deep sea drilling project initial reports volume 59 · example of both initial volcanism of an...
TRANSCRIPT
30. PETROLOGY AND GEOCHEMISTRY OF ARC THOLEIITES ON THE PALAU-KYUSHURIDGE, SITE 448, DEEP SEA DRILLING PROJECT LEG 59
Robert B. Scott, Department of Geology, Texas A&M University, College Station, Texas
INTRODUCTION
This chemical and petrologic study of rocks from Site448 on the Palau-Kyushu Ridge is designed to answersome fundamental questions concerning the volcanicorigin of remnant island arcs. According to the recon-struction of the Western Pacific prior to about 45 m.y.ago (Hilde et al., 1977), the site of the Palau-KyushuRidge was a major transform fault. From a synthesis ofexisting geological and geophysical data (R. Scott et al.,this volume), it appears that the ridge originated by sub-duction of the Pacific plate under the West PhilippineBasin. Thus the Palau-Kyushu Ridge should be a primeexample of both initial volcanism of an incipient arcformed by interaction of oceanic lithospheric plates andremnant-arc volcanic evolution. The Palau-KyushuRidge was an active island arc from about 42 to 30 m.y.ago, after which initiation of back-arc spreading formedthe Parece Vela Basin (R. Scott et al., this volume;Karig, 1975a). This spreading left the western portion ofthe ridge as a remnant arc that separates the WestPhilippine Basin from the Parece Vela Basin. In spite ofnumerous oceanographic expeditions to the PhilippineSea, including the two previous DSDP Legs 6 and 31(Fischer, Heezen et al., 1971; Karig, Ingle et al., 1975),and even though the origins of inter-arc basins havebeen linked by various hypotheses to that of remnantisland arcs (Karig, 1971, 1972, 1975a, and 1975b; Gill,1976; Uyeda and Ben-Avraham, 1972; Hilde et al.,1977), very little hard data are available on inactive rem-nant arcs.
Continental orogenic belts contain complexes such asthose in Newfoundland and the Caucasus Mountainsthat have been interpreted as structural fragments ofmarginal basin-magmatic arc complexes. Thus recogni-tion of processes involved in continental-crust genera-tion requires a thorough understanding of the genesis ofboth remnant arcs and marginal basins. To this endseveral questions must be addressed: (1) Is the Palau-Kyushu Ridge actually a remnant arc? (2) If so, doesthis remnant arc have any petrologic features unique tothe related marginal-basin formation? (3) Does the arcrepresent the initial phases of magmatic-arc evolution;that is, does volcanism on the Palau-Kyushu Ridge havean arc-tholeiitic character followed by progressivelymore calc-alkalic volcanism on the West Philippine andMariana Ridges? Or is the initial arc-tholeiite-magmaticphase repeated each time a new arc is formed after sun-dering of the previous arc? (4) Is it possible to use thecombination of petrologic and geomorphologic charac-teristics of the Palau-Kyushu Ridge to specify how and
where the magmatic arc was sundered? (5) What is thetemporal relationship between the cessation of arc vol-canism and the sundering of the arc (with subsequentback-arc basin formation)? (6) Is there any indicationthat immature, largely submarine arcs have establishedwidespread hydrothermal activity that may be the pre-cursor of ore-deposition mechanisms characteristic ofmature island arcs?
The most extensive petrologic studies of Palau-Kyushu Ridge rocks prior to Leg 59 were reported bythe International Working Group on the IGCP Project"Ophiolites" (1977), which used suites of samples col-lected during cruise 17 of the Dmitry Mendeleev be-tween June and August of 1976. Rocks were collectedfrom two dredge stations (1397 and 1396) located on thesteep western face of the Palau-Kyushu Ridge (see Site448 report for a location map, this volume.). The rockscollected included subalkalic basalts, basaltic andesites,andesites, two-pyroxene gabbros, and low-grade pum-pellyite-prehnite or greenschist-facies hydrothermalmetamorphic rocks. These volcanic rocks are reportedto be highly vesicular, typical of shallow-marine or evensubaerial volcanism. Two analyses of typical basalticandesites are presented in Table 1 for comparison withour own findings. The International Working Groupconcluded that the highly vesicular nature and thepetrologic trends of these rocks are typical of island-arcmagmatic products. The high Fe/Mg ratios and low Ni,Cr, Sr, and Ba contents are suggestive of arc tholeiites.The high K is probably the result of sea-water weather-ing. The low Ba and Sr and high FeO/FeO + MgO ratioalso rule out the possibility that these may be calc-alkalic series differentiates even though the Al valuesare high. Results from these dredged samples also sug-gest that volcanic activity ceased by the end of theOligocene.
Table 1. Two analyses of typical basaltic andesites from IGCP Project"Ophiolites," Palau-Kyushu Ridge, 1977.
SiO2 TiO2 AI2O3 FeOa MnO MgO CaO Na2θ K2OSamples (wt. %)
Site 1396(Rock 1-2)Site 1397(Rock 2-1)
52.3
54.2
1.10
1.15
19.4
19.0
9.44
8.99
0.13
0.22
2.73
2.50
8.17
8.50
3.64
3.51
1.00
0.45
Site 1396(Rock 1-2)Site 1397(Rock 2-1)
B
<IO
<IO
Ba
70
40
F
400
200
Li
21.5
4.5
Sr(ppm)
340
270
Zn
173
123
Cu
190
85
Co
100
100
Ni
45
40
Cr
320
260
FeO* stands for total Fe expressed as FeO.
681
R. B. SCOTT
Studies of volcaniclastic debris drilled at DSDP Hole296 (Ingle, Karig, et al., 1975) indicate that major vol-canism with an accumulation rate of 140 m/m.y. wanedduring the late Oligocene on the northern part of thePalau-Kyushu Ridge. Here abundant pyroclastic debrisalso indicates a shallow-marine or subaerial origin ofthe volcanic material; the Leg 31 shipboard scientiststheorized that the Palau-Kyushu Ridge was an active arcthat became inactive and subsequently submerged as theShikoku and Parece Vela Basins began to form. Mag-netic anomaly identification in the Shikoku Basin(Watts and Weissel, 1975; Shih, in press) suggests that25 m.y. ago, at the time of cessation of volcanism at Site296, the Shikoku Basin began to open; this was some 5m.y. after cessation of volcanism at Site 448 and the in-itial opening of the Parece Vela Basin.
METHODSAnalytic methods used in this study include atomic-absorption
spectrophotometry (AAS), instrumental neutron-activation analysis(INAA), x-ray fluorescence (XRF), electron-microprobe analysis, anda variety of colorimetric techniques. A Perkin-Elmer model 306 atTexas A&M was used for major-element analyses by AAS with thefollowing precisions in one standard deviation of the oxide: AI2O3 =3.5%, FeO* = 2.2%, MgO = 2.0%, CaO = 2.0%, Na2O = 1.2%,K2O = 1.4% and MnO = 2.0% (throughout this chapter, FeO* istotal Fe expressed as FeO). The XRF data were collected at the Uni-versity of Birmingham as part of the data made available for ship-board scientific use. The ARL-EMX electron microprobe at theUniversity of Texas at Austin was used for probe work on glasses andcrystalline phases; one-standard-deviation precision show that SiO2 =0.7%, TiO2 = 3.8%, A12O3 = 0.5%, FeO* = 1.2%, MnO = 7.9%,MgO = 1.6%, CaO = 0.8%, Na2O = 1.7%, and K2O = 12.5%.Data reduction was done by a modified Bence-Albee program de-signed for a Hewlett Packard 9825 (Tiezzi and Scott, in press). TheTRIGA reactor, a germanium-lithium (GeLi) detector, and a 4096channel Canberra-Scorpio analyzer system at Texas A&M Universitywere used for the INAA studies; one-standard-deviation precisions in-dicate that Fe = 1.8%, Co = 0.9%, Cr = 1.3%, Hf = 12%, Sc =0.8%, La = 1.5%, Ce = 10%, Sm = 2.1%, Eu = 3.0%, Tb = 10%,Yb = 10%, and Lu = 15%. Colorimetric methods were used forwhole-rock SiO2 and TiO2 analyses (with a standard deviation of0.8% for SiO2 and 6.0% for TiO2).
RESULTSTable 2 lists in stratigraphic order preliminary major-
element data on whole-rock samples (by AAS and XRF)and on glasses (by microprobe) for rocks cored at Site448. Analyses of clasts from volcaniclastic sedimentaryunits inserted between flows are included in the table.Trace-element data by XRF (see Mattey, et al., thisvolume) will be referred to in the discussion but will notbe reported here. Major-element analyses of individualcrystalline phases by microprobe are presented in Table3, and the INAA trace-element and rare-earth-element(REE) data are compiled in Table 4.
DISCUSSION
Petrography
A clear petrographic distinction between the flowsand volcaniclastic sedimentary units of the remnant arcsin the Philippine Sea and the flows of the interveningmarginal basins is evident, even without chemical analy-sis. Detailed petrographic descriptions are given in the
Site 448 chapter for the Palau-Kyushu Ridge, in the Site451 chapter for the West Mariana Ridge, and in Sites449 and 450 chapters for the Parece Vela Basin. Thestratigraphy of flows and volcaniclastic units in Site 448are also displayed in figures and discussed in detail inthe Site 448 chapter. Only a short summary is givenhere.
The basalts of the West Philippine Basin basementand the Parece Vela Basin basement are commonlyplagioclase-olivine-spinel phyric and relatively nonvesic-ular; in contrast, the basaltic flows and clasts on the tworidges are commonly plagioclase-clinopyroxene-ortho-pyroxene phyric and have a pronounced vesicularity (upto 30%) typical of shallow, more volatile eruptions ex-pected in island arcs. Although fresh olivine is rare inthe remnant-arc units (it is found by probe analysis onlyin Unit 6), there are altered relic phenocrysts that are in-terpreted as olivine pseudomorphs in Units 8, 11, 22,and 28. A coarsely crystalline clast in Unit 10 (tuffs andvolcaniclastic breccias) contains hornblende, plagioc-lase, quartz, magnetite, and ilmenite but no K-feldsparand probably is a hornblende diorite. This clast, how-ever, has a chemistry (discussed later) characteristic ofarc tholeiites. The petrographic data from rocks studiedat Site 448 therefore support the contention that theyhave arc-tholeiite affinities.
Another petrographic feature that distinguishes be-tween ridge and basin volcanic rocks is the degree ofalteration. Except for the glass margins of pillow lavas,the basalts on the ridges are severely altered and containabundant zeolites and clay minerals (see Aldrich et al.,this volume). A hydrothermal-metamorphic gradientexists in the flows and volcaniclastic breccias that mayreflect a geothermal gradient 7 to 9 times the normalgradient; certainly such a thermal regime is in keepingwith an island-arc environment. Although the heat flowin the adjacent back-arc basins is slightly high (Watan-abe et al., 1977), there is no indication of abnormalhydrothermal alteration effects upon the basin basalts.Probably the severe alteration of ridge basalts is a func-tion of the high geothermal gradient, the high porosityin the flows, and the numerous permeable volcaniclasticunits separating the flows.
We made an attempt, based upon petrographicfeatures, to distinguish among flows, sills, and dikes atSite 448. We tentatively identified 4 sills (Units 33, 35,45, 47) and 6 dikes (Units 26, 27, 28, 37, 49 and 51) fromthe total of 26 recognized igneous units. As expectedfrom physical models of island-arc construction, the in-trusive units are toward the base of the pile, have adistinctly lower vesicularity, and average only 6 metersthick—significantly thinner than the 15-meter-thickflows. There are several examples of clear-cut dike rela-tionships where laminated tuffs and massive volcani-clastic breccias are cut by igneous intrusions at attitudessignificantly greater than those of bedded units aboutand below the igneous unit. In some cases it is possibleto identify a unit as a dike, even though contacts are ab-sent, when flow structures of vesicles or phenocrysts arenot parallel to bedding of adjacent units. This is particu-larly useful because of incomplete recovery of core. Sills
682
PETROLOGY AND GEOCHEMISTRY OF ARC THOLEIITES
Table 2. Major-element whole-rock and glass chemistry of flows, dikes, sills, and clasts in breccias be-tween flows from Site 448.
Samplea
(by unit)
Unit 6 (flow)448-37-1 (32)448-37-1 (52)448-37-1 (52)448-37-1 (52)448-39-1 (75)448-39-1 (75)448-39-1 (75)448-39-1 (116)448-39-2 (32)448-40-1 (68)448-40-1 (82)
Unit 8 (flow)448-43-1 (119)448-43-2 (21)448-44-1 (103)
Unit 9 (flow)448-47-2 (22)
448-47-2 (22)448-47-2 (22)448-48-1 (77)448-48-1 (119)448^8-2 (18)448-48-2 (93)448-48-3 (22)
Unit 10 (clast)448-50-3 (80)448-50-4 (50)
Unit 11 (flow)448-51-4(38)448-52-1 (80)448-53-1 (2)448-53-1 (2)448-53-1 (2)448-53-2 (58)448-53-2 (59)
Unit 12 (clast)448-55-1 (110)448-57-1 (33)448-57-2 (16)448-57-2 (102)448-58-2 (58)448-58-2 (49)
Unit 13 (flow)448-58-3 (77)448-58-3 (113)448-58-3 (113)448-58-3 (113)448-58-4 (138)448-59-1 (62)448-59-1 (62)448-59-1 (62)448-59-2 (61)448-59-3 (51)448-59-3 (51)448-59-3 (74)448-61-3 (55)
Unit 13 (flow)448A-15-1 (30)448A-15-1 (30)448A-15-1 (30)448A-15-1 (123)448A-15-3 (143)448A-16-2 (71)448A-16-2(71)448A-16-2(71)
Type of Analysis'3
X
aPIP2a
PIP2X
X
aX
X
aX
a
PIP2X
aaX
X
aa
A
X
a
PIP2aX
aaaX
aa
X
aPIP2X
a
PIP2X
PIP2X
X
aPIP2X
X
a
PIP2
SiO2
43.250.851.651.850.351.051.049.449.350.749.8
48.551.249.6
50.9
52.853.047.950.750.447.749.8
60.047.2
49.449.250.651.952.449.849.2
50.648.749.248.950.351.0
49.953.755.454.148.052.955.455.650.754.554.850.050.2
52.654.954.749.450.252.255.255.2
Tiθ2
1.541.521.441.501.581.511.431.641.631.571.52
1.601.571.54
1.48
1.231.321.601.531.571.551.52
1.301.45
1.281.211.271.111.011.191.15
1.050.931.210.810.971.30
1.121.041.141.091.021.161.121.171.091.091.161.101.12
1.031.111.091.111.081.061.121.02
A12O3
14.213.513.013.0
12.7312.9412.88
13.814.012.912.7
13.413.614.0
14.1
13.412.714.613.913.514.114.3
12.816.6
13.413.913.613.314.314.514.7
14.516.116.717.114.314.2
14.513.213.613.314.213.513.313.516.814.213.216.415.0
15.1513.513.815.916.415.713.513.4
FeO*
13.2514.8515.5815.6214.5515.0215.2413.5813.7413.5913.52
14.5314.3714.00
13.73
11.8813.9914.6115.1514.1614.2613.94
11.1111.30
14.3912.6612.4313.4513.2013.5612.12
6.3910.8112.2810.4712.8511.36
11.2912.1612.3712.6010.2411.7012.4312.5310.8112.6612.6910.4811.01
10.9612.0812.2211.0210.7011.3612.3012.40
MnO
0.220.29
_ c—
0.33——
0.220.220.290.24
0.280.170.23
0.31
—
0.230.240.260.250.23
0.240.16
0.220.240.26
——
0.220.20
0.110.070.150.220.280.27
0.190.17
——
0.170.21
——
0.19——
0.170.15
0.16——
0.200.180.17
—
—
MgO
4.674.274.044.024.324.044.124.644.144.394.67
3.594.054.63
3.77
3.143.203.243.283.813.374.48
2.374.48
5.114.564.334.624.484.875.19
2.296.564.895.375.293.92
4.524.323.653.643.885.253.673.694.473.663.704.286.30
4.053.563.554.684.024.073.693.71
CaO
10.279.038.608.608.278.848.719.969.60
10.208.82
8.638.349.54
8.52
8.057.989.458.198.27
10.299.53
5.727.93
9.299.695.298.788.919.03
10.23
4.329.567.83
10.569.278.38
9.247.728.358.11
12.748.198.118.459.807.838.07
10.278.84
8.418.268.03
10.1210.168.168.188.20
Na 2O
2.932.582.492.492.292.462.592.903.172.552.72
2.812.552.95
2.26
2.382.073.112.772.613.002.85
2.773.93
2.402.802.412.462.622.293.33
4.532.572.952.452.342.64
3.392.422.432.804.562.792.532.742.982.402.233.583.44
2.732.231.943.473.162.382.702.79
K2O
0.721.370.380.381.320.420.390.790.800.591.61
1.310.630.57
1.40
0.380.410.860.510.590.610.68
0.461.17
1.101.141.230.370.330.510.47
4.251.091.360.560.501.79
1.651.570.440.521.201.380.490.450.830.450.441.191.41
1.490.410.440.941.391.260.490.44
683
R. B. SCOTT
Table 2. (Continued).
Samplea
(by unit)
Unit 14 (flow)448A-16-3 (10)448A-17-1 (87)448A-18-1 (77)448A-18-1 (112)448A-18-1 (112)448A-18-1 (112)448A-18-1 (112)448A-20-1 (118)448A-20-4 (4)
Unit 15 (clast)448A-21-1 (56)448A-21-1 (77)448A-22-1 (7)448A-25-1 (23)
Unit 16 (flow)448A-26-2 (32)448A-26-2 (60)
Unit 18 (flow)448A-28-1 (52)448A-28-1 (96)
Unit 20 (flow)448A-32-1 (47)448A-33-1 (134)448A-33-2 (6)
Unit 22 (flow)448A-36-5 (53)448A-36-7 (130)
Unit 23 (flow)448A-37-2(141)448A-37-2 (145)448A-38-1 (99)
Unit 24 (flow)448A-39-2 (140)448A-40-1 (33)
Unit 26 (dike)448A-41-1 (17)448A-41-1 (90)
Unit 27 (dike)448A-41-2 (12)448A-41-2 (12)448A^l-2 (58)448A-41-4 (4)448A-41-4 (16)
Unit 31 (flow)448A-44-1 (34)448A^4-2 (55)448A^5-1 (43)448A^5-2 (13)448A-45-2 (22)448A^6-2 (82)
Unit 33 (sill)448A-47-1 (72)448A-47-1 (72)448A-47-2 (108)448A-47-2 (106)448A-47-3 (138)448A^7-3 (138)448A-47-3 (138)
Unit 35 (sill)448A-50-1 (47)448A-50-1 (95)448A-50-1 (99)448A-50-2 (35)
Type of Analysis"
X
\
\
PIP2P3P4X
X
aaaX
aX
X
a
X
aa
a
aX
a
X
a
X
a
a la 2X
X
a
aX
aaX
X
aaX
X
a la 2a 3
aaaX
SiO2
50.049.750.855.455.455.255.951.249.8
51.850.449.949.8
47.547.7
49.048.2
51.249.649.7
49.950.6
50.950.049.6
49.850.2
49.750.3
49.550.049.950.550.4
54.154.553.654.452.154.7
53.553.948.355.054.753.254.1
53.053.753.453.3
Tiθ2
1.09.15.12.11.14.13.20.08.29
1.031.171.211.28
1.311.22
1.131.01
1.411.361.33
1.010.95
1.481.631.51
1.361.31
1.251.30
1.281.271.141.431.30
1.321.491.411.331.361.21
1.381.421.551.231.361.581.27
1.371.421.391.45
A12O3
15.915.514.513.413.513.413.714.214.0
14.514.614.614.1
15.813.2
15.412.9
13.414.615.2
14.013.5
12.812.612.0
13.214.9
13.614.2
13.413.212.613.113.4
13.913.013.413.813.913.5
13.813.413.913.513.313.515.2
13.612.412.912.8
FeO*
11.0011.4911.4512.6312.6312.5012.6612.2911.35
10.1211.5111.1311.35
11.7311.72
12.611.23
12.6614.1712.52
10.7911.56
12.2913.0712.06
12.8313.14
12.1011.22
11.5611.7212.9212.7412.56
11.7711.6311.7711.5811.2611.87
12.9712.9914.3212.4012.5012.9712.01
12.9813.2412.2412.49
MnO
0.190.200.21
————
0.210.21
0.130.590.070.21
0.210.37
0.200.26
0.180.260.28
0.040.26
0.240.220.27
0.220.48
0.210.27
0.280.270.200.220.28
0.340.220.160.100.190.18
0.170.120.250.180.100.080.14
0.320.360.340.24
MgO
4.964.114.143.663.623.653.654.664.41
2.974.093.666.21
6.046.47
7.029.22
6.195.875.47
6.678.18
5.806.436.35
5.614.71
6.704.31
6.446.317.467.267.44
5.865.855.264.506.165.71
4.324.073.845.594.373.424.36
6.025.045.335.44
CaO
9.799.728.938.338.258.308.298.328.88
7.937.987.769.32
9.079.93
6.146.38
8.298.808.75
10.0710.33
7.299.016.40
9.218.23
9.158.44
9.088.449.028.948.08
6.647.557.508.029.457.75
7.086.189.527.747.066.756.92
7.588.227.718.33
Na2O
3.173.003.142.642.422.722.672.993.04
3.343.053.473.52
3.563.48
2.442.58
3.102.722.74
2.062.62
2.322.761.93
4.552.49
2.642.53
2.282.532.793.022.82
2.362.942.402.522.462.81
2.703.612.922.842.972.782.76
2.752.262.472.55
K 2 O
1.051.331.820.450.470.470.502.141.66
0.881.191.250.63
1.101.63
1.182.03
1.470.670.65
0.160.16
1.031.163.73
1.110.57
1.281.75
0.451.750.810.500.45
0.120.360.170.180.200.34
0.300.560.790.320.280.360.30
0.320.280.330.30
684
PETROLOGY AND GEOCHEMISTRY OF ARC THOLEIITES
Table 2. (Continued).
Samplea
(by unit)
Unit 37 (dike)448A-51-1 (85)448A-51-1 (85)448A-51-1 (85)448A-51-1 (85)448A-51-1 (85)448A-51-2 (134)448A-51-3 (135)
Unit 39 (flow)448A-52-1 (77)448A-52-2 (128)
Unit 41 (flow)448A-54-3 (81)448A-54-3 (115)448A-56-1 (31)
Unit 43 (flow)448A-57-1 (96)
Unit 47 (sill)448A-59-2 (85)448A-59-2 (126)
Unit 49 (dike)448A-62-1 (60)448A-62-1 (60)448A-62-2 (7)
Unit 51 (dike)448A-65-1 (81)448A-65-2 (130)448A-66-2 (125)448A-66-2 (125)
Type of Analysis'3
aPIP2P3P4\a
X
a
aX
a
X
X
a
a la2a
\a
a la 2
SiO2
54.342.442.954.254.454.854.2
51.151.6
53.854.154.3
55.2
51.553.2
52.753.053.6
52.252.651.952.7
TiO2
1.331.791.861.311.311.271.34
1.481.51
1.321.251.28
1.28
1.681.52
1.571.631.49
1.171.221.261.31
A12O3
12.312.011.712.713.012.812.1
12.912.6
13.513.613.5
12.9
13.513.3
12.112.812.4
13.313.013.613.2
FeO*
12.8719.3720.7314.4214.1112.2711.38
12.9212.91
11.2211.7111.46
12.12
11.9511.32
12.4812.949.80
12.0712.5211.4511.99
MnO
0.29
——
0.200.17
0.220.09
0.180.210.10
0.18
0.210.18
0.120.180.10
0.220.200.050.08
MgO
4.865.226.093.233.335.975.41
6.486.36
4.295.784.70
6.01
5.805.33
5.095.015.58
5.915.535.415.43
CaO
7.308.208.207.678.127.697.36
8.767.82
8.838.588.62
7.76
9.547.36
9.517.909.55
9.048.238.618.25
Na2O
2.211.431.372.632.892.582.53
2.482.70
2.532.612.56
2.63
2.922.50
2.462.662.59
2.632.502.373.24
K2O
0.290.380.420.420.340.250.34
0.100.41
0.270.310.33
0.37
0.380.19
0.250.420.14
0.190.200.230.38
a Sample designations give the hole number-core number-section, and center of the sample interval (in parentheses)in stratigraphic order. Unit numbers are the same as those in the Site 448 chapter.
b x i X R F data collected at the University of Birmingham by John Tarney, David Mattey, and Nicholas Marsh foruse by shipboard scientists on Leg 59; a = atomic absorption and colorimetric methods; p = glass microprobedata; subscripts 1, 2, 3, and 4 refer to different areas analyzed by probe or by atomic absorption and colorimetricmethods at the same interval.
c Dash (—) indicates that no analysis was made.
are more difficult to recognize. Where upper and lowercontacts show abnormally thin quenched margins andusually an absence of glass, sills may be expected. Someintrusive units, however, can apparently form glassymargins against cool country rock, as shown by the thinglassy margin of the basaltic dike of Unit 37. In onecase, the chilled margins of a sill have been symmetri-cally deformed to produce drag features on the top andbottom, presumably created by repeated injections ofmagma. In another case, a symmetrical differentiationof plagioclase laths at the top and bottom of a sill sug-gests that symmetrical boundary-flow conditions ex-isted. In either case it is difficult to envision mechanismsthat might create such symmetry in an extrusive flow.
Chemistry
Whole-rock chemistry (Table 2) of nonglass flow in-teriors by AAS and XRF shows a high degree of scatter;attempts to characterize the units using Kuno's (1965)SiO2-alkali (Na2O + K2O) plot suggested that alkali-olivine, high-alumina, and tholeiitic basaltic magmas allcoexisted in one 600-meter sequence of lavas. Thisunlikely possibility was supported by Miyashiro's (1974)SiO2-FeO*/MgO plots, in which trends of analyses fell
into both the calc-alkalic and the tholeiitic fields. TheAFM plot of these data is somewhat more discrimina-tory (Fig. 1) but still shows considerable scatter betweenthe Skaergaard tholeiitic trend and those of Asama andAmagi calc-alkalic trends. If only probe analyses offresh glasses are plotted, however, a tight group near theSkaergaard trend line is produced (Fig. 2) with anelongation toward the alkali apex. An enlargement ofthat group (Fig. 3) from 5% to 25% alkalis, 60% to80% FeO*, and 15% to 35% MgO shows that eachvolcanic unit has very distinct alkali-enrichment trends.Such trends might be created by two scales of crystalfractionation: (1) crystal fractionation on a unit-widescale or (2) local crystal fractionation within thequenched pillow margins. The tholeiitic character ofthese rocks is further confirmed by the Kuno plot (Fig.4) and the Miyashiro plot (Fig. 5) of glass compositions.Obviously the quenched glassy margins of the pillowsrepresent a reasonable approximation of a closed systemeven in otherwise extensively altered rocks (Scott andHajash, 1976).
Comparisons among the tholeiitic mid-ocean-ridgebasalts, arc tholeiites, calc-alkalic basalts, and thedrilled Palau-Kyushu Ridge rocks using both major-
685
R. B. SCOTT
Table 3. Major-element chemistry of phases within flows, sills, and dikes from Site 448.
Samplea
448-37-448-37-448-39-448-39-448-39-448-39-
(52)i(52)2(75)i(75)2
(75)3(75)4
448-53-1 (2)448-58-3 (113)!448-58-3 (113)2448-59-1 (62)
448A-18-1 (112)i448A-18-1 (112)2448A-41-4 (16)i448A-41-4 (16)2
448A-50-1 (99)i448A-50-1 (99)2448A-51-1 (85)448A-56-1 (31)!448A-56-1 (31)2
448A-59-2 (126)i448A-59-2 (126)2448A-62-2 (7)i448A-62-2 (7)2
448-58-3 (113)!448-58-3 (113)2448-58-3 (113)3448-59-1 (62) i448-59-1 (62)2
448A-15-1 (30) i448A-15-1 (30)2448A-51-1 (85)i448A-51-1 (85)2
448-37-1 (52)
448-50-3 (80) M T a
448-50-3 (80)iLa
448A-41-4 (16)IL448A-50-1 (99) I Li488A-50-1 (99) I L 2
448A-51-1 ( 8 5 ) M T
448A-56-1 (31) I L
448A-59-2 (126)ILi448A-59-2 (126) I L 2
448A-62-2 (7) I L
448-50-3 (80) i448-50-3 (80)2
448-37-1 (52)A
448-37-1 (52)H
448-37-1 (52)L
448-39-1 (75)Ai448-39-1 (75)Hi448-39-1 (75) A 2
448-39-1 (75) L 2
448-39-1 (75) A 3
448-47-2 (22)A
448-50-3 (80)A
448-50-3 (80)H
448-50-3 (80)L
448-53-1 (2) A
448-58-3 (113)Ai448-58-3 (113)H2448A-58-3(113)L2
448-59-1 (62) A i448-59-1 (62) A 2
448-59-3 (51)A
Unit
666666
10121212
14142727353537414147474949
1212121212
12123737
6
99
2735353741474749
99
666666668999
10121212121212
SiO2
50.9
50.551.150.652.350.652.251.852.4
51.153.151.150.753.053.349.451.451.851.549.155.855.4
52.7—
54.154.8
53.9
51.752.6
36.59
0.88
0.00
2.093.561.070.132.490.420.170.24
52.450.9
47.846.348.046.846.448.648.8—
48.253.151.557.143.448.747.649.346.045.945.2
Tiθ2
0.58
0.580.560.561.430.460.360.340.31
0.320.340.360.310.420.300.790.350.340.350.690.360.51
——
0.290.17
0.18
0.360.28
0.00
4.6618.71
15.4314.9815.099.13
13.6312.5115.5018.30
0.040.21
0.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.00
A12O3FeO* MnO
Clinopyroxenes
3.23
2.963.123.073.303.554.482.242.20
1.982.052.092.032.372.323.892.492.312.103.592.353.22
15.1313.7513.9317.0814.5714.9612.937.05
11.9810.02
11.5111.2911.379.12
10.7510.4116.839.43
10.0010.8213.0710.7211.08
e
_———————
—0.030.280.330.300.540.260.290.290.320.320.29
Orthopyroxenes
1.67—
2.141.14
1.68
1.861.70
0.00
0.74
1.28
3.933.213.154.433.003.823.203.20
0.441.82
28.1129.7423.4230.4831.1129.0728.64
—28.1326.8227.0723.7033.7732.8033.2532.2132.9332.7932.20
16.3618.4015.8117.7118.13
17.9017.5321.3820.25
Olivine
32.65
Opaques
81.53
77.46
75.6079.1378.3880.2178.8077.5878.3977.86
14.0813.44
3lagioclase
———
—————————————
—
——
0.550.60
—
—
0.420.370.390.300.330.350.420.46
—
————————————————_—
MgO
18.8917.9816.9118.0114.5115.8816.7616.9116.1116.88
16.0416.0717.1317.6016.8416.8616.5716.4316.6716.9015.4416.9117.00
27.8725.7725.0525.0226.02
24.2627.3118.9320.33
30.76
0.39
0.33
0.340.170.222.950.490.990.800.29
11.7612.40
————————————————_—
CaO
15.4517.6016.6113.3915.2814.4115.4920.5617.9518.15
18.2718.6918.2219.7419.1719.1914.5519.7620.2218.9518.1618.2018.00
2.322.644.102.282.10
2.222.044.885.02
—
—
0.160.100.080.200.130.360.100.38
21.8518.97
15.1516.3514.1214.8515.2613.6613.6212.2712.019.31
10.985.62
13.3016.3516.7915.9616.2316.1316.32
Na2O
0.200.100.140.090.190.160.110.090.150.30
0.150.160.240.210.230.200.180.190.190.220.220.200.24
0.140.120.000.00
0.000.000.160.10
—
—
0.180.030.000.000.010.050.010.21
0.160.30
2.792.253.133.062.743.483.653.833.715.915.067.733.402.091.692.541.972.092.13
K2O
0.070.050.020.060.110.030.020.020.040.05
0.010.030.000.000.010.000.050.010.030.000.020.040.00
—0.050.010.040.02
0.050.050.060.08
—
—
0.080.000.040.020.050.590.030.01
0.060.15
0.070.000.080.080.090.080.030.080.160.150.120.710.070.050.040.070.070.070.13
En/Fs/Wob
49/22/2947/20/3346/21/3348/26/2643/24/3346/24/3048/21/3247/11/4245/19/3647/16/37
45/18/3745/18/3746/18/3648/14/3846/16/3846/16/3845/26/2946/15/3945/15/4046/17/3743/21/3647/17/3647/17/36
68/24/868/27/572/24/468/27/569/27/4
68/28/471/25/455/35/1058/32/10
Fo c
63
An d
75807172756867646346542768818477828180
686
PETROLOGY AND GEOCHEMISTRY OF ARC THOLEIITES
Table 3. (Continued).
Samplea
448A-15-448A-15-448A-15-448A-15-448A-15-448A-15-448A-18-
1 (30)Li1 (3O)A1
(3O)H1(30)H2
1 (30)A2
1 (30)L2I ( H 2 ) A I
448A-18-1 (H2)X2448A-18-1 (112)A3
448A-41-4 (16)LI448A-41-4 (16)AI448A-41-4 (16)HI448A-41-4 (16)A2
448A-41-4 (16)L2
448A-41-4 (16)H2448A-50-448A-50-448A-50-448A-50-448A-50-448A-51-448A-51-
(99)A1(99)H1(99)L1
(99)L2
(99)H2(85)A1(85)A2
448A-56-1 (31)AI448A-56-1 (31)A2448A-59-2 (126)A1
448A-59-2 (126)L1
448A-59-2(126)H1
448A-59-2 (126)A2
448A-59-2 (126)L1
448A-59-2 (126)H2
448A-62-2 (7)H 1448A-62-2 (7)Li448A-62-2 (7)Ai448A-62-2 (7)A2448A-62-2 (7)L2448A-62-2 (7)H 2
448-37-1 (52)448-58-3 (113)
448A-16-2 (71)448A-18-1 (112)
Unit
121212121212141414272727272727353535353537374141474747474747494949494949
612
1214
SiO2
47.047.842.546.547.847.043.346.244.051.948.9—
47.751.648.149.951.955.751.249.052.847.444.844.348.251.646.547.449.946.445.448.746.754.257.352.7
59.259.4
59.359.4
TiO2
0.000.000.000.000.000.000.000.000.000.000.00—
0.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.00
0.000.00
0.000.00
A12O3 FeO* MnO
Clinopyroxenes
33.5033.6333.0234.3333.5933.4433.7033.1034.7030.0832.14
—32.9130.6033.5232.9731.8029.2131.5433.6428.5232.1934.4034.6431.4229.6432.5631.6529.6332.7433.4732.0133.0834.4532.0435.52
18.6618.71
18.6718.73
—
—
0.83—
0.77——
1.01
—0.88—
1.290.900.650.710.82
———
0.78
0.79——
Zeolites
-
_
—
—
——
——
0.01—
0.04——
0.01
0.00—
0.020.010.010.000.02—————
0.00
—0.01——
-
_
—
MgO
———————
0.15—
0.13——
0.15——
0.16—
0.200.110.120.160.14—————
0.16
—0.13——
-
_
—
CaO
12.5015.7617.0316.8815.3511.7516.3416.3917.3914.5616.8016.1317.1514.5118.1816.2117.7510.0914.5117.4713.3716.5619.4719.4416.8515.2817.8115.9113.2117.4317.7413.5016.1715.2212.6916.50
0.020.04
0.070.05
Na2O
3.622.491.671.792.353.402.171.981.453.582.482.131.993.861.492.512.134.873.842.004.131.940.840.792.382.774.381.904.131.941.283.961.932.874.442.03
4.645.53
4.835.90
K2O
0.170.070.040.090.100.260.050.040.050.130.090.080.080.160.160.050.040.300.100.060.100.060.040.060.060.050.090.020.100.060.020.070.090.090.140.07
7.957.58
7.617.59
En/Fs/Wob
677785837864808287697880826786788252678364829393797183766283886582746181
a Sample numbers refer to hole-core-section and center of the sample interval (in parentheses) in stratigraphic order.Subscripts 1, 2, and 3 refer to different crystals analyzed by probe within the same sample; subscripts A, H, L refer toaverage, high, and anorthite contents in plagioclases, respectively; MT refers to magnetite, and II to ilmenite.
" The En/Fs/Wo column for pyroxenes is the mole per cent of enstatite, ferrosilite, and woUastonite respectively.*\ The Fo for olivines is the mole per cent forsterite.d The An column for plagioclases is the mole per cent anorthite in the plagioclases.e Dash (—) indicates that no analysis was made.
and trace-element criteria are of value, particularly inlight of the severe alteration of Site 448 ridge rocks.Table 5 is a brief summary of discrimination criteria.There is no question that the flows, dikes, and sills areall arc tholeiites. The high Fe/Mg, low A12O3, inter-mediate K2O, intermediate Ba, and low Sr, Ni, Cr, andlight (L) REE/heavy (H) REE values are the principaldistinguishing features of arc tholeiites. No sample fallsoutside the trend except one clast—Sample 448A-21-1(56)—which has a normalized La/Lu ratio of 1.15 andan average enrichment relative to carbonaceous chon-drites of about 30 (Fig. 6). The hornblende diorite thatis also somewhat enriched in REE has a normalizedLa/Lu ratio of 0.84, clearly of arc-tholeiitic character.The slight LREE-enrichment that is related to overallREE-enrichment in these two rocks is probably a resultof crystal fractionation rather than calc-alkalic origins,because the heavy REE are also enriched (Jakes andGill, 1970). Migdisov (this volume) has studied the tuffsin Units 4 and 5 of Hole 448 and reports major-element
evidence that is suggestive of calc-alkalic volcanism. Hefound samples with the following average chemicalcharacter: A12O3 = 13.7%, Ba = 49ppm, andSr = 169ppm. These values are impressively close to the averagevalues for all the flows at Site 448 (A12O3 = 14.0%, Ba= 49 ppm, Sr = 169 ppm); therefore these tuffs can beconsidered to be arc-tholeiitic series differentiates. Noevidence exists at Site 448 that volcanism on the Palau-Kyushu Ridge ever reached the calc-alkalic stage beforevolcanism began to wane in the middle late Oligocene.However, the data reported by the International Work-ing Group on the IGCP Project "Ophiolites" (1977) in-dicate that the basaltic andesites do appear to have calc-alkalic affinities. For example, one sample has 19.4%A12O3, 70 ppm Ba, and 340 ppm Sr. Although the Bavalue is somewhat low for typical calc-alkalic rocks, theA12O3 and Sr values do fit. It may be tentatively con-cluded that, at least locally, some calc-alkalic volcanismoccurred on the Palau-Kyushu Ridge, even though con-clusive evidence is wanting. In contrast, the XRF data
687
R. B. SCOTT
Table 4. Rare-earth element and other trace-element chemistry of whole rocks from Site 448.'
Sampleb
448-39-1 (75)448-48-1 (119)448-50-3 (80)448-53-1 (2)448-55-1 (110)
448A-15-1 (30)448A-21-1 (56)448A-26-2 (32)448A-28-1 (96)448A-33-1 (134)448A-36-5 (53)448A-37-2 (141)448A^0-l (33)448A^1-1 (90)448A-41-2 (12)!448A-45-1 (43)448A-47-3 (138)!448A-50-1 (95)448A-51-3 (135)448A-52-2 (128)448A-54-3 (81)448A-59-2 (126)448A-62-2 (7)448A-65-2 (125)
Unit
689
1011
12151618202223242627313335373941474951
La
4.510.217.97.88.2
10.432.86.64.87.66.06.3
10.66.04.75.58.16.17.37.87.25.99.06.0
Ce
12.112.118.413.511.5
14.825.110.610.410.713.09.8
11.97.89.8
11.412.511.111.111.410.98.6
11.39.8
Sm
2.023.174.432.612.56
3.028.152.181.482.581.902.583.082.211.822.031.032.542.722.502.582.512.062.27
Eu
0.971.371.381.160.94
1.132.701.020.831.130.801.121.151.020.881.081.121.091.121.031.051.090.740.87
Tb
0.380.600.170.440.36
0.45130
0.380.270.430.350.550.490.420.280.370.520.430.460.470.420.510.350.41
Yb
2.53.74.22.71.6
3.15.22.31.43.21.93.33.02.61.82.43.22.93.02.73.72.92.02.9
Lu
0.720.700.840.370.25
0.450.980.380.350.480.320.510.550.440.380.460.490.510.530.430.490.500.370.45
Co
4234344420
35333737
1154649423346373741333244493031
Cr
35106
2310
251330498
44.770122529381226632614362621
FeO*
14.6915.2911.2012.586.49
11.0910.2811.8211.2514.3710.8812.4813.3411.4111.8711.4512.4913.4111.5013.1011.3611.629.72
11.41
Sc
4344354334
38454243443843394041434142263741433642
Hf
1.051.092.261.080.80
1.571.680.640.650.810.880.971.060.540.600.650.960.890.900.640.960.710.760.71
a All elements are reported as ppm except Fe, which is per cent FeO.Sample numbers refer to the hole-core-section and center of the sample interval (in parentheses). Subscript 1refers to the area analyzed indicated in Table 2.
MgO50
% MgO
Figure 1. AFM plot (Na2O + K2O, FeO*, MgO wt. °7o, where FeO* istotal Fe expressed as FeO) of XRF data for Site 448. (Hole 448samples of igneous units are shown by asterisks, Hole 448 clastsfrom volcaniclastic breccia units by pluses, Hole 448A samples ofigneous units by dotted circles, and Hole 448A clasts from volcani-clastic breccia units by ×'es. The trend lines for the tholeiiticSkaergaard [SK] intrusion and the calc-alkalic Asama and Amagivolcanoes are shown for comparison.)
from Site 451 on the West Mariana Ridge indicate astrong calc-alkalic character: Ba averages 185 ppm; Sr,525 ppm; A12O3, 18.2%. Thus it appears that someevolutionary trends exist in the progression of arc activi-ty from the Palau-Kyushu to West Mariana ridges.
MgO40 50 60
% MgO
Figure 2. AFM plot of probe data for fresh glasses from Site 448. (Thedifferent symbols refer to volcanic Units 6 [indicated by *], 8 [ + ],10 [ 0 ] , 12 and 14 [×] , and 37 [•]. Note that the grouping has aslight alkali-enrichment trend.)
The serious effect of pervasive alteration in Site 448rocks is even more exaggerated by normative plots:when XRF data are plotted on the normative oli-vine-diopside-quartz diagram of Shibata (1976), varia-tions occur ranging from highly quartz normative tonepheline normative along a silica control trend, withsome scattering into the Mg-rich part of the olivine field(Fig. 7). The redistribution of silicon and enrichment of
688
PETROLOGY AND GEOCHEMISTRY OF ARC THOLEIITES
MgO
% MgO
Figure 3. Expanded AFM plot of 5% to 25% alkalis, 60% to 80%FeO*, and 15% to 30% MgO of data shown in Figure 2.
5 r
54 56
SiO.
Figure 4. Plot Siθ2 versus Na2θ + K2O (wt. %) for fresh glass probeanalyses from Site 448. (Fields of alkali-olivine basalt, high-alumina basalt, and tholeiitic basalt and differentiates are shownafter Kuno [1965].)
magnesium in basalts during hydrothermal reactionwith sea water (Hajash, 1975) is well established andprobably is responsible for most of the variationsobserved. Obviously, this alteration makes normativediagrams petrologically useless. A signficant contrastoccurs with the fresh glass analyses, which clustertightly between the 1 atm (Fig. 8) and the naturaloceanic-tholeiite two-pyroxene cotectic. Also, the posi-tion of the glasses about the two-pyroxene cotectic con-forms with the petrographic observations of coexistingpyroxenes and rare olivine; apparently the liquid com-positions have all left the olivine-clinopyroxene-ortho-
50
FeO*/MgO
Figure 5. Plot of FeOVMgO versus Siθ2 (wt. %) for fresh glass probeanalyses from Site 448. (The slope of the line separatingcalc-alkalic [CA] and tholeiitic [TH] differentiation trends isshown according to Miyashiro [1974].)
pyroxene reaction point and any olivine phenocryststhat remain must be relic and in disequilibrium.
Confirmation of a clinopyroxene-orthopyroxenecontrol on liquid fractionation can be found in thebehavior of Co and Cr. Because the trend of the two-pyroxene cotectic is toward the quartz apex, the degreeof pyroxene fractionation can be estimated. Both Crand Co are preferentially removed by pyroxenes, whichdepletes residual liquids in these trace metals; thisbehavior is demonstrated in Figure 9, which showsdecreasing Cr and Co contents and an increasingpercentage of normative quartz in the olivine-diopside-quartz diagram.
Microprobe studies of the pyroxenes in the glassesand holocrystalline rocks indicate that the compositionsof coexisting ortho- and clinopyroxenes (tie lines) follow
689
R. B. SCOTT
Table 5. Comparison of selected trace- and major-element criteria for discrimination among tholeiiticmid-ocean-ridge basalts, arc-tholeiite series differen-tiates, and calc-alkalic series differentiates.
Elementsa
Fe/MgAI2O3V0K2O%Ba ppmSr ppmNi ppmCr ppmLREE/HREE
MORBb
1.1116.60.07
10110150400< l
CAC
1.417.71.25
300400
2050
- 1 0
ATd
<2.015.60.43
502002020
< l
Site 448
4.613.80.45
491691425
< l
40
10
a The Fe/Mg and K data for Site 448 are from probeanalyses; Al and trace-element material from XRFdata; and REE information from INAA data.
" MORB = mid-ocean-ridge basalts, from Johnson(1979), Bass et al. (1973), and Carmichael et al.1974).
c CA = calc-alkalic series differentiates, from Jakesand Gill (1970) and Taylor (1969).
° AT = arc-tholeiite series differentiates, from Jakesand Gill (1970).
Clast
La Ce Sm Eu Yb Lu
Figure 6. Rare-earth elements (REE) normalized to carbonaceouschondrites versus atomic sequence. (The basalt data [22] fall into anarrow range indistinguishable from those reported by Jakes andGill [1970]. The hornblende diorite clast—Sample 448-50-3,[80]—and another clast—Sample 448A-21-1 [56]—have higheroverall REE contents and very small LREE enrichment that are theresult of crystal fractionation rather than the influence of acalc-alkalic magmatic source.)
well recognized trends. Figure 10 shows the actual com-positions determined in glasses: orthopyroxenes withabout 2 to 2.5 mole percent wollastonite exist betweenabout 78 and 71 mole percent enstatite, whereas pigeon-itic pyroxenes with 10 mole percent wollastonite arefound below about 68 mole percent enstatite. With theexception of one sample, a distict immiscibility field ex-ists between low-Ca pyroxenes and augite that containsat least 26 mole percent wollastonite. The one olivine isshown with a tie line to coexisting augite. Figure 11generalizes these fractionation trends and shows themuch more limited field of clinopyroxene in holocrys-talline rocks with a much larger immiscibility field, asexpected.
The compositions of plagioclases are shown inFigures 12 and 13; the degrees of zoning in the units areindicated by brackets. Note that the plagioclases inglasses (Fig. 12) are considerably more sodic than are
Figure 7. Normative olivine-diopside-quartz diagram after Shibata(1976), showing the compositions of samples analyzed by XRF.(The 1 atm, 5 kb, and the natural oceanic tholeiitic cotectics forolivine-diopside, hypersthene-diopside, and olivine-hyperstheneare indicated. The three samples on the olivine-diopside join ac-tually fall in the nepheline normative field.)
Figure 8. Normative olivine-diopside-quartz diagram after Shibata(1976), showing the compositions of fresh glasses by probeanalysis (same diagram as Fig. 7).
those holocrystalline rocks (Fig. 13); the highly frac-tionated diorite is included with the glassy rocks. Adistinct and progressive increase in solid solution ofK-feldspar in the plagioclase, with increasing Na-feldspar content, is apparent in Figure 12.
In all the whole-rock and microprobe chemicalstudies of individual phases, no obvious trends or pat-terns are present that can be correlated with strati-graphic position. The petrology of the Palau-Kyushu
690
PETROLOGY AND GEOCHEMISTRY OF ARC THOLEIITES
50
εI 30oO•D
C
ü
10
50 70%Norm QinOI-Di-Q
Figure 9. Concentration of Cr and Co versus the percentage of nor-mative quartz in the olivine-diopside-quartz diagram. (The inversetrace-metal to normative-quartz relationship is interpreted as theresult of progressive pyroxene-crystal fractionation. The Co andCr data were measured by INAA, and the normative-quartz con-tent of the glasses was calculated from probe data.)
10 20 30 40 50 60 70 80 90 100
Figure 11. Generalized fractionation trends of pyroxenes in glass andholocrystalline rocks (xln rx) (same diagram as Fig. 10).
10 20 40 50 60 70 80 90 100
Figure 12. Normative ternary feldspar diagram, showing composi-tions of plagioclase from glasses. (The brackets show the degree ofzoning of plagioclases for the indicated units. Although holocrys-talline, the diorite clast—Sample 448-50-3, [80]—is included, be-cause it represents a crystalline product of a more fractionatedmember of the arc-tholeiitic series.)
Figure 10. Composition of pyroxenes analyzed by microprobe inglasses plotted on the normative pyroxene diagram. (Solid lines arepyroxene tie lines, and the dashed line is the augite-olivine tie line,if the enstatite-ferrosilite join represents the forsterite-fayalitejoin.)
691
R. B. SCOTT
10 20 30 40 50 60 70 80 90 100
Figure 13. Normative ternary feldspar diagram, showing the composi-tions of plagioclase from holocrystalline rocks. (Brackets show thedegree of zoning of indicated units.)
Ridge can readily be summarized as being predominate-ly an arc-tholeiitic series that is controlled principally bypyroxene-plagioclase fractionation. Local calc-alkalicvolcanism may exist, but more conclusive evidence mustbe sought before this possibility is accepted.
Clearly more extensive studies of glasses and phasesshould be done to identify detailed mechanisms of frac-tionation. Careful correlations of the chemistry of thePalau-Kyushu Ridge with the West Mariana Ridge andthe presently active Mariana Arc should also be done.For example, trace-element data do not even exist forthe compositions of the active volcanoes in the MarianaArc, it is not possible to determine whether they arecalc-alkalic or not. Detailed petrochemical studies ofexposed volcanic units on the arc must also be con-ducted to characterize the petrologic evolution of thefrontal arc since 40 m.y. ago; these results could becompared to results of studies of the remnant arcs toderive an integrated petrologic evolution of the islandarc-back basin system.
ACKNOWLEDGMENTS
This paper is Texas A&M University Geodynamics Research Pro-gram Contribution No. 10.
REFERENCES
Bass, M. N., Mobely, R., Rhodes, J. M., et al., 1973. Volcanic rockscored in the central Pacific, Leg 17: Deep Sea Drilling Project. InWinterer, E. L., Ewing, J. I., et al., Init. repts. DSDP, 17:Washington (U.S. Govt. Printing Office), 429-446.
Carmichael, I. S. E.j Turner, F. J., and Verhoogen, J., 1974. IgneousPetrology: New York (McGraw-Hill).
Fischer, A. G., Heezen, B. C , et al., 1971. Init. Rept. DSDP, 6:Washington (U.S. Govt. Printing Office).
Gill, J. B., 1970. Geochemistry of Viti Levu, Fiji and its evolution asan island arc. Contrib. Mineral. Petrol., 27:179-187.
, 1976. Composition and age of Lua Basin and Ridgevolcanic rocks: implications for evolution of an interarc basin andremnant arc. Bull. Geol. Soc. Am., 87:1384-1395.
Hajash, A., 1975. Hydro thermal processes along mid-ocean ridges :an experimental investigation. Contrib. Mineral. Petrol., 53:205-226.
Hilde, T. W. C , Uyeda, S., and Kroenke, L., 1977. Evolution of thewestern Pacific and its margins. Tectonophysics, 30:145-165.
Ingle, J. C , Karig, D. E., et al., 1975. Site 296. In Karig, D. E., Ingle,J. C , Jr., et al., Init. Repts. DSDP, 31: Washington (U.S. Govt.Printing Office), 191-275.
International Working Group on the IGCP Project "Ophiolites,"1977. Initial report of the geological study of the ocean crust of thePhilippine Sea floor. In Bogdanov, N. (Ed.), Bolletino del Grupodi Lavoro aulle Ofioliti Mediterranee, (Vol. 2): Bologna (Estrattoda Ofioliti), 137-168.
Jakes, P., and Gill, J., 1970. Rare-earth elements and the island-arctholeiite series. Earth Planet. Sci. Lett., 9:17-28.
Johnson, J. R., 1979. Transitional basalts and tholeiites from the EastPacific Rise, 9°N. J. Geophys. Res., 84:1635-1651.
Karig, D. E., 1971. Site surveys in the Mariana area (Scan IV). InFischer, A. G., Heezen, B. C , et al., Init. Repts. DSDP, 6: Wash-ington (U.S. Govt. Printing Office), 681-689.
, 1972. Remnant arcs. Bull. Geol. Soc. Am., 83:1057-1068., 1975a. Basin genesis in the Philippine Sea. In Karig, D. E.,
Ingle, J. C , et al., Init. Repts. DSDP, 31: Washington (U.S. Govt.Printing Office), 857-879.
., 1975b. Origin and development of marginal basins in thewestern Pacific. /. Geophys. Res., 76:2542-2560.
Karig, D. E., Ingle, J. C , Jr., et al., 1975. Init. Repts. DSDP, 31:Washington (U.S. Govt. Printing Office).
Kuno, H., 1965. Fractionation trends of basalt magmas in lava flows./. Petrol., 6:302-321.
Miyashiro, A., 1974. Volcanic rock series in island arcs and activecontinental margins. Am. J. Sci., 274:321-355.
Scott, R. B., and Hajash, A., 1976. Initial submarine alteration ofbasaltic pillow lavas. Am. J. Sci., 276:480-501.
Shibata, T., 1976. Phenocryst-bulk rock composition relations ofabyssal tholeiites and their petrogenic signficance. Geochim.Cosmochim. Ada, 40:1407-1417.
Shih, T. C , in press. Magnetic lineations in the Shikoku Basin. InKlein, G. deV., Kobayashi, K., et al., Init. Repts. DSDP, 58:Washington (U.S. Govt. Printing Office).
Taylor, S. R., 1969. Trace-element chemistry of andesite and associ-ated calc-alkaline rocks. In McBirney, A. R. (Ed.), Proceedings ofthe Andesite Conference, Bulletin 65: Oregon (Dept. of Geologyand Mineral Industries), pp. 43-63.
Tiezzi, L. J., and Scott, R. B., in press. Crystal fractionation in acumulate gabbro, Mid-Atlantic Ridge, 26°N. Lat. /. Geophys.Res.
Uyeda S., and Ben-Avraham, Z., 1972. Origin and development ofthe Philippine Sea. Nature Phys. Sci., 240:176-178.
Watanabe, T., Langseth, M. G., and Anderson, R. N., 1977. Heatflow in back-arc basin of the western Pacific. In Talwani, M., andPitman, W. C , III (Ed.), Island Arcs, Deep-Sea Trenches, and
Back-arc Basins: Washington (Am. Geophys. Union), pp. 137-161.
Watts, A. G., and Weissel, J. K., 1975. Tectonic history of the Shi-koku marginal basin. Earth Planet. Sci. Lett., 25:239-250.
692