18. BASALT DRILLED AT THE VEMA FRACTURE ZONE, DSDP LEG 39

Enrico Bonatti and Keith Hartman,Lamont-Doherty Geological Observatory of Columbia University, Palisades, New York

Fabrizio Innocenti, Instituto di Petrografia, Universita di Pisa, Pisa, Italy,and

Robert Kay, Department of Geology, Cornell University, Ithaca, New York

ABSTRACT

An attempt to drill into lower crust at the foot of the northernwall of the Vema fracture valley resulted in recovery of basaltcobbles beneath about 170 meters of sediment. Maximum pene-tration into the basalt cobble zone was 13 meters. Data obtained byoptical microscopy, electron probe, and major and rare-earthelement analysis indicate that the basalts are olivine tholeiites andrepresent a magma type commonly erupted along mid-oceanicridges. The basalts were probably erupted originally at the axialsegment of the Mid-Atlantic Ridge. The basaltic cobble zoneprobably represents talus material channeled into the valley fromthe upper levels of the valley wall.

INTRODUCTION

The main objective of drilling close to the foot of thenorth wall of the Vema fracture valley was to attemptto sample a section of the lower oceanic crust. Thisattempt was based on the following two assumptions:(a) a partial section of more or less "normal" oceaniccrust is exposed at the north wall of the Vema FractureZone valley (where by "normal" crust we mean crustoriginally emplaced in the active, axial zone of the Mid-Atlantic Ridge); (b) by drilling close to the foot of thenorth wall one would bypass a thickness of crustroughly equal to the height of the wall (about 2 km),thus allowing one to drill directly into lower crust.Assumption (a) is supported by several lines of evidencewhich have been discussed by Bonatti and Honnorez(1976). Assumption (b) may also be valid, with somelimitations, as suggested by the fact that a variety ofgabbros and metagabbros, thought to represent lowercrustal material, have been dredged from the Vemanorth wall (Bonatti and Honnorez, 1976). The drillingattempt was done about 30 to 40 kilometers west of thenorthern axial Mid-Atlantic Ridge segment, close tothe foot of the north wall (Figure 1), allowing for a sedi-ment column above basement thick enough to permitdrilling into hard rock. No previous detailed site sur-veys were available, and the limited time allotted to thissite did not permit any extensive survey from theGlomar Challenger. Figure 2 shows the location of thedrilling sites relative to the north wall of the VemaFracture Zone. Three attempts were made. The firstone (Hole 353, 10°54.90'N, 44°02.25'W) had to beinterrupted after drilling into about 380 meters of sedi-ment and before reaching basement, because of amechanical failure. In the second attempt (Hole 353A)

basement was reached after going through a sedimentsection 168 meters thick. A third attempt (Hole 353B)was made a few hundred meters from Hole 353B(Figure 2); basement was reached below about 170meters of sediment. In both Holes 353A and 353B thebasement consisted of cobbles of fresh basalt. Thisbasalt rubble layer was penetrated for 13 meters at Hole353A, and for about 10 meters at Hole 353B. Noserious attempt at deeper penetration was made due totime limitations.

DESCRIPTION OF THE BASALT

The cobbles recovered at Holes 353A and 353B rangein diameter from five to about 15 cm; some smallerfragments were also recovered. Some of the cobblesappear to have been partially smoothed by the action ofthe drill and were probably part of larger cobbles. Thecobbles consist exclusively of a dark, fresh, micro-crystalline, and glassy basaltic rock. Thin sections ofseveral samples from both holes were microscopicallystudied. The texture of the rocks varies from aphyric tosparsely phyric. Typically, the samples consist ofclusters of elongated platelets and needles of calcicPlagioclase (frequently with skeletal habit), and abun-dant isolated partly euhedral crystals of olivine, par-tially altered to smectites and chlorites, in a very finegroundmass containing secondary phases, opaques,and glass. The various samples from both holes displaya range of textures and degree of crystallinity, but theessential mineralogic characteristics appear to besimilar in all the samples. One sample (DSDP 39-353B-1-1, 85-92 cm) was studied in more detail. This samplehas partly euhedral and skeletal microphenocrysts ofolivine (ranging from F075-85) scattered within acicularor felty aggregates of Plagioclase (ranging from AΠÓO-65)

507

E. BONATTI, K. HARTMAN, F. INNOCENTI, R. KAY

12°

10°

Mn

i

1 T

VEMA F.Z.0102030

km

DSDP 39

AXIAL VALLEY

TRANSVERSERIDGE

VALLEY

A B

45°W

basalt + metabasalt only

serpentinite + basalt

44° 43Q 42° 40°

| = serpentinite + gabbro + basalt

= serpentinite only

D = serpentinite + gabbro

Figure 1. Simplified morphology of the Vema Fracture Zone (after Van Andel et al, 1967), showing the approximatelocation of drilling sites, Leg 39 and other sites where basalts were dredged.

10° 55'

10° 50'

slope here = 21

northernh wall

VEMA FRACTURE VALLEY

X DR fix

® satellite fix

depths in fathoms

i

Location of DSDP Leg 39 holesVema Fracture Zone (F. McCoy)

44° 05' 44° 00' 43° 55'

Figure 2. Location of Holes 353, 353A, and 353B relative to the north wall of the Vema fracture valley. Compilation by F.McCoy.

508

BASALT DRILLED AT VEMA FRACTURE ZONE

in a groundmass of fine skeletal Plagioclase crystals,opaques, and partly devitrified glass. A distinct olivine-plagioclase paragenesis was noted; namely, olivinecrystals were observed as nuclei for the radiatingPlagioclase variolitic aggregates (Figure 3), and in somecases, cross-cutting the Plagioclase crystals. Theseobservations suggest that crystallization of olivinepreceded that of Plagioclase. Electron probe data forsome olivine and Plagioclase crystals are reported inTable 1. Table 2 shows the results of major and rare-earth elemental analysis and CIPW norms of thissample. Major elements were determined by classicalmethods, rare-earth elements by mass spectrometry, asdescribed in Kay (1970). Shown are also major elementdata for two other samples from Hole 353B, obtainedin the U.S.S.R. by G.C. Kazpe (personalcommunication).

TABLE 1Major element composition and QPW norms

for basalt Sample 353B1-1, 85-93 cm,Concentration of REE also shown.

i ^

§ t ••

^j ,/p *

i

i

' »..<

j

j

\

f

» 5

I

f f

.#;

"y•

\ '

•

0.2 mm

Figure 3. Olivine crystal as nucleus for radiating Plagioclaselaths. Sample 353-1-1, 85-93 cm.

DISCUSSION

The mineralogy and chemistry of the basalts drilledat Holes 353A and 353B are very similar to that ofseveral other basalts obtained on the northern wall ofthe Vema fracture valley, at various distances from theaxial segment of the Mid-Atlantic Ridge, and to basaltsobtained within the axial rift valley itself (Bonatti, etal., in preparation). All of these basalts would fall in theolivine tholeiite or quartz tholeiite field of Yoder and

SiO 2

A1 2O 3

F e 2 O 3

FeOMgOCaON a 2 OK 2 OH 2 O +

TiO 2

p 2 o 5

MnOTotal

OqOrAbAnDiHy01NeMt11ApLaCeNdSmEuDyErYbLuBaBa/Ce

Composition (in

353B-14

48.6014.933.516.348.10

11.322.810.14

1.651.76

0.200.16

99.52

αpw

_0!83

23.7827.7221.8112.302.57

_5.093.340.442.929.669.603.401.255.503.353.100.464.760.49

353B-5a

49.9716.034.175.538.08

10.262.720.180.31

1.770.170.19

99.38

2.051.06

23.0231.0115.0017.15

—_

6.053.360.37

---—----—--

aAnalyses performed by G. C. Kazpe

%)

353B-12a

49.6215.833.976.277.87

10.822.660.150.25

1.770.160.19

99.95

1.210.89

22.5130.8217.5216.90

—_

5.763.360.35

--——--—--——

U.S.S.R.

Tilley (1964) normative classification (Figure 4). Thechondrite-normalized rare-earth elements pattern fromHole 353B shows light REE depletion typical of ocean-ridge basalts (Figure 5).

The petrochemistry of basalts from the northern wallof the Vema fracture valley (including those drilled inDSDP Leg 39), is consistent with the assumption thatthese basalts were originally emplaced at the Mid-Atlantic Ridge axial segment. Basalts with alkalineaffinities and/or a more differentiated chemistry wouldbe expected in the case of within-fracture-zone, "leaky"volcanism (Melson et al., 1967; Honnorez and Bonatti,1970; Barberi et al., 1975). The hypothesis that thebasalts from the Vema fracture north wall wereoriginally emplaced at the Mid-Atlantic Ridge axialsegment is supported by indications of an increase inage of these basalts with distance from the Mid-Atlantic Ridge axial rift valley (Bonatti and Honnorez,

509

E. BONATTI, K. HARTMAN, F. INNOCENTI, R. KAY

TABLE 2Electron probe analyses of Plagioclase and olivine crystals

in basalt Sample 353B-1-1, 85-93 cm.

Elements

SiO2

TiO2

A12O3

FeOMnOMgOCaONa2OK2OTotal

SiO2

A12O3

Cr2°3FeOMgOCaOTotal

53.140.11

29.290.820.030.47

12.133.750.02

99.78

40.180.580.07

18.5240.07

0.43100.23

54.980.15

27.190.840.020.46

12.443.700.01

99.80

40.540.220.08

14.7444.070.26

99.91

Probe Analysis Results (in %)

Plagioclase

53.550.14

29.150.720.010.33

10.974.660.03

99.58

Olivin<

40.160.220.07

13.8644.69

0.2299.24

53.320.13

28.550.900.020.45

12.543.880.02

99.81

40.310.270.07

13.9144.72

0.2799.56

53.400.10

28.531.020.030.49

12.264.070.03

99.95

40.470.240.07

13.4744.93

0.2499.43

54.440.15

27.161.220.020.51

11.604.320.04

99.48

54.850.17

27.301.170.020.52

11.894.040.02

100.00

1976). Following this assumption, and given aspreading rate from the northern Mid-Atlantic Ridgeaxial segment of about 1.0-1.2 cm/year (Van Andel et

al., 1971; Bonatti and Honnorez, 1976), the age of thebasalt drilled at Holes 353A and 353B should be closeto 3 m.y. We interpret the basalt cobbles sampled at

Ne Di

RIDGE AXIS BASALTSOFFSET ZONE BASALTS (NORTH WALL)TRANSVERSE RIDGE BASALTS

ALKALIFIELD

OL Hy

Figure 4. Q-Di-Hy-Ol-Ne normative diagram for basalts from the Vema Fracture Zoneincluding sample 353-1-1, 85-93 cm. Norms were obtained assuming Fe2O3 = 1.5%.(from Bonatti et al, in preparation).

510

BASALT DRILLED AT VEMA FRACTURE ZONE

10

78A

THOLEIITIC MAGMA SERIESJuan de Fuca Ridge

THOLEIITIC BASALTGorda Ridge

FERRO BASALT

Sm Eu Dy

Figure 5. Chondrite-normalized rare earth pattern for basalt3533-1-1, 85-93 cm, compared with pattern for Gordaand Juan de Fuca ridge basalts (from Bonatti et al, inpreparation).

Holes 353 A and 353B as part of a zone of talus materialwhich was fed into the Vema fracture valley from theupper part of the northern wall.

Dredge hauls collected from the slopes of thenorthern wall in the vicinity of the drilling sites indicatethat serpentinized peridotites are exposed on the wall.These dredge hauls include Station A-II-20-8 fromMelson and Thompson (1971), located about 5kilometers west of the drill sites, and station P7OO3-O9from Bonatti and Honnorez (1976) located about 30kilometers west of the sites. Gabbros and metagabbroswere recovered at several sites on the north wall(Bonatti and Honnorez, 1976). Given the presence ofserpentinite on the wall, it is surprising that no frag-ments of serpentinite or of other plutonic rocks werefound in the talus material drilled at Site 353. Apossible explanation of these findings is that the talusfrom the upper basaltic portion of the northern wall isfed to the transverse valley preferentially along south-facing shallow channels or "canyons." Some indicationof the existence of such shallow channels can bee seenin van Andel et al. (1971) topographic map of the areaunder discussion. It is possible therefore that the talusmaterial is not homogeneously distributed along the

foot and the lower part of the slope but that it is thickerbeneath the channels, and thin or absent away from thechannels. If this interpretation is correct, the relativelythick talus encountered at Site 353, containing noserpentinite or other plutonic rocks, coupled with thepresence of serpentinite close-by on the wall, mightindicate that the drilling was done unforeseeably in anarea close to the exit of one of the channels feedingtalus to the Vema transverse valley. Post-site evaluationof topographic lines indeed suggests that the drillingwas done in one such area, as shown in Figure 1.

These considerations suggest that the original objec-tive of drilling at Site 353, namely, to obtain a section ofthe lower oceanic crust, bypassing the upper crustalbasaltic layer in the Vema Fracture Zone, is still aviable objective, despite our failure to accomplish itduring Leg 39. A successful accomplishment of drillinginto section of oceanic lower crust at the VemaFracture Zone will require, in addition to detailed pre-liminary surveys of the area, a time expenditure at thesite larger than that afforded in Leg 39.

ACKNOWLEDGMENTSFigure 2 was compiled by F. McCoy. Research was

supported by NSF/IDOE and ONR.

REFERENCESBarbed, F., Bonatti, E., Marinelli, G., and Varet, J., 1974.

Transverse tectonics during the split of a continent: datafrom the Afar Rift: Tectonophysics, v. 23, p. 17-29.

Bonatti, E. and Honnorez, J., 1976. Sections of the Earth'scrust in the equatorial Atlantic: J. Geophys. Res., v. 81,p. 4104-4116.

Bonatti, E., Honnorez, J., Innocenti, F., and Kay, R., inpreparation. Basalts from the Vema Fracture Zone (NorthAtlantic).

Honnorez, J. and Bonatti, E., 1970. Nepheline gabbro fromthe Mid-Atlantic Ridge: Nature, v. 228, p. 850-852.

Kay, R., 1970. The rare earth geochemistry of alkalinebasaltic volcanics: Ph.D. Thesis, Columbia Univ., 179 p.

Melson, W.G., Jarosewich, E., Cifelli, R., and Thompson,G., 1967. Alkali olivine basalt dredged near St. Paul'sRocks, Mid-Atlantic Ridge: Nature, v. 215, p. 381-382.

Melson, W.G., and Thompson, G., 1971. Petrology of atransform fault zone and adjacent ridge segments: Phil.Trans. Roy. Soc. London, v. 268, p. 423-441.

Van Andel, T.H., Corliss, J.B., and Bowen, V.T., 1967. Theintersection between the Mid-Atlantic Ridge and theVema Fracture Zone in the north Atlantic: J. Marine Res.,v. 25, p. 343-349.

Van Andel, T.H., Von Herzen, R.P., and Phillips, J.D., 1971.The Vema fracture zone and the tectonics of transverseshear zones in oceanic crustal plates: Marine Geophys.Res., v. 1, p. 251-283.

Yoder, H.S., and Tilley, C.E., 1962. Origin of basalticmagmas: an experimental study of natural and syntheticrock systems: J. Petrol., v. 3, p. 342-390.

511