20. GEOCHEMISTRY OF BASALTS FROM DEEP SEA DRILLING …

20. GEOCHEMISTRY OF BASALTS FROM DEEP SEA DRILLING PROJECT HOLES 556-5641

N. E. R. Drake, J. M. Rhodes, and L. K. Autio, Department of Geology and Geography,University of Massachusetts2

ABSTRACT

The nine holes (556-564) drilled during DSDP Leg 82 in a region west and southwest of the Azores Platform (Fig. 1)exhibit a wide variety of chemical compositions that indicate a complex petrogenetic history involving crystal fractiona-tion, magma mixing, complex melting, and mantle heterogeneity. The major element chemistry of each hole exceptHole 557 is typical of mid-ocean ridge basalts (MORBs), whereas the trace element and rare earth element (REE) abun-dances and ratios are more variable, and show that both depleted Type I and enriched Type II basalts have been eruptedin the region.

Hole 556 (30-34 Ma), located near a flow line through the Azores Triple Junction, contains typically depleted ba-salts, whereas Hole 557 (18 Ma), located near the same flow line but closer to the Azores Platform, is a highly enrichedFeTi basalt, indicating that the Azores hot-spot anomaly has existed in its present configuration for at least 18 Ma, butless than 30-34 Ma.

Hole 558 (34-37 Ma), located near a flow line through the FAMOUS and Leg 37 sites, includes both Type I and IIbasalts. Although the differences in Zr/Nb and light REE/heavy REE ratios imply different mantle sources, the (La/Ce)ch (> 1) and Nd isotopic ratios are almost the same, suggesting that the complex melting and pervasive, small-scalemantle heterogeneity may account for the variations in trace element and REE ratios observed in Hole 558 (and FAMOUSsites).

Farther south, Hole 559 (34-37 Ma), contains enriched Type II basalts, whereas Hole 561 (14-17 Ma), located fur-ther east near the same flow line, contains Type I and II basalts. In this case, the (La/Ce)ch and Nd isotopic ratios aredifferent, indicating two distinct mantle sources. Again, the existence along the same flow line of two holes exhibitingsuch different chemistry suggests that mantle heterogeneity may exist on a more pervasive and transient smaller scale.(Hole 560 was not sampled for this study because the single basalt clast recovered was used for shipboard analysis.)

All of the remaining three holes (562, 563, 564), located along a flow line about 100 km south of the Hayes FractureZone (33°N), contain only depleted Type I basalts.

The contrast in chemical compositions suggests that the Hayes Fracture Zone may act as a "domain" boundary be-tween an area of fairly homogeneous, depleted Type I basalts to the south (Holes 562-564) and a region of complex,highly variable basalts to the north near the Azores hot-spot anomaly (Holes 556-561).

INTRODUCTION

During DSDP Leg 82, nine holes (556-564) were drülednear several spreading flow lines perpendicular to theMid-Atlantic Ridge in a region west and southwest ofthe Azores Platform (Fig. 1). The northern flow linesare on either side of the Pico Fracture Zone and passthrough the Azores Triple Junction (Holes 556 and 557)and the FAMOUS and Leg 37 sites (Hole 558). The south-ern flow lines straddle the Hayes Fracture Zone (Holes559-561 and 562-564) 650 km south of the Azores Tri-ple Junction. Drilling perpendicular to the ridge axisin crust aged 14-18 Ma (Anomalies 5-6) and 30-37 Ma(Anomalies 12-13) provided an opportunity to examinegeochemical changes with time at a particular ridge seg-ment. Figure 1 shows the position of the drill sites andsummarizes the geochemical characteristics of the ba-salts from each hole. No analyses from Hole 560 are in-cluded because only one basalt clast was recovered, whichwas used for shipboard analysis. Further details on lo-cations and recovery are given in the site chapters of thisvolume.

1 Bougault, H., Cande, S. C , et al., Init. Repts. DSDP, 82: Washington (U.S. Govt.Printing Office).

2 Address: Department of Geology and Geography, University of Massachusetts, Am-herst, Massachusetts 01003.

40° N

30

40° W

Figure 1. Location and chemical characteristics of basalts sampled onDSDP Leg 82.

The task of this study was to examine geochemicalvariations over the past 35 Ma in order to evaluate theroles played by crystal fractionation, magma mixing,complex melting, and large-scale or small-scale mantleheterogeneity in the petrogenesis of Leg 82 basaltic mag-mas. Such information is fundamental to a broader un-derstanding of the composition and evolution of theoceanic mantle and crust.

421

N. E. R. DRAKE, J. M. RHODES, L. K. AUTIO

The 96 samples collected for this study were analyzedfor major and trace elemental abundances by X-ray flu-orescence (XRF) techniques (Appendixes A and B). Sev-en samples were selected for rare earth element (REE)analysis by instrumental neutron activation analysis(INAA) (Table 1). The chemical abundances and ratioswere used to determine the chemical groups that in turnwere used to evaluate the complex processes involved inbasalt petrogenesis in the Leg 82 area. The geochemicaldata were also used to examine the important questionof whether the Azores hot-spot anomaly consists of asingle mantle plume (Schilling, 1975) or represents sam-pling of pervasive, small-scale mantle veining (Hanson,1977; Zindler et al., 1979). The role of fracture zones (inthis case, the Hayes Fracture Zone at 33 °N) as possibledomain boundaries between regions of mantle with dif-ferent chemical signatures was also considered (Melsonand O'Hearn, 1979; Bougault and Treuil, 1980).

Throughout the text, tables, figures, and appendixeswe have used our own simplified sample numbering sys-tem. The first part of the number refers to the hole num-ber (556-564) and the second part represents the sam-ples in order of increasing depth within basement (i.e.,the number 1 refers to the sample nearest the top ofeach hole). The DSDP designation for each sample canbe found in the second to fourth rows at the top of Ap-pendix A, Tables 1-7, which list all of the major andtrace element abundances for each sample by hole num-ber in order of increasing depth within basement.

METHODS

Samples were crushed in a tungsten carbide shatterbox at the Ron-ald B. Gilmore X-Ray Facility in the Department of Geology and Ge-ography at the University of Massachusetts in Amherst. The major el-ement data were obtained by XRF on fused glass disks, prepared byfusing the sample with a lanthanum-bearing lithium tetraborate flux(Norrish and Hutton, 1969). The trace elements (Rb, Sr, Y, Ga, Zr,Nb, Zn, Ni, Cr, and V) were determined by XRF analysis on pressedpowder pellets. Corrections were made for nonlinear backgrounds,tube contamination, and interelement interferences (Norrish and Chap-pell, 1967). Corrections for matrix effects were based on a modifica-tion of the Compton scattering method (Reynolds, 1967). Additionaltrace element and REE data (La, Ce, Nd, Sm, Eu, Yb, Lu, Hf, Sc)(Table 1) were obtained by INAA in Fred Frey's laboratory at the Cen-

ter for Geoalchemy in the Department of Earth and Planetary sci-ences at the Massachusetts Institute of Technology, by the methods ofJacobs and others (1977).

Average values for BCR-1 analyzed as an unknown by XRF aregiven in Table 2 with one standard deviation. Each element was ana-lyzed twice for each sample. Error values labeled Leg 82 in Table 2 arebased on these replicate analyses and apply directly to the Leg 82 ba-salts under the given operating conditions. The values were calculatedusing the formula:

- x )2/(/i - 1),

where x and x are observation and mean for replicate analyses of thesame element in a single sample and n is the number of samples ana-lyzed.

Mg'-values were calculated as molecular Mg/(Mg + Fe2+), wheremolecular Fe has been proportioned as Fe3 + /(Fe2 + + Fe3 + ) = 0.1(Basaltic Volcanism Study Project, 1981). Fe2O3* and FeO* refer tototal Fe expressed as Fe3+ or Fe2 + , respectively. Major elements areexpressed as weight percent (wt.%) and trace elements in parts permillion (ppm).

BASALT CHEMISTRY

Hole 556Hole 556 is located near Anomaly 12 (30-34 Ma)

about 80 km north of the Pico Fracture Zone on a flowline extending through the Azores Triple Junction (Fig. 1).The 16 basalt samples analyzed from 177 m of base-ment may be classified as two distinct, but stratigraphi-cally interlayered basalt types. Group I is the more prim-itive of the two, with moderately high MgO (7.4 wt.%)and CaO (13.3 wt.%) contents, Mg'-values (0.66), andCaO/Al2O3 ratios (0.83) (Appendix A, Table 1; Appen-dix B, Table 1; Figs. 2-4). Ni (100 ppm) and Cr (340ppm) contents are the lowest of all Leg 82 basalts andthe Sr (99 ppm) contents are relatively low also (Figs. 5,6). Incompatible element abundances, TiO2 (1 wt.%),Zr (48 ppm), Y (22 ppm), and Nb (1.8 ppm) as well asZr/Y ratios (2.5) (Fig. 7), are low relative to other ocean-floor basalts (Melson et al., 1977; Basaltic VolcanismStudy Project, 1981).

Within Group I, two subgroups can be recognized;IA (556-1, 2, 7, 8, 9, 11, 13-16) and IB (556-10, 12, 17).Subgroup IA has lower Ni (99 versus 110 ppm) contents,Mg'-values (0.64 versus 0.67), and CaO/Al2O3 ratios

Table 1. Rare earth element and trace element abundances (ppm) in selected Leg 82 basalts.

Univ. Mass, sample numberHoleCore-Section(interval in cm)Sub-bottom depth (m)

LaCeNdSmEuYbLuScHf(La/Ce)c h

(La/Sm)c h

(La/Yb)c h

558-155827-3

112-114409.13

6.4214.712.02.811.012.420.38

34.11.971.161.251.61

558-255828-1

55-58415.07

3.067.59.72.630.963.060.43

40.11.701.090.630.61

558-755833-2

144-147462.46

9.7919.417.13.091.102.360.34

36.12.011.351.742.51

558-1355838-1

97-99505.48

5.3313.010.72.120.792.120.34

35.91.501.091.381.52

561-25611-14-9

411.57

13.9926.820.0

3.681.222.880.44

36.92.201.392.092.94

561-6

2-282-84416.83

2.878.29.83.021.413.780.55

42.52.100.930.520.46

563-2

24-122-25366.24

1.935.67.82.080.822.670.41

41.41.370.920.510.49

ML-76

9.6723.226.84.791.702.150.30

30.53.431.111.112.73

NB-STD

10.7327.716.83.721.212.320.489.95.761.021.582.80

N

0.330.880.600.1810.0690.200.034

Note: ch = chondrite-normalized ratios. ML-76 = Mauna Loa sample used as University of Massachusetts internal standard. NB-STD= M.I.T. internal standard. N refers to values used to normalize REE abundances to chondrites.

422

GEOCHEMISTRY OF BASALTS

Table 2. Precision of whole-rock basaltchemistry analyses.

Elements BCR-1

Major element (wt.%)

SiO2

TiO2

A12O3

Fe2O3*MnOMgOCaONa2OK2OP2O5

54.51 (0.17)2.25 (0.01)

13.52(0.10)13.36 (0.04)0.19(0.01)3.33 (0.04)6.93 (0.03)3.30 (0.28)1.70(0.01)0.38 (0.01)

Trace element (ppm)

RbSrYGaZrNbZnNiCrV

46.4 (0.3)326 (1)33.5 (0.3)22.2 (0.6)

194.5 (1.8)12.5 (0.4)

128 (2)22.8 (9.7)

5.7 (1.4)360(4)

Leg 82 error

(0.07)(0.03)(0.05)(0.04)(0.01)(0.04)(0.19)(0.18)(0.01)(0.001)

(0.20)(0.58)(0.28)(1.49)(0.55)(0.46)(0.56)(6.60)(1.28)(2.39)

Note: BCR-1 is aU.S.G.S. standard analyzedwith every X-ray fluorescence run. Meanand 1 standard deviation (expressed inwt.% or ppm) are given. The column la-beled Leg 82 error refers to the precisionof analysis at the Leg 82 concentrations(see text) and is given as 1 standard devia-

(0.83 versus 0.87), along with higher incompatible ele-ment abundances, TiO2 (1.02 versus 0.86 wt.%), Zr (58versus 48 ppm), and Y (23 versus 20 ppm), than Sub-group IB. However, the Sr (99 versus 100 ppm) contentis the same for both subgroups (Fig. 6). The three samplesfrom Subgroup IB and one sample from Subgroup IA(556-7) are unusually high in CaO (14 wt.%) (Fig. 3) forocean-floor basalts (Melson et al., 1977; Basaltic Vol-canism Study Project, 1981). These samples have few orno phenocrysts and no obvious carbonate veining or al-teration.

Group II, consisting of four samples (556-3 through556-6) has lower MgO (6.75 wt.%), CaO (11.7 wt.%),Ni (89 ppm), and Cr (250 ppm) contents, Mg'-values(0.58), and CaO/Al2O3 ratios (0.70), along with greaterZr/Y ratios (2.9) and incompatible element abundances—TiO2 (1.45 wt.%), Zr (90 ppm), Nb (2.5 ppm), and Y(30 ppm)—than Group I. Group II also contains 100ppm of Sr, the same amount as in Group I (Fig. 6).

Both groups have high Zr/Nb ratios (>30) (Fig. 8)and are typical depleted Type I ocean-floor basalts ac-cording to the classification scheme of Bryan and others(1976). The compositional differences between the twogroups coupled with a constant Zr/Nb ratio (Fig. 8) arequalitatively consistent with a comagmatic relationshipbetween the two groups. Approximately 50% crystal frac-tionation involving olivine, plagioclase, and clinopyrox-ene is required to generate Group II basalts from GroupI basalts.

11

10

Hole 556

O Hole 558

A Hole 559

Hole 561

• Hole 562

D Hole 563

0 Hole 564

(17)0

40 50 60 70 80Zr (ppm)

90 100 110 120

Figure 2. MgO versus Zr. Enclosures and Roman numerals refer tochemical groups within each hole. (All of the samples from Holes559, 563, and 564 belong to Chemical Group I.) Arabic numbers inparentheses refer to the University of Massachusetts sample num-bers (hole is indicated by shape of data point; see Appendix A ta-bles for the DSDP sample designations for the samples).

It is unlikely, however, that this process is one of sim-ple crystal fractionation. First, the observed phenocrystassemblages of only plagioclase ± olivine are not con-sistent with fractionation dominated by clinopyroxene(as is suggested by chemical considerations). Second,the fact that Ni and Cr abundances are approximatelythe same for the two groups, each with different amountsof incompatible elements (Fig. 5), is consistent with theevolution of both groups from a common parental mag-ma. Such inconsistencies may be explained by a combi-nation of fractional crystallization and mixing of con-sanguineous magmas (Rhodes et al., 1979). This inter-pretation receives support from the repetitive interlayer-ing of Groups I and II. It is also possible to interpret thegeochemical differences in terms of variable degrees ofpartial melting of a common source (Langmuir et al.,1977; Wood, 1979). In this case, the more enriched

423


12 13CaO (wt.%)

Figure 3. TiO2 versus CaO. Symbols and enclosures are the same as inFigure 2.

1.0

0.9 -

oCM

< 0.8

0.7 -

0.6

1

IB/•^

— / • I

1 I

- 1

1

1

IA

/T\/ ? — » >

ru m

a /

1

1 1 1

I

\ • ^ " ^ '•y• no) j

• • ./& N O_/ / •

Λ^•" ""•£ 1IIB/ !(1) UV

II

I I I

(17)

0

s.. • ,J

~-^^

1 1

I

s \ A )w

1 140 50 60 70 80 90 100 110 120

Zr (ppm)

Figure 4. CaO/Al2O3 versus Zr. Symbols and enclosures are the sameas in Figure 2.

Group II would have been produced by a lesser degreeof melting. However, a plot of Sr versus Zr (Fig. 6)shows that both groups have about the same concentra-tion of Sr (100 ppm) for very different concentrations ofZr. This relationship cannot be produced by differentdegrees of partial melting, but can be the result of frac-tionation/mixing processes.

Hole 557

Hole 557 is located near the same flow line as Hole556, but near Anomaly 5D (18 Ma) closer to the AzoresTriple Junction (Fig. 1). Only 3 m of basement werepenetrated before the hole was abandoned because ofinclement weather. Our one sample of this core, 557-1,falls into Group II of the shipboard studies (Site 557chapter, this vol.). It has extremely high incompatible el-ement abundances, TiO2 (3.67 wt.%), Zr (216 ppm), Y(41 ppm), Nb (26 ppm) as well as high Fe2O3* (16.5wt.%) and Sr (299 ppm), but very low MgO (5 wt.%),Ni (38 ppm), and Cr (19 ppm) abundances and a lowMg' -value (0.41) (Appendix A, Table 1). Thus, 557-1 isone of the more highly evolved or enriched basalts de-scribed to date from the ocean floor (Melson et al.,1977; Basaltic Volcanism Study Project, 1981). TiO2 andFeO* abundances in 557-1 are similar to those of the en-riched FeTi basalts of the Galapagos Spreading Center(Anderson et al., 1975; Schilling et al., 1976; Melson etal., 1977). Because of its extremely different abundancesand ratios, Hole 557 data are not plotted on Figures 2-8.

The Zr/Nb ratio (8) and Zr/Y ratio (5) indicate that557-1 is a Type II ocean-floor basalt (Bryan et al., 1976).The Zr/Y ratio is similar to that observed in Holes 407,408, 410A, and 413 of DSDP Leg 49 (Tarney et al.,1979), but is higher than the Zr/Y ratios (2-4) observedin the other Leg 82 basalts (Fig. 7). It is likely that thisbasalt was derived from an unusually enriched source.If these basalts were derived from primitive Type II ba-salts similar to Group IV in Hole 558 (Appendix A,Table 2), the observed increase in incompatible elementabundances would suggest as much as 70% crystal frac-tionation of olivine, plagioclase, and clinopyroxene. How-ever, the lack of clinopyroxene phenocrysts and the closecorrelation of TiO2 and Y abundances to other incom-patible element abundances (indicating that clinopyrox-ene has not been a major fractionating phase) suggestthat 557-1 is enriched by derivation from an enrichedsource rather than by evolution from a more typical ocean-floor basalt. CIPW norm calculations also indicate amore moderate state of evolution (557-1 is only slightlyquartz normative).

Hole 558

Hole 558 is located between Anomalies 12 and 13(34-37 Ma) about 50 km south of the Pico FractureZone on a flow line passing through DSDP Leg 37,Hole 335 (approximately 16.5 Ma), and the FAMOUSarea (Recent) (Fig. 1). The 16 samples collected from153.5 m of basement are compositionally the most com-plex of the Leg 82 basalts. The data indicate that fourdistinct basalt groups and two subgroups have been sam-pled (Appendix A, Table 2 and Appendix B, Table 2).

424


Figure 5. Zr versus Ni. Symbols and enclosures are the same as in Figure 2.

140

120

100

40

2080 90 100 110 120 130 140 150 160

Sr (ppm)170 180 190 200 210 220

Figure 6. Zr versus Sr. Symbols and enclosures are the same as in Figure 2. Lines represent Zr/Sr ratios.

All of the chemical groups in this study correspondto shipboard groups except that three samples (558-1,558-10, and 558-11) are included together in Group I be-cause, although not stratigraphically contiguous, theyhave similar incompatible element abundances and ra-tios (Appendix A, Table 2; Figs. 7, 8).

Group I has moderate abundances of MgO (8.8 wt.%),CaO (11.7 wt.%), Sr (139 ppm), Ni (169 ppm) and Cr(344 ppm), Mg' -values (0.65), and CaO/Al2O3 ratios(0.77) (Appendix B, Table 2; Figs. 2-6). The abundancesof incompatible elements are moderate as well, TiO2(1.26 wt.%), Zr (79 ppm), Nb (9.8 ppm), Y (23 ppm), asare the Zr/Y ratios (3.5) (Appendix B, Table 2; Fig. 7).Sample 558-1 has the highest MgO (9.5 wt.%), Ni (277ppm), and Cr (510 ppm) abundances in Group I and thelowest CaO (10.6 wt.%) contents and CaO/Al2O3 ratio(0.73) (Appendix A, Table 2; Figs. 2-5), which could bedue to olivine accumulation.

Group II, consisting of one sample, 558-2, also con-tains moderate abundances of major and trace elements;MgO (8.3 wt.%), CaO (12.4 wt.<Fo), Ni (170 ppm), Cr(400 ppm), and CaO/Al2O3 ratio (0.78) (Appendix A,

Table 2; Figs. 2-5). The Mg'-value is high (0.67), where-as the Sr content (95 ppm) is the lowest of all Hole 558basalts (Fig. 6). Incompatible element abundances aresomewhat low: TiO2 (1.2 wt.%), Zr (66 ppm), Nb (2.3ppm), and Y (26 ppm). The fact that Group II has thehighest Y content and lowest Zr/Y ratio (2.6) for thishole (Fig. 7) precludes the possibility that Group II evolvedfrom any of the other samples in Hole 558 by fractiona-tion of clinopyroxene.

Group III consists of seven samples (558-3 through558-9). It has the lowest CaO (11.6 wt.%), Ni (160 ppm),and Cr (368 ppm) and highest Sr (173 ppm), TiO2 (1.35wt.%), and Zr (88 ppm) abundances of all samples inHole 558 (Appendix B, Table 2; Figs. 3, 5, 6). Yet the Niand Cr contents are still fairly high, as are MgO con-tents (8.7 wt.%) and Mg'-values (0.66) (Figs. 2, 5). Thissuggests that the samples from Hole 558 are among themost primitive of Leg 82 basalts, even though Group IIIis the most evolved group at this site.

Within Group III are two subgroups. The single sam-ple (558-3) in Subgroup IIIA differs from the other sixsamples in Subgroup IIIB; it has lower MgO (8.2 versus

425


70 80 90Zr (ppm)

100 110 120

Figure 7. Y versus Zr. Symbols and enclosures are the same as inFigure 2. Lines represent constant Zr/Y ratios.

22

20

18

16

14

50 60 70 80 90Zr (ppm)

100 110 120

Figure 8. Nb versus Zr. Symbols and enclosures are the same as inFigure 2. Lines represent constant Zr/Nb ratios.

8.9 wt.%) contents and Mg'-values (0.64 versus 0.67)and greater abundances of incompatible elements, TiO2(1.45 versus 1.31 wt.%), Zr (93 versus 82 ppm), Nb(15.3 versus 13.6 ppm), and Y (24 versus 21 ppm) (Ap-pendix B, Table 2). The similar CaO/Al2O3 ratios (0.77)and Ni (157 versus 162 ppm) and Cr (356 versus 379ppm) contents for both subgroups (Figs. 4, 5) imply thatSubgroups IIIA and IIIB may have either fractionatedalong closely parallel trends from slightly different pa-rental magmas or undergone mixing during the frac-tionation process. Because IIIA contains about the sameamount of Sr (170 versus 176 ppm) and more Zr thanIIIB (Fig. 6), a melting trend is not possible, whereasmixing during fractionation is possible. Although theZr/Y ratios are the same for both subgroups (3.8) (Fig. 7),the abundance of Y is higher for IIIA than IIIB, alsosupporting a more complex relationship than simple crys-tallization dominated by clinopyroxene fractionation.

Group IV, with five samples (558-12 through 558-16)has high MgO (7.5-10.8 wt.%), Ni (185-230 ppm) andCr (442-482 ppm) contents, and Mg'-values (0.62-0.71),with moderate CaO (12.1 wt.%) and Sr (112 ppm) con-tents and CaO/Al2O3 ratios (0.76) (Appendix B, Table 2;Figs. 2-6). These basalts exhibit low incompatible ele-ment abundances; TiO2 (< l wt.%), Zr (55 ppm), Nb(7.6 ppm), and Y (19 ppm) and Zr/Y ratios (3.0) (Fig. 7).These abundances and ratios make Group IV the mostprimitive of the Leg 82 basalts. The primitive nature ofGroup IV basalts is also shown by the trends seen in Fig-

ures 2, 3, and 5 of Zr versus MgO, CaO, and Ni, respec-tively, which indicate olivine-controlled fractionation.

Within Group IV, Samples 558-13 and 558-16 arepossible candidates for primary ocean-floor basalts. Theyhave the highest abundances of MgO (10.8 and 9.9 wt.%),Ni (230 and 214 ppm) and Cr (449 and 441 ppm), andMg'-values (0.71 and 0.69) of all Leg 82 basalts (except558-1 for Ni) and low incompatible element abundances(Appendix A, Table 2). They are chemically similar toother proposed primary ocean-floor basalts such as thosefrom DSDP Leg 3 (Frey et al., 1974), the FAMOUS area(Langmuir et al., 1977; White and Bryan, 1977), and theMid-Atlantic Ridge at 45°N (Melson et al., 1977; Rhodeset al., 1979; Basaltic Volcanism Study Project, 1981).

It seems unlikely that the basalts from these fourgroups are comagmatic. Basalt samples from all fourgroups, which have comparable MgO and Ni abundances,Mg'-values, and CaO/Al2O3 ratios, have markedly dif-ferent abundances of Zr and other incompatible ele-ments (Figs. 2, 4, 5). In addition, these four chemicalgroups all have different Zr/Nb ratios (Fig. 8): GroupsI, III, and IV have low values (8, 6, and 7.5, respective-ly), typical of Type II basalts (Bryan et al., 1976), butGroup II is somewhat anomalous with a Zr/Nb ratio ofabout 20 (Fig. 8), similar to many basalts sampled onthe Juan de Fuca Ridge (Liias and Rhodes, 1982).

REE data were obtained for four representative sam-ples, one from each group (Table 1; Fig. 9A). Samples558-1, 558-7, and 558-13 (Groups I, III, IV, respective-

426


50 -

50I I I I I I I I I I I

La Ce Sm EuREE atomic number

Yb Lu

Figure 9. A. Chondrite-normalized rare earth element (REE) abun-dances for selected basalts in Hole 558. Sample numbers are thesimplified numbers used in this study. Complete DSDP designa-tions are located at the top of Appendix A, Table 2. B. Chondrite-normalized REE abundances for selected basalts in Holes 561 and563. Sample numbers are the simplified numbers used for thisstudy. Complete DSDP designations are located at the top ofAppendix A, Tables 4 and 6.

ly) all show LREE-enriched patterns with chondrite-nor-malized (ch) ratios: (La/Sm)ch > 1 (1.25-1.8) and (La/Yb)ch > 1 (1.5-2.5) (Table 1). In marked contrast, Sam-ple 558-2 (Group II) has a LREE-depleted pattern with(La/Sm)ch < 1 (0.63) and (La/Yb)ch < 1 (0.61). Al-though the (La/Sm)ch ratios are variable, the (La/Ce)ch

ratios are > 1 (1.09-1.35) for all four groups, similar toother basalts along this flow line, such as those fromDSDP Leg 37 (Blanchard et al., 1976) and the FAMOUSarea (Langmuir et al., 1977; White and Bryan, 1977).

Hole 559

Hole 559 is located between Anomalies 12 and 13(34-37 Ma) midway between the Oceanographer andHayes fracture zones (Fig. 1). All nine samples from the

63 m of basement are similar in composition (AppendixA, Table 3; Appendix B, Table 3) and compose a singlechemical group consisting of enriched Type II basalts(Bryan et al., 1976) with typical Zr/Nb ratios (7-7.5)(Fig. 8). These basalts have some of the lowest CaO(11.7 wt.%) contents and CaO/Al2O3 ratios (0.76) andhighest incompatible element abundances—TiO2 (1.54wt.%), Zr (107 ppm), Y (28 ppm), and Nb (15 ppm)—ofall Leg 82 samples except 557-1 (Appendix A, Tables 1,3; Appendix B, Table 3; Figs. 3, 4, 7, 8). However, theMgO (7.1 wt.%) and Ni (135 ppm) contents and theMg'-value (0.59) are higher than expected relative to theconcentration of incompatible elements (Figs. 2, 5).

The basalts from Hole 559 show stratigraphically in-terlayered variations in MgO, FeO*, and Mg'-values.Samples 559-2, 559-3, 559-5, 559-7, and 559-8 have lowerMgO abundances (6.5 versus 7.5 wt.%) and Mg'-values(0.56 versus 0.62) and higher Fe2O3* (11.6 versus 10.5wt.%) abundances than Samples 559-1, 559-4, 559-6,and 559-9 (Appendix B, Table 3; Fig. 2). This interlayer-ing involving only the elements Mg and Fe could be theresult of periodic replenishment of a multiply saturatedmagma by a more-primitive, olivine-rich melt (Rhodeset al., 1979). The relatively high Ni values (Fig. 5) alsosuggest that these basalts are enriched in olivine relativeto other similarly evolved basalts (Basaltic VolcanismStudy Project, 1981).

The incompatible element abundances (Appendix B,Table 3) are consistent with approximately 50% crystalfractionation from a more primitive Type II magma sim-ilar to the Group IV basalts sampled at Hole 558 (Ap-pendix B, Table 2). Low CaO/Al2O3 ratios (0.76) (Fig. 4)coupled with high Zr/Y ratios (3.8) (Fig. 7) point to-wards clinopyroxene involvement in the fractionationprocess. However, clinopyroxene is notably absent fromthe observed phenocryst assemblage. The discrepancybetween the chemistry and petrography, already encoun-tered in Holes 556, 557, and 558, is a common problemfor MORB (Dungan and Rhodes, 1978; Basaltic Vol-canism Study Project, 1981) and points to a complexpetrogenesis involving fractionation and mixing processes.

Sample 559-1 has the highest Y (31.6 ppm) abun-dance and the lowest Zr/Y ratio (3.43) of Hole 559 ba-salts (Fig. 7) and therefore could not have fractionatedfrom compositions similar to the other samples in Hole559. Possibly the observed anomalous behavior of 559-1is a result of mantle heterogeneity or complex meltingprocesses.

Hole 561Hole 561 is located near Anomaly 5E (14-17 Ma)

east of Hole 559 and along the same flow line (Fig. 1).The 15.1 m of basement is represented by nine samplesthat fall into two stratigraphically distinct groups. Group Iis represented by two samples in the upper 0.7 m of thecore, and Group II by seven samples in the lower 14.4 m.(Appendix A, Table 4; Appendix B, Table 4).

The upper basalts (Group I) have greater MgO (8.1versus 7.4 wt.%), CaO (12.2 versus 11.6 wt.%), Sr (215versus 89 ppm), and Ni (176 versus 130 ppm) abun-

427


dances and Mg'-values (0.64 versus 0.59), but lowerFe2O3* contents (9.9 versus 11.5 wt.%) than the lowerbasalts (Group II).

Zr/Nb ratios are strikingly different for the two groups(Fig. 8). Group I, with low Zr/Nb ratios (<5), is a TypeII basalt (Bryan et al., 1976), whereas Group II, whichhas higher Zr/Nb ratios (>25), is a Type I basalt. Tworepresentative samples, 561-2 of Group I and 561-6 ofGroup II, were analyzed for REE (Table 1; Fig. 9B).Sample 561-2 is LREE enriched with chondrite-normal-ized ratios > l : (La/Ce)ch = 1.39, (La/Sm)ch = 2.09,and (La/Yb)ch = 2.94, all typical of LREE-enrichedType II ocean-floor basalts (Bryan et al., 1976). Sample561-6, however, is a typical LREE-depleted Type I ocean-floor basalt, with chondrite-normalized ratios < 1: (La/Ce)ch = 0.93, (La/Sm)ch = 0.52, and (La/Yb)ch = 0.46.This is in marked contrast to Hole 558 where the (La/Ce)ch ratio is fairly constant ( > l ) even though varia-tions in the (La/Sm)ch ratios occur between Groups Iand II (Table 1; Figs. 9A and 9B).

Group I has greater Zr/Y (4 versus 2.5) (Fig. 7) andZr/TiO2 (90 versus 55) ratios and Zr abundances (100versus 75 ppm) than Group II, but lower Y (25 versus30 ppm) and TiO2 (1.2 versus 1.4 wt.%) contents (Ap-pendix B, Table 4). These features, combined with thedifferent Zr/Nb and (La/Ce)ch ratios, suggest that twodistinct parental magmas are required, which is furthersupported by the two trends of decreasing Ni abundancesat different Zr concentrations (Fig. 5). The Nd isotopicratios (Jenner et al., this vol.) are distinctly different forthe two groups and also indicate two distinct sources.

Hole 562

Hole 562 is located in 14-17 Ma crust near Anomaly5D along a flow line about 100 km south of the HayesFracture Zone (Fig. 1). Sixteen samples from 90.1 m ofbasement fall into two stratigraphically separated chem-ical groups (Appendix A, Table 5; Appendix B, Table 5).Group I is represented by six samples in the upper49.5 m of core and Group II by ten samples in the lower40.6 m. Both groups have Zr/Nb ratios >20 (Fig. 8)and are depleted Type I MORB (Bryan et al., 1976).

Group I, which is higher stratigraphically and there-fore more recently erupted, is the more evolved of thetwo groups. It has lower CaO (11.6 versus 12.9 wt.%)and MgO (7.0 versus 7.2 wt.%) abundances, CaO/Al2O3

ratios (0.74 versus 0.79), and Mg'-values (0.57 versus0.61) (Figs. 2-4). Abundances of incompatible elements,TiO2 (1.6 versus 1.2 wt.%), Zr (96 versus 70 ppm), andY (35 versus 27 ppm), and Zr/Y ratios (2.8 versus 2.6)(Fig. 7), are correspondingly higher in Group I thanGroup II. However, Ni (109 versus 108 ppm), Cr (231versus 257 ppm), and Sr (100 versus 98 ppm) abun-dances are similar for the two groups (Figs. 5, 6).

Within Group II, A12O3 increases and CaO/Al2O3 ra-tios decrease with decreasing Zr (Fig. 4) as a consequenceof plagioclase accumulation. The most plagioclase phyricsamples occur at the high Al2O3-low Zr end of thetrend.

The above-noted increase in incompatible elements isconsistent with about 22-25% crystal fractionation. The

small increase in Zr/Y ratios from Group II to Group Iand the fact that TiO2 and Y are behaving less incom-patibly than Zr and Nb indicate clinopyroxene on the li-quidus. Based on these relations and the trends shownin Figure 7, the two groups could be comagmatic, withGroup I basalts having evolved from a magma with GroupII composition.

However, as in Hole 556, the Ni abundances for thetwo groups are approximately the same, although abun-dances of ZR (and other incompatible elements) are dif-ferent (Appendix B, Table 5; Fig. 5). These Ni-Zr rela-tionships for the two basalt groups may be the result ofthe mixing of consanguineous magmas (Rhodes et al.,1979), or alternatively, fractionation of distinct parentalmagmas produced by slightly different amounts of melt-ing of a similar source. Once again, the two chemicalgroups have the same Sr concentrations with different Zrconcentrations (Fig. 6). This precludes partial melting andsupports the idea of fractionation and magma mixing asproposed for Holes 556 and 558 in this study.

Hole 563

Hole 563 is located near Anomaly 13 (34-37 Ma)west of Hole 562 and near the same flow line south ofthe Hayes Fracture Zone (Fig. 1). The eight samples tak-en from 18 m of basement compose a single chemicalgroup. These samples have the following major and traceelement abundances: MgO (7.7 wt.%), CaO (13 wt.%),Fe2O3* (10.1 wt.%), Ni (103 ppm), and Cr (239 ppm),Mg'-values (0.63), and CaO/Al2O3 ratios (0.8) (Appen-dix A, Table 6; Appendix B, Table 6). Abundances of Sr(84 ppm) and incompatible elements, TiO2 (1 wt.%), Zr(50 ppm), Y (23 ppm), and Nb (2.3 ppm), and Zr/Y ra-tios (< 2.5) are low.

As in Hole 559, the samples in Hole 563 show simi-lar stratigraphically interlayered variations in MgO andFe2O3* abundances and Mg'-values (Appendix A, Table 6;Fig. 2). Samples 563-1, 563-2, 563-4, 563-6, and 563-8have lower MgO (7.5 versus 8 wt.%) abundances andMg'-values (0.62 versus 0.64) and greater Fe2O3* (10.3versus 9.9 wt.%) abundances than Samples 563-3, 563-5, and 563-7. The interlayered fluctuations in these ele-ments could imply periodic replenishment of an evolv-ing magma reservoir by more primitive melts (Rhodes etal., 1979).

As in Hole 562, plagioclase accumulation is indicatedboth petrographically and chemically. In samples withsuch accumulations, MgO and Zr abundances and CaO/A12O3 ratios decrease whereas A12O3 increases with in-creasing amounts of plagioclase (Figs. 2, 4).

The compositional data (Appendix A, Table 6) indi-cate that Hole 563 contains some of the more primitiveof the Leg 82 samples. Unlike most Leg 82 basalts, how-ever, clinopyroxene fractionation does not appear to beimportant in the petrogenesis of Hole 563 basalts, anobservation further supported by the Zr/Y ratios, whichremain roughly constant (<2.5) or decrease slightly withincreasing Zr (Fig. 7).

One sample, 563-2, was analyzed for REE. It yieldedchondrite-normalized ratios < 1: (La/Ce)ch = 0.92, (La/Sm)ch = 0.51, and (La/Yb)ch = 0.49 (Table 1; Fig. 9B).

428


The resulting LREE-depleted pattern, along with Zr/Nb ratios >20 (Fig. 8), mark Hole 563 samples as de-pleted Type I basalts (Bryan et al., 1976). Similar resultswere obtained for other basalts collected south of theHayes Fracture Zone.

Hole 564Hole 564 is located near Anomaly 13 (34-37 Ma)

south of the Hayes Fracture Zone, about 10 km northof Hole 563 (Fig. 1). Twenty samples were taken from81 m of basement. These basalts exhibit high incompati-ble element abundances, TiO2 (1.5 wt.%), Zr (96 ppm),Y (33 ppm), and Nb (5 ppm); low abundances of CaO(11.8 wt.%) and Fe2O3* (11.9 wt.%); and moderate MgO(7.2 wt.%) and Sr (105 ppm) contents, Mg'-values(0.57), and CaO/Al2O3 ratios (0.80) (Appendix A, Table7; Appendix B, Table 7). These compositional charac-teristics suggest that Hole 564 basalts are fairly evolved,possibly with clinopyroxene on the liquidus.

Sample 564-17 deviates from the compositional rela-tionships exhibited by the other basalts from this holeand may constitute a separate group or subgroup. Itsdistinctly lower MgO content (5.2 wt.%) and Mg' -value(0.47) and higher Fe2O3* content (13.1 wt.%) suggestthat 564-17 is much more evolved than the other sam-ples. Yet the greater CaO content (13.8 wt.%) and CaO/A12O3 ratio (0.90) and much lower SiO2 content (47 wt.%)of this basalt are characteristic of more primitive basalts(Basaltic Volcanism Study Project, 1981). The higherK2O (0.62 versus 0.33 wt.%), Rb (15 versus 6 ppm), andSr (115 versus 105 ppm), coupled with the other majorelement characteristics, may indicate that extreme sea-water alteration has occurred.

In the plot of Ni versus Zr (Fig. 5), the Ni content de-creases markedly with almost constant Zr in all Hole564 samples, suggesting that olivine is the only liquidusphase. However, the Ni values (117 ppm) appear to betoo high relative to the low CaO contents and CaO/A12O3 ratios and high incompatible element abundancesnormally associated with a more evolved magma. As inthe other holes, this discrepancy in the chemistry im-plies a complex petrogenesis.

There is a pronounced and systematic change in ba-salt composition with depth. MgO, Ni, and Cr abun-dances and Mg' -values all decrease downhole, whereasFe2O3* and incompatible element (Nb, Zr, Y, and TiO2)abundances increase (Appendix A, Table 7). The grad-ual changes of concentration with depth suggest that theearlier erupted magma was more evolved than the morerecently erupted magma. Such changes are commensu-rate with the gradual emptying of a zoned magma cham-ber or the relatively rapid replenishment of a slowlyevolving magma by more primitive melt.

The most striking aspect of chemical variation withdepth is the change in Zr/Nb ratio, which is greater(>20) near the top of the hole, but gradually decreasesdownhole to < 15 (Appendix A, Table 7; Fig. 8). How-ever, this appears to be mostly the result of changingNb. On the other hand, Zr/Y ratios remain essentiallyconstant (2.8-3.0) with the downhole increase in Zr andY (Fig. 7). Zr/Nb ratios are unlikely to change during

crystal fractionation or more complex processes involv-ing both fractionation and mixing of consanguineousmagmas. Therefore, it seems probable that changes inthe Zr/Nb ratio result from the progressive mixing of aprimitive but Nb-enriched melt with one that was evolvedbut had lower Zr/Nb ratios.

Again, Sample 564-17 is different from the others; ithas a lower Zr/Y ratio (2.7) and greater Y content (36ppm) (Fig. 7). This inverse relationship is similar to thatexhibited by Sample 559-1. There is no obvious explana-tion for this anomalous behavior, and further investiga-tion is warranted. Two representative samples, one fromthe top and one from the bottom of Hole 564, will beanalyzed for REE and isotopes.

DISCUSSIONNumerous isotopic and chemical studies have been

made of basalts along the Mid-Atlantic Ridge from theAzores Triple Junction to south of the Hayes FractureZone (Fig. 1) (Schilling, 1975; White et al., 1976;O'Nions et al., 1977; Tarney et al., 1979; White andSchilling, 1978; Zindler et al., 1979; Dupré and Allègre,1980; White and Hoffman, 1982). These studies haveshown that the basalts from this area exhibit composi-tional diversity ranging from primitive to highly evolvedand LREE-depleted to LREE-enriched (Types I and IIof Bryan et al., 1976).

The holes sampled during Leg 82 (Fig. 1) contain sev-eral types of basalts that exhibit a variety of chemicalcharacteristics and may occur together within a givenhole (Table 1; Appendix A, Tables 1-7; Appendix B, Ta-bles 1-7). Most of these chemical groups are uniform incomposition, and, in general, simple crystal fractiona-tion cannot be readily documented by chemical varia-tion within groups. The exceptions are the olivine frac-tionation trend observed in Hole 558 (Group IV) (Figs.2, 3, 5) and the plagioclase accumulation trends ob-served in Hole 562 (Group II) and Hole 563 (Figs. 2, 4).

Chemical variation, however, does exist between groupswithin a given hole. In Holes 556, 558, and 562 (Fig. 1),the decreasing CaO/Al2O3 ratios and increasing Zr/Yratios (Figs. 4 and 7) seem to indicate that one groupmay have evolved from another by multiply saturatedcrystal fractionation. However, several lines of evidenceindicate that the chemical variation between groups isnot compatible with simple crystal fractionation. First,in these same three holes there is a discrepancy betweenestimates of the state of magmatic evolution based onmajor and compatible trace elements (e.g., MgO, CaO,Ni, Cr contents, Mg'-value, CaO/Al2O3 ratios) and thatbased on incompatible elements (TiO2, Zr, Nb, Y) (Ap-pendix A, Tables 2, 3, 5; Figs. 2-5). This discrepancy isclearly evident in Zr versus MgO and Ni (Figs. 2, 5),where associated groups in Holes 556, 558, and 562 havesimilar MgO and Ni abundances, but markedly differ-ent Zr contents. This phenomenon may imply that (1)each of the groups evolved from slightly different paren-tal magmas or (2) the groups within each hole are re-lated to a common parental magma by interplay of mix-ing and fractionation (Rhodes et al., 1979). The latterhypothesis receives further support from the existence

429


within Holes 556, 558, and 562 of different chemicalgroups with the same Sr but different Zr contents (Fig.6). This cannot be explained by partial melting.

The stratigraphic interlayering of different chemicalgroups within the same hole (Holes 556 and 558) (Ap-pendix A, Tables 1 and 2) indicates that similar magmacompositions were episodically erupted. This supportsthe involvement of complex fractionation-mixing pro-cesses in the petrogenesis of these basalts. Further sup-port comes from interlayered variations in Mg'-valuesin Holes 559 and 563 (Fig. 2), which could be due to re-peated influxes of a more primitive melt into a moreevolved magma (Rhodes et al., 1979). Further evidencefor fractionation-mixing is the commonly observed dis-crepancy between the chemical criteria that appear to re-quire clinopyroxene fractionation and the absence ofclinopyroxene phenocrysts in any of the Leg 82 samples.This is a common problem in ocean-floor basalt petro-genesis (Dungan and Rhodes, 1978; Basaltic VolcanismStudy Project, 1981) and is probably indicative of mix-ing between more-primitive and more fractionated mag-mas (Rhodes et al., 1979).

The systematic downhole trends in major and traceelement abundances in Hole 564 (Appendix A, Table 7;Fig. 8) from relatively primitive to more evolved basaltsis compatible with mixing of distinct magmas rather thansimple crystal fractionation (Rhodes et al., 1979; Batizaand Johnson, 1980).

To gain a better understanding of Leg 82 basalt pet-rogenesis, it is important to assess the roles of mantleheterogeneity, complex melting, and melt segregation inthe generation of the chemical diversity observed bothwithin and between holes, especially Holes 558 and 561(Table 1; Figs. 1-9; Appendix A, Tables 2, 4). For exam-ple, basalts with distinctly incompatible trace elementratios Zr/Y, Zr/Nb, and LREE/HREE (Figs. 7-9B)sampled within a single drill hole (Holes 558 and 561)have been cited as evidence for large-scale mantle hetero-geneity involving a "depleted" upper mantle with mantleplumes upwelling from a more "fertile" deeper mantlelayer (Schilling, 1973; Schilling et al., 1976; White andSchilling, 1978). Alternatively, mantle heterogeneity mod-els involving mantle "blobs" (Schilling, 1975; Schillinget al., 1982), or pervasive small-scale veining or metaso-matism of the upper mantle (Hanson, 1977; Zindler etal., 1979) have also been proposed. Complex meltingand melt segregation in an otherwise homogeneous man-tle (Langmuir et al., 1977; Wood, 1979) may cause simi-lar differences in incompatible element and REE abun-dances, but not in ratios of highly incompatible ele-ments (La/Ce)ch or Nd isotopes.

To evaluate these models it is necessary to develop aconcept of the scale of heterogeneity with respect toboth time and distance. The studies referenced abovesuggest the possibility of mantle "domains." For exam-ple, it appears that there is a boundary at about 33°N inthe vicinity of the Hayes Fracture Zone separating de-pleted (Type I) MORBs (Bryan et al., 1976) to the southfrom enriched (Type II) basalts associated with the topo-graphic high of the Azores Platform (White and Schil-ling, 1978; Bougault and Treuil, 1980).

In this study, basalts from holes drilled south of theHayes Fracture Zone (Holes 562, 563, 564) (Fig. 1) aredepleted in incompatible trace elements as expected. How-ever, north of the Hayes Fracture Zone, basalts are bothenriched (Holes 557 and 559) and depleted (Hole 556) inincompatible elements. At two sites (Holes 558 and 561),both types occur within the same hole (Figs. 1, 8, 9A,9B). As other studies have shown (Blanchard et al.,1976; Langmuir et al., 1977; Tarney et al., 1979; Wood,1979; Wood et al., 1979; Zindler et al., 1979), this closespatial association of enriched and depleted basalts isnot uncommon. The occurrence of depleted basalts inolder Hole 556 (30-34 Ma) and enriched basalts in youn-ger Hole 557 (18 Ma) along a flow line from the Azoreshot spot (Fig. 1) suggests that mantle anomalies are tran-sient features, perhaps adding support to the mantleblob hypothesis (Schilling, 1975; Schilling et al., 1982)or small-scale mantle veining (Hanson, 1977; Zindler etal., 1979). In Hole 561, Groups I and II are not strati-graphically interlayered and have distinctly different ra-tios for the highly incompatible elements, Zr/Nb and(La/Ce)ch, in addition to differences in Zr/Y, (La/Sm)ch,and (La/Yb)ch ratios (Table 1; Figs. 7, 8, 9B). Thesetraits suggest that Hole 561 basalts are derived from twodistinct sources. Nd isotopic data (Jenner et al., thisvol.) and geochemical data point to an LREE-enrichedsource for Group I and an LREE-depleted source forGroup II (Table 1; Fig. 9B). The scale of the sourceheterogeneity cannot be pinpointed, although the pres-ence within a single hole of magmas from two differentsources may imply pervasive, small-scale heterogeneitysuch as mantle veining (Hanson, 1977; Zindler et al.,1979). The occurrence of only enriched basalts in olderHole 559 (34-37 Ma) along the same flow line as youn-ger Hole 561 (14-17 Ma) (Fig. 1) could also be the resultof a more transient mantle anomaly.

As noted above, variable incompatible and REE ra-tios within the same hole (such as Holes 558 and 561)may indicate either mantle heterogeneity or complex melt-ing of a single source. If complex melting is involved,the ratios of the very highly incompatible elements, suchas Zr/Nb and especially (La/Ce)ch, should remain thesame, whereas other ratios such as Zr/Y, (La/Sm)ch,and (La/Yb)ch may vary (Langmuir et al., 1977; Wood,1979). This is seen in Hole 558 where the (La/Ce)ch ra-tios are about the same (1.09-1.35) for all four groups(Table 1; Fig. 9A). In contrast, other incompatible andREE ratios vary from LREE enriched (Groups I, III,and IV) to LREE depleted (Group II) (Table 1; Figs. 7,8, 9A; Appendix B, Table 2).

The data are not compatible with a petrogenesis in-volving only complex melting because Zr/Nb ratios varyconsiderably between Groups I, III, and IV (approxi-mately 6-9) and Group II (20) (Fig. 8). Therefore, man-tle heterogeneity may be indicated as well. The Nd iso-topic ratios for Hole 558 basalts are almost identical forall four groups (Jenner et al., this vol.), so large-scaleplume-related mantle heterogeneity involving two dif-ferent sources does not seem likely.

A more plausible possibility is smaller-scale mantleveining and/or metasomatism (Hanson, 1977; Zindler

430


et al., 1979) that may generate the highly variable in-compatible element ratios and crossing REE pattern ob-served in Hole 558, along with the constant (La/Ce)ch

and Nd isotopic ratios. Similarly, the repetitive interlay-ering of diverse magmas within the same hole does notsupport large-scale mantle heterogeneity. Variation intrace element ratios and crossing REE patterns appearto be restricted to regions of high volcanic activity alongelevated ridge segments in hot-spot areas. Other exam-ples are the Juan de Fuca Ridge (Liias and Rhodes,1982); Iceland (Wood, 1979; Zindler et al., 1979; Schil-ling et al., 1982); FAMOUS (Langmuir et al., 1977; Whiteand Bryan, 1977); and Hole 558 (site chapter, Site 558,this vol.). The observation that such highly variable ba-salt types appear to be confined to hot-spot areas sug-gests small-scale heterogeneity as well as complex melt-ing.

Basalts in Hole 556 have a drastically different chem-ical signature from the basalts in Hole 557 (AppendixA, Table 1) even though both holes were drilled on thesame flow hne through the Azores Triple Junction (Fig. 1).This chemical difference, along with the abrupt changesin the Zr/Nb ratios and REE patterns between chemicalgroups within Holes 558 and 561 (Figs. 8, 9A, 9B), iscounter to the more gradual transitions expected betweenType I and II basalts in plume or mantle blob situations(Schilling, 1975). The basalts from Leg 82 appear tohave undergone a complex petrogenesis involving small-scale mantle heterogeneity (veining or metasomatism),complex melting, and mixing-fractionation processes.

CONCLUSIONS

1. There is no evidence for simple crystal fractiona-tion within or between chemical groups, except for Hole558 (Group IV), Hole 562 (Group II), and Hole 563 asdiscussed above.

2. The concept of widespread, complex fractionation-mixing processes is supported by several lines of evidence:(1) similar MgO, Ni, and Sr abundances with differentamounts of incompatible elements in different chemicalgroups within the same hole; (2) episodic eruptions ofsimilar magmas within the same hole; (3) stratigraphicinterlayering of basalts with changing Mg'-values inHoles 559 and 563, suggesting that multiply saturatedmagmas were periodically replenished by more primitivemelt; and (4) discrepancy between chemical composi-tion requiring clinopyroxene fractionation and the lackof clinopyroxene phenocrysts in the samples, a widelyobserved problem in ocean-floor basalt petrogenesis.

3. The Hayes Fracture Zone appears to be a "do-main" boundary between the depleted Type I basalts inthe south and the highly variable basalts in the area be-tween the Hayes Fracture Zone and the Azores TripleJunction (Fig. 1).

4. Hole 557 is one of the most highly enriched and/or evolved basalts sampled in the Atlantic Ocean (oranywhere else) (Melson et al., 1977; Basalt VolcanismStudy Project, 1981). The composition of Sample 557-1most closely resembles the FeTi basalts at the GalapagosSpreading Center (Schilling et al., 1976; Melson et al.,1977).

5. Although incompatible and REE ratios vary fromdepleted to enriched in Hole 558, (La/Ce)ch ratios (1.09-1.35) (Table 1; Fig. 9A) and Nd isotopic ratios (Jenneret al., this vol.) are nearly the same for all four groups inHole 558. These constant ratios suggest that complexmelting of a single source is probably the main processgenerating the observed compositional variations, al-though the varying Zr/Nb ratios (Fig. 8) may imply sourcevariability on a small scale.

6. Variations in trace element ratios and crossing REEpatterns appear to be restricted to the region of high vol-canic activity along elevated ridge segments in a hot-spot area (in this case, the Azores Platform) (Fig. 1).

7. Two distinct sources are indicated for Hole 561 be-cause the two (noninterlayered) chemical groups havedistinctly different, highly incompatible element ratios,including (La/Ce)ch (Table 1; Fig. 9B) and 143Nd/144Nd

(Jenner et al., this vol.).8. The concept of more pervasive, small-scale mantle

heterogeneity such as veining and/or metasomatismrather than a large-scale mantle plume or mantle blobsreceives support from several lines of evidence: (1) re-striction of high variability in trace element and REE ra-tios to hot-spot areas (Holes 556-561); (2) abrupt changesin Zr/Nb ratios (Fig. 8), REE patterns (Table 1; Figs. 9Aand 9B), and Nd isotopic ratios (Jenner et al., this vol.)between different groups within the same hole (Holes558 and 561); (3) drastic differences in composition andincompatible element ratios between the older, depletedbasalts in Hole 556 (30-34 Ma) and the younger, highlyenriched FeTi basalt in Hole 557 (18 Ma) (Appendix A,Table 1), which are about 22 Ma apart along the flowline passing through the Azores Triple Junction (Fig. 1);and (4) the existence of enriched basalts in older Hole559 (34-37 Ma) and of both enriched and depleted ba-salts in younger Hole 561 (14-17 Ma) along the sameflow line between the Oceanographer and Hayes frac-ture zones (Fig. 1).

9. The Azores hot-spot anomaly has existed in itspresent configuration for at least 18 Ma (Hole 557), butless than 30-34 Ma (Hole 556) (Fig. 1).

ACKNOWLEDGMENTS

We wish to thank the following people for their help in completingthis study: F. Frey for use of the MIT INAA laboratory and P. Ila forsample preparation and instruction in using the lab; Marie Litterer fordrafting the figures; Gerda Kunkel for typing the tables; and KarenThatcher of Words-Worth for typing the manuscript and additionaltables.

We also wish to express our appreciation to John F. Bender andJohn Longhi for their many helpful suggestions and meticulous carein reviewing this paper.

This research was funded by NSF Grant #7826330.

REFERENCES

Anderson, R. N., Clague, D. A., Klitgord, K. D., Marshall, M., andNishimori, R. K,. 1975. Magnetic and petrologic variations alongthe Galapagos Spreading Center and their relation to the Galapa-gos melting anomaly. Geol. Soc. Am. Bull., 86:683-694.

Basaltic Volcanism Study Project, 1981. Basaltic Volcanism on theTerrestrial Planets: New York (Pergamon Press), pp. 132-160.

Batiza, R., and Johnson, J. R., 1980. Trace element and isotopic evi-dence for magma mixing in alkalic and transitional basalts nearthe East Pacific Rise at 8°N. In Rosendahl, B.R., Hekinian, R., et

431


al., Init. Repts. DSDP, 54: Washington (U.S. Govt. Printing Of-fice), 63-69.

Blanchard, D. P., Rhodes, J. M., Dungan, M. A., Rodgers, K. V.,Donaldson, C. H., Brannon, J. C , Jacobs, J. W., and Gibson, E.K., 1976. Chemistry and petrology of basalts from Leg 37 of theDeep Sea Drilling Project. J. Geophys. Res., 81:4231-4246.

Bougault, H., and Treuil, M., 1980. Mid-Atlantic Ridge: zero agegeochemical variations between Azores and 22CN. Nature, 286:209-212.

Bryan, W. B, Thompson, G., Frey, F. A., and Dickey, J. J., 1976. In-ferred settings and differentiation in basalts from the Deep SeaDrilling Project. J. Geophys. Res., 81:4285-4304.

Dungan, M. A., and Rhodes, J. M., 1978. Residual glasses and meltinclusions in basalts from DSDP Legs 45 and 46: evidence formagma mixing. Contrib. Mineral. Petrol., 67:417-431.

Dupré, B., and Allègre, C. J,. 1980. Pb-Sr-Nd isotopic correlation andthe chemistry of the North Atlantic mantle. Nature, 286:17-22.

Frey, F. A., Bryan, W. B., and Thompson, G., 1974. Atlantic Oceanfloor: geochemistry and petrology of basalts from Legs 2 and 3 ofthe Deep Sea Drilling Project. J. Geophys. Res., 79:5507-5527.

Hanson, G. N., 1977. Geochemical evolution of the sub-oceanic man-tle. J. Geol. Soc. London, 134:235-253.

Jacobs, J. W., Korotev, R. L., Blanchard, D. P., and Haskin, L. A.,1977. A well-tested procedure for instrumental neutron activationanalysis of silicate rocks and minerals. J. Radioanal. Chem., 40:93-114.

Langmuir, C. H., Bender, J. F., Bence, A. E., Hanson, G. N., andTaylor, S. R., 1977. Petrogenesis of basalts from the FAMOUSarea: Mid-Atlantic Ridge. Earth Plant. Sci. Lett., 36:133-156.

Liias, R. A., and Rhodes, J. M., 1982. Does a mantle plume influencebasalt composition along the Juan de Fuca Ridge? EOS, Trans.Am. Geophys. Union, 63:1153-1154. (Abstract)

Melson, W. G., Byerly, G. R., Nelen, J. A., O>Hearn, T., Wright, T.L., and Valuer, T. L., 1977. A catalog of the major element chem-istry of abyssal volcanic glasses. Smithson. Contrib. Earth Sci.,19:31-60.

Melson, W. G., and O'Hearn, T., 1979. Basaltic glass erupted alongthe Mid-Atlantic Ridge between 0-37°N; relationships between com-position and latitude. In Talwani, M., Harrison, C. G., and Hayes,D. E. (Eds.), Deep Drilling Results in the Atlantic Ocean: OceanCrust: Washington (Am. Geophys. Union), pp. 249-261.

Norrish, K., and Chappell, B. W., 1967. X-ray fluorescence spectrom-etry. In Zussman, J. (Ed.), Physical Methods in DeterminativeMineralogy: New York (Academic Press), pp. 161-214.

Norrish, K., and Hutton, J. T., 1969. An accurate X-ray spectrographicmethod for the analysis of a wide range of geological samples.Geochim. Cosmochim. Ada, 33:431-453.

O'Nions, R. K., Hamilton, P. J., and Evensen, N. M., 1977. Variationin 143Nd/144Nd and 87Sr/86Sr ratios in oceanic basalts. Earth Plan-et Sci. Lett., 34:13-22.

Reynolds, R. C , 1967. Matrix corrections in trace element analysis byX-ray fluorescence: estimation of the mass absorption coefficientby Compton scattering. Am. Mineral., 48:1133-1143.

Rhodes, J. M., Dungan, M. A., Blanchard, D. P., and Long, P. E.,1979. Magma mixing at mid-ocean ridges: evidence from basaltsdrilled near 22°N on the Mid-Atlantic ridge. Tectonophysics, 55:36-61.

Schilling, J.-G., 1973. Iceland mantle plume: geochemical study ofReykjanes Ridge. Nature, 242:565-575.

, 1975. Azores mantle blob: Rare earth evidence. Earth Plan-et. Sci. Lett., 25:103-115.

Schilling, J.-G., Anderson, R. N., and Vogt, P., 1976. Rare earth, Fe,and Ti variations along the Galapagos spreading centre, and theirrelationship to the Galapagos mantle plume. Nature, 261:108-113.

Schilling, J.-G., Meyer, P. S., and Kingsley, R. H., 1982. Evolution ofthe Iceland hot spot. Nature, 296:313-320.

Tarney, J., Wood, D. A., Varet, J., Saunders, A. D., and Cann, J. R.,1979. Nature of mantle heterogeneity in the North Atlantic: evi-dence from Leg 49 basalts. Proc. Second M. Ewing Symp.,pp. 285-301.

White, W. M., and Bryan, W. B., 1977. Sr-isotope, K, Rb, Cs, Sr, Baand rare-earth geochemistry of basalts from the FAMOUS area.Geol. Soc. Am. Bull., 88:571-576.

White, W. M., and Hofmann, A. W , 1982. Sr and Nd isotope geo-chemistry of oceanic basalts and mantle evolution. Nature, 296:821-825.

White, W. M., and Schilling, J.-G., 1978. The nature and origin ofgeochemical variation in Mid-Atlantic Ridge basalts from centralNorth Atlantic. Geochim. Cosmochim. Ada, 42:1501-1516.

White, W. M., Schilling, J.-G., and Hart, S. R., 1976. Evidence forthe Azores mantle plume from strontium isotope geochemistry ofthe central North Atlantic. Nature, 263:659-663.

Wood, D. A., 1979. Dynamic partial melting: its application to thepetrogenesis of basalts erupted in Iceland, the Faeroe Islands, theIsle of Skye (Scotland) and the Troodos Massif (Cypress). Geo-chim. Cosmochim. Ada, 43:1031-1046.

Wood, D. A., Tarney, J., Varet, J., Saunders, A. A., Bougault, H.,Joron, J. L., Treuil, M., and Cann, J., 1979. Geochemistry of ba-salts drilled in the North Atlantic by IPOD Leg 49: implicationsfor mantle heterogeneity. Earth Planet. Sci. Lett., 42:77-97.

Zindler, A., Hart, S. R., Frey, F. A., and Jakobsson, S. P., 1979. Ndand Sr isotope ratios and rare earth element abundances in Reyk-janes Peninsula basalts: evidence for mantle heterogeneity beneathIceland. Earth Planet. Sci. Lett., 45:249-262.

Date of Initial Receipt: 18 November 1983Date of Acceptance: 4 April 1984

432

APPENDIX AChemical Compositions of Basalts from Leg 82

Note to Appendix A tables: Chemical group refers to chemical groups discussed in text. Fe2O3* and FeO+ refer to total Fe expressed as Fe3+ andFe 2 + , respectively. Calculation of Mg'-value discussed in text (Mg/[Mg + Fe2]). Small discrepancies in totals may be attributed to rounding.

Table 1. Holes 556 and 557.

Univ. Mass, sample no.

Core-Section(interval in cm)Sub-bottom depth (m)Chemical group

Major elements (wt.%)

SiO2

TiO2

A12O3

Fe2O3MnOMgOCaONa2OK2OP2O5

556-1Hole 556

2-174-76462.25

IA

50.551.05

15.639.540.157.58

12.522.010.350.12

556-2Hole 556

2-2141-144464.43

IA

50.221.05

15.968.720.157.51

13.312.020.200.12

556-3Hole 556

2-526-28467.77

II

50.041.49

16.0510.910.176.64

11.722.830.310.15

556-4Hole 556

3-366-69474.77

II

50.031.41

15.4611.420.186.92

11.742.580.370.16

556-5Hole 556

5-199-102490.01

II

50.211.49

15.9710.830.176.58

11.572.460.270.16

556-6Hole 556

6-564-67504.66

II

50.691.39

16.339.750.176.84

11.803.090.140.14

556-7Hole 556

6-6108-111506.60

IA

49.501.00

16.288.430.157.59

14.392.590.240.11

556-8Hole 556

7-1128-132508.30

IA

50.121.03

15.8910.080.167.17

12.742.600.290.11

556-9Hole 556

7-383-86510.85

IA

50.021.03

16.118.980.177.33

13.362.510.350.11

556-10Hole 556

8-213-16517.65

IB

50.370.88

15.858.330.147.87

14.252.380.270.11

556-11Hole 556

9-464-67530.16

IA

50.851.08

15.389.490.157.60

12.572.760.330.11

556-12Hole 556

10-363-66537.65

IB

49.710.83

17.797.550.127.19

14.162.540.250.10

556-13Hole 556

10-472-75539.24

IA

50.260.99

16.388.720.157.09

13.532.330.380.12

556-14Hole 556

11-2121-124545.73

IA

50.411.02

15.499.190.167.57

13.382.260.380.10

556-15Hole 556

12-354-57555.56

IA

50.330.97

16.288.910.147.58

12.902.210.330.11

556-16Hole 556

12-427-30556.78

IA

50.460.99

16.009.110.157.26

12.882.300.530.11

556-17Hole 556

16-2110-113586.1

IB

50.100.88

15.648.590.167.54

14.272.090.240.11

557-1Hole 557

1-150-53461.02

II

47.873.67

12.7316.510.225.01

10.002.660.520.41

Trace elements (ppm)

RbSrYGaZrNbZnNiCrV

Mg' -valueFeO*CaO/Al2O3

Zr/NbZr/Y

99.50 100.27 99.98 100.44 100.32

7.110523.415.762

2.37795

311241

0.648.580.80

27.22.67

2.410423.915.264

2.375

178277238

0.657.850.83

27.72.67

5.510132.716.797

2.29284

23555

0.579.820.73

43.92.96

7.19930.716.1892.4

8488

251238

0.5710.280.76

37.02.89

4.310133.116.4872.6

8788

251238

0.579.750.72

37.52.95

1.89830.717.0882.7

9797

280250

0.618.770.72

32.62.86

2.310122.614.7562.2

73145366215

0.667.590.88

25.52.49

6.39922.215.458

1.26891

331208

0.619.070.80

48.42.62

4.19723.315.0562.0

7481

373218

0.648.080.83

28.22.42

3.29920.414.048

2.168

106284221

0.687.500.90

23.12.38

10.799.324.315.260.4

1.980.173.393.8

253.2

0.648.540.82

31.82.49

3.899.118.216.146.4

2.476.190.0

381.2201.9

0.686.790.80

19.32.55

10.797.221.515.458.1

1.777.390.2

404.6229.4

0.647.850.83

34.22.70

9.4103.623.015.556.0

1.269.073.0

239.6228.0

0.648.270.86

46.72.43

5.795.721.814.755.6

1.071.091.6

387.7222.9

0.658.020.79

55.62.55

14.494.121.315.058.2

1.565.491.2

395.4224.7

0.648.200.81

38.82.73

2.898.520.114.350.4

1.669.4

135.7291.6231.5

0.667.730.91

31.52.51

10.8299.0

41.022.6

216.226.2

113.338.318.7

300.0

0.4014.860.618.35.27

Note: See note at beginning of Appendix A.


Table 2. Hole 558.

Univ. Mass, sample no.Core-Section(interval in cm)Sub-bottom depth (m)Chemical group


SiO2

TiO2

A12O3

Fe2O3MnOMgOCaONa2OK2OP2°5

Total


RbSrYGaZrNbZnNiCrV

Mg'-value

FeOCaO/AI2O3

Zr/NbZr/Y

558-127-3

112-114409.13

IA

50.601.25

14.6410.320.159.53

10.622.210.260.15

99.72

3.6128.922.215.775.87

8.6483

277.4509.7195.6

0.67

9.290.738.783.42

558-228-1

55-58415.07

II

51.091.16

15.828.800.168.26

12.402.150.170.12

100.14

2.594.725.61666.53

3.1090.33

169.8400.5238.3

0.67

7.920.78

21.462.60

558-330-1

115-118433.67IIIA

50.321.45

15.1610.460.158.17

11.582.300.460.21

100.27

9.717024.215.692.8015.2589.45

157.5356.3257.9

0.63

9.410.766.093.83

558-432-2

130-132453.31

IIIB

50.331.35

15.409.360.148.67

11.892.240.400.19

99.97

2.5179.521.815.184.6213.7585.61

170.6367.0242.8

0.67

8.420.776.153.88

558-532-4

73-75455.74

IIIB

50.421.30

14.869.830.178.97

11.632.130.310.17

99.80

6.6174.821.615.482.6014.4983.87

156.2388.9249.8

0.67

8.850.785.703.82

558-632-5

101-103457.51

IIIB

49.941.30

15.309.910.148.78

11.702.350.410.19

100.04

6.8176.421.21481.5013.3981.33

157.9371.6233.9

0.66

8.920.766.093.84

558-733-2

144-147462.46

IIIB

50.381.31

14.859.830.179.01

11.551.840.300.18

99.40

5.4174.521.815.482.1913.4187.74

156.1384.8243.7

0.67

8.850.786.133.77

558-833-3

61-64463.13

IIIB

50.291.30

14.929.860.169.05

11.511.880.300.18

99.45

6.817321.114.882.7913.9898.53

155.9398.3252.0

0.67

8.870.775.923.92

558-933-3

139-141463.90

IIIB

49.931.31

15.349.720.158.70

11.742.360.380.18

99.81

6.9175.221.216.280.2412.6584.74

175.0362.8233.4

0.66

8.750.776.343.78

558-1035-3

107-109481.58

IA

49.431.22

15.0310.140.178.92

12.132.520.350.16

100.06

6142.222.514.875.999.53

87.02185.8380.5239.9

0.66

9.120.817.973.38

558-1136-1

74-78487.26

IB

49.221.31

15.5410.430.187.82

12.262.280.410.20

99.63

7.7145.92415.385.6411.3487.44

152.5308.5246.4

0.62

9.390.797.553.57

558-1237-1

89-92496.41

IV

49.350.98

16.159.900.148.33

12.042.160.290.14

99.48

7111.218.514.755.407.57

78.63208.3458.4197.6

0.65

8.900.757.322.99

558-1338-1

97-99505.48

IV

49.120.95

15.709.870.17

10.8011.712.090.220.13

100.76

3.7108.2

18.515.254.767.24

73.14230.1

449.4190.5

0.71

8.880.757.562.96

558-1438-2

116-118507.17

IV

49.450.99

16.1510.100.168.33

12.212.120.230.13

99.87

3.9114.119.214.357.297.75

82.6219.6481.5211.5

0.65

9.090.767.392.98

558-1539-1

111-113510.12

IV

48.891.02

16.6110.480.177.54

12.922.320.300.14

100.38

4.9116.219.215.457.867.74

88.65184.88476.6224.4

0.62

9.430.787.483.01

558-1639-211-13

510.62IV

49.610.96

15.609.870.189.89

11.712.170.290.13

100.40

5109.118.214.355.747.44

72.26213.6441.5191.6

0.69

8.880.757.493.06


Table 3. Hole 559.

Univ. Mass, sample no.Core-Section(interval in cm)Sub-bottom depth (m)


SiO2

TiO2

A12O3

Fe2O3*MnOMgOCaONa2OK2O

P2O5

559-11-1

117-119239-18

49.231.61

16.1911.300.176.45

11.732.760.620.25

559-22-3

111-113251.12

50.621.52

15.499.930.127.84

11.572.700.350.20

559-34-1

121-124257.72

49.901.54

15.3510.440.167.56

11.782.640.380.21

559-45-1

50-53265.51

49.501.54

15.4011.210.196.69

11.942.770.440.21

559-56-2

14-16275.65

49.671.54

15.3910.590.157.28

11.692.390.440.20

559-67-1

111-114284.12

48.481.59

15.8611.710.196.38

12.352.470.420.24

559-77-3

18-20286.19

49.681.54

15.3110.770.157.46

11.352.540.510.20

559-87-3

103-106287.19

49.971.52

14.9710.800.167.52

11.242.590.390.21

559-98-3

6-10295.08

48.651.50

15.1412.020.246.77

11.982.330.690.20

Total


Rb

SrYGaZrNbZnNiCrV

Mg' -value

100.30 100.34 99.96 99.89 99.35 99.69 99.51 99.38 99.52

9.9167.931.613.3

108.415.5

176.0103.9163.8291.9

5.9162.527.115.7

105.414.490.4

211.2254.7246.5

5.2162.828.315.8

107.015.098.3

142.0267.3266.8

6.5166.628.016.0

105.715.2

102.7126.9260.9276.9

7.4161.827.817.2

107.514.791.0

127.3277.1267.1

6.9171.429.216.9

111.615.696.6

113.2276.1291.9

7.0161.828.216.5

106.815.496.5

121.0247.3253.8

7.0161.127.615.7

107.015.088.3

108.9245.5252.7

17.0169.027.117.2

103.114.583.6

134.4253.6253.6

0.56 0.63 0.61 0.57 0.60 0.55 0.60 0.61 0.55

FeO*CaO/Al2O3

Zr/NbZr/Y

10.170.726.993.43

8.940.757.323.89

9.390.777.133.78

10.090.786.953.78

9.530.767.313.87

10.540.787.153.82

9.690.746.943.79

9.720.757.133.88

10.820.79

7.113.80

Note: See note at beginning of Appendix A. All samples belong to Chemical Group I.

434


Table 4. Hole 561.

Univ. Mass. sample no.Core-Section(interval in cm)Sub-bottom depth (m)Chemical group

561-1HI-CC

3-58.60

I

561-21-14-9

411.57I

561-31-1

84-87412.36

II

561-41-2

24-26413.25

II

561-52-1

100-103415.52

II

561-62-2

82-84416.83

II

561-73-1

9-13417.61

II

561-83-1

76-79424.28

II

561-93-2

77-80425.79

II


SiO2

TiO2

A12O3

Fe2O3*MnOMgOCaONa2OK2O

P2O5

Total


Rb

SrYGaZrNbZnNiCrV

Mg' -value

49.711.23

15.499.970.167.93

12.332.480.370.23

49.751.21

15.269.870.158.18

12.102.510.360.23

50.451.37

15.2911.080.187.34

11.662.720.340.13

50.131.35

15.0511.610.187.48

11.312.570.300.14

50.251.33

14.9611.820.187.08

11.422.58054

0.13

49.991.35

15.2810.940.207.85

11.662.550.280.13

50.111.34

14.881.890.187.22

11.492.590.500.13

50.091.39

15.1011.390.177.44

11.692.530.290.13

50.051.37

14.7911.410.187.67

11.832.240.260.13

99.89 99.62 100.47 100.12 100.28 100.23 100.33 100.22 99.91

5.421625.515.6

10221.985

166416240

5.521525.514.8

10021.693

186413236

6.810032.318.176

3.5104120228297

6.98831.417.6752.1

109123245305

13.88730.717.4712.6

96126222292

6.59031.717.4732.6

106159237302

12.18631.016.873

3.1

96114217290

6.68632.617.0772.0

101128236313

5.28532.417.7763.0

101138238311

0.62 0.65 0.59 0.59 0.57 0.61 0.57 0.59 0.60

FeO*CaO/Al2O3

Zr/NbZr/Y

0.970.804.73.99

8.880.794.63.92

9.970.77

21.82.34

10.450.75

35.22.38

10.640.76

26.92.31

9.840.76

28.02.32

10.700.77

23.72.36

10.250.77

38.62.37

10.270.80

25.22.33


Table 5. Hole 562.

Univ. Mass, sample no.Core-Section(interval in cm)Sub-bottom depth (m)Chemical group


SiO2

TiO2

A12O3

Fe2O3MnOMgOCaONa2OK2OP2O5

Total


RbSrYGaZrNbZnNiCrV

Mg'-value

FeOCaO/Al2O3

Zr/NbZr/Y

562-1HI-CC16-19243.23

I

49.231.57

15.7111.900.186.71

11.642.290.250.15

99.63

2.610536.117.596.5

3.1103.9109.2241.8313.8

0.55

10.710.74

31.12.67

562-22-3

8-10253.09

I

49.911.58

15.3111.790.187.24

11.532.420.310.14

100.41

5.696.635.117.397.3

3.995.6

107.6236.5298.9

0.57

10.610.75

24.92.77

562-34-1

75-77268.76

I

49.831.60

15.2111.750.197.41

11.422.550.130.17

100.26

1.994.43517.892.4

3.890.1

106211.1284.1

0.58

10.570.75

25.62.78

562-45-1

64-66277.65

I

50.321.58

15.9710.700.166.73

11.852.520.210.16

100.19

2.399.534.917.396.4

3.495.8

119.5235.4286.7

0.58

9.630.74

28.42.76

562-55-4

29-31281.80

1

49.521.54

15.8211.250.177.05

11.532.700.100.15

99.83

1.2102.134.417.193.4

3.286.6

105.1215.3274

0.58

10.120.73

29.22.72

562-66-3

135-138290.37

I

49.321.54

15.6911.540.207.03

11.842.630.350.16

100.30

6.7103.632.917.993

3.1106.2109.2246.6286.1

0.57

10.380.75

30.02.83

562-76-4

6-10290.58

I

50.191.20

17.278.950.136.94

12.902.140.180.13

100.02

3.191.926.516.169

2.583

169.5244.9250

0.63

8.050.75

27.62.60

562-87-2

56-58297.07

I

49.981.20

16.2210.340.167.19

12.312.010.220.12

99.75

3.597.526.316.367.52.2

80.1105.7242.8243.3

0.60

9.300.76

30.72.55

562-98-1

132-134301.33

I

49.671.25

16.389.890.167.28

12.632.170.210.13

99.79

3.996.427.716.572.12.7

88.1102262267

0.62

8.900.77

26.72.60

562-108-1

144-146301.45

I

49.431.30

15.5010.170.177.61

13.232.200.250.13

99.98

4.5108.429.416.174.32.5

93.1121.6267.8278.4

0.62

9.150.85

29.72.53

562-118-2

25-29306.77

II

48.901.16

17.119.910.157.37

12.971.970.210.13

99.88

3.390.425.115.565.72.3

81.298.7

248.6243.4

0.62

8.920.76

28.62.62

562-129-1

82-84304.77

II

48.331.19

16.739.890.187.28

13.852.040.220.13

99.84

4.1101.32616.167.8

3.083.6

109.6251.2254.6

0.62

8.900.83

22.62.61

562-139-1

145-147305.46

II

49.781.28

15.7510.570.177.16

12.442.040.380.14

99.69

8.2114.428.716.973.42.3

93.9112.5265.1282.7

0.60

9.510.79

31.92.56

562-1410-1

90-93313.92

II

49.121.26

16.0111.070.187.09

12.962.10.320.14

100.24

6.195.12816.673.32.7

99.3104.2264.2277.3

0.58

9.960.81

27.12.62

562-1510-3

52-55316.54

II

49.721.29

15.3810.600.177.82

12.452.020.260.13

99.84

4.689.628.516.5742.5

95.2124.4268.1279.9

0.62

9.540.81

29.62.60

562-1611-1

115-117323.16

II

49.181.15

17.439.810.166.52

12.921.910.250.12

99.46

4.794.525.215.965

2.881.198.2

250.6244.2

0.59

8.830.74

23.22.58


435


Table 6. Hole 563.

Univ. Mass sample no.Core-Section(interval in cm)Sub-bottom depth (m)

563-123-1

115-117365.66

563-224-1

22-25366.74

563-324-1

57-60367.09

563-424-2

93-95368.94

563-524-2

123-127369.94

563-624-47-10

371.09

563-725-1

36-38373.87

563-825-2

115-117376.16


SiO 2

TiO 2

A1 2O 3

F e 2 O 3

MnOMgOCaON a 2 OK 2 O

Φ9.731.02

15.7810.180.177.61

13.201.960.260.09

49.470.97

16.4010.260.157.47

13.071.880.320.10

49.761.05

15.269.980.158.04

12.941.970.260.10

49.230.97

16.0910.430.177.54

13.002.290.290.09

49.981.05

15.409.890.168.03

13.012.210.250.09

49.580.99

16.0210.070.177.35

13.411.930.310.10

49.790.98

16.379.720.167.80

12.972.200.200.09

49.961.01

15.8510.350.177.55

13.062.200.270.11

Total 99.99 100.11 99.52 100.09 100.06 99.93 100.27 100.53


Rb

SrYGaZrNbZnNiCrV

Mg'-value 0.63 0.62 0.64 0.61 0.64 0.62 0.64 0.62

5.586.723.715.155.9

2.681.3

110.3330.2293.4

7.483.322.815.952.5

2.1178.097.97

322.8279.8

5.28624.915.659.2

2.981.195.3

338.1306.4

6.882.122.815.755.3

1.778.0

103.8326.1285.5

582.525.115.4

60.022.6

96.2106.9350.9310.7

6.882.322.914.954.7

2.578.9

106.8318.7286.1

2.582.621.615.454.5

2.277.8

106.9322.6272.9

5.783.723.615.554.9

2.679.196.7

324.9298.0

FeOCaO/Al 2 O 3

Zr/NbZr/Y

9.160.84

27.952.36

9.200.80

24.892.30

8.980.85

20.412.38

9.390.81

32.532.43

8.900.84

23.152.40

9.060.84

21.882.39

8.750.79

24.772.52

9.510.82

21.122.33

Note: See note at beginning of Appendix A. All samples belong to Chemical Group I.

Table 7. Hole 564.

Univ Mass, sample no.Core-Section(interval in cm)Sub-bottom depth (m)


SiO2

TiO2

A12O3

Fe2O3MnOMgOCaONa2OK2O

P2O5

Total


RbSrYGaZrNbZnNiCrV

Mg' -value

FeOCaO/Al2O3

Zr/NbZr/Y

564-1Hl-CC11-15

282.61

49.351.48

15.7711.550.196.13

12.822.510.440.16

100.40

7.5115.131.717.493.7

4.1101.3122.7277.4352.0

0.54

10.390.81

22.92.96

564-21-1

129-131285.30

50.211.43

14.9411.290.187.52

11.892.470.350.14

100.42

6.4106.8

32.216.191.3

3.995.3

138.5261.5346.6

0.59

10.160.80

23.42.83

564-31-3

38-40287.39

49.861.40

14.7111.580.177.59

11.632.760.360.15

100.22

6.0101.831.816.490.9

4.494.2

112.8258.2339.1

0.59

10.420.79

20.72.86

564-42-23-5

294.54

49.571.44

15.2211.600.187.00

12.152.390.390.16

100.10

5.8103.632.317.292.44.7

101.4121.1258.3348.9

0.57

10.440.80

19.72.86

564-52-3

18-20296.19

50.581.47

15.1710.110.177.77

12,052.260.200.15

99.93

2.7108.133.718.094.3

3.998.1

154.4242.0339.1

0.63

9.100.79

24.22.80

564-fi3-2

35-37303.86

49.621.45

14.9211.710.207.32

11.922.820.380.14

100.47

6.3105.532.817.295.4

5.0110.1104.5256.1349.2

0.58

10.540.80

19.12.91

564-74-2

37-41312.89

49.871.46

14.8911.920.207.32

11.762.310.360.15

100.26

5.8103.032.516.692.8

4.695.2

104.1233.6340.8

0.58

10.730.79

'20.22.86

564-84-4

140-142316.91

49.881.45

14.7611.820.167.54

11.522.560.360.17

100.22

7.299.932.216.492.9

4.998.2

117.8240.9338.4

0.58

10.640.78

19.02.88

564-95-2

89-91322.40

50.391.46

14.6911.820.177.48

11.522.480.190.16

100.34

2.699.532.717.295.5

4.2107.5117.5238.6339.1

0.58

11.130.78

22.72.92

564-105-4

24-27324.76

49.441.51

14.9912.370.196.78

11.762.580.400.17

100.19

7.0105.633.617.298.1

5.2102.1113.7240.9360.2

0.55

10.67.0.78

18.92.92

564-116-1

15-18329.17

50.541.48

14.6411.860.177.24

11.422.120.190.16

99.83

2.3100.933.416.694.6

5.491.8

118.6221.4335.2

0.57

10.670.78

17.52.83

564-126-4

48 52334.00

49.421.44

14.3812.310.198.43

11.282.050.240.16

99.92

4.9104.831.916.493.6

4.592.2

144.2214.1325.9

0.60

11.080.78

20.82.93

564-136-5

32-34335.33

50.171.50

14.6411.950.187.48

11.532.080.330.17

100.03

5.7101.233.017.096.6

4.994.3

123.0222.3340.2

0.58

10.750.79

19.72.93

564-147-1

12-15338.14

49.771.49

14.511.960.177.42

11.452.100.410.16

99.42

7.2103.633.317.297.3

5.298.4

119.8236.2349.4

0.58

10.760.79

18.72.92

564-157-3

13-15341.14

50.631.52

14.7311.560.186.99

11 672.150.230.16

99.82

2.6102.134.217.199.1

5.696.7

121.0235.3346.7

0.57

10.400.79

17.72.90

564-168-2

7-10348.59

49.871.48

14.5611.900.177.54

11.402.220.420.16

99.73

7.5103.232.516.297.4

5.597.6

118.6232.3346.0

0.58

10.710.78

17.73.00

564-179-1

18-20356.19

47.241.52

15.1613.070.195.20

13.S32.260.620.21

99.31

14.7114.635.619.396.1

5.885.3

103.9217.2383.7

0.47

11.760.91

16.62.70

564-189-1

140-143357.42

49.111.53

15.0812.010.186.86

11.762.660.360.17

99.73

6.9106.933.417.798.9

6.0111.2139.7222.8362.1

0.56

10.810.78

16.52.96

564-199-2

128-130358.79

48.991.58

15.0412.530.186.70

11.862.280.350.17

99.67

6.2107.934.117.1

10O.55.7

109.7121.2215.1371.3

0.54

11.280.79

17.62.95

564-209-3

68-71359.70

49.341.55

14.9612.560.206.81

11.872.440.330.17

100.23

6.3107.233.817.4

101.36.7

107.3117.3233.7365.3

0.54

11.300.79

15.13.00

Note: See note at beginning of Appendix A. All samples belong to Chemical Group 1.

436


APPENDIX BAverage Compositions of Basalts from Leg 82

Note to Appendix B tables: N refers to number of samples in each chemical group. Fe2O3* and FeO* refer to total Fe expressed as Fe3 +

Calculation of Mg'-value is discussed in the text (Mg/[Mg + Fe2]). Small discrepancies in totals may be attributed to rounding.Fe2

Table 1. Hole 556. Table 2. Hole 558.

Chemical groupN

Major elements (wt.

SiO2

TiO2

A12O3


I12

%)

50.110.90

15.998.400.157.44

13.262.350.330.11

IA9

50.271.02

15.949.120.157.43

13.162.360.340.11

IB3

50.060.86

16.498.160.147.53

14.232.340.250.11

II4

50.231.45

15.9510.730.176.75

11.712.740.270.15

Total 99.04 99.90 100.17 100.16


RbSrYGaZrNbZnNiCrV


Zr/NbZr/Y

6.49922.415.148.4

1.873

100340229

0.668.170.83

33.72.57

7.39922.715.258.5

1.77399

342230

0.648.250.83

36.42.58

3.3100

19.614.348.4

271

111319218

0.677.340.87

24.62.57

4.710031.816.692.8

2.59089

250248

0.589.660.73

37.82.92

Chemical groupN

III7

HIA1

IIIB6

Major elements (wt.•Vo)

SiO2

TiO2

A12O3


Total


IV5

49.751.26

15.0710.300.178.76

11.672.340.360.17

51.091.16

15.828.800.168.26

12.402.150.170.12

50.271.35

15.1410.110.168.52

11.632.220.410.20

50.321.45

15.1610.460.158.17

11.582.300.460.21

50.221.31

15.119.750.168.86

11.672.130.350.18

49.280.99

16.0510.060.168.90

12.112.120.260.13

99.85 100.14 100.01 100.27 99.75 100.06

RbSrYGaZrNbZnNiCrV


Zr/NbZr/Y

5.813922.915.379.2

9.887.2

169344243

0.659.270.788.103.46

2.59525.616.066.5

3.190

170400238

0.687.920.78

21.462.60

8.2017322.815.487.614.488

160368250

0.669.100.776.083.83

9.717024.215.692.815.289

157356259

0.649.410.776.093.83

6.717621.4515.282.313.687

162379243

0.678.790.776.063.83

4.9112

18.714.856.27.5

79205461203

0.669.050.767.453.00

Note: See note at beginning of Appendix B. Note: See note at beginning of Appendix B.

437


Table 3. Hole 559. Table 4. Hole 561. Table 5. Hole 562.

Chemical groupN


SiO2

TiO2

A12O2


49.521.54

15.4610.970.177.11

11.732.580.470.21

Total 99.77


RbSrYGaZrNbZnNiCrV

Mg' -valueFeOCaO/Al2O3

Zr/NbZr/Y

8.116528.316.0

106.915.0

103135258267

0.599.880.767.093.78

Note: See note at begin-ning of Appendix B.

Chemical groupN

Major elements (wt.

SiO2

TiO2

A12O3


I2

ft)

49.731.22

15.389.920.168.06

12.212.500.360.23

II7

50.151.36

14.9911.450.187.44

11.582.540.360.13

Total 99.75 100.18


RbSrYGaZrNbZnNiCrV


Zr/NbZr/Y

5.521525.515.2

100.921.769

176414238

0.648.920.804.653.95

8.38931.717.474.42.7

102130232301

0.5910.300.77

28.52.34

Note: See note at beginning of Ap-pendix B.

Chemical groupN

Major elements (wt.

SiO2

TiO2

A12O3


I6

ft)

49.691.57

15.5911.490.187.03

11.642.520.230.16

II10

49.431.23

16.3810.210.167.23

12.872.060.250.13

Total 100.10 99.95


RbSrYGaZrNbZnNiCrV


Zr/NbZr/Y

3.910034.217.595.7

3.496

109231291

0.5710.340.74

28.22.76

4.69827.116.270.22.6

88108256262

0.619.110.79

27.82.59

Note: See note at beginning of Ap-pendix B.

Table 6. Hole 563.

Chemical groupN


SiO2TiO2

A12O3


49.691.00

15.9010.110.167.67

13.082.080.27

0.10

Total 100.06


Table 7. Hole 564.

Chemical group IN


SiO2

TiO2

AI2O3Fe2O3

MnOMgOCaONa2OK2OP2O5

49.691.48

14.8911.870.187.16

11.852.380.350.16

Total 100.01


RbSrYGaZrNbZnNiCrV


Zr/NbZr/Y

5.68423.415.455.92.3

81103329292

0.639.160.82

24.62.39

RbSrYGaZrNbZnNiCrV


Zr/NbZr/Y

6.110533.017.195.6

5.099

117238349

0.5710.690.80

19.42.90



438

Documents

20. GEOCHEMISTRY OF BASALTS FROM DEEP SEA DRILLING …