related to spreading centers of the eastern …35. mineralogy and geochemistry of ore minerals from...

35. MINERALOGY AND GEOCHEMISTRY OF ORE MINERALS FROM BASALTSRELATED TO SPREADING CENTERS OF THE EASTERN PACIFIC

WITH SPECULATIONS ON ORE-FORMING PROCESSES

Ed L. Schrader1 and W. J. Furbish, Department of Geology, Duke University, Durham, North Carolina

ABSTRACT

Two opaque phases, Ti-magnetite and pyrite, are ubiquitous inthe basalts retrieved from the eastern Pacific on DSDP Leg 54. Addi-tional, but minor, phases observed in reflected light include pyrrho-tite, chalcopyrite, ilmenite, hematite, goethite, spinel, and amor-phous manganese oxides.

The dominant mode of occurrence for the Ti-magnetite and allsulfide phases is as grains, crystals, or spheres enclosed in basalt glassor interstitial to crystalline silicates. Ti-magnetites are present mainlyas skeletal crystals whose grain size appears directly proportional tothe degree of undercooling of the lava. The sulfides occur generallyas monomineralic or polymineralic spheres apparently separated im-miscibly from the silicate melt. Sulfide veining and replacement ofsilicates and primary oxides takes place in several samples and sug-gests a secondary (e.g., hydrothermal) period of ore mineral deposi-tion. The veins are dominantly pyrite, with very minor chalcopyriteand/or pyrrhotite in a few cases.

The samples studied and published research results support theconcept of spreading centers as loci for certain types of metallogene-sis. Ore bodies related to divergent plate margins may include mas-sive sulfides, stratiform complexes, and hydrothermal manganeseoxide deposits, all of which are related to the types of magmatic andsecondary processes which affected the basalts recovered on Leg 54.

INTRODUCTION

Increasing interest in the formation of ore depositsalong active plate margins (Schrader et al., 1977; Bonat-ti et al., 1976; Keays and Scott, 1976; Rona, 1978) hasstimulated research into the chemistry and petrographyof opaque phases in oceanic basalts. The diverse tec-tonic settings drilled in the eastern Pacific on Leg 54 ofthe Deep Sea Drilling Project provide an unusual oppor-tunity for the study of ore mineral genesis near oceanicspreading centers. The findings of Schrader et al.(1977), Schrader et al. (Picritic basalts etc., this volume),and the present study lead us to speculate upon the ori-gin of mafic cumulate ores, Co-Ni sulfide ores associ-ated with hybrid basalts, and low-temperature Fe-Mnoxides. All of these deposits may be associated with thetectonic processes of magmatic upwelling, positioningof quasi-static axial magma reservoirs beneath oceanridges, and the actual process of sea-floor accretion.

A variety of opaque phases occur in the basaltsrecovered on Leg 54. The dominant ore minerals arepyrite and Ti-magnetite. Skeletal to subhedral magne-tites have a size range of < 1 µm to > 100 µm. Pyrite oc-curs as disseminated sulfide globules in basaltic glass

1 Present address: Department of Geology and Geography, TheUniversity of Alabama, University, Alabama.

and as the main component of secondary sulfide veins.Additional sulfides in the globules are pyrrhotite, chal-copyrite, and Co-Ni phases. Chalcopyrite and pyrrho-tite are the only sulfides other than pyrite found in theveins. Other ore minerals in the basalts are spinel, il-menite, hematite, and goethite. It is evident from the di-verse mineralogy discovered in these basalts that a vari-ety of magmatic, deuteric, weathering, and hydrother-mal events were responsible for the formation andgrowth of these ore phases.

PETROLOGIC AND CHEMICAL RESULTSTable 1 details the size, occurrence, and relative

abundance of the ore minerals in the 154 samples westudied. Both shipboard and land-based opaque micro-scopic identifications, X-ray diffraction examinations,and electron microprobe analyses were made to obtainthese data. Samples were selected from basalts from sixsites cored near the East Pacific Rise crest in the vicinityof the Siqueiros fracture zone, and from two sites nearthe Galapagos Rift (Figure 1).

Representative electron microprobe analyses of se-lected silicates and other phases are presented in Table2. Of the analyzed phases, all of the silicates werephenocrysts or agglomerated grains (clinopyroxenes);chalcopyrite and iron oxide grains occurred as vein fill-ings in a silicate matrix.

789

E. L. SCHRADER, W. J. FURBISH

TABLE 1Occurrences of Ore Minerals in Leg 54 Drill Holesa

Sample(Interval in cm) <20 µm

Ti-Mt>20µm

Pyrite

Spheres Other Cpy Po Gt Sp Co-Ni

Hole 420

13,CC, 50-5714-1,0-814-1, 10-1715-1, 15-1915-1,22-2715-1, 69-7216-1, 2-10

Hole 4212-1, 12-152-1, 15-172-1, 19-243-1, 28-353-1, 71-753-1, 102-1053-1, 115-1223-1, 145-1484-1, 4-74-1, 32-36

Hole 4227-1, #27-1, #47-1, #7E7-1, #117-2, #38-5, #58-5, #149-1, #4C9-2, #69-3, #6C9-4, #3B9-4, #13A9-5, #610-1, #3

Hole 424

4-6, 112-1165-1, 9-145-1,31-345-1,59-675-1, 126-1295-2, 90-935-3, 22-255-4, 10-136-1, 29-326-1, 50-536-1, 107-1106-1, 138-1406-2, 32-356-2, 107-1096-3, 21-247-1,0-107-1, 13-15

Hole 424A

3,CC, 10-134-1, 3-54-1,45-47

CCCCccc

ccc—ccc——-

c—-——————————c

cVR

cc—————c—————c

VR

cC-VR

C

/R--__——

-—C——-

ccc

cccccccccccc—

——-ccccc-ccccc——

—

ccc

VRVRR*VR

VRVR-

VRR-CR-CR-CVRRR

VRRRRRRRRRRRRR

VR

CVRC-RC-RC

cR-CR-CR-CC

R-CC

cccR

VR

CC-VR

C

VRVR_R

VRVR

RR

VR—

VRVRRR

R-CC-R

RR-CR-CR-CR-CR-CC

cc

R-CR-CR-CR-CR

_VRVRVRVRVRRR

VRR

R-CR-CR-CR_—

VR_

VR

—VR—___—

——_—_——_—

VR—_——

VR———

VR-——

_————-————

VR_—__—

—

——————

—————————

VR_

_—————————

VR_—__——_———

—VR———

—

VR

VR

VRVR

VR

VR

VR

790

EASTERN PACIFIC ORE MINERALS

TABLE 1 - Continued

Sample(Interval in cm) <20 µm

Ti-Mt Pyrite>20µm Spheres Other Cpy Po Gt Sp Co-Ni

Hole 424B5-1, 10-145-1, 44-485-1,67-705-1, 116-1195-2, 1-65-2, 52-566-1, 0-76-1, 110-1206-1, 7-156-1, 15-236-1, 30-356-1, 55-636-1, 75-806-1, 95-1016-1, 101-1106-1, 138-144

Hole 424C

2-1, #22-1, #42-1, #62-1, #73-1, #13-1, #23-1, #4

Hole 4257-1, 43-477-1, 81-857-1, 129-1317-2, 4-67-2, 52-547-2, 55-597-2, 60-637-2, 115-1188-1, 30-358-1,44-478-1, 70-738-1,93-968-1, 105-1108-1, 130-1348-1, 145-1489-1, 26-299-1, 59-669-1, 99-1019-1, 106-1109-2, 67-699-2, 107-1109-2, 145-1479-3, 3-59-3, 62-659-3, 69-72

Hole 427

9-1, 0-79-2, 22-259-4, 29-319-4, 137-1399-5, 78-8010-1,64-6710-2, 12-1410-3, 102-10610-4, 53-5511-1,27-3011-1,47-49

VR-RC—

c—c

VRR

VRC—C

c—c-

cc

VR-CCC

c

cccc————-——-————ccccc-——-

c_—-————--c

_—c-c——-——c——c—c

_-———c-

_-——cccc

C-R

ccccccc———-—cccc

_ccccccccc_

VRR-CR

VRRR_

—VRVRC

VRVR-R

RC

R-CR-

VRC

cR-C

VRCCR_RRRRCC

R-CR

VRVRVRVRVR___

VRVRRR

_RRRRRRRRR

VR

_VRVRVRR

VRVR_

VRVRVR_

VRVR_-

VRVRVR

VR-C_

VRVR

R-CVRRC

VRR

R-CR-C

RRRR

R-CVRR-C

RVR

VR-CVRVRR

R-CRRR

RR-CR-CR-CR-CR-CR-CR-CR-CR-CVR

— —

_ _— _— —

_ __ ___ _

_ __ __ _-

_ __ _— —_ _

_ _-

_ _— _— —

VR

_ —

— —— —_ __ _— —— —— —

VR_ __ _— —— —— —

VR— —_ —_ _-

_ _VR VR_ __ _

— —— —— __ __ __ _

VRVR

VRVRVRVR

VR

VR

VR

VR

VR

VR

VR

VR

VR

791


TABLE 1 - Continued

Sample(Interval in cm)

Hole 428

5-3, 123-1265-4, 15-185-4, 36-395-4, 81-846-1, 18-216-1, 74-796-1, 127-1306-2,10-126-2, 116-118

Hole 428A

1-1, 21-231-1, 92-941-2, 4-71-2, 31-321-3, 65-681-4, 2-41-4, 40-431-4, 70-722-1, 27-302-1, 85-873-1, 20-233-1, 134-1374-1, 50-534-1, 114-1175-1, 21-245-1, 115-1185-2, 12-155-2, 93-955-2,114-1175-3, 31-345-3, 123-1265-4, 89-926-1,52-536-1, 100-1037-2, 6-97-2, 147-150

Hole 429A

1-1, 60-632-1,41-442-1, 125-1283-1, 103-1053-1, 123-1253-1, 128-1302-1,9-112-1, 88-902-2,9-11

Ti-Mt

<20µm

_C-—————R

R-CVRR-C—-—R--CC-----——-————-L

—

VR

CC

c-——

R-CCC

>20µm

C—C

ccccc-

_c—ccc—cc—-ccccccccccccccc

_——ccc——-

Pyrite

Spheres

RCCRRCCR

VR

VRVRR

VRRR—R

VRVRR——-————-——_

VRVRVRVR

VRVRVRCCC

VR_

VR

Other

________-

VRR_RR__R

R-CR—

R-CR-CC

ccccccRRCR_

c

R_____—_

VR

Cpy Po

_ __ —— —— * —— —_ —— —— _-

_ _

_ _— _— _— —_ __ —_ __ _

_ _— _— —— —

VRVR VR_ —— —— —— —

VRVR

_ _VRVR

_ _— —

_ _— —— —-

11

_—————

_-

______—_—————

VR———

—___—_

VRVR

————___—-

Gt Sp Co-Ni

_ _ __ _ __ _ __ _ __ _ __ _ __ _ __ _ _R - -

_ _ __ _ __ _ __ _ __ _ __ _ __ _ __ _ __ _ _

R-C_ _ __ _ __ _ _

VR_ _ __ _ __ _ __ _ __ _ __ _ __ _ __ _ __ _ __ _ __ _ __

R<:

c_ _ __ _ __ _ __ _ _

R-C VRR

R-VR -

Note:C = common (>2%).R rare (<2%).VR = very rare (<1%).Ti-Mt = titanium magnetite.Cpy = chalcopyrite.Po = pyrrhotite.II ilmenite.Gt = goethite and amorphous, hydrated iron oxides.Sp = spinel.Co-Ni = cobalt and nickel sulfide phases mineralogically indistinguishable.

See Initial Core Descriptions for further information on rock types and grain sizes of silicates.

792

10°N

- j

9°N

SIQUEIROS FRACTURE ZONE

PT 4 SITES(419 423. 426 429)

105°W

Figure 1. Location of Deep Sea Drilling Project for Leg 54.

104°W


TABLE 2Microprobe Analysis Data (wt. %)

for Silicates and Selected Other Phases

Quantitative Microprobe Analyses of SilicatesSample

(Interval in cm) SiO2 CaO Na2O AI2O3 FeO MgO K2O TiO2 MnO

420-15-1, (piece 9)Plagioclase:

CoreRim

421^-1, 32-36Plagioclase:

CoreRim

Clinopyroxene

422-7-1,90-95Plagioclase:

CoreRim

Clinopyroxene

423-5.CCPlagioclase:

CoreRim

Clinopyroxene

424-6-2, 107-109Plagioclase:

CoreRim

Clinopyroxene

425-9-1, 59-66Plagioclase:

CoreRim

427-9-4, 24-31Plagioclase:

CoreRim

Clinopyroxene

428A-7-1, 15-18Plagioclase:

CoreRim

429A-2-2, 9-11Plagioclase:

51.9251.64

14.0613.78

52.82 13.4152.62 13.3850.97 18.58

53.00 12.4055.24 10.6051.13 19.04

53.13 13.2351.23 14.47

51.56 13.6555.53 11.0350.15 19.55

51.01 14.6651.76 13.82

53.26 12.5753.44 12.0251.34 19.09

53.00 12.0957.14 8.65

3.833.26

3.653.750.21

4.28

5.07

0.24

3.953.530.26

3.635.060.17

3.123.68

4.954.350.25

4.455.68

29.0829.78

29.3129.39

3.12

28.8227.36

2.75

29.0630.40

3.58

29.0827.08

2.70

30.0228.99

27.9728.43

1.67

28.2526.16

0.74

0.74

0.260.24

0.84 0.230.84 0.208.37 16.62

0.76 0.160.87 0.108.51 16.77

0.73 0.240.72 0.208.57 17.34

0.71 0.220.78 0.07

10.56 15.32

0.03

0.02

0.02

0.03

0.030.05

0.04

0.03

0.010.02

0.02

0.03 0.02

0.040.080.99 0.22

0.080.06 0.010.91 0.20

0.010.030.83 0.25

0.081.03 0.27

0.60

0.72

0.21

0.23

0.65 0.110.68 0.10

10.12 16.74

0.780.63

0.170.05

0.06

0.04

0.08

0.070.69 0.28

CoreRim

51.8052.08

14.0213.86

3.563.11

29.1029.79

0.63 0.29 0.010.67 0.25 0.05

Semiquantitative Energy-Dispersive Microprobe Analyses (± 5 wt. %)

Sample(Interval in cmii Fe Cu i3 Si Co Ni Mn

0.06

425-7-2, 4-6Chalcopyrite

428A-5-1, 115-118Chalcopyrite

424A-6-l,7-15Fe oxide

428A-4-1,114-117Fe oxide

429-2-1,41-44Fe oxide

420-15-1, 15-19Pyrite sphere

424A-5-1, 67-70Pyrite veinlet

429A-3-1,128-130Pyrite spherePyrite veinlet

33.2 30.2 34.1 - -

32.1 30.4 34.8 - -

39.9 - a

38.2 - a

45.9 - 51.1 -

- a - 14.2

_a _a _a

_a _a _

46.8 - 52.0 - -

45.8 -46.1

54.9 -53.7 -

Average Compositions of Selected Opaque Phases

Fe Co Cu Ni

PyriteChalcopyritePyrrhotite

SpheroidalIntergrowths

Iron-richCopper-rich

Oxides

Ti-magnetitesIlmenite

Spinels

443058

Fe

4540

FeO

7146

FeO

12

0.10.20.2

Co

0.10.3

TiO2

2552

TiO2

0.4

0.332 0.3

Tr

Cu Ni

Tr 0.219 0.7

A12O3

2.00.2

A12O3

42

533639

S

5235

V 2 O 3

0.50.2

Cr2O3

26

(21 analyses)(7 analyses)(13 analyses)

(16 analyses)(12 analyses)

MnO

0.60.5

MgO

20

MgO

0.40.7

(25 analyses)(4 analyses)

(2 analyses)

Element present but comprises less than 5 wt. °,

Of the minerals with average compositions in Table2, ilmenite and pyrite compositions are nearly stoichio-metric; the pyrrhotites fluctuate only slightly in ironcontent; and the Ti-magnetites have a representativecomposition similar to that reported by Ade-Hall et al.(1976). Co-Ni phases (similar to those described byKeays and Scott, 1976) were analyzed as well; theirsmall size, however, prevented obtaining meaningfulquantitative data, and in all cases the surrounding pyrite(or pyrrhotite) was also excited by the electron beam.

CaO and Na2O values of the feldspars listed in Table2 confine most of these phenocrysts to compositions be-tween An60 and An70. The plagioclase phenocrysts arericher in SiO2 and NaO2 than those reported in picritesfrom the Siqueiros fracture zone (see Schrader et al.,this volume). Iron oxide compositions are similar tothose reported by Bonatti et al. (1976), but are lower incopper. The trace amounts of nickel in the chalcopyritemay reflect residual concentrations of this metal in thehydrothermal fluids derived from magmatic crystalliza-tion. However, since the chalcopyrite-pyrite vein inwhich this material is found is of yet undeterminedorigin, the nickel could also represent leaching ofunderlying basalts by chemically active ascending solu-tions (see Bonatti, 1978).

It may be fortuitous that the copper-containing ironoxide (Sample 428A, 4-1, 114-117 cm) stratigraphicallyis within 10 meters of a section containing chalcopyrite.However, the oxidation (hydrothermal or deuteric) ofthe copper-containing portions of overlying basalt mayhave released metallic ions into solution which werefixed in iron oxide fracture fillings below. Although notdirectly correlative, Bonatti et al. (1976) proposed ox-idation of FeCuS2 as a mechanism for the formation ofcopper-rich iron hydroxides.

Hole DescriptionsIn this section the mineralogy of each drill hole is

summarized and a general description of the occurrenceand chemistry of each of the minerals is provided.

Holes 420 and 420A

Opaque minerals generally constitute up to 2 per centof each sample. Four opaque phases were identified: Ti-magnetite, ilmenite, pyrite, and chalcopyrite. Skeletal toanhedral Ti-magnetite is by far the most abundantopaque phase, with the ilmenite occurring as exsolutionlamellae in only a few grains. Pyrite occurs in two mor-phologies — as discrete sulphide spheres floating in aglassy matrix, and as irregular globules associated withthe oxide phases and smectite. The occurrence of pyriteand smectite which rim the vesicles (Figure 2) may be in-dicative of deuteric alteration. Chalcopyrite occurs asan additional phase in two of the pyrite spheres.

Hole 421

There are two. dominant opaque minerals in thebasalts of Hole 421: Ti-magnetite and pyrite. The oxide

794


Figure 2. Vesicle rimmed by pyrite and smectite in fine-grained basalt. Sample 420-15-1 (Piece 9); reflectedlight; ×16.

is much more abundant than the sulfide, constitutingover 90 volume per cent of the opaque portions in allrocks. The Ti-magnetites are skeletal to anhedral andgenerally are less than 20 µm in diameter. The overallabundance of opaques does not exceed 2 to 3 per cent inany sample and is never less than 1 per cent. The sulfidehas two distinct morphologies; (1) spheroidal and (2) ir-regular masses. The irregular masses of sulfide accum-ulations are either associated with the Ti-magnetites oragglomerated around vesicles. The sulfide spheres con-tain cobalt-nickel phases as exsolved lamellae in twosamples.

Hole 422In general, the opaques are continuous in mineral-

ogy, morphology and relative abundance throughoutthis hole. The overall grain size of the Ti-magnetite islarge, the crystals having an average diameter of about60 to 80 µm; some of the larger skeletal Ti-magnetiteeven approach 150 µm in diameter. The oxide morphol-ogy is skeletal-subhedral to skeletal-euhedral (Figure 3).No exsolved ilmenite is present. Sulfides are more com-mon than at Site 420 and Hole 421; they are dominantlypyrite with cobalt- and nickel-rich phases as exsolutionproducts in smaller (5 to 10-µm) sulfide spheres. Thelarger pyrite masses occur either as an apparent replace-ment of feldspars or as the final product of crystalliza-tion; they are interstitial to the feldspars (Figure 4) andclinopyroxene. The grain size of the opaques increasestowards the middle of the sections, with Samples 422-8-5(Piece 14) and 422-9-1 (Piece 4C) being the coarsest. Sul-fide abundance also increases with increasing grain size.These factors indicate a closed or partially closed sys-tem in which volatiles were retained and cooling wasrelatively slow. One specimen [Sample 422-9-4 (Piece13A)] contained a very few small chromium spinels ad-jacent to plagioclase crystals that were included in mag-netites. Sample 422-10-1 (Piece 3) is not at all similar tothe rest of the rocks, and appears almost identical in

Figure 3. Stacked, twinned tetrahedra of skeletal Ti-magnetite with blebs of anhedral pyrite. Sample 422-7-2 (Piece 3); reflected light; × 8.

Figure 4. Pyrite sphere in glass interstitial to feldsparlaths. Sample 422-7-1 (Piece 11); reflected and trans-mitted light; ×16.

opaque abundance and morphology to the samplesfrom Site 420 and Hole 421.

Hole 424The ore minerals of this hole are dominantly skeletal,

titanium-rich magnetites (20 to 200 µm) (Figure 5) andspheroidal pyrite masses (5 to 50 µm) (Figure 5). Pyr-rhotite was identified in sulfide spheres of two samples.The overall size of the opaques at Hole 424 is muchlarger than at previous holes and than in other holes atSite 424, except for two specimens, Samples 424-4-6,112-116 cm, and 424-7-1, 13-15 cm. These two rocks areprimarily glass; the opaques are very small (1 to 2 µm)and constitute less than 0.5 per cent of the rocks, asagainst 6 to 12 per cent of the other samples. Thegenerally large grain size apparently reflects the fairlymassive nature of the cooling units recovered.

795


Figure 5. Rectilinear arrangement of skeletal Ti-magne-tite crystals. Sample 424-5-1, 31-34 cm; reflectedlight; ×16.

Figure 6. Chalcopyrite rind on pyrite-pyrrhotite sphere.Sample 424-6-1, 138-140 cm; reflected light; × 16.Contact is highlighted on photograph.

Sulfides are larger and generally more abundant inthese samples than in the rocks from Sites 420 through422, indicating that the cooling history of the Hole 424rocks was such that volatiles were retained and suffi-cient time for crystal growth was available. Thus, thecooling was probably more gradual than for the aphyricbasalts and, possibly, the restraining pressures weregreater.

The opaques are interstitial to the silicates, and mustbe considered to have crystallized last; however, twostages of opaque crystallization can be seen. The largermagnetites and sulfides are generally not associated withglass, and are concentrated along plagioclase/clino-pyroxene boundaries. Smaller (5 to 10-µm) magnetitesare apparent in the rare residual pockets of glass. Addi-

tional evidence for two cycles of opaque formation isderived from the following textural relationships. Thelarger magnetites have sulfide rims, and evidently crys-tallized prior to the pyrites (Figure 7); however, whensulfide spheres are adjacent to glassy sections, the pyriteis rimmed with fine-grained Ti-magnetite (Figure 8) ofthe size generation found in the glass. Also, larger mag-netites often are accompanied by sulfide spheres flat-tened against them (Figure 9). The implication is thatthe sulfide was still liquid after the magnetite crystal-lized. Thus, Ti-magnetite can be seen to have crystal-lized at two discrete intervals relative to the sulfides.

Holes 424 A, B, and CThe ore minerals in these holes are identical, textural-

ly and mineralogically. The basalts cored here are eitherportions of small pillows or thin flows. The Ti-magne-tites are skeletal to anhedral in habit and generally areless than 20 jam in diameter. Smaller grains of anhedralTi-magnetite constitute the only opaque in the glassy,aphyric samples from these locations. Sulfides occur asspheres and as irregular masses in the more crystallinesamples. The larger sulfide masses are anhedral tosubhedral crystal groups associated with patches of thegroundmass which has been altered to a layer-latticesilicate. This same assemblage also fills vesicles. Amor-phous iron oxyhydroxides are often associated withthese vesicle fillings. This hydrated phase has crystalliz-ed to hematite in several cases (Figure 10). The restrictedoccurrence of larger pyrite grains to altered areas mayimply a hydrothermal origin for this form of the sulfide.

Pyrite spheres are common and range from 1 to 10µm in size. Many of these spheres were slightly de-formed as they adhered to the solid Ti-magnetites dur-ing cooling as at Site 424. Often, one side of a sphere ap-pears to have been flattened against an already crys-tallized magnetite.

Figure 7. Subhedral Ti-magnetite with pyrite rim (indi-cated by arrows). Sample 424-2-6, 112-116 cm; re-flected light; ×16.

796


Figure 8. Two sulfide spheres. Larger sphere is pyriterimmed by Ti-magnetite; smaller, unrimmed sphereis pyrrhotite. Sample 424-6-2, 32-35 cm; reflectedlight; ×16. Arrow indicates rim.

Figure 10. Vesicles filled with smectite cores, hematiteintermediate zones, and amorphous Fe oxide outer-most rinds. Sample 424B-6-1 (Piece 16); reflectedlight; ×8.

Figure 9. Pyrite spheres flattened against subhedral Ti-magnetite. Both opaque phases occur in interstitialglass. Sample 424-4-6, 112-116 cm; reflected light;×16.

Hole 425

The ore minerals found at this hole are of twoorigins— primary magmatic and hydrothermal. The pri-mary minerals comprise Ti-magnetite and sulfide spheres.The hydrothermal minerals are pyrite and chalcopyrite.These exogenetic sulfides reside in veins (Figure 11),fracture fillings, and subhedral masses (Figure 12) thatcross-cut crystalline and glassy components of the ba-salt. The modes of sulfide emplacement include voidfilling and direct replacement of the primary magmaticconstituents. Where a veinlet crosses through a largemagnetite crystal, the sulfide has replaced the oxide.Thus, Fe+3 has been reduced in the presence of somehydrothermal catalyst, probably hydrogen sulfide.

Figure 11. Pyrite veining silicates and replacing magne-tite. Sample 425-9-2 (Piece 16); reflected light; × 16.Py = pyrite, Mt = magnetite.

Some secondary iron oxides and clay minerals also oc-cur in these veins.

Ti-magnetite dominates the primary ore minerals.The sizes of these skeletal crystals range from 1 µm inSample 425-9-1, 106-110 cm (Piece 15) to over 35 µm inSamples 425-8-1, 44-46 cm (Piece 6) and 425-9-1, 126-129 cm (Piece 5). The oxide phases generally comprise85 to 90 per cent of the opaques; the latter generallyconstitute from 3 to 7 per cent of the samples.

From textural criteria, this general paragenetic se-quence can be established:

Magnetite — magmatic pyrite — hydrothermalpyrite and chalcopyrite — iron oxides

The relative volumes of the chalcopyrite and iron ox-ides are subordinate to the volume of the other hydro-thermal phase, pyrite.

797


Figure 12. Subhedral to anhedral pyrite replacing sili-cate glass. Sample 425-8-1 (Piece 12); reflected light;×16. Py = pyrite.

Figure 13. Ilmenite lamellae in Ti-magnetite. Sample427-10-4, 53-55 cm; reflected light; × 16; arrow indi-cates ilmenite.

Hole 427

This hole yielded basaltic samples with the mostdiverse opaque mineralogy retrieved on Leg 54. Asusual, skeletal (anhedral to subhedral) Ti-magnetitedominates the opaques in total volume. These oxidescomprise about 90 to 95 per cent of the opaques in anygiven sample, while opaques generally make up between5 and 10 per cent of an entire sample. Except forSamples 427-9-1, 0-7 cm (Piece 1) and 427-11-1, 47-49cm (Piece 4), the Ti-magnetites are large, with crystalsaveraging 30 to 40 µm in diameter. The Ti-magnetitesare pervasively fresh in all specimens but one, Sample427-10-4, 53-55 cm (Piece IE). In this rock, ilmenite hasexsolved from the granular, anhedral Ti-magnetite crys-tals, probably as a result of deuteric oxidation-alter-ation (Figure 13). The ilmenite laths are from 1 to 3 µmwide and may reach lengths of over 20 µm. These lathsgenerally are well-spaced and parallel, never abundantenough to form the rectilinear trellis pattern seen inmany basalts.

Vesicles in many of the samples are filled with calcite,iron oxides (goethite), pyrite, and Ti-magnetite (Figure14). At this time, the origin of the vesicle fillings isundetermined but could be either deuteric or hydrother-mal. Pyrrhotite occurs as an exsolution product in py-rite lying along the vesicle walls. The paragenesis ofthese fillings is texturally determinable as:

Ti-magnetite — pyrite and pyrrhotite — calcite (fans)— calcite (drusy) — calcite (mosaics) — goethite.

The goethite appears simply to be an oxidation productof the earlier produced iron-bearing species. Additional-ly, sulfide veining of silicates occurs as at Hole 425(Figure 15).

As in previously sampled areas, grain size of theopaques was proportional to crystallinity of the rockand overall grain size of the silicate phases. The coarsestopaques were near the middle of the hole. Very finegrained opaques were observed in the sample taken justabove the base of the hole.

Figure 14. Ti-magnetite, pyrite, and pyrrhotite rim cal-cite filled vesicle. Sample 427-9-2, 22-25 cm reflectedand transmitted light; ×8;Mt = Ti-magnetite; Po =pyrrhotite; Py = pyrite; Ct = calcite.

Holes 428 and 428AThe oxide and sulfide mineralogy of these holes is

nearly identical. Large Ti-magnetites >20 µm dominatethe opaques volumetrically (Figure 16 and Figure 17);however, pyrites are common as spheres, individualsubhedral grains, and as veins in Hole 428A. The sulfideveins also contain chalcopyrite. Pyrrhotite is commonas an exsolution product in a few of the larger sulfidespheres in Hole 428A.

Pockets of silicate glass, now altered to a clay, are ac-companied by subhedral pyrite and oxidized Ti-magne-tite (Figure 18). It is possible that iron derived from thereaction of the magnetite with hydrothermal solutions,rich in SO4~

2, could be the source of such pyrite. A hy-pothetical reaction is:

Fe3O4 + 2(SO4-2) = 2(FeOOH) + FeS2 + 2O2.

798


Figure 15. Pyrite veining and replacing silicate glass.Sample 427-9-4, 29-31 cm; reflected light; ×8.

Figure 16. Large, unaltered skeletal Ti-magnetite. Sam-ple 428A-5-1, 115-118 cm; reflected light; ×8.

Although it might seem unlikely that free oxygen is pres-ent during the formation of pyrite in this environment,iron oxides and pyrite appear to have been depositedcontemporaneously in the areas of silicate alteration.Whether theoretically favorable or not, a natural mech-anism must exist for the coprecipitation of these species.

The authors have observed several sulfide vein depos-its on lands that have obviously undergone coprecipita-tion of pyrite and iron oxides, both during orogenesisand subsequent weathering (Schrader and Furbish,1976).

Hole 429ASulfide spheres are more common in this hole than at

many of the previous holes; however, no multiplicity of

Figure 17. Skeletal Ti-magnetite in interstitial glass. Al-so sulfide spheres and second-generation Ti-magne-tite in glass. Sample 428-5-4, 15-18 cm, reflected andtransmitted light; × 16. Mtl = first generation Ti-magnetites; Py = Py sphere; Mt2 = second genera-tion Ti-magnetite.

Figure 18. Silicate glass, altered to smectite, veined bypyrite. Sample 428A-5-3, 123-126 cm; reflected andtransmitted light, × 16.

phases exists in the spheres. Pyrite is the only recogniz-able sulfide.

The five uppermost polished sections are oxidizedand altered. Groundmass minerals, feldspar, and glasswere replaced by these oxides and calcite. The opaquesin these areas are apparently oxidized; rim staining sur-rounds remnant Ti-magnetites. Small sulfides veinletsappear to have been emplaced after the oxidation andcross-cut other secondary minerals as in Holes 428 and428A (e.g., clays and chlorite). Some of the iron oxy-hydroxides are crystallized into hematite and goethite

799


which are evident in the polished samples (Figure 19).Cr-spinel is present as 5 to 10-µm euhedral crystals inone glassy sample (Figure 20).

Ti-magnetites dominate the opaques in size andoverall abundance. The morphology of these oxides isgenerally skeletal-dendritic to skeletal-subhedral. Theindividual crystal size within a particular sample is pro-portional to the degree of undercooling. The first sixsamples (from the top of the core downward) werequickly quenched, and thus the size of the Ti-magnetitesis small — generally 10 µm to 15 µm. The basalts com-prising the lower samples obviously were allowed tocool more slowly and the Ti-magnetites average about25 µm in diameter.

Descriptive Mineralogy

TitanomagnetitesThe dominant opaque in volume (<1% to 6%)

and size in all of the basalts was found to be Ti-magnetite. These oxides vary from subhedral grains lessthan 1 µm in diameter, through an entire size range andvarious morphologies, to euhedral tetrahedra of dia-meters greater than 100 µm. As previously stated,microprobe analyses indicate a composition of (0.38 to0.4 Fe3O4) (0.6 to 0.62 FeTiO2). This formula approx-imates the composition reported by Ade-Hall et al.(1976) for the same phase in DSDP basalts from theNazca plate.

The size and crystal morphology of the Ti-magnetitesappear to have been governed by the extent of super-cooling of the melt. The basalt samples containingclinopyroxene and plagioclase mosaics as the ground-mass apparently have cooled relatively slowly. The Ti-magnetites in these basalts are large (25 to 100 m indiameter) subhedral to euhedral crystals. The degree of

Figure 19. Irregular replacement of silicate glass by he-matite (in core) and smectite/amorphous iron oxides(darker rim). Sample 429A-4-1, 45-47 cm; reflectedlight; ×8. Hm = hematite.

Figure 20. Cr-spinel, surrounded by devitrified halo,floating in a glassy matrix. Sample 429A-2-1, 9-11cm; transmitted light, × 8.

skeletal development in these crystals may be a functionof cooling (e.g., crystallization), time, and the availabil-ity of chemical constituents in the proximal melt. Theglassy basalt samples contain very small (< 5 µm in di-ameter) and morphologically diverse Ti-magnetites.Multiply twinned, skeletal tetrahedra are common.Sizes may be as small as 0.5 µm in diameter. Abundanceis directly proportional to size. Samples that consist to-tally of glass have no primary oxides at all. These sam-ples do not, however, differ in bulk TiO2 and total FeOcontent from more crystalline adjacent samples. Thus,the necessary metallic constituents were available forcrystallization of Ti-magnetites. Two limiting factorsmay have inhibited the growth of these oxides:

(1) Insufficient time for nucleation and growth dueto rapid chilling of melt (we have shown that these ox-ides are among the final phases to crystallize); and

(2) Degassing of the rapidly cooling melt has left in-sufficient oxygen for the oxidation of the metals dis-solved in the silicate melt.

Very little oxidation or secondary alteration of themagnetites was observed in the samples studied. In nosample was oxidation common. This evidence leads usto the same conclusion as postulated by Grommé et al.(1969); the short period between eruption and samplinghas precluded extensive secondary reactions.

Deuteric alteration in the form of ilmenite exsolutionwas observed in six samples. These occurrences are sig-nificant, since this phenomenon is not common in sub-aqueously erupted tholeiites (Kitazawa and Ade-Hall,1972). No correlation exists between the abundance ofcracks and ilmenite lamellae as reported by Ade-Hall etal. (1976). However, the ilmenite occurs in basalt that ismore vesiculated than surrounding samples. This rela-tionship suggests that the evolution of certain magmaticgases (such as hydrogen) favored the oxidation of theprimary Ti-magnetites. The removal of H and OH byvesiculation would produce a localized environmentfavoring oxidation reactions in the remaining melt (Satoand Wright, 1966; and Ade-Hall et al., 1976).

800


Cr-SpinelThese minerals were seen in only a few samples

(Table 1); however, the presence of this phase has sig-nificant ramifications. The oxides typically occur in moreprimitive basalts than those retrieved on Leg 54 (Schra-der et al., 1977).

The spinels are optically and chemically similar tothose in picritic basalts from the Siqueiros transformfault. The crystals are euhedral to subhedral, showingno evidence of physical abrasion; however, some re-sorption has occurred at the juncture of crystal faces(Figure 20). In transmitted light the spinels are orange-red; in reflected light they appear tan to grey. Sizesrange from 5 µm to 25 µm in diameter.

The spinels in these basalts are probably xenocrystsfrom more mafic magmas. Spinels, orthopyroxenes,and Mg-rich olivines are typical of cumulate assem-blages in the near-static magma reservoirs beneath ac-tive transform faults (Schrader et al., Picritic basaltsetc., this volume). Thus, spinels and other dense phasespresent in evolved basalts cored or dredged from risecrests may indicate the influence of transform faults orother through-going structures on magmatic ascent.Deeply penetrating structures may tap primitive mag-mas which evolve during ascent, retaining spinel xeno-crysts.

IlmeniteThe presence of ilmenite was restricted to a few

deuterically altered Ti-magnetites, except in Sample420-14-1, 10-17 cm. This specimen has a few ilmeniteblades clustered around a large vesicle. The orientationof the blades is radial to the gas bubble. This spatialrelationship was probably generated and controlled bythe rapid evolution of gases during vesiculation. Thestreaming of hydrogen out of the immediate area sur-rounding the vesicle created a micro-environment thatwas sufficiently oxidizing for ilmenite to form. Aftervesiculation and ilmenite crystallization, small Ti-mag-netites nucleated and grew in the same area but were notoxidized.

Iron OxidesAmorphous-hydrated iron oxides, goethite, and he-

matite were identified in samples of basalt retrieved onLeg 54. Apparently the hematite was dehydrated fromeither goethite or amorphous iron oxides. The hematitegenerally occurs as the central portion of a vesicle fillingor as the core of a pocket of silicate glass altered to clayand amorphous iron oxide.

The iron oxides were generated in two distinct en-vironments. Some vesicles and other open spaces werefilled or coated with amorphous or finely crystallizedhydrated iron oxides (Figure 21). These coatings prob-ably precipitated from seawater reacting with surround-ing basalts or from diluted and cooled hydrothermalfluids. No alteration of silicate phases is associated withthese oxides.

As previously discussed, another class of iron oxidesoccurs only in veins that are associated with alterationof the silicate phases. Often these oxides are associated

Figure 21. Amorphous iron-manganese oxides liningwall of an empty vesicle. Sample 424C-2-1 (Piece 7).Transmitted light; × 16.

with euhedral to subhedral, secondary pyrite crystals.Glass and olivine are two phases that appear to be selec-tively altered to clay, sulfide, and iron oxide (Figure 22).Associated with the alteration of glass and olivine tolayer lattice silicates is the replacement of feldspar (andsurrounding glass) by low-Mg calcite. Since this form ofcalcite does not directly precipitate from sea water(Ronald D. Perkins, personal communication), the mecha-nism generating this style of alteration must be eitherdeuteric alteration/hydration or secondary hydrother-mal interaction with the basalt.

SulfidesTwo major periods of sulfide deposition occurred in

the basalts retrieved on Leg 54:1) Primary — globular segregations and crystal

euhedra; and,2) Secondary — veins and disseminated crystals in an

altered matrix.The primary sulfides appear to have crystallized

directly from the silicate melt. Sulfide spheres, similarto those described by Mathez and Yeats (1976), andMoore and Calk (1971), are present in a majority of thesamples and constitute volumetrically 80 to 85 per centof all the sulfides in all samples. Pyrite is by far thedominant mineral phase in these spheres. Chalcopyrite,Co-Ni phases (probably pentlandite and cobaltite) andpyrrhotite also occur in the sulfide globules. Approx-imately 80 per cent of the spheres are monomineralic,being composed of only pyrite. The remainder consistof a variety of assemblages (e.g., pyrite-chalcopyrite;pyrite-pyrrhotite; pyrite-pyrrhotite-(Co, Ni) phase; pyr-rhotite-chalcopyrite). The phases are intimately inter-grown and exhibit textures typical of mutual exsolution.Flame-like exsolution textures in the pyrrhotite are simi-lar to pentlandite-pyrrhotite intergrowths which occurin various Canadian Co-Ni ore deposits (Stanton,1972). Since most of the exsolved phases are from 1 to

801


Figure 22. Replacement of olivine phenocryst by smec-tite (core), pyrite (high reflectivity mineral) and ironoxide (dark lining between matrix and pseudomorph).Sample 421-4-1, 32-36; reflected light; ×8.

0.5 µm in width, microprobe analyses were not accurateor reproducible.

The secondary sulfides are presumed to have beenhydrothermal and/or deuteric processes. Most of thesulfide veining consists of pyrite. In a few veins, pyrite isaccompanied by pyrrhotite and chalcopyrite segrega-tions. No Co-Ni phases occur as vein fillings. The sub-sidiary phases are not exsolved from the pyrite as in thespheres but occur as small (1 =µm) blebs floating in thelarger pyrite masses, suggesting a low-temperature sul-fide immiscibility.

Sulfide veins replace silicates and Ti-magnetite. Ad-ditionally, primary or deuteric sulfide accumulationsappear to have replaced some of the feldspar pheno-crysts during crystallization of the melt.

SPECULATIONS ON ORE DEPOSITSFORMED AT TRAILING PLATE EDGES

Axial magma reservoirs at active spreading centersoffer a greater variety of tectonic and physicochemicalsettings for the emplacement of ores than many otherenvironments. The variety of ore deposits that couldtheoretically be formed within this tectonic regime isquite diverse; however, from our observations and pre-viously documented chemical affinities, only a few typesof ore deposits should be expected to occur regularly attrailing plate edges. Some of these deposites are:

(1) Disseminated copper sulfides and massive sul-fides;

(2) Chromite and magnetite pods or zones in strati-form mafic and ultramafic complexes;

(3) Co-Ni-Cu sulfide vein systems; and(4) Bedded iron-manganese oxides.

Disseminated Copper Sulfides and Massive SulfidesDisseminated copper deposits are distinguished from

typical "massive sulfide" ore systems in that the major

zone of ore formation comprises tiny veinlets — ~5mm wide (ore stringers), mineralizing the parent intru-sion or country rock. There is no central mass ofchemically precipitated sulfide ore below the stringerzone as seen in massive sulfides. The sulfide oresassociated with the ultramafic sequence on Cyprus havebeen described erroneously as disseminated sulfides,whereas they actually represent small volcanogenicmassive sulfide systems (personal communication, M. J.Holtzclaw, Chevron Resources Company).

Although Bonatti et al. (1977) suggested that por-tions of the Mid-Atlantic Ridge might lie above dissemi-nated sulfide deposits, the total weight of evidence isagainst such a possibility. Disseminated copper sulfidesystems have a marked preference for intermediate tosilicic volcanics (andesites to alkali-rhyolites) (Sillitoe,1973; Lowell and Guilbert, 1970; and Wallace et al.,1968). The mineralized veins are usually dominated byfree-quartz, late-stage, volatile-containing minerals (to-paz, fluorite), and Fe-Cu-Mo sulfides. Nickel and cop-per are usually lacking in these ores (Stanton, 1972).Additionally, the host intrusions for such mineraliza-tion are thought to have vented subaerially (Sillitoe,1973). The pressure constraints associated with the vent-ing are apparently important in the deposition of ore-bearing phases.

The source of mineralization is not primary magmat-ic or deuteric fluids. Rather, the sulfides are generallythought to be transported by cyclic meteoric systems(formed as a result of caldera subsidence) that interactwith intrusions and wall rock. Ring fractures borderingcalderas localize and focus these reactive solutions inlow-grade disseminated sulfide deposits.

Along rise crests, the major structural features tendto parallel the spreading center and generally give wayto through-going features such as axial grabens (orhorsts) and transforms (Rosendahl et al., 1976). Insteadof focusing and providing a locus for mineralizingfluids, these features would tend to disperse and dilutesuch solutions except in special cases (Rona, 1978).

The location of the mounds hydrothermal field nearthe Galapagos spreading center indicates the downwardmovement of cold sea water nearer the rise crest (Lons-dale, 1977). Heated reactive sea water then migratesaway from the crest along fissures and ascends at thesite of the iron-manganese oxide mounds. This sug-gested path of fluid flow is opposite to that described byBonatti (1978).

Although the formation of disseminated sulfides asdescribed here is not yet recognized at rise crests, thesame processes invoked for this type of deposition couldin unusual circumstances be effective in transportationand localization of metal and sulfate ions on a largeenough scale to produce typical massive sulfide depos-its. In general, massive sulfide systems are thought to beassociated with two tectonic regimes: (1) subduction andback arc spreading-related vulcanism (Urabe and Sato,1978); and (2) incipient and aborted rift systems withrestricted sea-water circulation (Campbell et al., 1978).Since massive sulfide deposits usually replace, and inter-calate with, laterally contemporaneous marine and/orpyroclastic sediments, the rise crest is not an ideal en-

802


vironment for their formation. The sediment blanket isthin and the crust is being constantly moved away fromthe locus of hydrothermal activity. This explanationruns parallel to that espoused by Lonsdale (1977) for thelack of hydrothermal deposits along the axis of theGalapagos rift system. More plausible tectonic settingsfor disseminated sulfide deposits are related to, and par-tially responsible for, the development of high-rankmagmatic differentiation suites associated with islandarc chains and volcanic seamounts. These settings de-mand static volcanic systems.

The processes of island arc vulcanism and magmaticevolution beneath seamounts are longer lived and geo-chemically more efficient in volatile concentration thanthe upwelling of basalts near rise crests. Althoughmassive and disseminated sulfide formation is not pre-cluded at spreading centers, other environments appearmore receptive.

Stratiform Deposits

The ultimate source and the causes of rhythmic layer-ing in large layered deposits of chromite, magnetite,pyroxenite, and anorthosite (such as the Bushveld andTransvaal Complexes in Rhodesia and the StillwaterComplex in the United States) have long been discussed.The near steady-state nature of an axial rise crest mag-ma chamber as described by Rosendahl et al. (1976)would offer an excellent environment for the evolutionof such layered sequences; these might contain, for ex-ample, podiform chromite deposits (Thayer, 1964).

Rosendahl et al. (1976) proposed that such a reservoirunderlies the East Pacific Rise near the Siqueirostransform fault. The seismic data indicated systematicvariation within the magma chamber, corresponding toless dense layers toward the top and more dense layersnear the base. Actual dimensions of the chamber werenot derived.

Hodges and Papike (1976) recognized the efficiencyof cumulus processes in forming gabbros and perido-tites retrieved at DSDP Site 334. Their data indicate along period of slow crystal accumulation and coolingbeneath an Atlantic spreading center.

Several samples retrieved on Leg 54 contain olivineand spinel chemically and mineralogically similar to pic-rites of the Siqueiros fault zone. Schrader et al. (1977)and Schrader et al. (Picrites, etc., this volume) have sug-gested that picritic basalts dredged from the Siqueirosfracture were erupted from an intermediate to deep zonewithin the axial magma chamber. This zone would liebetween the less dense (feldspar-rich) and more dense(olivine and pyroxene-rich) regions comprising the res-ervoir stratigraphy. The olivine, orthopyroxene, andspinel contents of the Siqueiros picrites may indicatemineral segregation by gravity settling of these densephases. Thus, eruptive basalts with unusually high con-centrations of these as cumulus minerals may indicateincipient gravitative differentiation that has occurred insuch stratified magma reservoirs (see Figure 23).

Since the axial reservoirs are intersected by centralgraben (or horst) faults and penetrated by cross-cuttingfaults in the Siqueiros transform system, a variety of

Feldspar flotation

Incipientcumulates^.

Cumulates

Figure 23. Diagrammatic cross section through magmachamber underlying the East Pacific Rise as de-scribed by Rosendahl et al. (1976). Seismic responseindicated: a less dense layer near the apex, corre-sponding to flotation of feldspar in a silicate melt, anintermediate density zone, and a dense layer near thebase (perhaps lower in melt content). We posit thatthe segregations are due to gravitative response ofmore dense phases (olivine, spinel, orthopyroxene)and, thus, their cumulation at the base. Black dia-monds in the figure represent cumulate phases andtheir relative abundance in the various density zonesas defined by seismic investigations.

controls are available to create rhythmic banding. Thecontinued but sporadic movement along transform faultscould be the control needed to induce repetitive magmapulses within the affected magma chamber. Such faultsalso allow cooling sea water access to the deep crust.The mineral phases crystallizing within a magma at anyone instant could then be related to the amount of time(coincident with strain accumulation along such faults)that the magma has resided in the chamber, and the lossof heat that it has therefore experienced, since its previ-ous pulse of magma. Longer residence periods wouldallow greater accumulation of dense silicate phases andrelated ores which would be laterally accreted as oceanicLayer 3. Repetitive pulses of primitive magma wouldrestrict cryptic variation (Wager, 1968), perhaps allow-ing rhythmic repetition of olivine- and chromite-richlayers.

Co-Ni-Cu Sulfide VeinsIn samples of basalt retrieved during Leg 54 no veins

containing Ni or Co sulfides were recognized; however,these metals were seen to be common constituents of thesulfide spheres dispersed in the glass matrix of manysamples. The average crustal concentration of Co andNi (25 and 75 ppm, respectively) is significantly belowthe average concentrations of these cations in basalts(Co = 48 ppm, Ni = 150 ppm) (Krauskoph, 1967). It isobvious that the ongoing processes of magmatic evolu-tion which occur near the East Pacific Rise (Batiza etal., 1977) may cause these "incompatible" metals tobecome residually concentrated in highly differentiatedbasalts or released into a coexisting brine during crystal-lization of the basalt. Additionally, McBirney and Wil-liams (1969) have shown that these trace constituentscould be concentrated by magmatic hybridization of atholeiitic basalt. The alkaline basalts and other effusive

803


rocks described from the Galapagos Islands (McBirneyand Williams, 1969) in several cases are significantlyenriched in Co, Ni, and Cu over the relatively moreprimitive basalts in the vicinity.

Known ore deposits comprising Co-Ni sulfides arefrequently associated with slightly hybridized or frac-tionally evolved mafic suites. Examples include the as-sociations of noritic gabbros (Sudbury, Ontario), ultra-mafics (Lynn Lake, Manitoba), and diabases (Cobalt,Ontario).

Variations in the generally static condition of magmareservoirs underlying the East Pacific Rise could inducetypes of basalt hybridization similar to those associatedwith the forementioned deposits. If the magma chamberis isolated from a continual source of replenishing meltfrom the mantle, evolution of the remaining melt wouldproceed along lines favoring production of a volatile-rich, incompatible metal-rich, and alkali-rich residualmelt. This isolation could allow the systematic concen-tration of these trace constituents in a brine phase whichis evolved during final cooling and solidification. Such afluid, migrating through fissures and cracks generatedby cooling contractions, would serve as a mineralizingsource for the basalt and surrounding strata. The occur-rence of such evolved sequences of basalts and basalticdifferentiates is common in the seamounts, volcanicislands, and sea-floor ridges of the eastern Pacific.

A variety of Canadian and Mediterranean cobalt-nickel sulfide deposits occur in greenstone belts. Sovietgeologists (Yuri Bilibin and others) in the mid-1940'srecognized these belts as remnant "mobile" zones. Wenow interpret these belts as possible plate junctures andfossil spreading centers. Thus the environments of oce-anic vulcanism associated with divergent plate marginsare possibly the setting for localization of such cobalt-nickel vein systems.

Bedded Iron-Manganese Deposits

Iron-manganese deposits of the mounds hydrother-mal field were sampled during Leg 54. These depositsare being actively formed adjacent to the Galapagosrift.

The mounds hydrothermal field was discovered dur-ing a deep-tow survey south of the Galapagos rift in1972 (Klitgord and Mudie, 1974), and was subsequentlystudied in detail during a second deep-tow survey (Lons-dale, 1977). The mounds are conical in section and elon-gate in plan and are arranged in parallel chains of 1 to2 km in length. Lonsdale (1977) also states that themound chains overlie small bedrock faults and fissures.He suggests that these openings are the channels forascending hydrothermal fluids which are responsible forthe oxide deposition.

The general basement structure in the area of themounds comprises normal faults dipping northwardtoward the base of the spreading center (Klitgord andMudie, 1974). These faults give rise to low-relief abyssalhills and valleys parallel to the axis of the major riftsystem.

Preliminary, shipboard XRF whole-rock analyses ofthe basement basalts recovered on Leg 54 showed them

to be highly fractionated, Fe-rich tholeiites (SiO2 ~51%; A12O3 ~ 13.5%; FeO*2 ~ 14.2%; MgO ~ 4.8%;and, CaO — 10.4%). The cored samples comprise fine-to coarse-grained plagioclase and clinopyroxene basalts.Glassy margins are rare, and cooling units can only beestimated by grain size variations.

The general megascopic appearance is one of fresh,fine-grained basalts with 1- to 2-cm alteration rinds ad-jacent to fractures and joint surfaces. Vesicles can beseen to be filled with blue-green clay minerals, iron ox-ides, and calcite. Needle-like crystals, perhaps zeolites,are also present in a few small (~ 1-mm) vesicles.

Two mounds were drilled as well as a reference sec-tion between them. The recovered sediment cores con-sisted of interbedded Fe-Mn-rich zones and calcareous,foraminifer-nannofossil ooze. The Fe-Mn zones domi-nate the upper 20 meters of core and cap the mound sur-faces. The metal oxides appear amorphous upon micro-scopic examination (Figure 24).

It has been reported that many of the bedded manga-nese ore deposits are associated, perhaps accidentally,with volcanic activity (Hewitt, 1966). Many occurrencesof this type of ore contain various amounts of volcanicash and other debris (Stanton, 1972).

The conspicuous occurrence of manganese ores withcarbonates and the low concentration of iron in manymanganese ores can be explained by this sea floor, effu-sive hypothesis. The interbedded siliceous carbonateunits in manganese ore deposits might correspond to thediatom-bearing carbonate oozes lateral to the deposits.The efficiency of sea-floor hydrothermal systems toconcentrate manganese relative to iron is well docu-mented (Hekinian, Rosendahl et al., 1978).

iDSDP424:

/ "1

•1-1

, J1680X OX. CRUST

Figure 24. Amorphous manganese oxides at sediment/water interface of Hole 424. Oxides are the platy ag-gregates. Recent fossils comprise the remainder ofthe sample. Sample 424-1-1, 1-2 cm; scanning elec-tron micrograph; × 1680.

Expressing total iron content.

804


Chemical analyses of the mud and oxides/hydroxidescomprising the mounds are reported in another chapterof this volume (see Schrader et al., Geochemistry ofsediments).

These bedded, volcanogenic manganese oxide depos-its may be analogs to such deposits as the manganesebeds in Central Kazakhstan, Russia (Shatsky, 1964)which contain up to 20 per cent Mn intercalated withsilica-rich layers and pseudomorphs after carbonateminerals. Further study of the chemistry, mineralogy,and modes of deposition of these deposits in futureDSDP cruises is certainly warranted.

CONCLUSIONS

A variety of ore minerals are present in tholeiites ofthe eastern Pacific. Both primary magmatic and second-ary ore minerals occupy open spaces and replace thesilicate phases of the basalts. The opaque phases aredominated by titanium-rich magnetites with subordi-nate pyrite. Sulfides occur as magmatic segregationsand in hydrothermally deposited veins. Some ore depos-its that are typically associated with diabases, green-stone belts, and ultramafics were likely formed at oce-anic spreading centers. Three types of deposits seem tobe at least circumstantially indicated by the mineralsand textures observed in the samples studied. Thesedeposits are:

1) Stratiform (cumulate) complexes;2) Vein-type Ni-Co ± Cu systems; and,3) Bedded manganese oxides.

ACKNOWLEDGMENT

The authors gratefully acknowledge critical review of thismanuscript and stimulating discussions offered by Dr. M. A.Goldstein, Chevron Resources Company, Denver, Colorado.

REFERENCES

Ade-Hall, J. M., Fink, L. K., and Johnson, H. P., Petro-graphy of opaque minerals, Leg 34. In Yeats, R. S., Hart,S. R., et al., 1976. Initial Reports of the Deep Sea DrillingProject, v. 34: Washington (U. S. Government PrintingOffice), p. 349-362.

Batiza, R., Rosendahl, B. R., and Fisher, R. L., 1977, Evolu-tion of oceanic crust 3. Petrology and chemistry of basaltsfrom the East Pacific Rise and Siqueiros transform fault. / .Geophys. Res., v. 82, p. 265-276.

Bonatti, E., Guerstein-Honnorez, B. M., and Honnorez, J.,1976. Copper-iron sulfides mineralizations from the equa-torial Mid-Atlantic Ridge. Econ. Geoi, v. 71, p. 1515-1525.

Bonatti E., 1978, The origin of metal deposits in the oceaniclithosphere. Sci. Am., v. 238, p. 54-61.

Campbell, F. A., Ethier, V. G., Krouse, H. R., and Both, R.A., 1978. Isotopic composition of sulfur in the Sullivanorebody, British Columbia. Econ. Geol., v. 73, p. 246-268.

Grommé, C. S., Wright, T. L., and Peck, D. L., 1969. Mag-netic properties and oxidation of iron-titanium oxides inAlae and Makaophui Lava Lakes, Hawaii. J. Geophys.Res., v. 74, p. 5277-5294.

Hekinian, R., Rosendahl, B. R., Cronan, D. S., Dimetriev,Y., Fodor, R., Goll, R. M., Hoffert, M., Humphris, S. E.,Mattey, D. P., Natland, J., Petersen, N., Schrader, E. L.,

Srivastava, R. K., and Warren, N., 1978. Hydrothermaldeposits and associated basement rocks from the Galapa-gos spreading center. Oceanologica Acta, v. 1, p. 473-482.

Hewitt, D. F., 1966, Stratified deposits of the oxides and car-bonates of manganese. Econ. Geol., v. 61, p. 431-461.

Hodges, F. N. and Papike, J. J., 1976. DSDP Site 334: Mag-matic cumulates from oceanic Layer 3. /. Geophys. Res., v.81, p. 4135-4151.

Keays, R. R. and Scott, R. B., 1976. Precious metals in ocean-ridge basalts: Implications for basalt as source rocks forgold mineralization. Econ. Geol., v. 71, p. 705-720.

Kitazawa, K. and Ade-Hall, J. M., 1972. Magnetic and opaquemineralogical properties of basalts from DSDP Legs II, III,V, and VI. Am. Geoph. Union Trans., v. 53, p. 974 (Ab-stract).

Klitgord, K. D. and Mudie, J. D., 1974. The Galapagos spread-ing centre: A near-bottom geophysical survey. Geophys. J.R. Astron. Soc, v. 38, p. 563-586.

Krauskopf, K. B., 1967. Geochemistry: New York (McGraw-Hill).

Lonsdale, P., 1977. Deep-tow observations at the moundshydrothermal field, Galapagos Rift. Earth Planet. Sci.Lett., v. 36, p. 92-110.

Lowell, J. D. and Guilbert, J. M., 1970. Lateral and verticalalteration-mineralization zoning in porphyry ore deposits.Econ. Geol., v. 65, p. 373-408.

Mathez, E. A. and Yeats, R. S., 1976. Magmatic sulfides in ba-salt glass from DSDP Hole 319A and Site 320, Nazca Plate.In Yeats, R. S., Hart, S. R., et al., Initial Reports of theDeep Sea Drilling Project, v. 34: Washington (U. S. Govern-ment Printing Office), p. 363-374.

McBirney, A. R. and Williams, H., 1969. Geology and petrol-ogy of the Galapagos Islands. Geol. Soc. Amer. Memoir, v.118, p. 192.

Moore, J. G. and Calk, L., 1971. Sulfide spherules in vesiclesof dredged pillow basalt. Am. Mineralogist, v. 56, p. 476-488.

Rona, P. A., 1977. Criteria for recognition of hydrothermalmineral deposits in oceanic crust. Econ. Geol., v. 73, p.135-160.

Rosendahl, B. R., Raitt, R. W., Dorman, L. M., Bibee, L. D.,Hussong, D. M., and Sutton, G. H., 1976. Evolution ofoceanic crust, 1, A geophysical model of the East PacificRise crest derived from seismic refractions data. J. Geo-phys. Res., v. 78, p. 5294-5304.

Sato, M. and Wright, T. L., 1968. Oxygen fugacities directlymeasured in magmatic gases. Science, v. 153, p. 1103-1105.

Schrader, E. L. and Furbish, W. J., 1976. Correlations ofuranium and trace metal concentrations in mineralized rhy-olites from Nevada. Geol. Soc. Am. Abstracts, v. 8, p.1089-1090.

Schrader, E. L., Rosendahl, B. R., and Furbish, W. J., 1977,Primative Picrites from the Siqueiros Fracture Zone, EastPacific Rise. Ibid., v. 9, p. 1162.

Shatsky, N. S., 1964. On manganiferous formations and themetallogeny of manganese, Paper I. Volcanogenic-sedi-mentary manganiferous formations. Int. Geol. Rev., v. 6,p. 1030-1056.

Sillitoe, R., 1973. The tops and bottoms of porphyry copperdeposits. Econ. Geol., v. 68, p. 799-815.

Stanton, R. L., 1972. Ore Petrology, New York (McGraw-Hill).

Thayer, T. P., 1964. Principal features and origin of podiformchromite deposits, and some observations on the Guleman-Soridag district, Turkey. Econ. Geol., v. 59, p. 1497-1524.

Urabe, T. and Sato, T., 1978. Kuroko deposits of the KosakaMine, Northeast Honshu, Japan-Products of submarine

805


hot springs on Miocene sea floor. Econ. Geoi, v. 73, p. Wallace, S. R., Muncaster, N. K., Janson, D. C , McKenzie,161-179. W. B., Bookstrom, A. A., and Surface, V. E., 1968. Multi-

Wager, L. R., 1968. Rhythmic and cryptic layering in mafic pie intrusion and mineralization at Climax, Colorado. Inand ultramafic plutons. In Hess, H. H. and Poldervaart, Ridge, J.E. (Ed.), Ore Deposits of the United States, 1933-A. (Eds.), Basalts: New York (Wiley Interscience), v. 2, p. 1967: Am. Inst. Mining Metall. Petroleum Engnrs., v.573-622. 1, p. 605-640.

806

related to spreading centers of the eastern …35. mineralogy and geochemistry of ore minerals from...

Documents