Table 10).

Neutron Activation Analyses

Table 11 presents the results of neutron activation analyses ofsulfide minerals hand-picked from samples of massive sulfide. Com-parative data on the chemical composition of minerals from oceanicsulfide deposits of other regions (EPR and MAR) are also included.It was possible to separate only pyrite without visible contaminationfrom most of the samples. We also managed to pick sphalerite andpyrrhotite out of one of the samples from zone 2 in Hole 856H.

Sphalerite is richer in minor elements relative to other minerals ofthe sulfide deposit. In Sample 139-856H-5R-1, 52-53 cm, fromwhich different minerals were analyzed, most of the elements reachtheir maximum concentrations in sphalerite. One exception is Co,which is preferentially enriched in pyrrhotite. This is consistent withCo enrichment in bulk samples of pyrrhotitic sulfide from zone 4 inHole 856H. Also enriched in pyrrhotite, compared to other minerals,are Cs, Se, Hg, and Yb. Pyrite is characterized by maximum concen-trations of U and Sm, relative to other sulfides.

DISCUSSION

Comparison of Chemical Composition with OtherSeafloor Sulfide Deposits

The average chemical data from Site 856 cannot be directly com-pared with data for other oceanic massive sulfide deposits that weresampled only from the surface. The sharp downward decrease inconcentrations of most elements in the Middle Valley cores is the mainreason that the averages for Zn, Cu, and most of the minor elements arelow (Table 6). Most of the gangue-related elements have the samerange of concentrations as those of other seafloor deposits. Mg and Caare higher than in most of the deposits, however. Guaymas Basin in theGulf of California, where massive sulfides also overlie sedimentarycover, is the only site for which the published data on these elementsare comparable to those for the cores from Middle Valley.

The absence of any significant enrichment of the Middle Valleydeposit in elements common to oceanic massive sulfides from sedi-mentary environments, such as Pb, As, and Sb, is especially notewor-thy. This absence of sediment-derived elements holds true even for thesurficial part of the deposit, where element concentrations increase.

361

S. KRASNOV, T. STEPANOVA, M. STEPANOV

Table 5 (continued).

Core, section,

interval (cm)

139-856B-

3H-5, 110-113

139-856C-

1H-1,48-50

139-856D-

1H-1, 99-102

1H-2, 64-67

1H-4, 62-64

1H-5, 113-116

139-856-

1H-1, 106-108

1H-2, 84-86

139-856F-

2X-CC, 1-3

2X-CC, 32-34

139-856G-

1R-1, 145-147

lR-4,42^4

2R-1,67-69

2R-3, 29-31

2R-5, 64-66

3R-1, 61-64

4R-1,49-51

4R-1,75-77

5R-1,47-49

6R-2, 101-103

6R-3,95-97

7R-1,38^H)

7R-4, 67-69

139-856H-

1R-1, 108-110

1R-2,4-7

1R-3, 59-62

1R-CC, 3-5

2R-l,8-10

3R-1,34-37

3R-2, 67-70

3R-3, 14-16

4R-1, 18-20

4R-2, 144-146

5R-1,52-53

6R-1,24-26

7R-1,80-82

8R-1, 124-126

9R-1,80-82

10R-1, 13-15

11R-1, 49-51

13R-1,48^9

14R-l,4-6

15R-1,60-62

16R-1, 79-81

17R-1, 35-36

Ti(ppm)

22

43

12

17

34

8

83

22

36

42

30

169<l1146

44

75

350

40

44

72

180

103

8

7

29

22

32

110

56

120

32

46

50

50

16

120

24

120

42

48

140

22

100

Pb(ppm)

120

534

132

74

34

60

98170

<IO54

136

21010020

118

66

92

61

<IO120

75

110

130

30620

154

3400

<IO37

12

25

<IO<IO<IO<IO26

110

130

100

58

100

29

58

<IO<IO

Sn(ppm)

20.0

5.4

53.0

46.0

14.0

7.8

100.0

27.0

15.0

5.6

48.0

28.0

22.0

45.0

26.0

12.0

23.0

6.0

5.0

5.8

3.0

5.06.0

26.0

20.0

14.0

60.0

31.0

15.0

15.0

10.0

9.1

4.0

4.0

5.55.06.5

5.9

6.0

8.3

10.0

7.6

5.55.75.0

Bi(ppm)

<0.5

0.9

1.2

1.1

1.1

2.0

<0.5<0.5

0.7

1.0

1.5

<0.51.2

<0.5

1.6

1.8

2.2

1.0

0.5

<0.5

<0.5

2.6

2.2

<0.5

1.9

<0.51.1

1.0

1.0

1.5

2.2

3.1

2.5

1.00.8

1.3

2.3

<0.5<0.5

6.2

12.0

12.0

5.5

9.510.0

Ge(ppm)

15.0

<l.O

4.0

4.0

<l.O

<l.O

9.0

9.0

1.0

2.5

1.0

<l.O

3.0

<l.O

2.03.4

<l.O

<l.O

<l.O

20.0

8.2

9.3

10.0

1.0

5.0

<l.O

5.0

6.3

<l.O

<l.O

<l.O

<l.O

<l.O

<l.O

<l.O

<l.O

16.0

12.0

11.0

14.0

9.0

<l.O

<l.O

<l.O

3.0

Tl

(ppm)

5.1

1.0

3.13.7

1.6

4.0

9.2

5.5

3.02.0

7.3

4.9

5.0

4.0

9.2

2.2

<l.O

<l.O

<l.O

7.9

14.0

4.5

4.4

10.0

6.1

6.7

8.8

<l.O

<l.O<l.O

<l.O

<l.O

<l.O

<l.O

2.8

4.3

5.5

7.4

6.7

5.5

<l.O

<l.O

<l.O

<l.O

<l.O

V(ppm)

<5

<5

<5<5

<5

<5

11

14

<5<5

<5

<5<5o<5

10

<5

<5

<5

<5

<5

o<5

<5

<5

<5

<5

<5

<5

18

11

14

<5

<5

<5<5<5

<5

<5

<5

<5

<5

12

<5

11

Li(ppm)

<30

<30

40<3030

40

<30

<30

<30

<30

<30<3040

<30

40

<30

<30<30<30<30<30<30<30

<30<30<30

<30

<3()

<30

<30

<30

<.Vi

<30

<30<30<30

<30

<30

<30

<30

<30

<30

<30

<30

<30

Note: n.a. = not analyzed.

The evident diversity in geochemical associations between thedrilled deposit and previously studied sulfide deposits of sediment-free ridge crests (Fig. 13) may be easily attributed to the change fromthe surface collection of samples to vertical coring. In oceanic sul-fides from sediment-free ridges (Fig. 13B), sampled by conventionalmethods, the main gangue-mineral elements do not form separategroups, but enter, along with Zn, a single large association corre-sponding to relatively low-temperature surficial parts of deposits.Thus, in the Middle Valley sulfides, elements can be divided intomore groups than in the sulfides of sediment-free ridges.

Ore aggregates belonging to the two main types commonly aresampled from the surficial parts of deposits in sediment-free environ-ments. Cu-enriched aggregates come from high-temperature chim-neys and from some central parts of mature sulfide mounds (Krasnovet al., 1992). In contrast, aggregates enriched in Zn and gangueminerals belong to moderate-temperature chimneys and to the periph-eral parts of large mounds, such as in the TAG active mound in theAtlantic (Lisitzin et al., 1990). In the Middle Valley deposit, sampledby deep coring, we found a greater variety in the types of sulfide andgangue-mineral aggregates, even in the upper part of the cored holes,which produces the observed diversity of geochemical associations.

REE patterns in the upper part of the Middle Valley deposit (Fig.12) on the whole resemble those of the MAR sulfides from hydrother-mal fields at 26°N and 24.5°N (Krasnov et al., 1992). In pyriticsulfides of zone 3 the REE pattern is more similar to those of depositsfromtheEPRatl3°N.

Iron concentrations in sphalerites from Site 856 (11 to 16 wt% Fe)are lower than the values reported by Koski et al. (1985) for sulfidedeposits of Guaymas Basin (15 to 24 wt% Fe). Iron concentrations insphalerites of massive sulfides sampled from the surface on sediment-free ridges are variable, even within individual deposits. At 44°N onthe Juan de Fuca Ridge, the Fe concentrations are between 1 and 17wt% in type A sulfides and <3 wt% in type B sulfides (Koski et al.,1984). At 13°N on the EPR, Fe concentrations in zinc sulfides varyover a wide range, from low values (2 to 6 mol% FeS) in surficialdendritic aggregates, to 35 mol% FeS in the Cu-enriched inner parts ofchimneys (Fouquet et al., 1988). At 21°N on the EPR, the iron contentsvary from 2 to 22 wt% Fe at different types of vents (Oudin, 1983).

Sphalerite is depleted in most minor elements relative to sphal-erites from other oceanic massive sulfide deposits (Table 11). Thistendency is most evident for Au, Ag, As, Sb, and Cr. It corresponds tothe general depletion of the cores in minor elements in comparisonwith other deposits, as observed in the bulk samples (Table 5). Theexceptions are U and Ni: their concentrations in sphalerite from theMiddle Valley cores are higher than in those from other sites. Thepyrites do not exhibit the same tendency, however, as their minorelement concentrations are mostly at the same levels shown by pyritesfrom other regions.

Copper concentrations in Cu-Fe sulfides (32.6 to 34.1 wt% Cu inchalcopyrites and about 23 wt% Cu in isocubanites) are comparableto those in the same minerals from 21°N on the EPR (Oudin, 1983),13°N on the EPR (Fouquet et al., 1988), and 44°N on the Juan de FucaRidge (Koski et al., 1984). Oudin (1983) reported copper concentra-tions of 31.6 to 33.4 wt% Cu in chalcopyrites and 21.39 to 22.95 wt%Cu in chalcopyrrhotites (isocubanites) from 21°N on the EPR.

Model of Massive Sulfide Formation

No significant amount of sediment was obtained in cores from thesulfide deposit at Site 856, although the core recovery at this site wasrelatively low (20% and 33% in the two deepest sulfide holes). Theabsence of sediment throughout the section was confirmed by logging(Shipboard Scientific Party, 1992). Thus, the sulfide body underinvestigation in this study, like all other sulfide deposits studiedpreviously in the ocean (e.g., Krasnov et al., 1992), must have formedabove the seafloor. In Middle Valley, the accumulation of Pleistocenesediments flanking the deposit was concurrent with its growth. How-ever, sulfide deposition could have been interrupted during relativelyshort episodes. While sulfide formation was interrupted, the surfaceof the deposit was covered with sediments, represented now by thin,highly altered interlayers.

Evidence of pyrrhotite replacement by pyrite, described herein,shows that pyrrhotite was the main primary mineral of the coredmassive sulfides. Duckworth et al. (this volume) arrived at the sameconclusion, based on detailed mineralogical investigations. The forma-tion of pyrrhotite instead of pyrite as the main sulfide mineral is typicalof oceanic sites of massive sulfide deposition, where solutions passthrough sedimentary cover in the course of convective circulation, asat Guaymas Basin in the Gulf of California (Koski et al., 1985), and atEscanaba Trough on the Gorda Ridge (Koski et al., 1988).

It is generally accepted that strongly reducing conditions, with lowfθ 2 and fS2, are needed for the crystallization of pyrrhotite in oceanichydrothermal systems (Koski et al., 1985,1988). In black smokers atsediment-free ridges, pyrrhotite appears in certain regions (e.g., at21°N on the EPR) to be the main constituent of the "smoke," as afine-grained precipitate formed by crystallization in the solutionswithin their sub-bottom pathways (Haymon and Kastner, 1981). Onsedimented ridges, the additional reduction of hydrothermal solutions

362

MASSIVE SULFIDE DEPOSIT, SITE 856

through solution-sediment interaction causes the deposition of largepyrrhotite masses.

The sedimentary pile is usually a sink for the main ore metals anda source for Ca in these environments. In Guaymas Basin, theserelationships result in a low percentage of precipitated sulfides, espe-cially Cu and Zn sulfides, in hydrothermal deposits (Koski et al.,1985). In other oceanic sites, as in hydrothermal deposits of GordaRidge and the Okinawa Trough, Pb, As, Sb, and Sn are extracted fromsediments and enriched in the massive sulfides (Table 6). Some ofthese metals may form discrete minerals in the sulfide deposits (Koskiet al., 1988; Halbach et al., 1989). The structures and textures ofpyrrhotite-rich massive sulfides of zone 4 in the lower part of Hole856H are similar to those previously described from Guaymas Basin(Koski et al., 1985), Gorda Ridge (Koski et al., 1988), and the surfi-cial part of the Middle Valley deposit (Goodfellow and Blaise, 1988).Therefore, the sulfides of zone 4 are interpreted to represent primaryaggregates, which, for some reason, escaped the later alteration thataffected the rest of the sulfide body.

The Middle Valley deposit differs greatly in its chemical composi-tion from deposits of some other sediment-covered spreading centers,such as Gorda Ridge and the Okinawa Trough (Table 6). At MiddleValley, the sulfide deposits contain only traces of Pb, As, and Sb, whichare believed to have been extracted from the sediments by hydro-thermal solutions. Concentrations of As and Pb in sulfide deposits ofGuaymas Basin are in the same range as those observed in the surficialpart of the Middle Valley deposit. It is possible that the temperatures ofthe hydrothermal solutions that formed the deposit under study werenever high enough in the shallow part of the convective system toextract significant amounts of metals from the sediments.

Prominent in the vertical element distribution of the drilled de-posit is the enrichment of its upper part (zone 1) in Zn, Si, Al, Ca, Mg,and a wide range of trace metals. Most of these elements (Ag, As, Au,Ba, Mn, Mo, Ni, Sb, Sn, Sr, Ta, Tl, La, Sm, and Tb) reach theirmaximum values at the top of the deposit, corresponding to subzone1A of Hole 856H. In contrast, a peak of Cu and Se enrichment ispresent in the lower part of zone 1 (Fig. 11).

Zn enrichment, characteristic of the uppermost parts of ancientmassive sulfide deposits on land, is generally attributed to the zonerefining process (Franklin et al., 1981; Eldridge et al., 1983). The innerstructure of mature deposits is determined by the progressive upwardreplacement of lower temperature minerals, including Zn sulfides, byhigh-temperature Cu and Fe sulfides during growth of the sulfidebodies. In typical ancient deposits, Cu is enriched in the intermediateparts below zones of Zn enrichment; the central parts consist of thehighest-temperature, coarse-grained Fe sulfides. Edmond (1980) wasthe first to discuss the zone refining process in connection with modernoceanic massive sulfide ore formation. Hannington et al. (1986) attrib-uted gold enrichment in the surficial parts of large deposits on theMid-Atlantic Ridge to the same process.

Zone refining is the most probable origin for Zn-enriched subzone1A and Cu-enriched subzone IB of the Middle Valley deposit. Thesecond zone of Cu enrichment, in the lower part of Hole 856H (Fig.11), may be the result of a decline of hydrothermal activity that led tocrystallization of lower-temperature chalcopyrite in interstitial spacesof the earlier-formed, highest-temperature Fe sulfides.

The vertical distribution of trace elements in pyrites (Table 11) isbasically similar to that in the bulk sulfides. Thus, the zone refiningprocess was not confined to the low-temperature minerals, whichcontain most of the trace elements, from lower to upper levels of thedeposit. Iron sulfides were also recrystallized, with the trace metalsleached from them at lower levels.

Although sphalerite shows maximum enrichment in the majorityof minor elements (Table 11), it is not the only (and for some ele-ments, even the main) host. Considering the dominance of iron sul-fides in the deposit, iron sulfides should be significant hosts for mostof the minor elements. However, the probable existence of small

inclusions in the mineral grains limits the use of INAA data to definethe actual mineral phases that contain the minor elements.

Almost all of the geochemical correlation groups, shown on Fig-ure 13A, include elements (Fe, Cu, Zn, Ba, Al, Si, and Ca) that evi-dently form the host minerals for minor elements of the same groups.Group 5, which contains the largest number of elements, including Sband As, presumably is associated with sulfosalts. Hannington et al.(1986) discovered the important role of sulfosalt minerals in concen-trating certain minor elements, including Au, in oceanic massivesulfide deposits.

Secondary Alteration of the Deposit

The main features of the distribution of metals in Hole 856H arerepeated in Hole 856G, except for the enrichment of Zn and associ-ated metals in the uppermost part (Fig. 11). The apparent differencesbetween the two holes in the extent of replacement of primary pyr-rhotite by pyrite did not influence the metal zoning. The verticaldistribution of metals was determined by zone refining that took placeduring formation of the sulfide deposit, mostly in its later stages. Thepervasive processes of pyrrhotite replacement occurred later, whenthe deposit had already formed, and did not affect the lowest part ofthe deposit that was reached by drilling (zone 4 in Hole 856H), andonly partly affected zone 2.

In zone 3 (Holes 856G and 856H), pyrrhotite was replaced byalmost pure pyrite. Such a replacement process requires an additionalsource of sulfur, and reflects more oxidizing conditions compared tothose of primary pyrrhotite formation. In the overlying magnetite-bearing zones, oxidation was much more intense.

A positive correlation is evident between pyrite and magnetiteconcentrations in magnetite-bearing zones from the studied cores(Shipboard Scientific Party, 1992). These two minerals presumablyformed as a result of a single process, when sulfur and oxygen wereadded to the system. Additional sulfur was fixed in pyrite, while oxy-gen and excess iron gave rise to formation of magnetite.

According to Duckworth et al. (this volume), sulfur is isotopi-cally heavier in pyrite from Middle Valley than in pyrrhotite be-cause of the involvement of seawater sulfate in secondary pyriteformation. The talc found in massive sulfide at different levels (fromzone 3 and upward) is considered in similar geologic environmentsas evidence for the involvement of seawater as a source for the Mg(Koski et al., 1985).

Sphalerite redeposition during secondary alteration may explainthe uniform iron concentration (11 to 16 wt% Fe) in the analyzedsphalerites (Table 7). This contrasts strongly with the widely varyingFe concentrations in Zn sulfides from the surficial parts of most otheroceanic sulfide deposits (see above). The low Fe concentration in thesample from zone 3 may indicate higher S2 fugacities of the solutionsduring alteration of this zone (e.g., Scott and Kissin, 1973).

Only preliminary assumptions can be made concerning the geom-etry of the secondary hydrothermal convection. Below we discuss thevariety of indirect evidence that favors a shallow, sill-driven, second-ary convective system.

The small but distinct, positive anomaly of sediment-derived met-als (Pb, Sb, As, Ge, and Tl) in the sulfides of the most strongly alteredzone 3 is significant (Fig. 11). Considering the mineral chemistry andelement correlations, this anomaly cannot be attributed simply to theassociation of these metals with pyrite, particularly as pyritic subzoneIB lacks signs of their enrichment. Enrichment of zone 3 in sediment-derived metals during secondary alteration is therefore probable.

Zone 3 is also the only part of the sulfide deposit that shows apositive Eu anomaly and no dominance of LREEs over HREEs (Fig.12). Europium is the only rare earth element subject to reduction tothe 2+ form. In this valence it preferentially fractionates into plagio-clase during magmatic processes, replacing Ca. Plagioclase pheno-crysts that form in magmatic systems also preferentially concentrate

363

S. KRASNOV, T. STEPANOVA, M. STEPANOV

Depth(mbsf)

Fe

MgO

Depth

2 ^ 3

Depthfmbβl) - *

10 15% 0J I 0

2000 ppm o so

Depth(mbsf) - I

Depth N .(mbsf)

Mn

Depth

SiO2

Figure 11. Element distribution vs. depth for massive sulfide samples from Site 856. Solid circles connected by heavy lines correspond to samples from Hole 856H,open circles connected with thin lines are from Hole 856G, right-pointing triangles are from Hole 856B, left-pointing triangles from Hole 856C, asterisks fromHole 856D, squares from Hole 856E, and crosses from Hole 856F.

heavy REEs (Graf, 1977). Thus, the supply of additional Eu andHREEs into hydrothermal solutions, concentrated in the massive sul-fides, is usually explained by alteration of plagioclase.

Hole 856B, drilled in sediments immediately adjacent to the sul-fide deposit, shows intense plagioclase alteration at a depth of morethan 70 mbsf (Shipboard Scientific Party, 1992). The high Ca concen-tration in pore waters in this hole also supports the existence of thisprocess. Thus, plagioclase in the sediments flanking the sulfide de-

posit is the most probable source for the additional Eu and HREEenrichment noted in our massive sulfide samples.

It is difficult to determine whether the two types of sulfide altera-tion, namely in zone 3 and in the upper part of the deposit, resultedfrom a single process. At elevated temperatures, heated seawaterevolved to form a hydrothermal solution due to interaction with sedi-ments. Most probably, this solution entered the sulfide deposit at dif-ferent stratigraphic levels, passing through permeable turbidite lay-

364

MASSIVE SULFIDE DEPOSIT, SITE 856

Depth

Depth

Figure 11 (continued).

ers. Such layers readily channelize solution flow in the Middle Valleyhydrothermal systems (ODP Leg 139 Scientific Drilling Party, 1992).Notably, the negative sulfate anomaly at 78 mbsf in Hole 856B, whichcorresponds to the base of zone 3 in Hole 856H, may be evidence forlateral flow (Shipboard Scientific Party, 1992). The ascending flow ofthese secondary solutions passed through the pores and cracks in thesulfide deposit.

At higher levels, under mild geothermal conditions, the seawaterthat entered the sulfide deposit must have been less geochemicallyevolved. It probably retained part of its original sulfate, which causedstrong oxidation of the sulfides. The more evolved solutions of thedeeper, higher-temperature part of the sedimentary column are re-sponsible for the alteration seen in zone 3. In these latter solutions, the

retained seawater sulfur was reduced to S from interaction with thesediments. The solutions also are likely to have retained some sea-water Mg, which was fixed in the talc of zone 3 and in the upper partof the deposit.

The scale of the alteration suggests that secondary convection mayhave been driven by the intrusion of one of the sills common in thesedimentary sequence at Site 856. Sill intrusions, related to the for-mation of Bent Hill, occurred later than the formation of the sulfidedeposit (Stakes and Franklin, this volume). In Hole 856B there is nosill at a depth corresponding to the base of sulfide reworking in Hole856H. Yet this fact does not contradict the model, and the absence ofa correlation of sills between Holes 856A and 856B (ShipboardScientific Party, 1992) still permits a relatively small lateral continu-

365

S. KRASNOV, T. STEPANOVA, M. STEPANOV

Lg

1.0 •

0.8

0.6 -

0.4 •

0.2 -

SampleChondritβ

La Ce Sm Eu

0.8

0.6 -I

0.4

0.2

0

Zone 2

La Ce Sm Eu Tb Yb

1.0 •

0.8 -

0.6 -

0.4 -

0.2 -

0

La Ce Sm Eu Tb Yb

0 8 -

0.6 -

0.4 -

0.2 •

Zone 4

La Ce Sm Eu Tb Yb

1.4 -

1.2-

1.0-

0.8 -

0.6-

0.4 •

0.2-

'' \ MAR

N

s Av / \

** v~»/ / \\13'N EPR / A

×^ -*——*— _̂ / A

La Ce Pr Nd Sm Eu Tb Yb

Figure 12. Average chondrite-normalized REE patterns concentrations in bulkmassive sulfide samples from four different zones of Hole 856H, and from theMid-Atlantic Ridge (MAR) and East Pacific Rise (EPR). Chondrite valuesfrom Taylor and McLennan (1985).

ity of the sills in the subsurface. A sill intrusion could heat secondaryhydrothermal solutions in its vicinity to higher temperatures than anydistant magmatic source. This would explain the later mobility ofsediment-derived metals that remained relatively immobile duringthe earlier formation of the main part of the sulfide deposit.

Secondary hydrothermal systems are also associated with the sillsthat penetrate the sedimentary cover in Guaymas Basin, Gulf ofCalifornia (Gieskes et al., 1982). These sill intrusions caused expul-sion of the pore waters that formed special types of hydrothermalprecipitates on the seafloor, such as talc-sulfide deposits (Lonsdale etal., 1980). The formation of larger carbonate-sulfate-sulfide chim-neys associated with sill-derived mounds near the axis of GuaymasBasin has also been attributed by some authors to secondary sill-driven hydrothermal systems (Lisitzin et al., 1990). Although therelative proportions of pore water and seawater participation in sec-ondary hydrothermal activity may be different, on the whole webelieve that the Guaymas Basin model is applicable to Middle Valley.

A schematic reconstruction of the cored massive sulfide deposit isshown in Figure 14. It presents the formation of the deposit by anupflow of hydrothermal solutions, genetically related to a relativelydeep-seated magmatic body, and its subsequent partial alteration bylateral flow of secondary solutions, related to a sill. In reality theprocess was undoubtedly not that simple, and the differences betweenthe cores from Holes 856G and 856H suggest a complex three-dimensional geometry of secondary alteration of the deposit.

SUMMARY AND CONCLUSIONS

Primary pyrrhotite in a large, inactive massive sulfide deposit,drilled at Site 856 in Middle Valley, formed above the sedimentarycover, probably from hydrothermal solutions with low fθ 2 and fS2,which are typical of sedimented ridges. The sediments, however, didnot contribute significant quantities of Pb, Sb, and As to the hydro-thermal solutions or to sulfide minerals, as at hydrothermal systemson Gorda Ridge and the Okinawa Trough. One possible reason forthis lack of sedimentary contribution is that the temperature of thesolutions in the upper part of the Middle Valley hydrothermal systemwas too low to extract metals from the sediments.

In terms of ore petrology, the drilled deposit has a complex innerstructure. The correlation of sulfide facies between the two drilleddeep holes, 856G and 856H, is not always possible. Essentially, Znand almost all of the trace metals are scarce in the inner part of thesulfide deposit. Zn and trace metal concentrations increase in theouter part of the deposit (to about 28 mbsf in Hole 856H), as a resultof zone refining.

The zone refining process of progressively upward replacement ofcomparatively low-temperature mineral phases by higher temperatureminerals took place during later stages of the growth of the deposit.This process led to the mobilization of trace elements contained withinthe low-temperature phases. Apart from maximum Zn and trace metalenrichment in the surficial part of the deposit, the zone refining causedCu enrichment in the subsurface (18 to 28 mbsf in Hole 856H). Similarvertical variations in Zn and Cu concentrations are typical of ancientmassive sulfide deposits on land. Copper enrichment in the deepestdrilled part of the deposit most likely formed during the waning stageof high-temperature hydrothermal activity.

Two types of alteration of primary pyrrhotite were connected withan intense post-depositional hydrothermal reworking process: (1) al-most complete pyritization of pyrrhotite in the lower central part of thecored section (48 to 75 mbsf in Hole 856H) and (2) replacement ofpyrrhotite by pyrite-magnetite aggregates in the upper part. Primarypyrrhotite is only preserved in the lowest drilled part of the deposit.

The presence of talc in secondary reworked sulfides shows thatseawater, evolved to different extent, participated in both types ofalteration. Seawater sulfate took part in the strongly oxidizing altera-tion in the upper part of the deposit. Below, under higher temperature

366

MASSIVE SULFIDE DEPOSIT, SITE 856

conditions, more geochemically evolved seawater probably con-tained sulfur mainly in reduced form that caused pyritization.

The secondary convection systems could have been driven by heatfrom one of the sills that intruded the upper part of the sedimentarycolumn following the main stage of sulfide formation. Evolved seawa-ter entered the deposit laterally at different levels, possibly throughpermeable turbidite layers. A small but definite enrichment of second-ary pyritic sulfide in sediment-derived elements such as Pb, Sb, and As,formed by lateral flow in the vicinity of the sill. The temperature herewas sufficiently high for extracting metals from the sediments on theflanks of the sulfide deposit. The absence of LREE enrichment and theexistence of a positive Eu anomaly in the pyritic sulfide may be due torelease of additional Eu and HREE into secondary solutions fromhydrothermally decomposed plagioclase in the turbidites.

ACKNOWLEDGMENTS

The German Agency for Academic Exchange (DAAD) funded theresearch stay of the first two authors at Marburg University, duringwhich polished sections were prepared, described, and photographed,along with statistical processing of geochemical data. We are mostgrateful to Prof. Werner Tufar and the staff of his laboratory inMarburg for organizing this visit and for their constant support of ourwork. We also would like to thank Dr. Andreas Schaper for taking theSEM photos, and Drs. Ye. Yel'yanova and Ye. Baranov for helpfuldiscussions. We appreciate the participation of O. Khaleneva, whodrafted the figures.

REFERENCES*

Baedecker, P.A., and McKown, D.M., 1987. Instrumental neutron-activationanalysis of geochemical samples. U.S. Geol. Surv. Bull, 177O:H1-H14.

Chernyakhovskaya, V.Z., and Osyko, Y.P., 1983. Catalogue of Standard RockSamples: Moscow (VIMS), (in Russian)

Davis, E.E., Goodfellow, W.D., Bornhold, B.D., Adshead, J., Blaise, B.,Villinger, H., and Le Cheminant, G.M., 1987. Massive sulfides in asedimented rift valley, northern Juan de Fuca Ridge. Earth Planet. Sci.Lett., 82:49-61.

Edmond, J.M., 1980. Ridge crest hot springs: the story so far. Eos, 61:129-132.Eldridge, C.S., Barton, P.B., Jr., and Ohmoto, H., 1983. Mineral textures and

their bearing on formation of the Kuroko ore bodies. In Ohmoto, H., andSkinner, B.J. (Eds.), The Kuroko and Related Volcanogenic MassiveSulfide Deposits. Econ. Geol. Monogr., 5:241-281.

Fouquet, Y., Auclair, G., Cambon, P., and Etoubleau, J., 1988. Geologicalsetting and mineralogical and geochemical investigations on sulfide de-posits near 13°N on the East Pacific Rise. Mar. Geol., 84:143-178.

Franklin, J.M., Lydon, J.W., and Sangster, D.F., 1981. Volcanic-associatedmassive sulfide deposits. Econ. Geol., 75th Anniver. Vol., 485-627.

Gieskes, J.M., Kastner, M., Einsele, G., Kelts, K., and Niemitz, J., 1982.Hydrothermal activity in the Guaymas Basin, Gulf of California: a synthe-sis. In Curray, J.R., Moore, D.G., et al., Init. Repts. DSDP, 64: Washington(U.S. Govt. Printing Office), 1159-1167.

Goodfellow, W.D., andBlaise,B., 1988. Sulphide formation and hydrothermalalteration of hemipelagic sediment in Middle Valley, northern Juan de FucaRidge. Can. Mineral., 26:675-696.

Graf, J.L., 1977. Rare earth elements as hydrothermal tracers during theformation of massive sulfide deposits in volcanic rocks. Econ. Geol,72:527-548.

Halbach, P., Nakamura, K., Washner, M., Lange, J., Sakai, H., Kaselitz, L.,Hansen, R.D., Yamano, M., Post, J., Prause, B., Seifert, R., Michaelis, W.,Teichmann, F, Kinoshita, M., Marten, A., Ishibashi, J., Czerwinski, S., and

Blum, N., 1989. Probable modern analogue of Kuroko-type massivesulfide deposits in the Okinawa Trough back-arc basin. Nature, 338:496-499.

Hannington, M.D., Peter, J.M., and Scott, S.D., 1986. Gold in sea-floorpolymetallic sulfide deposits. Econ. Geol, 81:1867-1883.

Haymon, R.M., and Kastner, M., 1981. Hot spring deposits on the East PacificRise at 21°N: preliminary description of mineralogy and genesis. EarthPlanet. Sci. Lett., 53:363-381.

Honnorez, J., Mevel, C , and Honnorez-Guerstein, B.M., 1990. Mineralogyand chemistry of sulfide deposits drilled from hydrothermal mound of theSnake Pit active field, MAR. In Detrick, R., Honnorez, J., Bryan, W.B.,Juteau, T., et al. Proc. ODP, Sci. Results, 106/109: College Station, TX(Ocean Drilling Program), 145-162.

Kappel, E.S., and Franklin, J.M., 1989. Relationships between geologic devel-opment of ridge crests and sulfide deposits in the northeast Pacific Ocean.Econ. Geol, 84:485-505.

Koski, R.A., Clague, D.A., and Oudin, E., 1984. Mineralogy and chemistry ofmassive sulfide deposits from the Juan de Fuca Ridge. Geol. Soc. Am. Bull,95:930-945.

Koski, R.A., Lonsdale, P.F., Shanks, W.C., Berndt, M.E., and Howe, S.S.,1985. Mineralogy and geochemistry of a sediment hosted hydrothermalsulfide deposit from the southern trough of Guaymas basin, Gulf ofCalifornia. J. Geophys. Res., 90:6695-6707.

Koski, R.A., Shanks, W.C., III, Bohrson, W.A., and Oscarson, R.L., 1988. Thecomposition of massive sulfide deposits from the sediment-covered floorof Escanaba Trough, Gorda Ridge: implications for depositional processes.Can. Mineral, 26:655-673.

Krasnov, S.G., Cherkashev, G.A., Ainemer, A.I., Grintal, E.F., Grichuk, D.V.,Davydov, M.P., Poroshina, I.M., Stepanova, TV, Sudarikov, S.M., Bo-chek, L.I., Datzenko, V.A., Dubinin, Y.P., YeFyanova, Ye.A., Zairi, N.M.,Kolosov, O.V., Mironov, Yu.V, Popov, V.Y., Andreeva, I.A., Vaganov, P.A.,German, N.Y., Gurevich, N.I., Kreyter, I.I., and Maslov, M.N., 1992.Hydrothermal Sulfide Ores and Metalliferous Sediments of the Ocean: St.Petersburg (Nedra). (in Russian)

Lisitzin, A.P., Bogdanov, Y.A., and Gurvich, Y.G., 1990. Hydrothermal De-posits of Oceanic Rift Zones: Moscow (Nauka). (in Russian)

Lonsdale, P.F., Bishoff, J.L., Burns, V.M., Kastner, M., and Sweeney, R.E.,1980. A high-temperature hydrothermal deposit on the seabed at a Gulf ofCalifornia spreading center. Earth Planet. Sci. Lett., 49:8-20.

ODP Leg 139 Scientific Drilling Party, 1992. Hot rocks and massive sulfide:northern Juan de Fuca Ridge. Eos, 73:193-198.

Oudin, E., 1983. Hydrothermal sulfide deposits of the East Pacific Rise (21°N).Part 1: descriptive mineralogy. Mar. Min., 4:39-72.

Scott, S.D., and Kissin, J., 1973. Sphalerite composition in the Zn-Fe-S systembelow 300°. Econ. Geol, 68:475^179.

Shipboard Scientific Party, 1992. Site 856. In Davis, E.E., Mottl, M.J., Fisher,A.T., et al., Proc. ODP, Init. Repts., 139: College Station, TX (OceanDrilling Program), 161-281.

Taylor, S.R., and McLennan, S.M., 1985. The Continental Crust: Its Compo-sition and Evolution: Oxford (Blackwell Scientific).

Vaganov, P.A., 1975. On the possibility of using a monostandard in neutron-activation analysis of rock samples. Sci. Notes Leningr. State Univ., Ser.Phys. Geol. Sci., 25:125-127. (in Russian)

* Abbreviations for names of organizations and publications in ODP reference lists followthe style given in Chemical Abstracts Service Source Index (published by AmericanChemical Society).

Date of initial receipt: 1 December 1992Date of acceptance: 30 August 1993Ms 139SR-230

367

S. KRASNOV, T. STEPANOVA, M. STEPANOV

Table 6. Comparison between bulk chemical composition of oceanic massive sulfides from Middle Valley and other locations.

Location

Mid-Atlantic Ridge26°N24°30'N

East Pacific Rise13°N21°N9°40'N22°S

Juan de FucaAxial SeamountCleft Segment

Explorer RidgeGalapagos RidgeGuaymas BasinGorda RidgeOkinawa Trougha

Middle ValleySurface samplesb

Cores (this study)AverageMedianStd. dev.

Fe

30.2517.55

26.1323.1233.0228.68

5.0324.7421.9130.3735.6136.79

8.59

30.30

39.5142.08

9.67

Cu

1.6216.25

1.910.613.591.30

0.610.0564.834.940.551.091.40

0.40

0.430.320.36

Zn

1.294.06

2.365.931.992.80

28.8411.48

1.651.171.191.57

23.36

2.47

1.410.532.02

S

34.27

38.7138.00

227

42.5529.0752.2026.3732.75

26.00

38.0735.759.85

Mg

0.068

0.1600.022

20.72

0.36

0.04

2.92

1.921.261.88

Al

0.19

0.330.160.19

0.170.08

0.450.07

0.05

1.56

0.430.081.68

Si

1.55

1.031.290.96

10.254.172.996.913.81

4.80

8.13

3.801.245.87

Ca

0.25

0.34

0.16

0.17

0.112.66

0.06

0.32

0.660.141.12

Ba

0.0870.053

0.0500.069

0.0291.640

0.1700.2003.360

1.97

0.190.030.43

Mn

0.0350.026

0.0920.0290.010

0.0390.120

0.0230.1550.0050.180

0.092

0.0330.0190.043

Pb

446262

1501161102

21481820275226711

912015.94%

400

15256

534

Cr

14.827.9

15.7

12.312.0

54.9

26.020.019.1

Co

103.815.9

88.644.122.3

10.55.4

94.867.0

25.0

43.411.173.9

Ni

38.045.0

2.7

25.8

25.429.7

95.0

32.427.522.8

As

67.561.5

51.3430.5

711.2420.7241.0123.9476.4

4436.13.08%

227.0

51.334.559.4

Table 7. Correlation coefficients for the elements of the Middle Valley massive sulfides calculated using the data on the 40 bulk samples analyzed(Table 5).

FeGeSCuCoBiZnCdPbBaSrHfZrTlSbAgAuSnMoMnAsAlScThLaCeSmEuTbYbNiRbCsSeHgTaSiMgCaV

0.580.35

-0.31-0.33

-0.34

-0.31-0.38

-0.37

-0.37

-0.34-0.59-0.83-0.56

Ge0.61

0.52

0.39

0.31

-0.51

S

0.52

0.52

-0.54

-0.65-0.60-0.31

Cu0.540.48

Co0.88

-0.30-0.33

BiZn

0.750.72

0.35

Cd

0.31

Pb

-0.32 -0.41 0.320.330.340.320.53

0.460.52

0.52

0.400.34

0.37

0.41

0.39

Ba0.88 Sr0.38 0.38 Hf0.46 0.54 0.48 Zr

Tl0.56 Sb0.31 0.46 Ag0.49 0.75 0.66 Au

0.77 0.55 0.65 Sn0.36 0.42 0.55 Mo

0.37 0.58 0.56 0.60 0.66 Mn0.59 0.59 0.36 0.38 0.38 As

0.42 0.65 0.590.40

0.310.36 0.32 0.32 0.53

0.38 0.34 0.35 0.47 0.440.37

0.46 0.60 0.45 0.490.38 0.40

0.33 0.33

0.39 0.44 0.44 0.420.38

0.39

0.37

0.34

0.31

0.38 0.54

-0.45 0.31

0.35 0.42

0.35

-0.33 0.34-0.34

0.39 0.38

0.43

0.41

0.35

Note: Only coefficients significant at the 95% confidence level are shown.

368

MASSIVE SULFIDE DEPOSIT, SITE 856

Table 6 (continued).

Location Se Sr Mo Cd Sb Ti Sn Ge Tl Au Sc Rb

Mid-Atlantic Ridges26°N24°30'N

East Pacific Rise13°N21 °N9°40'N22°S

Juan de FucaAxial SeamountCleft Segment

Explorer RidgeGalapagos RidgeGuaymas BasinGorda RidgeOkinawa Trougha

Middle ValleySurface samples*5

Cores (this study):AverageMedianStd. dev.

7.822.9

2.328.953.7

100.0

87.7

41.8134.0034.15

323.6

33.5

668.330.0

108.4

1200.0

213.6200.0129.5

115.1

20.124.8

110.7

35.710.4

119.4169.8

108.0

23.717.018.1

52.0

48.7493.2

28.2

549.5134.070.132.0

114.854.8

600.0

39.5

30.118.531.7

11.68.9

31.4

267.34.9

15.07.2

255.3

26.9

4.21.94.7

0.05

0.5

18.26.0

43.8

0.0270.012

0.010

0.002

0.006

0.0060.0040.006

10.0

10.0

10.0

12.5

14.98.7

13.6

0.15

3.3

0.64

2.351.103.17

10.0

10.0

4.3

5.4

9.7

36.8

2.5

3.392.902.95

48.642.7

38.6172.6

9.236.6

165.263.10.55

37.515.185.9

2500.0

8.70

3.543.002.98

1.0310.40

0.260.180.060.15

4.420.13

0.200.071.405.50

0.14

0.007 5.970.004 4.000.007 5.72

1.31.3

1.90.3

0.9

1.5

1.8

0.590.430.60

52.645.8

35.13.26

24.8

60.0

23.570.0

22.0322.0012.76

Table 7 (continued).

AlScThLaCeSmEuTbYbNiRbCsSeHgTaSiMgCaV

Al0.750.540.34

0.42

0.49

Sc0.670.62

0.54

0.39

0.50

0.39

Th0.310.44

0.32

0.60

0.38

La

0.37

-0.32

Ce

0.36Sm

0.600.320.42

Eu

0.43(14-1

0.32

Tb0.400.37

0.340.32

Yb

0.31Ni

0.44Rb

-0.32Cs

Se0.34 Hg

0.700.45

Ta0.60 Si

0.36Ca

0.41

369

S. KRASNOV, T. STEPANOVA, M. STEPANOV

Table 6 (continued).

Location

Mid-Atlantic Ridges26°N24°30'N

East Pacific Rise13°N21°N9°40'N22°S

Juan de FucaAxial SeamountCleft Segment

Explorer RidgeGalapagos RidgeGuaymas BasinGorda RidgeOkinawa Trougha

Middle ValleySurface samplesb

Cores (this study):AverageMedianStd. dev.

Cs

0.971.25

0.26

0.32

1.50

1.502.10

0.440.320.29

Hf

1.582.63

1.27

1.12

4.70

0.505.10

1.110.910.68

Ta

0.170.11

0.31

0.09

1.10

0.49

0.380.080.69

Th

0.560.65

0.84

0.56

1.00

0.85

0.360.300.29

La

1.346.92

0.97

0.851.82

4.10

4.101.70

6.10

1.671.301.45

Ce

10.624.4

3.0

0.513.0

10.2

1.88.9

7.25.08.2

Sm

0.710.87

0.38

0.260.58

2.00

1.80

1.191.100.81

Eu

0.750.76

0.91

0.010.51

1.23

1.30

0.460.400.25

Tb

0.240.14

0.19

0.050.22

0.49

0.39

0.230.200.19

Yb

0.320.47

0.73

0.500.60

0.47

0.74

0.780.700.68

Notes: Fe, Cu, Zn, S, Mg, Al, Si, Ca, and Ba in percent (%)• all other elements in parts per million (ppm). Analyticalresults, except for those indicated by footnotes a or , are represented by average values for Fe and S andmedian values for all other elements, from the review of Krasnov et al. (1992).

a Averages from five samples (Halbach et al., 1989).b Averages from 10 samples for Au (Davis et al., 1987).

Figure 13. A. Positive geochemical correlations in bulk sulfide samples from Site 856. Correlation coefficients greater than 0.8 are shown by triple lines, thosegreater than 0.7 by double lines, those greater than 0.6 by thick lines, and those greater than 0.5 by thin lines. Coefficients lower than 0.4, significant at the 95%confidence level, are shown selectively by dashed lines. Groups of associated elements are numbered from 1 to 8 (see text). B. Geochemical correlation of massivesulfide samples from sediment-free ridges (Krasnov et al., 1992). Conventions are the same as for (A).

370

MASSIVE SULFIDE DEPOSIT, SITE 856

Table 8. Representative microprobe analyses (wt%) of sphalerite.

Core, section,interval (cm) Fe Zn Cd Cu Total

Additionalcomments

139-856E-1H-1, 106-108 15.6 47.7 35.0 0.19 0.11

139-856F-2X-CC, 1-32X-CC, 1-3

139-856G-4R-1,49-516R-2, 101-103

139-856H-3R-2, 67-706R-1, 24-2615R-1, 60-62

14.4 49.8 33.8 0.20 0.0014.7 48.1 35.8 0.20 0.00

98.60

98.20 Inclusion in pyrite98.80 Grain with chalcopyrite inclusion

12.8 50.1 35.5 0.16 0.00 98.564.8 62.5 33.1 0.13 0.00 100.53

13.8 50.8 34.4 0.16 0.0115.2 50.7 34.1 0.17 0.0011.3 52.2 34.7 0.19 0.50

99.17 Inclusion in pyrite100.1798.89 Grain with chalcopyrite inclusion

Table 9. Representative microprobe analyses (wt%) of pyrite.

Core, section,interval (cm)

139-856E-1H-1, 106-108

139-856F-2X-CC, 1-32X-CC, 1-32X-CC, 1-32X-CC, 1-32X-CC, 1-3

139-856G-4R-1,49-514R-1,49-516R-2, 101-1036R-2, 101-1036R-2, 101-103

139-856H-3R-2, 67-706R-1, 24-269R-1, 80-8215R-1, 60-6217R-1, 53-36

Fe

46.30

46.8046.8046.0046.1046.00

46.0045.5045.7046.3046.70

46.4047.0047.0046.9046.70

S

53.30

53.5053.0052.8053.6052.80

53.4052.9053.0053.0052.60

53.2052.6053.0052.4052.80

Cu

0.08

0.010.000.010.000.00

0.070.000.000.010.01

0.010.010.020.020.03

As

0.00

0.000.000.050.070.00

0.000.000.010.100.08

0.000.000.050.000.00

Total

99.68

100.3199.8098.8699.7798.80

99.4798.4098.7199.4199.39

99.6199.61

100.0799.3299.53

Additionalcomments

Skeletal crystal

Center of a large grainMargin of the same grainSkeletal crystalCenter of a euhedral crystalMargin of the same crystal

Grain with magnetite intergrowthPyrite without intergrowthColloform pyriteCenter of a rounded grainMargin of the same grain

Grain with magnetite, chalcopyrite, and sphalerite intergrowthSkeletal crystal with sphaleriteFine-grained aggregateAnhedral grainAnhedral grain

Table 10. Representative microprobe analyses (wt%) of Cu-Fe sulfides.

Core, section,interval (cm)

Additionalcomments

Chalcopyrite139-856F-2X-CC, 1-3 30.6 33.2 36.2 100.0 Inclusion in pyrite

139-856H-3R-2,67-70 30.0 32.6 36.5 99.1 Inclusion in pyrite9R-1,80-82 30.8 33.0 36.8 100.6 Inclusion in pyrite15R-1,60-62 30.2 33.2 35.3 98.7 Intergrown with pyrite17R-1,35-36 30.6 34.1 34.8 99.5 Intergrown with pyrrhotite

Isocubanite139-856E-1H-1, 106-108 41.4

139-856G-4R-1,49-51 40.5

Note: Zn, As, Co, and Cd are below detection limits (<0.01%).

23.4

22.6

36.1

35.5

100.9 Intergrown with pyrite and sphalerite

98.6 Intergrown with chalcopyrite

Figure 14. Proposed mechanism of massive sulfide formation and secondary

alteration. 1 = unaltered and moderately altered sediments, 2 = turbidite layers,

3 = stockwork zone, and 4 = sill emplaced after formation of the sulfide deposit.

Zones inside the sulfide deposit: Po = primary pyrrhotitic sulfide, Py = secondary

pyritic sulfide, and Py + Mt = aggregates containing pyrite and magnetite as the

main secondary minerals. Paths of primary hydrothermal solutions are shown

by open arrows; those of secondary solutions by solid arrows.

371

Table 11. Trace element concentrations (Au in ppb;

Core, section,interval (cm)

139-856G-2R-3, 29-313R-1, 61-644R-1,75-776R-3, 95-977R-1,38-40

139-856H-lR-2,4-7lR-2,4-73R-1,34-373R-3, 14-164R-1, 18-205R-1,52-537R-1,80-828R-1, 124-1269R-1,80-8211R-1,49-5114R-1,4_6

19 samples fromsediment-free ridges

139-856H-5R-1,52-53139-856H-5R-1,52-53

22 samples fromsediment-free ridges

Mineral

Cubic, pyriteFine-grained pyriteFine-grained pyriteCubic pyriteFine-grained pyrite

Cubic, pyriteRhombohedral, pyriteFine-grained pyriteFine-grained pyriteFine-grained pyriteCubic, pyriteFine-grained pyriteFine-grained pyriteFine-grained pyriteFine-grained pyritePorphyroblastic pyrite

Pyrite

PyrrhotiteSphalerite

Zinc sulfides

Au

n.a.1145018

110

140230

12131374

;o3660

16

370

1736

470

all other elements in ppm) in sulflde minerals from massive sulfide samples, determined by INAA.

Ag

2.02.05.56.14.7

<l.O4.04K3.23.52.43.17.41.51.72.5

12.0

3.07.4

129.0

As

n.a.n.a.

1617

130

15060

4n.a

1023

1307866

2

26

27

146

Sb

11.04.72.71.21.0

7.46.00 40.3

0.30.46.5

3.70.3

10.5

0.30.7

27.7

Co

2.111.026.0

1.766.0

2.72.39.6

33.015.010.011.084.020.059.0

170.0

1910.9

12.04.5

113.0

Ni

1912434641

3512

<K)6730161566

1840

86

12140

174

Sc

0.200.931.200.800.60

0.330.300.701.601.50

( l O i l

0.700.400.400.500.70

2.80

0.6021.0

4.20

Cr

3336

1756

6837<<i351229<547191511

85

3950

127

Zr

250170n.a.16090

100<IOO

n a3001601102001208260

110

820

170640

2039

Cs

0.340.501.200.160.20

0.330.320.5 10.400.800.500.500.530.340.300.60

1.32

0.70<O.IO

3.74

Rb

3016

2736

<IO25164721313577323943

151

2585

128

Ba

2203602 10340310

5802500

6206505004802000.1%

660180440

976

90S20

1777

Hf

0.401.202.0I.I0.8

1.20.180.81.90.8n.a.0.8n<>2.10.50.9

4.0

<0.53.6

7.0

Ta

0.080.140.140.100.09

0.090.110.100.170.200.10

0.30<0.05

0.060.14

0.27

0.200.30

0.62

Th

0.10.30.40.50.3

0.3040 ^2.50.60.60.30.8

<O.l0.31.0

1.8

0.30.7

2.3

U

n.a.12.02.80.92.6

1.31.03 02.51.74.55.12.81.23.01.1

0.8

0.543.0

1.0

Se

6

216

3847n a79

10168

1242

26

13040

42

Hg

175

n.a.9

21

164

n a42n.a.22n.a.20101528

n.a.

5216

n.a.

La

n.a.4.62.41.0

<0.5

1.0<l.O

1 66.2l.l1.01.21.00.71.00.9

2.3

2.02.7

2.8

Ce

2.1n.a.7.0

17.014.0

6.913.09.0

15.014.03.0

20.0n.a.24.0n.a.n.a.

9.1

19.012.0

18.0

SIB

n.a.

1.00n.a.0.70

0.420.641 70n.a.

0.901.60

0.400.900.40

0.70

0.200.50

2.63

ELI

1.000.381.401.301.50

0.200.880 300.800.500.300.60n.a.n.a.0.400.40

0.87

0.300.70

2.75

Tb

0.280.23n.a.0.300.40

0.110.23n a0.34n.a.0.490.30n.a.0.150.400.5 1

0.45

0.51n.a.

1.18

Yb

0.390.400.801.00n.a.

0.460.781 501.000.701.801.301.501.00n.a.n.a.

1.14

7.201.10

2.52

AN

O1

Note: n.a. = not analyzed. Data on sulfides from sediment-free ridges are median values from previous studies (Krasnov et al., 1992).