13. MAJOR AND MINOR ELEMENTS IN HYDROTHERMAL AND … · MAJOR AND MINOR ELEMENTS IN HYDROTHERMAL...

13. MAJOR AND MINOR ELEMENTS IN HYDROTHERMAL AND PELAGIC SEDIMENTS OF THEGALAPAGOS MOUNDS AREA, LEG 70, DEEP SEA DRILLING PROJECT1

A. A. Migdisov, B. P. Gradusov, N. V. Bredanova, E. V. Bezrogova, B. V. Saveliev, and O. N. Smirnova, Institute ofGeochemistry and Analytical Chemistry, USSR Academy of Sciences, Moscow 117334, USSR

INTRODUCTION

Interaction between young basaltic crust and sea-water near the oceanic speading centers is one of the im-portant processes affecting the chemical composition ofthe oceanic layer. The formation of metalliferous hy-drothermal sediments results from this interaction.

The importance of the interaction between seawaterand basalt in determining the chemical composition ofpore waters from sediments is well known. The in-fluence of mineral solutions derived from this inter-action on ocean water composition and the significantflux of some elements (e.g., Mn) are reported by Lyle(1976), Bogdanov et al. (1979), and others. Metal-richsediments found in active zones of the ocean basins il-lustrate the influence of seawater-basalt interaction andits effect on the sedimentary cover in such areas. Therole of hydrothermal activity and seawater circulation inbasalts with regard to global geochemistry cycles hasrecently been demonstrated by Edmond, Measures, Mc-Duff, McDuff et al. (1979), and Edmond, Measures,Mangum (1979).

In the area of the Galapagos Spreading Center the in-teraction of sediments and solutions derived from inter-action of seawater and basalt has resulted in the forma-tion of hydrothermal mounds. The mounds are com-posed of manganese crusts and green clay interbeddedand mixed with pelagic nannofossil ooze. These moundsare observed only in areas characterized by high heatflow (Honnorez, et al., 1981) and high hydrothermal ac-tivity.

Hydrothermal mounds were studied by deep-tow(Sclater and Klitgord, 1973; Lonsdale, 1977), by the sub-mersible Alvin (Corliss et al., 1979), and by direct drill-ing during Leg 54 (Hekinian et al., 1978; Natland et al.,1979; Rosendahl, Hekinian et al., 1980) of the Deep SeaDrilling Project (DSDP). Green clay and Mn-crust sam-pled from Galapagos mounds are very similar to thoseobserved in the FAMOUS area (Hoffert et al., 1978,1980).

The first analytical data on samples obtained fromthese expeditions show a significant difference betweengreen clays and Mn crusts from the Galapagos hydro-thermal mounds and metalliferous sediments recoveredfrom the mid-oceanic ridges, especially the East PacificRise (Boström and Peterson, 1969; Bender et al., 1971;

Honnorez, J., Von Herzen, R. P. , et al., Init. Repts. DSDP, 70: Washington (U.S.Govt. Printing Office).

Sayles and Bishoff, 1973; Bonatti, 1975; Cronan, 1976,Migdisov, Girin et al., 1979).

The metalliferous sediments from mid-oceanic ridgesdirectly cover basaltic basement; they consist of Fe andMn hydroxides and Fe smectites and are enriched bymany minor and trace elements (Ba, Ni, Cu, Zn, Cd,Sb, Mo, REE, etc.). Th, Ce, Rb, Cr, and some otherelements were depleted in these sediments relative to thedeep oceanic clays. In contrast, hydrothermal sedimentsfrom the Galapagos mounds are poor in all these ele-ments (Hekinian et al., 1978; Hoffert et al., 1980) andare both vertically and laterally restricted (Honnorez etal., 1981).

Various hypotheses on the origin of the hydrothermalmounds sediments were proposed by Hoffert et al.(1980) and Dymond et al. (1980). They are as follows:

1) Deposition from solution, resulting in wide-spread-ing sedimentary layers.

2) Interaction of pelagic oozes in the mounds areasedimentary cover with upward percolating hydrother-mal solutions which replace the pelagic ooze by greenclays.

3) Upward flux of hydrothermal solutions from lo-cally permeable zones of basement through the sedimen-tary cover, with the formation of minerals resultingfrom an advancing oxidation gradient, without any in-teraction of solutions with pelagic sediments.

On Leg 70 of the Deep Sea Drilling Project (DSDP),a more detailed series of holes was drilled in 30 to 40meters of sedimentary cover, both on mounds and in anoff-mounds area. Regions of low heat flow withoutmound occurrences were also investigated. The hydrau-lic piston corer (HPC) was used, and detailed samplingof undisturbed sedimentary sections was accomplished.

Data obtained on Leg 70 (Honnorez et al., 1981) sug-gest that the hydrothermal sediments are local in re-gional extent, which allows us to exclude the first hy-pothesis. Some aspects of the third hypothesis also arequestionable given the data obtained from Leg 70. Forexample, differences in thickness of pelagic ooze frommounds, off-mounds, and low heat-flow area sedimen-tary sections were observed. The evidence of microfossiltests, replacement, and infilling by smectite (Honnorezet al., 1981; Schrader et al., 1980) also casts doubt onthe third hypothesis. However, data on the element dis-tribution in sediments indicate that the hypothesis of-fered by Dymond et al. (1980) may be correct—namely,element differentiation through a redox-potential gra-dient in sedimentary sections. The data of Schrader et

277

A. A. MIGDISOV ET AL.

al. (1980) maybe taken to suggest that in the moundssmectite textures have been produced both by replace-ment of pelagic sediments and by open space filling. Theauthors recognized "complete replacement of Forami-nifera tests" in all mounds holes of Leg 54. They notedthat this type of texture is not common statistically andis "often obscured by overgrowth of additional greenclay." Open space filling textures have been observed as"zoned and sequential infillings of void spaces," imply-ing "controlled deposition." The authors suggested thateither types of smectite formation are concurrent or theprocess of replacement may be assumed to be slightlylater. At the same time the relative role of both pro-cesses in smectite generation in mounds is not yet clear,and geochemical data supporting void infilling process-es are absent.

The problem of the redistribution of major elementsin the pelagic ooze during their dissolution and replace-ment has also not been resolved.

In light of these problems, the purpose of this chapteris the following:

1) To study changes with depth in the mineralogicaland chemical composition of sediments in mounds cores.

2) To compare and correlate changes in the distribu-tion of major and minor elements on mounds, off-mounds, and in low heat-flow areas.

3) To speculate on the origin of hydrothermal sedi-ments on the basis of obtained data.

METHODS OF ANALYSES

Prepared samples were washed to remove marine salts beforeanalysis. Different techniques were used for determination of majorand minor elements.

Data on major elements were obtained both by X-ray fluorescence(XRF) and chemical techniques. Samples were heated to 105°C priorto analysis. The chemical technique used, including determination ofvolatile components (H2O+, CO2) and the oxidation state of Fe, is dis-cussed by Migdisov, Girin et al. (1979).

XRF analyses were performed on a 28-channel PW-1600 "Phil-ips" machine. A description of the procedure is given in Tsameryan etal. (1980).

Data on the major elements, determined by two independent tech-niques, were very similar and are presented in Tables 1-6. Numbersrepresent the mean value of the two analyses.

Minor and trace elements were determined by XRF, atomic ab-sorption, and instrumental neutron activation. Cr and V were ana-lyzed by XRF during major element determination. Li, Rb, Sr, Ni,Zn, Cu, and Pb were analyzed by the atomic absorption method usingPerkin Elmer-405 and Per kin Elmer -603 machines. For Pb determina-tion, a graphitic cuvette HGA-76 B machine was used in some cases(Sedyh et al., 19áθ). Neutron activation analyses were carried out us-ing the technique described by Migdisov et al. (1980). A 40-cm3 Ge(Li)detector and Nokia LP 4900 channel analyzer were used. In somecases radiochemical concentrations of rare earth elements (REE) wereexamined.

Mineralogical and structural anomalies of the hydrothermal clayminerals were also studied. Measurements were made on a diffrac-tometer DRON-2, with Fe-radiation filtered with Ni using a commonsample treatment technique.

MAJOR ELEMENTS

The data obtained are shown in Tables 1-10 and Fig-ures 1-17. Figure 1 presents the distribution of some ele-ments with depth at two mounds sites, 507 and 509. Aninverse relationship in Fe and Mn concentrations existswithin the hydrothermal sediments. The iron-rich smec-

tites lack Mn, which is concentrated in the uppermostlayer of the mound in the form of Mn crusts. In turn,very little Fe is found in the manganese crusts.

There are some cases where full fractionation of Feand Mn is not observed. These are: (1) in oxidized smec-tites which are localized directly underneath the Mncrusts (these smectites contain much higher Mn concen-trations relative to deeper layers, but one cannot excludesome contamination by Mn oxides); (2) in some fine-grained smectites from burrows mixed with pelagic ooze(Sample 509B-5-2, 35-37 cm [1]). Mn enrichment ofpelagic oozes alternating with green clays and the lowFe2O3/MnO2 ratio (0.5) in these sediments must also benoted (Samples 509B-5-2, 35-37 cm [2] and 509B-5-1,86-90 cm); (3) in the oxidized Fe-Mn sediments browncarbonate and siliceous ooze with Fe-Mn nodules andcrusts occur in the uppermost horizons of the mounds.We noted the similarity of the composition of micro-nodules separated from the uppermost mound sedimentsto the Fe-Mn crusts and nodules from these oozes and tothe carbonate-free basis composition of the uppermostpelagic sediments in the mounds. We therefore classi-fied all these materials as "upper crusts" in contrast to"lower crusts" which occur farther down in the moundsection and are characterized by a lower Fe-Mn ratioand by lower iron, silica, and alumina concentrations.The least differentiated type of Fe-Mn oxide sedimentswas sampled as a dark brown nonconsolidated mud in alayer of pelagic ooze (Sample 509B-2-3, 12-14 cm) with-in the unit of the hydrothermal sediments. The Fe2O3/MnO2 ratio in the analyzed sample is approximately1/2.

In the pelagic oozes from the low heat-flow areas andthe off-mounds sedimentary sections, higher Fe and Mnconcentrations relative to normal oceanic carbonate sed-iments (Turekian and Wedepohl, 1961) are observed.Further, the Fe and Mn concentrations (on a carbonate-free basis) approach the Fe and Mn concentrationsfound in deep ocean red clays (Turekian and Wedepohl,1961; Strakhov, 1976; Migdisov, Bogdanov et al., 1979).The highest concentrations of these elements, especiallyMn, are found in the upper oxidized surface layers ofthe pelagic sediments (Lynn and Bonatti, 1965). Thelayers of pelagic ooze occurring between hydrothermallayers or from lower in the mounds sections have rela-tively higher concentrations of Mn and Fe than do theoff-mounds and low heat-flow, nonmounds, sedimen-tary sections (Tables 7-9).

At the same time, significant inhomogeneity of theelements in the pelagic ooze on the mounds (Tables 1-3)must be taken into account. For example, there are sam-ples (e.g., Sample 509-2-2, 127-131 cm) with high con-centrations of iron (15% Fe2O3) with average or lowconcentrations of Mn (0.34% MnO), and there are otherexamples (e.g., Sample 509B-5-2, 35-37 cm [2]) wherethe concentrations of these elements run opposite: e.g.,11% MnO and 5.76% Fe2O3.

Iron in hydrothermal nontronitic smetites is mainlyin an oxidized state (Rateev et al., 1980; Hoffert et al.,1980). However, the FeO concentrations in the smectites(Tables 1-3, 7-9) fluctuate from below the determina-

278

HYDROTHERMAL AND PELAGIC SEDIMENTS

Table 1. Major and minor elements in sediments from Hole 507D, DSDP Leg 70.

\sample

X*Component^

1-1,14-20 cm

v no. 13

Major element oxides (%)

SiO2

TiO2A12O3Fe2O3

FeOMnOMgOCaONa2OK2OP2O5H2O +

co2Total

9.4

3.82.2

Minor elements (ppm)

LiRbCsSrCoNiCrVCuZnSbPbSeLaCeSmEuTbYbLu

95

—70524

780——

185456

18.792.8

13122.0

——2.0

—

1-1,18-20 cm

no. 14

21.50.173.104.51

40.90a

2.208.481.530.820.2105.05

12.14

100.58

98

13.854525

5902660

170406

12.21.27.1

11.113.82.320.3900.4321.6900.300

2-1,18-19 cm

no. 15

2.50.070.514.10

39.74a

0.952.324.260.370.224

611.4

30010.5514031445011.7

<0.51.75.444.60.740.2700.2130.5600.095

2-1,22-26 cm

no. 16

48.950.050.85

35.520.420.5153.090.511.321.360.0966.650.08

99.41

1025

—104

3.61031

—3692

1.2182.85.775.60.980.444—0.9470.170

2-1,91-95 cm

no. 17

52.10.080.52

35.82

0.0823.490.322.511.340.069

————

10————

——————

—

2-2,6-10 cm

no. 18

53.40.080.65

} 35.98

0.0653.850.351.831.290.081

—————

10—————————————

4-2,130-135 cm

no. 24

53.650.090.38

32.150.880.0353.820.231.123.160.0584.89

100.46

662

1.8360.65

20—5

211.0

<0.51.00.741.470.1740.1070.0740.2130.062

5-1,78-80 cm

no. 25

57.300.41

10.68

J13.14

0.1383.241.162.202.610.112

4090

3.4270

127637

100153390

1.8—

1220.4433.524.180.9630.6391.9330.356

6-3,7-9 cm

no. 27

52.70.070.35

) 32.96

0.0555.050.381.512.370.078

————

7.320

—1425

—0.5————————

6-3,94-98 cm

no. 28

18.70.131.989.161.060.1642.04

33.470.260.500.1024.21

26.68

98.46

7.518—

10704.7

2221—

50980.6—4.7

14.59.62.50.64—1.3300.190

10-2,116-119 cm

no. 36

8.650.111.552.520.560.2100.81

46.180.300.340.0662.11

35.93

99.34

7.591.3

14964.9

2128—

51610.5—3.1

14.4610.051.640.3690.3511.1130.174

a Expressed as MnO2.

tion limit to greater than 3%. The highest FeO concen-tration is observed in a fine-grained smectite mixed withpelagic ooze. The fluctuation of FeO content in thesmectites is not random but shows a distinct trend, in-creasing from the top to the bottom of the hydrothermalunits. As shown in Figure 2, the degree of iron reductionin the smectites increases with depth in the cores. Com-paring Figures 2 and 3, we can conclude that at the off-mounds sites and in the low heat-flow areas, the in-crease in FeO/Fe2O3 with depth is very similar to this ra-tio increase of the smectites in the mounds. The areaswith high heat flow (Holes 509 and 509B) are assumedto have the greater gradient of FeO/Fe2O3, with the ra-tio increasing with depth.

Elements that do not depend on the redox gradientalso vary with depth in the mounds. For example, potas-sium content in smectites is shown versus depth in Fig-ure 4. Average concentrations of the major elementsin the smectites from the "upper," "middle," and"lower" parts of mounds hydrothermal units are givenin Tables 7 and 8. Although such a division of the smec-

tites into three groups is quite arbitrary and differentlimits for such division can be chosen, the increase inK2O with depth, along with changes in relative abun-dances of most of the major elements, suggests altera-tion in the compositions of the smectites.

XRD investigations of the smectite samples supportthe regularity of their alteration with depth. Rateev etal. (1980) have shown that the green hydrothermal claysare the K-Fe smectites close to nontronite. Smectite canbe defined "as a mixed-layer mineral in which typicallymontmorillonitic spaces filled with hydroxyl ions alter-nate with spaces filled with potassium." The authorssuggest that this may be the result of glauconitizationor celadonitization by influence of hydrothermal solu-tions. Fe- and K-rich glauconite-like mica in variableproportions tends to occur in the later stages of theseprocesses. Smectite samples analyzed in this work arecharacterized by higher Fe and K and lower Al content,with the lower smectites assumed to be in a higher levelof celadonitization, as compared to the previously citedwork. Hoffert et al. (1980) nevertheless concluded on

279


Table 2. Major and minor elements in sediments from Holes 507F, 507C, and 507H, DSDP Leg 70.

\Sample\

ComponenK

1-1,5-7 cm

no. 105


SiO2

TiO 2

A12O3

F e 2 O 3

FeOMnOMgOCaONa 2 OK 2 O

P2O5H 2 O +

co 2LOI

Total

17.90.132.502.21—2.92*0.88

36.750.640.330.1254.53

28.411.02

98.34


LiRbCsBaSrCoNi

CrCuZn

CdSbAsPbSeLa

CeSmEuTbYbLu

79.52.2

1291201

11_

15—

2250.0571.0——4.8

10.106.281.6120.4320.3931.2700.209

2-1,14-17 cm

no. 107

27.30.214.14

7 12

0.2792.02

29.580.890.560.119———

-

1617

——

98012

190_

23489

_—_

487

17.20—1.7000.450_1.300

—

Hole 507F

2-2,132-134 cm

no. 101

52.400.070.41

33.730.330.0303.480.181.141.950.0505.72——

99.49

642

1.269037

0.74.8

101929

0.270.7_

<0.50.71.27

11.900.9300.2140.1980.4510.069

3-3,41-45 cm

no. 103

14.450.122.112.570.700.2782.23

41.090.330.320.0743.27

31.79—

99.33

611

——

4109

107_

76184

___

303.8——1.450__1.360

—

6-3,55-59 cm

no. 108

9.550.112.001.760.580.1460.76

44.610.270.290.0782.29

35.001.43

98.87

890.6

1481314

4.0211850

125_0.43.1

443.3

10.158.801.2700.3010.2691.2740.175

Hole 5O7C

1-2,59-62 cm

no. 37

26.80.122.120.880.620.3410.86

33.870.380.290.0802.98

26.873.00

99.21

———

101010_____

1.6__4.88.70—1.650__1.040

—

1-4,141-145 cm

no. 39

23.900.183.982.220.970.2401.15

34.630.550.530.0853.46

26.971.78

100.64

913

1.279

10258.1

1341896

206

3.8_

305.7

10.307.521.5400.5010.3101.3000.233

Hole 507H

6-1,53-55 cm

no. 40

12.20.182.181.230.720.1380.78

42.850.230.260.0784.23

33.34—

98.42

7.59——

12586.0

66_

65150

————4.5——1.040——1.130

—

6-8,115-119 cm

no. 43

13.900.264.021.510.990.1721.04

41.790.560.650.1121.65

31.761.04

99.45

1014

1.26125

10887.2

52125775

_0.69.3—6.8

18.8015.022.1360.5530.4621.4040.213

Expressed as Mnθ2

the basis of X-ray diffractograms that the green clayshad a smectite structure and contained a mica phasewith a tf-spacing equal to 10Å.

Our XRD investigation has shown that the greenclays are represented by mixed-layered minerals whichcontain a high iron (Fe-montmorillonite-nontronite se-ries) three-layer structure. This conclusion is based on0̂60 spacing values of diffractograms which vary from

1.502 to 1.505Å (Fig. 5). The ratio of the secondary re-flection (002) of the smectite and the basal mica reflec-tion (001) is significant. There is 80-90% smectite in themixed-layer structure of Sample 66 which is from theupper smectite layers (2.8 m sub-bottom depth) of Sam-ple 509-1-2, 126-129 cm. An equal and possibly greatersmectite content is found in the upper green clay layersof Samples 507D-2-1, 22-26 cm and 91-95 cm (1.2-1.9m sub-bottom depth). Approximately half as much smec-tite (40-50%) is found within the mica-smectite mineralfrom Hole 507D (Table 1: no. 24; Sample 507D 4-2,130-135 cm), which was sampled at 12.8 meters sub-bottom. The minimum content of smectite structures(30-40%) and maximum mica-layer content are foundin the deepest hydrothermal sediment layer at 19.6 me-ters sub-bottom in Hole 5O9B. The structural alterationsobserved (Fig. 6) are in agreement with the increase of

K2O concentration with depth in the mounds sectionsfrom 1.3 to 5.0% (Fig. 4).

Some green clays from the mounds sections (Samples507D-5-1, 78-80 cm and 509B-3-2, 104-109 cm) whichalternated with pelagic-ooze layers differ significantlyfrom the Fe-rich and Al-poor clays previously discussed.They have a higher concentration of Al, and their com-positions are close to that of a beidellite-nontronitictype of smectite.

As was noted previously (Honnorez et al., 1981) adistinct gradient change of amorphous SiO2 concentra-tion with depth in all but Hole 509 and the off-moundssections was observed. A typical example of the sili-ceous microfossil test distribution through the sedimen-tary sections is shown in Figure 7. It can be seen thatthese tests are lacking in the lower horizons of themounds and that they are nearly absent in the hydro-thermal units. At the same time, the concentrations ofsiliceous tests within the sediments does not differ withdepth in the low heat-flow regions (Fig. 7). Their preser-vation here is much better than in the pelagic oozesfound in the mounds areas. CaCO3, which is the maincomponent in the pelagic sediments, is absent in green-clay layers (Tables 1-8). In addition, this mineral is ob-served only in fine-grained smectites which are mixed

280


with pelagic oozes. Minor amounts of Corg and Spyr areobserved in the mixed layers, but they are practically ab-sent in granular pure smectites.

One main peculiarity of the mounds hydrothermalsediments is that they are poor in alumina (Tables 7 and8). Boström and Peterson (1969) have also reported alu-minium-poor and Fe-Mn rich metalliferous sediments inareas of high heat flow near mid-ocean rises. Alumin-ium is one of the most stable elements under sedimen-tary conditions, and the change in the ratios of variouselements to alumina is used to estimate the input rate ofsediments from hydrothermal sources. Comparison ofthe element ratios to alumina in the mounds and off themounds to those which are not affected by hydrother-mal solutions (Hole 508) may give a rough estimation ofthe elemental input from hydrothermal solutions and anapproximate scale of the redistribution of the elements.

The results of this comparison show the distinct geo-chemical differences between hydrothermal sedimentsoccurring at different levels within the mounds sections.The upper crusts differ from the lower crusts and fromthe Fe-Mn oxide mud, but they have significant similar-ity to the uppermost oxidized layers of the off-moundpelagic oozes in the high heat-flow area (Fig. 8). The Fe-Mn oxide sediments differ from all Mn-crusts and fromoxidized off-mound oozes by their significant enrich-ment in Fe and Si. At the same time all oxidized sedi-ments of the high heat-flow area share a common pecu-liarity: They are strongly enriched in manganese in com-parison with the low heat-flow area pelagic oozes. Allthe hydrothermal crusts are enriched in Mn, K, Na, Mg,and, in some cases, in Fe and Si (for the Fe-Mn oxidesediment) relative to pelagic oozes.

The comparison of iron-rich smectites from differentlevels of the mounds sections with the mound pelagicoozes and the pelagic sediments from the low heat-flowarea is shown in Figure 9. High ratios of SiO2, Fe2O3,FeO, K, Na, and Mg to alumina in smectites are evident.There is also the enrichment in Fe2O3 and K in themound pelagic oozes (both in basal sediments and inoozes alternating with smectites). A high manganese toalumina ratio is characteristic of pelagic oozes inter-layering with hydrothermal sediments. The high alu-mina smectites, scattered as thin layers or patches inthese pelagic oozes, are similar to the host sediments inelemental ratios, but they differ significantly from hy-drothermal iron-rich and alumina-poor green clays. Inother words, interpreting obtained data relative to theleast mobile alumina, we can suggest significant input.

A more accurate estimation of elemental input orsubtraction caused by hydrothermal activity can be ob-tained by evaluating the absolute masses of elements oroxides in the mound sequence with respect to those inthe low heat-flow sedimentary sequences. In these esti-mates, the physical properties of sediments (porosityand density) which were ignored in calculations of ratiosto alumina have to be considered.

Unfortunately, data on physical properties were notavailable for all samples analyzed; thus we used ship-board measurements for intervals close to the analyzedsamples or average values for each type of sediment in

the hole if physical properties had not been measurednear our samples.

The results of the calculations are presented in Table10. They show a high input of silica in the hydrothermallayers. The mass of silica in smectite is up to five timesas large as in pelagic ooze found in the low heat-flowarea. At the same time, silicon is depleted up to two-thirds from the basal pelagic oozes in the mounds sec-tion. Silica flux is a maximum in the lower smectites anddecreases in the upper smectite layers. The same trendsexist for iron. Iron input in the lower smectites is 20times higher than in pelagic sediments. Both ferric andferrous iron supply is a maximum in these lower smec-tites. Estimates of iron mass in the mounds pelagic sedi-ments show a distinct increase. As a rule, the iron in-crease in pelagic layers alternating with smectites isgreater than in basal sediments. Within the smectites,alumina mass values differ from each other and fromthat of pelagics by a factor of three. In the upper smec-tites at Hole 509B, alumina mass values are the same asin pelagic sediments from low heat-flow areas. At thesame time the middle and lower smectites in Hole 5O9Band all the smectites in Hole 507D are depleted inalumina masses respective to low heat-flow sediments.The variations of alumina masses in smectites presum-ably result both from the significant role of the pore in-filling texture of smectites in some layers and from theinhomogeneity of alumina masses in parent pelagic oozesbeing replaced by smectite. Actually, pelagic ooze fromthe mounds sequences at Hole 507D contains only half asmuch alumina mass as low heat-flow pelagic sediments.

MINOR ELEMENTS

The distribution of minor elements and their rela-tions with major elements in hydrothermal sedimentsdiffers from pelagic ooze.

As a rule, element concentrations in pelagic sedi-ments are higher than in hydrothermal sediments, espe-cially when values obtained are calculated on a carbon-ate-free basis (Tables 7-9). It can be shown that only theupper crust and Fe-Mn oxide sediments are close to thepelagic sediments in some minor element concentra-tions. High alumina smectites also contain higher con-centrations of minor elements when compared to the hy-drothermal sediments.

Differences in relationships of elements are observedbetween pelagic and hydrothermal sediments. For ex-ample, in Figure 10 the Mn and Ni contents in the off-mounds pelagic ooze show a nice correlation. This rela-tionship is worse in mounds pelagic sediments and it isabsent in hydrothermal smectites and Mn crusts. Similartrends exist between Ni plotted against Sb (Fig. 11).Correlations between Ni, Co, and Cu are observed in allthe sediments. Zn correlates with Fe in Mn crusts (Fig.12). The relationship between K and Rb within hydro-thermal and pelagic sediments is shown in Figure 13.There is a nice correlation between these elements inpelagic oozes and in the smectites replacing them. Incontrast, this correlation is absent in Mn crusts.

The average concentrations of rare-earth elements(REE) in hydrothermal sediments are one order of mag-

281


Table 3. Major and minor elements in sediments from Hole 509B, Leg 70.

^\Sample

Component ^

1-1.40-42 cm

s. no. 96

Major element oxides (°7o)

Siθ2Tiθ2A12O3

Fe 2 O 3

FeOMnOMgOCaONa2θK2O

P2O5H 2O +CO 2

L.O.I.

Total

3.90.100.900.91

47.97a

1.493.61_0.490.105———

-


LiRbCsSrCoNiCrCuZnSbPbSeLaCeSmEuTbYbLu

1.91.97.7

89022

1461250

17315.4

<156.53.173.050.850.2990.2330.8050.134

1-1,84-87 cm

no. 99

17.30.163.073.70

9.81 a

1.2432.7

0.710.420.1475.62

27.43n.d.

99.31

810_

244010

14416

152220

108.54.5

10.791.50.430.401.20.18

1-1,108-110 cm

no. 64

10.50.132.085.46

36.08a

2.4615.76

1.270.410.186_

—

-

56

820_

17510

144220

_1.5

_

—

—

1-2,48-52 cm

no. 65

1.50.080.710.56

78.34a

3.421.681.531.140.1187.743.93

n.d.

100.80

< 278.8

6801460

_273710

< l1.62.623.50.780.1670.1410.4370.063

1-2,126-129 cm

no. 66

50.80.070.30

34.20

1.163.720.231.241.810.0735.88n.d.n.d.

99.48

530

_47

1.75.5

30_

270.51.70.50.450.880.160.0330.0270.0870.017

2-1,90-93 cm

no. 67

16.40.243.884.00

34.76a

3.029.351.160.700.160_

—

-

66—

54514

132307088

_< l

3.59.2

10.01.40.440.441.5

—

2-1,109-111 cm

no. 68

6.40.151.871.69

42.15a

3.446.531.050.470.130———

-

1014

_750

2815361

12298

8.2< l

10.4___——

—

2-1,124-126 cm

no. 100

49.500.161.39

29.10

4.06a

3.560.880.942.140.0657.110.68

n.d.

99.59

1140

—200

—12206155

—3.23.43.3—0.85——

—

2-2,40-42 cm

no. 69

45.100.224.26

23.02

9.88a

3.680.741.211.600.1298.101.69

n.d.

99.63

1534

1.7795

164840

13070

1.92.68.2

13.616.32.480.5720.4951.5040.296

2-2,112-115 cm

no. 70

51.90.244.12

28.4

0.2704.071.001.581.720.115———

-

1541

—270

—14205085

—<0.5

———~-————

2-2,127-131 cm

no. 71

35.550.325.519.330.690.2202.65

21.40.910.960.1107.41

14.91n.d.

99.97

2026

—730

—88

—78

270—————————

—

2-3,12-14 cm

no. 97

32.750.185.383.030.330.7051.04

28.241.211.430.0903.46

20.331.81

99.99

931

1.7876

3.4531548

1340.9

305.6

12.011.42.30.5720.4391.2740.218

2-3,16-18 cm

no. 72

23.650.080.82

17.76—

3S.00a

3.563.881.291.110.1878.644.43n.d.

100.41

810

—645156840

40321690

25.2< l

6.25.044.520.85——0.6120.090

Mn and Mnθ2

nitude less than their average concentrations in car-bonate-free pelagic ooze material (Tables 7 and 8). Forminimal concentrations of REE in hydrothermal smec-tites, they differ by about two orders of magnitude. TheREE distribution pattern in pelagic ooze and hydrother-mal sediments as a rule is much the same. The main dif-ference is the cerium deficiency. This peculiarity reflectsthe role of seawater in the REE pattern of this sedimentformation (Piper, 1974; Piper and Graef, 1974; Benderet al., 1971; Migdisov, Bogdanov et al., 1979). Smectitescontaining the lowest REE concentration are character-ized by different patterns (Figs. 14, 15). They are similarto basalt REE distribution patterns and are close to thesmectite found in veins and in basalt alteration zones(Fig. 16) recovered from the deep part of Hole,504B,Costa Rica Rift and described by Sharaskin et al. (inpress). These patterns are assumed to be caused by hy-drothermal solutions circulating through the basalts.

The minor element concentrations in the mounds sed-iments are localized at different depths in much thesame manner as the major element distributions—forexample, in the upper layer of the mounds sediments inHole 507D are concentrated sufficiently higher amountsof Ni, Co, Cu, Zn, Pb, and REE as compared with thelower Mn crusts. These concentrations also approachthe values found in the upper oxidized layer of pelagicooze (Tables 7 and 8).

The distribution of minor elements in the smectitesalso varies with depth. Rb content increases from the

top to the bottom of the hydrothermal unit in themounds sections. Figure 17 plots zinc content againstferrous iron, which also increases with depth. It shouldbe mentioned that higher concentrations of Zn, as wellas sulfide sulfur, occurred in smectites mixed with pe-lagic ooze. It is these varieties of smectite, characterizedby higher concentrations of these minor elements, whichare enriched in the pelagic sediments (Tables 1,3,7, and8). Their concentrations in mixed smectites are inter-mediate between pure smectite and pelagic sediments.

In a number of cases the data on minor element dis-tribution present evidence for redistribution of elementsnear the contact between smectite and pelagic sediment.For example, in Hole 509B, a smectite-infilled burrowsurrounded by pelagic ooze was studied (Table 3; Sam-ple 509B-5-2, 35-37 cm). Enrichment of iron, manga-nese (3% MnO), and a high concentration of Ni (600ppm) in the smectite was observed with more than 30%CaCO3. On a carbonate-free basis, Zn and Cu contentin the pelagic ooze near the contact with smectite is threeto four times and Ni concentration is nearly one-tenth aslarge as in the smectites. Enrichment of Co, Sb, andREE near the contact between the smectites and pelagicooze was also observed.

Inhomogeneity of minor element distribution in thepelagic ooze interlayers within hydrothermal zones andin the lower pelagic ooze units also suggests a redistri-bution of elements. For example, Ni content in themounds pelagic ooze differs from 5 to 180 ppm (on a

282


Table 3. (Continued).

3-1,86-89 cm

no. 73

52.450.080.45

32.10—0.5303.920.541.012.980.0486.130.26n.d.

100.50

359

514

11102424

—1.01.0——————

—

3-2,104-109 cm

no. 74

53.250.427.68

19.660.850.2124.640.981.101.940.1257.26n.d.0.53

98.65

2650

4408

5048

126217

1.00.9—

10.57.82.810.6550.5521.3870.238

3-3,17-22 cm

no. 75

53.100.080.69

31.060.850.0683.980.331.033.380.0655.810.19n.d.

100.64

970

64—8

301121

—<0.5

———————

—

4-1,46-51 cm

no. 98

52.800.080.60

31.980.560.2243.820.360.863.200.0305.850.21n.d.

100.57

575

60—

< 530

7.532

——_——————

—

4-2,98-103 cm

no. 76

54.100.233.29

26.40_0.0654.380.850.752.690.0806.50n.d.n.d.

99.34

1165

345_6

336653

0.87.87.0

101.9

__

—

4-321-26 cm

no. 77

53.150.080.22

32.10—0.0584.220.290.862.740.0345.91

n.d.n.d.99.66

<1260

473.86.4—

15190.7

<0.50.7——————

—

5-186-90 cm

no. 78

10.00.101.301.19_2.070a

0.9245.27

0.170.150.0752.17

35.10_

98.52

65

1470<5125

_83

160___——____

—

5-235-37 cm (1)

no. 79

33.750.111.54

16.560.943.255a

3.6717.320.552.350.0604.63

14.84_

100.02

1378

53016

6351081

1682.0_2.67.0

1.1__0.96

—

5-235-37 cm (2)

no. 80

10.500.162.141.370.392.690a

1.0942.66

0.320.280.0842.84

33.540.83

98.89

910

148012.568_

89175

1.6_5

10.99.52.00.42_1.27

—

5-380-82 cm

no. 81

47.850.081.00

26.911.700.0724.444.460.734.630.0505.643.24—

100.80

13115

2234.8

501054

1041.01.94.02.223.180.655__0.3830.091

6-335-37 cm

no. 82

8.800.061.851.070.390.5500.88

46.670.270.360.0801.68

37.17_

99.83

512

4593.7

< 53

2848

—30

1.811.073.931.75_

0.2931.2830.086

7-270-74 cm

no. 83

11.150.182.771.430.500.2400.88

44.140.280.360.0972.54

33.310.53

98.41

711

14645.8

18018.497

223—

304.8

1415.73.50.67_1.90.30

8-193-98 cm

no. 84

13.850.223.282.010.460.2321.30

41.440.360.430.1084.06

30.690.53

98.97

—

———————————_——

—

8-352-78 cm

no. 85

11.750.202.301.730.970.2451.29

44.460.300.320.0902.56

33.86_

100.08

11

10229.37.0

4250

1180.4

336.3

12.39.61.970.4840.4181.3640.222

Table 4. Major and minor elements in sediments from Hole 509, DSDP Leg 70.

^~\Sample

Component\

1-1,4-7 cmno. 47


SiO2

Tiθ2A12O3

Fe2O3

FeOMnOMgOCaONa2OK2OP 2 O 5

H2O +CO 2

L.O.I.

18.100.162.932.22—5.1300.86

35.230.520.350.1395.32

28.58

1-1,82-86 cm

no. 48

18.500.121.732.520.700.2881.12

38.230.310.270.1303.77

30.190.86

3-1,50-52 cm

no. 44

23.600.263.691.860.760.4841.21

34.980.480.420.1004.63

27.330.22

4-1,80-84 cm

no. 49

21.200.183.542.460.791.4201.36

34.640.530.460.0893.63

27.78

5-1,85-89 cm

no. 50

15.10.182.381.380.680.3460.91

41.160.350.320.0854.90

32.43

6-1,91-95 cm

no. 51

13.10.152.371.080.800.3110.87

42.510.370.280.1083.31

32.98

7-2,62-66 cm

no. 52

11.30.171.960.780.680.2030.75

43.670.410.260.0902.74

33.542.17

8-2,19-21 cm

no. 53

12.100.172.921.490.600.1940.78

42.860.330.330.1782.98

33.68—

Total 99.5


100.02 98.08 100.22 98.24 98.72 98.61

LiRbCsBaSrCoNiCrVCuZnSbSeLaCeSmEuTbYb

109

116025

245

124220

3.36.4

12.711.82.4

0.251.24

8152.3

901120

8.011124

105220

1.02.8

11.037.191.7220.3570.3231.313

1014

11807.5

145

106193

1223

12009

16521.616

122185

5.210.78.61.70.40.41.6

810

12905

10515

84175

910

14605

8215

72170

89.5

146010.77713

120126

1.75.07.77.91.60.40.41.46

1010

1.1151

14607.5

6719

105930.3

8.176.701.410.355

1.042

283


Table 5. Major and minor elements in sediments from Holes 508 and 508C, DSDP Leg 70.

\Sample

\Component^

1-1,39-43 cm

v no. 1

2-1,61-65 cm

no. 2

3-1,123-127 cm

no. 3

4-2,80-84 cm

no. 4

Hole 508

5-1,82-86 cm

no. 5

6-2,25-29 cm

no. 6

7-2,123-128 cm

no. 7

8-2,33-35 cm

no. 8

8-3,54-56 cm

no. 9

8-3,63-64 cm

no. 10

Hole 5O8C

1-1,24-28 cm

no. 11

1-2,15-19 cm

no. 12


SiO2

TiO2

A12O3

Fe2O3

FeOMnOMgOCaONa2OK2O

25H2O +CO2LOI

Total

17.80.121.813.110.600.3121.33

37.920.290.260.0993.13

29.223.56

98.92


LiRbCsBaSrCoNiCrCuZnSbSeLaCeSmEuTbYbLu

7.572.0

1001130

1014318.5

118215

1.247.89.01.5590.309

0.9070.234

27.50.194.71

}2.2,

0.9211.01

31.950.600.910.1163.50

25.20

98.82

19.40.193.75

2.58

0.3781.01

38.090.600.450.1163.0

30.0

99.57

21.80.122.42

1.71

0.2820.84

37.560.430.320.0963.02

29.560.66

98.82

7.28.5

126010

116

97174

411.2

2.0

1.67

12.70.152.371.240.420.3880.83

43.700.420.300.1062.75

34.45

99.67

9.70.132.01

0.2490.62

47.770.320.290.1140.50

37.60

100.51

16.200.142.631.140.490.1800.84

41.880.360.340.0992.55

31.491.71

100.05

9.510

11905.5

78

77139

3.89.4

1.63

18.650.163.252.520.700.1731.44

37.330.350.410.1043.43

28.072.21

98.79

11206.0

8615.287

1891.74.97.0

2.5

14.00.132.201.580.540.1531.23

44.610.870.130.1000.65

30.693.07

99.95

11706.5

55

1585

2.310.3

2.5

1.63

17.60.132.251.160.710.1404.47

41.640.260.040.1031.65

21 Al1.30

98.92

162.5

821130

5.55212.66

380.84.5

11.08.91.4160.4680.2901.2220.240

8.30.131.891.940.410.2141.62

45.790.150.270.1221.71

35.621.33

99.49

660.9

51780

3.6308.5

4948.5

3.06.34.61.086

0.2420.8510.171

7.80.131.712.71

0.2261.84

47.840.200.310.1660.5

37.37

100.81

carbonate-free basis [CFB], Ni content = 31-798 ppm),Co concentration varies from 4.6 to 58 ppm (28-257 onCFB), and zinc ranges from 48 to 223 ppm (295-989 onCFB). In all but the Mn-rich oxidized layers of themounds sediments, the distribution of the elements ismore homogeneous and varies more than two to fourtimes. The most homogeneous distribution of minor ele-ments is in the off-mounds area and in the low heat-flowareas.

The increase in the homogenization of elements in thepelagic oozes from the mounds to the off-mounds andto the low heat-flow areas is attributed to the decreasedinfluence of hydrothermal activity.

Data on minor element to alumina ratios (Figs. 8, 9)and the absolute masses of minor elements in the smec-tites and the pelagic oozes (Table 10) suggest an increaseand redistribution of a number of elements within thesediments studied. There are significant concentrationincreases in Ni, Co, Zn, Sb, Cu, and Cs relative toalumina in the upper manganese crusts (Fig. 8) in com-parison with pelagic ooze from the low heat-flow areas.Elements concentrated in minor amounts in the oxidizedpelagic ooze layer (Li, Rb, and Cr) and REE to aluminaratios are close to low heat-flow area sediments.

In the lower crusts the list of enriched elements isclearly different. Rb and Cr are enriched relative to alu-mina while the transition, rare-earth elements, and Seare not.

The Fe-Mn oxide sediments have a significant enrich-ment in Ni, Co, Zn, and Cu relative to alumina as wellas upper crusts and oxidized layers, but higher ratios ofRb and Cr to alumina are observed when comparing Fe-Mn oxide sediments to the pelagic oozes.

Minor element to alumina ratios (Fig. 9) and masscalculations (Table 10) in the smectites indicate a greatincrease of Li, Rb, Cs, and Sb in these sediments. De-pletion of Sr, Ni, Co, Cu, and Zn in mounds sedimentsare noted and observed both in the smectites and in thepelagic oozes which alternate with smectites. Redistribu-tion and local fluctuations of these elements are ob-served in the smectites which are mixed with pelagicoozes and in the basal pelagic oozes.

CONCLUSIONSSome conclusions follow from the data that we have

presented:1) Mn crusts and alumina-poor and iron-rich smec-

tites differ in major and minor element distribution

284


Table 6. Major and minor elements in sediments from Hole 510,DSDP Leg 70.

^ \ Sample

Components

2-2,145-149 cm

. no. 54


SiO2

TiO2A12O3

Fe2O3

FeOMnOMgOCaONa2OK2OP2O5H2O +

co2LOI

Total

27.050.112.100.960.610.2250.94

35.100.320.310.1023.31

27.131.04

99.31


LiRbCsSrCoNiCrCuZnSbSeLaCeSmEuTbYbLu

12.5141.2

10908

1321681

1981.04

10.15.91.1120.2650.2760.8500.135

2-6,80-84 cm

no. 56

20.60.132.500.950.550.2010.69

38.740.470.380.1162.71

30.680.72

99.44

912.5

12506.7

74

74136

—5.5

16.4—3.2

__1.76

—

5-5,133-137 cm

no. 60

23.400.101.230.290.610.1000.46

39.210.290.200.0402.90

30.780.60

100.21

6.56.5

12005.2

51

7793

3.49.8

2.3

1.4—

8-2,73-78 cm

no. 63

23.60.122.490.760.530.1581.05

37.330.320.270.1702.81

28.621.50

99.73

22100.9

11905.9

4514.263

1060.53.77.66.21.770.4110.3581.0740.181

from the pelagic sediments and surely have a differentorigin, but both must be attributed to hydrothermal ac-tivity.

2) A significant additional input of a number of ele-ments (Si, Fe, Mn, K, Rb, Li, etc.) to the mounds loca-tion are evident. There is a strong similarity between thislist of elements and those that enriched the thermal solu-tion sampled at the axis zone of Galapagos Rift (Ed-mond, Measures; McDuff, 1979; Edmond, Measures,Mangum, 1979).

3) The data strongly suggest interaction of pelagicooze with hydrothermal solutions. This interaction leadsto intensive redistribution of elements and to subtrac-tion of some elements from mounds sediments (Ni, Co,Cu, REE, Sr, etc.). These are the elements that are de-pleted in thermal solutions from the axis zone of Gala-pagos Rift.

4) Hydrothermally derived solutions rich in elementsappear to differentiate in the mound section. This mayresult from a redox gradient encountered during the up-ward percolation of hydrothermal solutions through thesediment cover.

5) The similarities in element distribution between up-per Mn crusts and Fe-Mn oxides and the off-mounds up-per oxidized layers of pelagic oozes (enrichment of Ni,Co, Cu, Zn, etc.) reflect the influence of seawater ontheir composition.

In light of the foregoing disscussion, it appears thatthe mounds form as the result of percolating solutionsrich in Si, Fe, and Mn which emanate from the basementthrough fracture zones and react with seawater and withthe overlying sediment layer. Available data give us noopportunity to resolve the problem of whether the inputof iron was synchronous with that of manganese orwhether the elements arrived separately. A great deal ofdata indicate a manganese influx to the water columnabove the Galapagos Spreading Center (Bolger et al.,1978; Lyle, 1976; Klinkhammer et al., 1977; Edmond etal., 1979 a,b). At the same time observations show thatsignificant iron input also takes place and that Fe and Mnfractionation occurs during oxidation in the seawater(Gordeev et al., 1979).

The occurrence of the Fe-Mn oxide sediment in themounds sections is assumed to reflect incomplete separa-tion of these elements, and the correlation of the redoxgradient with Mn and Fe separation suggests the possibil-ity of synchronous iron and manganese input.

Data on different ferrous iron masses at differentlevels in the mounds section show that iron arrives atthese levels in solution and is incompletely oxidized. Thisassumption agrees well with data from Edmond et al. onthe composition of hydrothermal solutions from the axisof the Galapagos Rift. From the data presented aboveand other work (Honnorez et al., 1981), it appears thatinfiltration of oxygen-bearing waters into a mound is oneof the main oxidizing factors. Proton formation and car-bonate dissolution follow from the reaction of ferrousiron oxidation and its combination with SiO2 and H2O. Itis obvious that carbonate dissolution leads to a large-scale redistribution of elements. It is likely that most dis-solved and redistributed elements are washed out ofmounds sediments by circulating water. Large amountsof oxygen are required for the precipitation of manga-nese in the form of todorokite, which was. fixed in thecourse of manganese and iron separation in the upperlayers of the mounds sequence. The uppermost crusts inoxidized pelagic ooze, which were subjected to contactwith seawater, have a similar elemental pattern as the pe-lagic oxidized layer. In contrast, the lower manganesecrusts have the characteristics of rapidly accumulatingmanganese formations (Moor and Vogt, 1976) and aresimilar to the smectites from the standpoint of minor ele-ment distribution. Fe-Mn oxide sediment is assumed tohave accumulated rapidly as a result of the simultaneousoxidation of both elements. Its composition demon-strates seawater influence. Direct deposition of this sedi-ment from open seawater is possible.

Comparison of the different types of Galapagos hy-drothermal mounds sediments with other products ofhydrothermal activity in oceans and ophiolite zones ofcontinents is interesting. The distribution of REE, re-flecting the influence of different source material, canbe chosen for this comparison. In Figure 16 is shown

285


Table 7. Average composition of hydrothermal and pelagic sediments (carbonate-free basis) in mounds and off-mounds sedimentary sections, Site507.

Hole

Lithology

Component

UpperMn

Crust

0-0.2


SiO2

Tiθ2A12O3

F e 2 O 3

FeOMnOMgONa 2 OK 2 O

P2O5

26.890.213.907.50

51.15a

2.752.001.030.26


LiRbCsCrCoNiCuSbZnSeLaCeSmEuTbYbLuNo. of samples

108

143228

759199

17482

5.8013.4014.602.450.490.522.060.382

5O7A

LowerMn

Crust

1.2

2.500.070.514.1

48.76a

0.954.260.370.22

_

611.404010.551441250

1.705.404.600.740.270.210.560.101

UpperSmectites

1.25

48.950.050.85

35.520.420.5153.091.321.360.10

1025

—

313.60

1036

1.292

2.805.805.600.980.44

—

0.950.171

5O7A, F

MiddleSmectites

1.9-5.2

52.600.080.53

34.940.330.0593.611.821.530.066

642

1.3100.74.8

190.7

290.61.27—

0.930.210.190.450.073

LowerSmectites

12.8-19.1

53.180.080.36

32.150.880.0454.441.322.770.068

662

1.820

0.606.209.501.20

235.00.741.470.170.11

—

0.210.062

507A

Fine-GrainedSmectite

Mixed withPelagic Ooze

Al-Smectites

Sub-bottom Depth (m)

20.0

48.810.345.16

23.912.770.435.320.681.310.27

2047

—

5512.3057

1311.60

25512.3037.8025

6.501.70—3.480.501

15.3

57.300.41

10.6813.14

—0.1383.242.202.610.112

4090

3.4371276

1531.8

39012.020.433.50

4.180.9630.6391.930.3561

LowerPelagic

Ooze

38.2

50.20.649.0

14.63.251.224.701.741.970.38

—

7.5016228

122296

2.9354

1883.958.30

9.522.142.046.461.011

507F

UpperOxide Layer

of PelagicOoze

0.0-0.7

53.930.397.536.66—8.802.651.931.000.38

2129

6.604533

——3.0

67814.5030.4018.904.801.301.183.830.631

MoundSlope

PelagicOoze

2.5-23.6

54.040.499.05

11.321.890.805.561.511.300.32

3342——

26306198

2.1795

1647.330.5

5.20.32—4.830.303

507C, H

Off-MoundPelagicOoze

2.1-31.1

59.860.64

10.264.842.780.703.181.401.450.30

3242

4.04625

280253

5.4498

1840.5038.60

5.301.71—4.130.704

Mn as MnO 2 .

North American shale normalized REE distributions inhydrothermal products. The lowest concentrations ofREE are characteristic of both vein smectites from ba-saltic basement and some samples of Galapagos moundsmectites. REE patterns of this product are shown inFigure 16 (no. 1 and 3) and are assumed to be attribu-table to the hydrothermal source which produces theseminerals. The same relationships are typical for someoceanic and ophiolitic formations (Fig. 16, no. 2, 4).These formations do not have any REE additions fromthe depositional environment. In contrast, the interac-tion of hydrothermally derived formations during theirdeposition with the surrounding environments tends tochange initial REE distribution. For example, much ofGalapagos hydrothermal smectites as well as lower man-ganese crusts have distinct cerium depletion and slightincreases in REE concentrations (Fig. 16, no. 5-7). Inthe upper crusts of the Galapagos mounds, REE con-centrations are further increased and the pattern of theirdistribution demonstrates a further similarity with sea-water (Fig. 16, no. 9). The same concentrations and pat-terns in REE (Fig. 16, no. 8) were observed in the hydro-

thermal pyrite that was oxidized by seawater near theMid-Atlantic Ridge (Bonatti, Honnorez-Guerstein et al.,1976).

The maximum REE content and the typical seawaterdistribution pattern are reported from the East PacificRise metalliferous sediments and Cyprus ochres andumbers (Piper and Graef, 1974; Piper, 1974; Robertsonand Fleet, 1976), which are supposed to be precipitateddirectly from open seawater (Fig. 16, no. 10, 11).

It appears that hydrothermally derived sedimentsfrom active ocean zones are close to ophiolitic zones inREE distribution. It is also suggested that the interac-tion of hydrothermal solutions with the surrounding en-vironments (e.g., seawater, pore water, sediments) leadsto the increase in REE concentration in the resultingproducts and to changes in the initial REE distributionpatterns.

This comparison also suggests that both replacementof sediments and minor scale infilling of voids withoutsignificant interaction with the sediments has takenplace during the formation of smectite in the Galapagosmounds.

286


Table 8. Average composition of hydrothermal and pelagic sediments (carbonate-free basis) in mounds and off-mounds sedimentary sections, Site509.

Lithology

Component

UpperMn

Crust

0-1.20


SiO2

TiO2

A12O3

Fe2O3

FeOMnOMgONa2OK2OP2O5

14.400.213.123.71—

55.37a

3.081.970.660.192



8.1107.7

4623.8

19814314.8

2118.2

13.612.52.110.640.591.990.295

LowerMn

Crust

1.20-2.70

1.550.080.710.50—

78.34a

3.421.481.140.118

<27—

701460279

371.65.34.991.10.1070.1410.5030.0611

Fe-MnOxide

Sediment

6.16

23.650.080.82

17.76

35.0Oa

3.561.291.110.187

810

4015684032125.2

6906.25.04.520.85

0.610.091

UpperSmectites

2.7-5.6

51.350.162.21

29.68

0.723.901.411.760.094

1036

25

9.8500.5

561.00.50.840.140.0330.060.135

2

Hole 509B

MiddleSmectites

8.2-11.8

52.780.080.57

31.400.850.303.951.023.180.056

664

2049.5

18

226.0

2

LowerSmectites

Fine-GrainMixed

SmectitesAl-

Smectites


12.2-20.0

52.680.121.28

29.341.190.1064.220.823.400.049

1081

25<3.7

< 1735.60.85

524.24.66.61.28

0.740.1334

18.1

49.750.162.27

24.411.394.7545.410.813.460.088

19115

1523.693.6

1192.9

2483.8

10.3

1.60

1.42

2

9.2

53.450.427.68

19.660.850.2124.641.101.940.125

2650

486.4

50126

1.0217

1110.57.82.810.6550.5521.390.2381

PelagicOoze in

HydrothermalUnit

5.8-18.1

57.670.499.299.681.223.693.691.691.840.23

3042

3.33628.9

296256

4.2480

13.244.441.1

8.402.081.594.881.094

LowerPelagicOoze

23.0-33.0

52.740.70

11.436.973.031.865.091.431.760.446

31.757.5

—98

109336277

1.861021.459.944.811.82.111.877.450.663

Hole 509

OxidizedPelagicOoze

0.09

55.840.427.007.261.16

11.933.051.240.940.41

27387.6

7948

526346

6.2671

13.63628.46.21.290.883.900.622


5.9-29

57.280.69

10.285.392.711.623.461.521.270.42

36455.0

6428

377382

4.6482

19.634.330.46.21.541.575.351.026

a Mn as MnO2.

ACKNOWLEDGMENTS

The authors gratefully acknowledge Drs. P. Borella, J. Honnorez,Yu. Balashov, and G. Zakariadze for discussions and critical reviews.We also express our appreciation to Rosemary Amidei and Drs. P.Borella and A. Sharaskin for their invaluable assistance in improvingthe English version of this manuscript.

REFERENCES

Bender, M., Broecker, W., Gornitz, V., et al., 1971. Geochemistry ofthree cores from the East Pacific Rise. Earth Planet. Sci. Lett., 12:425-433.

Bogdanov, Ju. A., Lisitzin, A. P., and Migdisov, A. A., 1979. O gen-ezise metallonosnich osadkov. In Smirnov V. I. (Ed.), Metal-lonosnye Osadki Yugo-Vostochnoi Chasti Tihogo Okeana: Mos-cow (Nauka), pp. 249-276.

Bolger, G. V., Betzer, P. R., and Gordeev, V. V., 1978. Hydrother-mally derived manganese suspended over the Galapagos SpreadingCenter. Deep-Sea Res., 25:721-733.

Bonatti, E., 1975. Metallogenesis at oceanic spreading centers. Ann.Rev. Earth Planet. Sci., 3:401-431.

Bonatti, E., Honnorez-Guerstein, M., Honnorez, J., and Stern, C ,1976. Hydrothermal pyrite concentrations from the RomancheTrench: Metallogenesis in oceanic spreading centers. Earth Planet.Sci. Lett., 32:1-10.

Bonatti, E., Zerbi, M., Kay, R.. et al., 1976. Metalliferous depositsfrom the Appennine ophiolites: Mesozoic equivalents of moderndeposits from oceanic spreading centers. Bull. Geol. Soc. Am., 87:83-94.

Bostròm, K., and Petersen, M., 1969. The origin of aluminium-poorferro-manganoan sediments in areas of high heat-flow on the EastPacific Rise. Mar. Geol., 1 All-Ml.

Corliss, J., Dymond, J. R., Gordon, L., et al., 1979. Submarine ther-mal springs on the Galapagos Rift. Science, 203:1073-1083.

Cronan, D., 1976. Basal metalliferous sediments from the Eastern Pa-cific. Bull. Geol. Soc. Am., 87:928-934.

Dymond, J., Corliss, J., Cobler, R., et al., 1980. Composition andorigin of sediments recovered by deep drilling of sediment mounds.In Rosendahl, B. R., Hekinian, R., et al., Init. Repts. DSDP,54: Washington (U.S. Govt. Printing Office), 377-385.

Edmond, J., Measures, C , McDuff, R., et al., 1979. Ridge crest hy-drothermal activity and the balances of the major and minor ele-ments in the ocean: The Galapagos data. Earth Planet. Sci. Lett.,46:1-18.

Edmond, J., Measures, C , Mangum, B., et al., 1979. On the forma-tion of metal-rich deposits at ridge crests. Earth Planet. Sci. Lett.,46:19-30.

Gordeev, V. V., Demina, L. L., and Konkin, A. M., 1979. Jelezo imarganetc gidrotermalnogo proishojdenija v glubinnich vodachTihogo Okeana (Galapagosskaja aktivnaja zona, depressija Hessa).In Smirnov, V. I. (Ed.), Metallonosnye Osadki Jugo-VostochnoyChasti Tihogo Okeana: Moscow (Nauka), pp. 224-236.

Gradusov, B. P., 1971. Zavisimost mejdu znachenijami d/n bazalnihrefleksov i sogerjanijami slyudistih paketov v smeshannosloinihobrazovanijah slyudo-smektitovogo tipa. Litologia i Poleznie Isk-opaemie, 6:152-155.

Hekinian, R., Rosendahl, B., Cronan, D., et al., 1978. Hydrothermaldeposits and associated basement rocks from the Galapagos Spread-ing Center. Oceanol. Acta, 1:473-482.

Hoffert, M., Perseil, A., Courtois, C , et al., 1980. Sedimentology,mineralogy and geochemistry of hydrothermal and sedimentarydeposits from Holes 424, 424A, 424B, 424C. In Rosendahl, B. R.,Hekinian, R., et al., Init. Rept. DSDP, 54: Washington (U.S.Govt. Printing Office), pp. 339-376.

287


Hoffert, M., Perseil, A., Hekinian, R., Choukroune, P., et al., 1978.Hydrothermal deposits sampled by diving saucer in transformfault " A " near 37°N on the Mid-Atlantic Ridge, FAMOUS area.Oceanol. Acta, 1:73-86.

Honnorez, J., Von Herzen, R. P., Barrett, T. J., Becker, K., Bender,M. L., Borella, P. E., Hubberten, H.-W., Jones, S. C , Karato,S., Laverne, C , Levi, S., Migdisov, A., Moorby, S., Schrader, E.L., et al., 1981. Hydrothermal mounds and young ocean crust ofthe Galapagos: Preliminary Deep Sea Drilling results, Leg 70.Bull. Geol. Soc. Am., 92:457-472.

Klinkhammer, J., Bender, M., and Weiss, R. F., 1977. Hydrothermalmanganese in the Galapagos Rift. Nature, 269(5626):319-320.

Lonsdale, P., 1977. Deep-tow observations at the mounds abyssal hy-drothermal field, Galapagos Rift. Earth Planet. Sci. Lett., 36,92-110.

Lyle, M., 1976. Estimation of hydrothermal manganese input to theoceans. Geology, 4(12):733-736.

Lynn, D., and Bonatti, E., 1965. Mobility of manganese in diagenesisof deep sea sediments. Mar. Geol., 3:457-474.

Migdisov, A. A., Bogdanov, Y. A., Lisitzin, A. P., et al., 1979. Geo-himija metallonosnyh osadkov. In Smirnov, V. I. (Ed.), Metallo-nosnye Osadki Yugo-Vostochnoi Chasti Tihogo Okeana: Moscow(Nauka), pp. 122-200.

Migdisov, A. A., Girin, Yu, P., Galimov, E. M., et al., 1980. Majorand minor elements and sulphur isotopes on the Mesozoic andCenozoic sediments at Sites 415 and 416, Leg 50, DSDP. InLancelot, Y., Winterer, E. L., et al., Init. Repts. DSDP, 50: Wash-ington (U.S. Govt. Printing Office), 675-689.

Migdisov, A. A., Miklishansky, A. Z., Saveliev, B. V., et al., 1980.Neutron activation analysis of rare earth elements and some traceelements in volcanic ashes and pelagic clays, DSDP, Leg 59. InKroenke, L., Scott, R., et al., (Ed.), Init. Repts. DSDP, 59: Wash-ington (U.S. Govt. Printing Office), 653-668.

Moor, W. S., and Vogt, P. R., 1976. Hydrothermal manganese crustsfrom two sites near the Galapagos Spreading Axis. Earth Planet.Sci. Lett., 29:349-356.

Natland, J., Rosendahl, B., Hekinian, R., et al., 1979. Galapagos hy-drothermal mounds: Stratigraphy and chemistry revealed by deepsea drilling. Science, 204:613-616.

Piper, D. Z., 1974. Rare earth elements in ferromanganese nod-ules and other marine phases. Geochim. Cosmochim. Acta, 38:1007-1022.

Piper, D. Z., and Graef, P., 1974. Gold and rare-earth elements insediments from the East Pacific Rise. Mar. Geol., 17:287-297.

Piper, D. Z., Veeh, H., Bertrand, W., et al., 1975. An iron-richdeposit from the north-east Pacific. Earth Planet. Sci. Lett., 26:114-120.

Rateev, M., Timofeev, P., and Rengarten, N., 1980. Minerals of theclay fraction in Pliocene-Quaternary sediments of the East Equa-torial Pacific. In Rosendahl, B. R., Hekinian, R., et al., Init.Repts. DSDP, 54: Washington (U.S. Govt. Printing Office),307-318.

Robertson, A., and Fleet, A., 1976. The origin of rare earth in metal-liferous sediments of the Troodos Massif, Cyprus. Earth Planet.Sci. Lett., 28:385-394.

Rosendahl, B., Hekinian, R., et al., 1980. Init. Repts. DSDP: Wash-ington (U.S. Govt. Printing Office).

Sayles, F., and Bischoff, J., 1973. Ferromanganoan sediments in theEquatorial East Pacific Rise. Earth Planet. Sci. Lett., 19:330-336.

Schrader, E., Rosendahl, B., Furbish, W., and Mattey, D., 1980. Min-eralogy and geochemistry of hydrothermal and pelagic sedimentsfrom the mounds hydrothermal field, Galapagos Spreading Cen-ter, DSDP, Leg 54. /. Sediment. Petrol., 50:917-928.

Sclater, J., and Klitgord, K., 1973. A detailed heat flow topographicand magnetic survey across Galapagos Spreading Center, at86°W. /. Geophys. Res., 78:6951-6975.

Sedyh, E. M., and Beliaev, Yu. I., 1980. Ustranenie vliyania osnovipri elektrotermicheskom atomno-absorbcionnom opredelenii Pb.Ag. Jurnal Analiticheskoj Himii, 12:2348.

Sharaskin, A. Ya., Migdisov, A. A., Rostshina, I. A., Miklishansky,A. Z., in press. Major and trace element chemistry of the Hole504B basalts and their alteration products (Costa-Rica Rift,DSDP, Leg 70). In Cann, J. R., Langseth, M. G., Honnorez, J.,Von Herzen, R. P., White, S. M., et al., Init. Rept. DSDP, 69:Washington (U.S. Govt. Printing Office).

Strakhov, N. M., 1976. Problemi Sovremennogo OkeanicheskogoLitogeneza, Trudy GIN (Vol. 292): Moscow (Nauka).

Turekian, K., and Wedepohl, K., 1961. Distribution of the elementsin some major units of the earth crust. Bull. Geol. Soc. Am.,72:175-192.

Table 9. Average composition of pelagicsediments from low heat-flow areas (car-bonate-free basis).

Component

Hole 508 Hole 510

Sub-bottom Depth(

0.4-32.5


SiO2

TiO2

A12O3

Fe 2 O 3

FeOMnOMgONa2OK 2 O

P 2 θ 5

59.450.539.565.501.691.0864.691.631.180.39



26.523.4

4.54723.8

285212

3.7452

15.531.828.1

3.810.9500.7813.5600.744

10

m)41.5-113.8

71.560.356.252.211.750.5112.341.080.880.32

37.232.3

2.94219.5

223224

2.0337

12.733.823.3

6.571.5461.4373.9570.6494

288

Figure 1. Distribution of Mn, Fe2O3, Ni, Cu, Zn, and Sb, recalculated on carbonate-free basis, Holes 507D, 507F, and 509B.


8" 10 12 14 16Sub-bottom Depth (m)

18 20 22

Figure 2. FeO/Fe2O3 ratio in smectites plotted against sub-bottomdepth in mounds sections. (Circles with dots = Hole 509B; doublecircles = Hole 507D.)

COO

Fo

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

-

-

-

-

0

t

El '

/

B f

_^ ^

i i i i ,

4×ç>> //f J

' • /

>

3 3

oCM

π 1 1 1 1 1 1 1 1 1 r

• i ' I I I L0 2 4 6 8 10 12 14 16 18 20 22


Figure 4. K2O concentration in smectites plotted against sub-bottomdepth in mounds sections. (Open circles = Hole 509B; solid cir-cles = Hole 507D.)

10 20Sub-bottom Depth (m)

30 40

Figure 3. FeO/FejC^ ratio in pelagic oozes in Holes 509B, 509, and508 plotted against sub-bottom depth. (Open squares = Hole 509;open circles = Hole 508; solid circles = Hole 509B.) (Data for thedeepest samples from Hole 508, in and near the lithified layer,were removed.)

290


Sample

66

27

24

9.8 -

° 9.5•o

9.2 -

o\81 >

-

-

1

\

\

\ 27

1

1

-

-

-

66

20 40 60Percent of Smectite Layers

80

Figure 6. Content of smectite layers in analyzed mixed-layered mica-smectite minerals. (Relationships between the d(k) of basal mica(001) and secondary smectite (002) reflections and the content ofsmectite layers are given by Gradusov, 1971.)

81

Figure 5. X-ray diffractograms of < 0.001-mm fractions of hydrother-mal green clays from Holes 5O7D and 5O9B. (Sample numbers inTables 1, 3 and text; a = air-dried samples; b = ethylene-glycolsaturated samples.)

291


0 20 40 (%) 0 20 40 (%) 0 20 40 (%)

I 1 2

14

l 16-9

18

20

22

24

26

28

30Hole507D

Hole508

Figure 7. Variation in content of siliceous microfossils in Holes 507D,507F, and 508 on the basis of smear-slide data. (See site summa-ries, this volume.)

La Yb Se Ni Cu Zn Sb

Figure 8. Element-to-alumina ratios in Mn crusts of mounds and inuppermost oxidized layers of pelagic oozes, normalized to ratios inaverage low heat-flow area sediments, Hole 508. (1 = lower Mncrusts; 2 = Fe-Mn oxide sediment; 3 = upper Mn crusts; 4 = up-permost oxidized layers of high heat-flow area pelagic oozes (off-mounds sediments); 5 = uppermost oxidized layers of low heat-flow area pelagic oozes.)

100-

. _ 5 I Average forHole 508 ooz

Figure 9. Element-to-alumina ratios in mounds smectites and pelagicoozes, normalized to ratios in average low heat-flow sediments,Hole 508. (1 = lower smectites; 2 = middle smectites; 3 = uppersmectites (some contamination by Mn crusts may take place); 4 =lower pelagic sediments in mound sections; 5 = pelagic oozes alter-nating with smectite in mounds sections; 6 = Al-rich smectite.)

292


Table 10. Average absolute masses of elements in mounds, off-mounds, and low heat-flow area sediments.

Hole

\iithology

Components

507A 509B

UpperSmectites

Major element oxides (mg/cm^)

SiO2

TiO2

A12O3

Fe 2O 3

FeOMnOMgONa2OK2OP2O5

Minor elements

LiRbCsSrCrCoNiCuSbZnSeLaCeSmEuTbYbLu

4550.467.91

3303.94.8

28.712.312.60.93

( I 0 " 4 '

93232

—967288

3393

33511.2

8562653.952.19.14.09—8.841.58

4771.5

20.6276

—6.7

36.313.116.40.87

Vo mg/cm•')

93335

—1469233

—91

4654.6

52194.77.81.30.310.561.25

—

507A 509B

LowerSmectites

8271.245.6

50013.70.70

69.020.543.1

1.1

93964

28560311

996

14818.7

3587811.522.92.61.71—3.270.93

4901.1

11.9273

11.10.99

39.27.6

31.60.46

93753—

1609233

<34< 158

3317.9

4843942.861.411.9——

6.901.20

507A 509B

MixedSmectites

83.00.588.79

40.64.70.739.11.22.20.45

3380—

4778932198

2222.7

4352164.342.611.12.84

—5.900.84

2450.80

11.2120

10.123.626.64.0

17.00.44

94566

_3842

73116

4606587

14.51218

1950.8

8.0

7.0—

5O7D 509B

Al-RichSmectites

96.80.69

18.0522.2—0.235.53.74.40.19

68152

5.7456

6320

128259

3.04659

2034.556.67.11.631.083.260.60

1871.47

26.968.8

3.00.74

16.23.96.80.44

91175

—4092

16822

175441

3.5760

3936.827.3

9.82.301.934.860.83

5O9B

PelagicOoze from

HydrothermalUnit

1201.03

19.320.1

2.57.77.73.53.80.48

63887

97147644

454400

6.5999

2891.885.317.36.04.59

10.102.30

507D 509B

MoundLowerPelagicOoze

46.70.598.4

13.63.01.14.41.61.80.36

4149

78278

15127

113275

2.7329

1778.154.3

8.91.991.906.010.94

55.20.76

12.07.33.21.95.31.51.80.49

3361—

6799103113352290

1.8637

2262.647.012.42.211.947.0

—

507C, H 509


1101.17

18.99.05.11.35.82.62.70.55

5876

7

8546

516466

9.9915

3374.671.1

9.73.15—7.601.28

88.81.07

15.98.44.22.55.42.42.00.65

5670

87903

9943

584592

7.1747

3053.2Al A

9.62.402.408.301.60

508

LowHeat-Flow

AreaSediment

1121.00

18.010.33.22.08.83.12.20.73

5044—

8845

536399

7.0850

2959.852.87.21.81.56.71.4

3 U U

800

700

600

- 5 0 0Q.a.2

300

200

100

n

®

-

-

_

® $®1

0 o8>

°o °^

Q/o0 a ®

Φ• i a ß 1 # 1 , 1 ,#

1 » 1 1 1 1

/

/

/ O

8

1 i r . 1

O

fCf

0®

I

I I I 1

1 1 1

-

0-

-

-0

-O O

0.02 0.05 0.1 0.2 0.5 1.0 2.0Mn (%)

5.0 10.0 20.0 50.0

Figure 10. Ni vs. Mn correlation in hydrothermal and pelagic sedi-ments of mounds areas. (Open circles = off-mounds and crossedcircles = mounds pelagic sediments; large open circles = Mn crustsand Fe-Mn oxide sediment; solid circles = Fe-rich smectites.)

100 200 300 400 500 600 700 800

Ni (ppm)

Figure 11. Sb vs. Ni correlation in hydrothermal and pelagic sediments(symbols as in Fig. 11).

293


6 0 0 -

5 0 0 -

4 0 0 -

N 300-

200-

100-

0.0 1.0 2.0 3.0 4.0 5.0 10.0Fe (%)

Figure 12. Zn vs. Fe correlation in manganese crusts and in Fe-Mnoxide sediment. (Open circles = Hole 507D; solid circles = Hole509B.)

1.0

0 . 5 -

0.2 -

0.1 -

0 . 0 5 -

0.02 -

0.01

1 1

SJ

I i

' , —O

/ y—

/

I I I 1

Sample I

9— — - Smectites

o o Mn Crusts

1 i

1314

16

-

La Ce Nd Sm Eu Tb Yb Lu

Figure 14. Distribution of rare earth elements in hydrothermal sedi-ments, Hole 5O7D, normalized on North American shale (NAS)distribution.

70

60

50

Q.

-£40cc

30

20

10

0

-

•

-

-

-

oO

θo°

o

o

o o|o

0 8 .

OO0 °

i

o

o .°o°o• .

I

90

•

o #

o

• I

115

-

-

-

-

-

Figure 13. Rb vs. K2O content in hydrothermal and pelagic sediments(symbols as in Fig. 11).

1.0

0.5

0.2

0.1

0.05

0.02

0.01

Sample

,-•66

— - Smectites

0 Mn Crusts

La Ce Nd Sm Eu Tb Yb Lu

Figure 15. Distribution of rare earth elements in hydrothermal sedi-ments, Hole 509B, normalized on North American shale (NAS)distribution.

294


0.01La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Yb Lu

Figure 16. NAS-normalized distribution of rare earth elements in hy-drothermally derived products from ocean spreading zone and con-tinental ophiolite sediments. (1 = smectites with low rare-earth ele-ment concentrations from Galapagos hydrothermal mounds (aver-age of three samples); 2 = Fe-rich hydrothermally derived productfrom northeast Pacific [Piper et al., 1975]; 3 = vein smectites fromHole 504B basalts, Costa Rica Rift (Sharaskin et al., in press);4 = Mn-enriched sediments, Malinelo deposit, Italy [Bonatti, Zer-bi, et al., 1976]; 5 = Fe-Mn oxide sediment from Galapagosmounds; 6 = lower Mn crusts of Galapagos mounds; 7 = Fe-richsmectites of Galapagos mounds; 8 = oxidized hydrothermal pyriteconcretions from Romanch deep (Bonatti, Honnorez-Guerstein, etal., 1976); 9 = upper manganese crusts of Galapagos mounds;10 = metalliferous sediments from the East Pacific Rise [Piper andGraef, 1974]; 11 = metalliferous sediments of Troodos massif,Cyprus [Robertson and Fleet, 1976].)

300

2 5 0 -

0.2 0.5 1.0FeO (%)

2.0 3.0 4.0 5.0

Figure 17. Zn vs. FeO content in mounds smectites carbonate-freebasis. (Open circles = Hole 5O9B; triangles = Hole 507D.)

295

13. MAJOR AND MINOR ELEMENTS IN HYDROTHERMAL AND … · MAJOR AND MINOR ELEMENTS IN HYDROTHERMAL...

Documents

Transcript of 13. MAJOR AND MINOR ELEMENTS IN HYDROTHERMAL AND … · MAJOR AND MINOR ELEMENTS IN HYDROTHERMAL...