Larsen, H.C., Duncan, R.A., Allan, J.F., Brooks, K. (Eds.), 1999 Proceedings of the Ocean Drilling Program, Scientific Results, Vol.

Larsen, H.C., Duncan, R.A., Allan, J.F., Brooks, K. (Eds.), 1999 Proceedings of the Ocean Drilling Program, Scientific Results, Vol. 163 9. LOW-PRESSURE MELTING STUDIES OF BASALT AND BASALTIC ANDESITE FROM THE SOUTHEAST GREENLAND CONTINENTAL MARGIN AND THE ORIGIN OF DACITES AT SITE 917 1 P. Thy, 2 C.E. Lesher, 2 and J.D. Mayfield 2 ABSTRACT A series of 1-atm. melting experiments on basaltic flows collected from Holes 918D and 989B in the oceanic succession of the East Greenland continental margin can be used to define possible phase equilibria and liquid lines of descent. A sample from Site 918 (Section 152-918D-108R-2) shows the melting order low-ca pyroxene (1153 C), augite (1182 C), olivine (1192 C), and plagioclase (1192 C). A sample from Site 989 (Section 163-989B-10R-7) melts in the order of low-ca pyroxene (1113 C), augite (1167 C), plagioclase (1177 C), and olivine (1184 C). In particular, the relatively early appearance of low-ca pyroxene distinguishes the melting relations for the oceanic succession from those observed for the basaltic continental succession at Site 917. A basaltic andesite flow from the Middle Series at Site 917 (Section 152-917A-27R-4) shows the melting order of low-ca pyroxene (1142 C), plagioclase (1173 C), and olivine (1173 C). This melting order is difficult to reconcile with the observed large compositional variations in SiO 2 and FeO for the Middle Series, which imply early magnetite fractionation. Major element considerations and rare-earth element modeling of the dacites of the Middle Series suggest that they formed by low extent of melting (<20%) of continental hydrated gabbroic or mafic amphibolite at pressures <8 kbar. These crustally derived melts represent possible contaminants of basaltic magmas of the Lower and Middle Series at Site 917. INTRODUCTION The East Greenland Early Tertiary rifted margin is composed of extensive seaward-dipping reflector sequences (SDRS) and continental flood basalts. The magmatic development during the continental rifting and initiation of oceanic spreading were investigated by researchers on both Leg 152 (Holes 915 919) and Leg 163 (Holes 988 990) of the Ocean Drilling Program. Hole 917A penetrated the early part of a subaerially erupted continental succession divided into a Lower Series of basalt and subordinate picrite, dominantly pahoehoe flows (444 m), and a Middle Series of basaltic and dacitic aa flows and minor tuffs (193 m) (Larsen, Saunders, Clift, et al., 1994). Werner et al. (1998) showed that the volcaniclastic deposits of the Middle Series represented primary or reworked tephra that originated from rare explosive hydroclastic and pyroclastic eruptions during the terminal phases of the effusive continental flood basalt succession. The continental succession was shown to have been derived from a relatively depleted mantle and was modified by crustal contamination (Fitton et al., 1998a, 1998b; Saunders et al., 1998). Thy et al. (1998) suggested that the assimilation of a hydrous silicic component suppressed plagioclase crystallization and permitted extended olivine fractionation and formation of accumulative picrites. Fitton et al. (1998a, 1998b) postulated that granulitic gneiss was the principal contaminant for the lower part of the Lower Series, while above this level, including the Middle Series, contamination by amphibolitic gneiss became increasingly important. The most heavily contaminated lavas were recovered from the Middle Series (Larsen, Saunders, Clift, et al., 1994). Fitton et al. (1998b) conjectured that the compositional variations up through the Lower and Middle Series at Site 917 reflected a temporal decline in the supply of primitive melts to the crust and a progressive shoaling of the site by magma storage and fractionation. 1 Larsen, H.C., Duncan, R.A., Allan, J.F., Brooks, K. (Eds.), 1999. Proc. ODP, Sci. Results, 163: College Station, TX (Ocean Drilling Program). 2 Department of Geology, University of California, Davis, CA 95616, U.S.A. thy@geology.ucdavis.edu The transition into the oceanic succession is represented by the Upper Series at Site 917 and is composed of basaltic and picritic flows compositionally and morphologically resembling the Lower Series flows (Larsen, Saunders, Clift, et al., 1994; Larsen and Saunders, 1998). The Upper Series is less contaminated than the Lower and Middle Series (Fitton et al., 1998a, 1998b; Saunders et al., 1998). The slight evidence of contamination and the mid ocean-ridge basalt (MORB)-like source characteristics of most lavas of the Upper Series lead Fitton et al. (1998b) to include the Upper Series in the synrift succession. Fram et al. (1998) showed that trace-element compositions of the succession of Upper Series lavas recorded a rapid removal of the thick continental lithosphere that accompanied plate separation and initiation of subaerial seafloor spreading in the region. This was reflected in an increase in the extent of partial melting and a decrease in pressure of melt separation from the mantle. The oceanic succession was drilled at Site 918 (and 915) in the main portion of seawarddipping reflectors. In contrast to the interpretation of Fitton et al. (1998b) and Fram et al. (1998), the boundary between the continental and oceanic successions was defined by Larsen and Saunders (1998) by the unconformity and thin sediment horizon between the Middle and Upper Series of Hole 917A. The principal scientific drilling objectives of Leg 163 were to sample the basal portion of the continental succession (Site 989), to drill into the breakup unconformity, and to document more completely the continent/ocean transition (Site 990). Thereby, Leg 163 would bridge the gaps in the record believed to exist between the Upper Series (Site 917) and lavas recovered at Sites 915 and 918 (Duncan, Larsen, Allan, et al., 1996). Only two units of aphyric basalt were recovered at Site 989. Based on flow composition and morphology, the Shipboard Scientific Party (Duncan, Larsen, Allan, et al., 1996) concluded that the lavas belonged to the oceanic succession and not to the Lower Series. Lesher et al. (Chap. 12, this volume) concluded that this unusual compound flow most likely represents a lava fan delta formed by multiple breakouts from a lava tube at a considerable distance from the eruption site at the axial rift. At Site 990, 13 units of aa and pahoehoe flows were recovered that also resembled the aphyric to phyric basalts of the oceanic succession recovered from Sites 915 and 918 (Duncan, Larsen, Allan, et al., 1996). At Site 988 we drilled north of the main 95

P. THY, C.E. LESHER, J.D. MAYFIELD transect into the featheredge of the seaward-dipping reflectors and sampled two flows. Most of the lava succession of the Southeast Greenland margin displays systematic compositional variations that have previously been related qualitatively to crystal fractionation and assimilation of primitive, mantle-derived melts (e.g., Fitton et al., 1998b; Larsen et al., 1998). Quantitative information on the low-pressure phase relations and liquid lines of descent of these primitive melts was provided in the experimental study of Thy et al. (1998). Here we present new low-pressure experimental data for a number of more evolved compositions belonging to the oceanic succession (Sites 918 and 989) and from the silicic flows of the Middle Series of the continental succession (Site 917). These results constrain the locations of low-pressure cotectics that permit assessments of fractionation assemblages involved in crustal differentiation and of the generation of the silicic lavas of the Middle Series. EXPERIMENTAL AND ANALYTICAL TECHNIQUES Selected Samples Three lava samples were selected for experimental study. They include Unit 41 from the Middle Series at Site 917 (Sample 152-917A-27R-2 [Piece 4, 38 43 cm]), Unit 12B from oceanic succession at Site 918 (Sample 152-918D-108R-1, [Piece 2B, 54 58 cm]), and Unit 1 from Site 989 (Sample 163-989B-10R7 [Piece 5A, 55 59 cm]). All samples in the following are referred to by their respective section identifications. The samples are all from relatively evolved, hypersthene normative basaltic flows with Mg/(Mg+Fe total ) atomic ratios of <0.54 (Table 1). The phase relations for the Site 989 sample were determined as part of a companion experimental study of nucleation and crystallization kinetics (Lesher et al., Chap. 12, this volume). The phase compositions for these experiments are reported in this study. We also relate the new data to our previous results on the magnesium-rich flows from Site 917, which included Unit 13 from the Upper Series (Section 152-917A-11R-4) and Unit 84 from the Lower Series (Section 152-917A-86R-7) (Thy et al., 1998). The selected sample from the Middle Series (Section 152-917A- 27R-2) is compositionally a basaltic andesite with 5.6 wt% MgO, while the two other samples (Sections 152-918D-108R-1 and 163-989B-10R-7) are basalts with MgO contents of 8 8.2 wt% (Table 1). The basaltic andesite sample (Section 152-917A-27R-2) contains small amounts (~2%) of plagioclase (An 53 65 ) and trace amounts of augite phenocrysts in a groundmass of fine-grained pyroxene, plagioclase, and mesostasis (Larsen, Saunders, Clift, et al., 1994; Demant, 1998). The sample from the oceanic succession (Section 152-918D- 108R-1) is aphyric with <1% plagioclase phenocrysts (An 72 86 ) and a fine-grained groundmass of plagioclase, augite, Fe-Ti oxide minerals, and mesostasis (Larsen, Saunders, Clift, et al., 1994). The near vent lava from Leg 163 (Section 163-989B-10R-7) is aphyric with a groundmass assemblage of plagioclase, augite, Fe-Ti oxide minerals, and mesostasis (Duncan, Larsen, Allan, et al., 1996; Lesher et al., Chap. 12, this volume). All samples are altered by very low-grade metamorphism and contain secondary assemblage of zeolites and clay minerals (Duncan, Larsen, Allan, et al., 1996; Demant et al., 1998). Despite the secondary replacements, the samples are relatively fresh and, in view of their aphyric to sparsely phyric nature, likely represent magmatic liquid compositions. Experimental Techniques The experimental techniques used in the present study are identical to those used by Thy et al. (1998), and only the main points need to be given. The rock samples were ground in a tungsten-carbide Table 1. Composition of starting materials. 152-917A-27R-2 (Piece 4, 38 43 cm) 152-918D-108R-1 (Piece 2B, 54 58 cm) 163-989B-10R-7 (Piece 5A, 55 59 cm) XRF* XRF Probe XRF* XRF Probe XRF* XRF Probe SiO 2 51.98 54.28 54.58 50.29 51.29 50.50 49.99 50.93 48.48 TiO 2 1.21 1.26 1.25 0.94 0.95 0.99 1.03 1.05 1.11 Al 2 O 3 14.82 15.48 15.82 13.56 13.77 13.94 13.41 13.66 13.95 Fe 2 O 3 10.27 3.61 6.11 FeO 1.54 8.81 7.01 FeO** 10.78 11.26 12.53 12.06 12.24 13.22 12.51 12.74 13.57 MnO 0.12 0.13 0.15 0.23 0.23 0.25 0.20 0.20 0.21 MgO 5.36 5.60 5.56 7.84 7.96 7.54 8.03 8.18 0.04 CaO 7.27 7.59 7.60 11.39 11.56 11.43 10.67 10.87 11.06 Na 2 O 3.28 3.43 3.34 2.01 2.04 1.84 2.11 2.15 0.24 K 2 O 0.80 0.84 0.77 0.08 0.08 0.14 0.12 0.12 0.13 P 2 O 50.12 0.13 0.08 0.09 0.09 0.06 0.10 0.10 0.03 LOI 2.84 1.31 1.33 Total 99.61 100.00 101.68 100.15 100.00 99.91 100.11 100.00 98.82 Mg# 0.470 0.442 0.537 0.504 0.534 0.514 V 239 323 355 Cr ND 114 58 Ni 52 101 81 Cu 78 221 186 Zn 143 81 102 Sr 501 72 78 CIPW weight norm Or 4.96 4.47 0.48 0.83 0.72 0.78 Ab 29.02 27.46 17.27 15.58 18.19 19.15 An 24.35 25.46 28.16 29.38 27.25 27.90 Hy 27.57 29.81 25.38 26.85 25.34 12.41 Di 10.41 9.32 23.74 22.48 21.57 22.02 Ol 0.00 0.00 2.95 2.86 4.70 15.19 Il 2.39 2.33 1.81 1.88 1.99 2.13 Ap 0.31 0.19 0.22 0.14 0.24 0.38 Cm 0.04 Notes: * = from Larsen et al. (1998) and Larsen et al. (this volume) as determined by XRF. = XRF analysis calculated anhydrous with all iron as FeO. = Electron microprobe analysis of superliquidus glass. FeO** = total iron as FeO. Mg# = Mg/(Mg+Fe) ratio with all iron as Fe 2+. LOI = loss on ignition. ND = no data. = CIPW weight norm calculated with all iron as FeO. 96

LOW-PRESSURE MELTING STUDIES AND DACITE ORIGIN shatterbox and subsequently by hand in an agate mortar under acetone to an estimated average grain-size below ~10 µm. The powder was mixed with polyvinyl alcohol and pressed into pellets and dried. These pellets were broken into ~50-mg pieces and fused to an 0.004- in Fe-Pt alloy suspension wire prepared by iron electroplating (Grove, 1981). The powder from the sample taken from Section 163-989B-10R-7 was pressed into 50-mg pellets without using an organic binder. The experimental charges were suspended in a 1-atm. vertical quench furnace at the run temperatures. Temperature was monitored by a Pt/ 90 Pt 10 Rh thermocouple (S-type) calibrated against the melting point of gold. The furnace atmosphere was controlled to the fayalitemagnetite-quartz oxygen buffer using a CO-CO 2 gas mixture and monitored by a solid ZrO 2 -ceramic oxygen probe calibrated against the Ni-NiO reaction. Run duration varied from 8 239 hr, generally increasing with decreasing melting temperature. The experimental products were quenched in air. A summary of the experimental conditions and results are given in Table 2. Analytical Techniques The phase compositions of the experimental products were determined using a Cameca SX50 electron microprobe operated with an accelerating voltage of 15 kv and a beam current of 10 na. Details of the analytical procedures can be found in Thy et al. (1998). A focused beam was used for minerals, while glasses were normally analyzed with a 10-µm broad beam to minimize volatilization of sodium. The small residual volumes of glass in some low-temperature experiments prohibited the use of a broad diameter beam, and these, therefore, were analyzed with a narrow focused beam. An internal glass standard prepared from international rock standard W-1 (Govindaraju, 1989) was analyzed concurrently and used to evaluate analytical precision and accuracy (Thy et al., 1998, table 1). The analyses were screened by compositional and stoichiometric criteria and are considered representative for the experiments. The average glass and mineral compositions for the individual experiments are reported in Tables 3, 4, 5, and 6. The iron in liquids has been redistributed between Fe 2 O 3 and FeO appropriate for the fayalite-magnetite-quartz oxygen buffer using the equations of Kilinc et al. (1983). The standard deviations of the replicate analyses are reported in Tables 3, 4, 5, and 6 and reflect the homogeneity of the phases. In particular, the results for plagioclase and augite for the low-temperature experiments reflect a high degree of heterogeneity and incomplete reaction of the starting material. Glass and olivine, on the other hand, are generally homogeneous. The modal proportions of the experimental phases were estimated by weighted mass balance using the compositions of the phases determined by electron microprobe (Bryan et al., 1969). All of the oxides were assigned a weighting factor of 1.0, except SiO 2 and Al 2 O 3, which were weighted by 0.4 and 0.5, respectively. Losses of sodium to the furnace gas and under the electron microbeam during analysis were estimated by including Na 2 O as a phase in the mass balance calculation. Table 2 presents the analyzed modes and shows that <20% by weight of Na 2 O is unaccounted for in a given experimental charge. EXPERIMENTAL RESULTS Phase Appearances and Proportions The results of the melting experiments are given in Table 2. Section 152-917A-27R-2 shows the melting order of low-ca pyroxene at Table 2. Experimental conditions and results. Run # Temp ( C) fo 2 log Phase assemblage Phase proportions Duration (hr) Gl Ol Pl Lpx Aug Σr 2 Na 2 O loss 152-917A-27R-2 (Piece 4, 38 43 cm) 27R2-6 1177 9.13 23 Gl 1.000 1.178 0.20 27R2-7 1168 8.64 21 Gl Ol Pl 0.862 0.032 0.106 0.369 0.30 27R2-9 1158 8.81 69 Gl Ol Pl 0.832 0.047 0.121 0.115 0.40 27R2-10 1148 9.25 68 Gl Ol Pl 0.752 0.065 0.183 0.170 0.30 27R2-11 1136 9.26 92 Gl Ol Pl Lpx 0.679 0.014 0.209 0.098 0.034 0.50 27R2-12 1130 9.39 71 Gl Ol Pl Lpx 0.574 <0.010 0.276 0.150 0.042 0.50 27R2-13 1120 9.55 91 Gl Ol Pl Lpx 0.504 <0.010 0.311 0.199 0.087 0.50 27R2-8 1111 9.58 68 Gl Ol Pl Lpx 0.470 <0.010 0.331 0.198 0.036 0.60 27R2-19 1101 9.65 239 Gl Ol Pl Lpx 0.425 0.066 0.370 0.139 0.408 0.50 152-918D-108R-1 (Piece 2B, 54 58 cm) 108R1-4 1197 8.66 20 Gl 1.000 1.187 0.20 108R1-5 1186 8.81 27 Gl Ol Pl 0.851 0.050 0.099 0.686 108R1-6 1177 9.13 23 Gl Ol Pl Aug 0.816 0.041 0.085 0.058 0.011 0.20 108R1-7 1168 8.64 21 Gl Ol Pl Aug 0.694 0.041 0.148 0.117 0.019 108R1-9 1158 8.81 69 Gl Ol Pl Aug 0.597 0.055 0.198 0.150 0.283 108R1-10 1148 9.25 68 Gl Ol Pl Aug Lpx 0.496 0.036 0.226 <0.010 0.253 0.020 108R1-11 1136 9.26 92 Gl Ol Pl Aug 0.390 <0.010 0.289 0.226 0.153 0.017 0.20 108R1-12 1130 9.39 71 Gl Ol Pl Aug Lpx 0.315 <0.010 0.315 0.218 0.176 0.005 0.20 108R1-8 1111 9.58 68 Gl Ol Pl Aug Lpx 0.117 0.020 0.410 0.154 0.299 0.129 0.10 108R1-21 1103 9.56 70 Gl Ol Pl Aug Lpx 0.221 <0.010 0.366 0.214 0.219 0.015 0.20 163-989B-10R-7 (Piece 5A, 55 59 cm) 989B-8 1186 8.83 8 Gl 1.000 989B-7 1182 8.81 22 Gl Ol 0.981 0.019 0.020 989B-4 1172 8.91 25 Gl Ol Pl 0.918 0.045 0.037 0.012 989B-10 1612 9.00 13 Gl Ol Pl Aug 0.825 0.069 0.097 0.009 0.018 989B-9 1158 9.07 20 Gl Ol Pl Aug 0.737 0.078 0.137 0.048 0.005 989B-1 1150 9.55 22 Gl Ol Pl Aug 0.697 0.078 0.143 0.082 0.218 0.10 989B-5 1125 9.48 27 Gl Ol Pl Aug 0.412 0.119 0.290 0.179 0.009 989B-3 1101 9.59 21 Gl Ol Pl Aug Lpx 0.263 0.121 0.356 0.036 0.224 0.002 0.10 Notes: Abbreviations of experimental phase assemblage are gl = glass; ol = olivine; pl = plagioclase; lpx = low-ca pyroxene; aug = augite; = phase not observed. Temp = the experimental temperature in C. The phase proportions are calculated by least squares mixing calculations, as explained in the text. Σr 2 = the sum of squares of residuals. The loss of Na 2 O has been estimated by including Na 2 O as a phase in the mixing calculations. Since the mineral phases were not analyzed for Runs # 10R7-1 and 108R1-8, appropriate compositions were used in the calculating of the mode. 97

98 Run # Temp ( C) N SiO 2 TiO 2 Al 2 O 3 FeO* MnO MgO CaO Table 3. Composition of experimental glass. Na 2 O K 2 O P 2 O 5 Na 2 O (wt% corr.) Mg# SD SiO 2 TiO 2 Al 2 O 3 FeO MnO MgO CaO Na 2 O K 2 O P 2 O 5 152-917A-27R-2 (Piece 4, 38 43 cm) 27R2-6 1177 6 54.58 1.25 15.82 12.53 0.15 5.56 7.60 3.34 0.77 0.08 3.54 0.442 0.27 0.06 0.09 0.18 0.05 0.12 0.12 0.08 0.05 0.02 27R2-7 1168 7 55.05 1.34 15.57 12.53 0.20 5.06 7.34 3.13 0.86 0.08 3.43 0.419 0.18 0.06 0.15 0.19 0.06 0.08 0.14 0.10 0.03 0.04 27R2-9 1158 5 55.94 1.45 14.92 11.90 0.16 4.66 7.15 3.05 0.95 0.10 3.45 0.411 0.08 0.07 0.07 0.36 0.01 0.06 0.11 0.13 0.06 0.03 27R2-10 1148 6 55.63 1.52 14.56 12.44 0.21 4.30 7.09 3.04 0.97 0.13 3.34 0.381 0.74 0.10 0.22 0.65 0.04 0.08 0.16 0.16 0.07 0.07 27R2-11 1136 8 56.19 1.67 14.20 13.15 0.20 4.19 7.12 2.99 1.04 0.13 3.49 0.362 0.28 0.13 0.17 0.26 0.04 0.07 0.10 0.11 0.04 0.04 27R2-12 1130 6 55.12 2.19 13.45 14.49 0.16 3.92 7.22 2.78 1.09 0.14 3.28 0.325 0.26 0.09 0.11 0.26 0.04 0.10 0.15 0.08 0.08 0.05 27R2-13 1120 6 55.37 2.17 13.15 14.90 0.17 3.57 7.12 2.79 1.20 0.17 3.29 0.299 0.26 0.05 0.11 0.37 0.03 0.07 0.13 0.10 0.03 0.05 27R2-8 1111 6 56.35 2.37 13.10 14.77 0.18 3.32 7.06 2.60 1.36 0.20 3.20 0.286 0.81 0.06 0.13 0.42 0.06 0.06 0.16 0.10 0.09 0.05 27R2-19 1101 6 58.38 2.27 13.25 12.04 0.16 2.58 5.95 2.46 1.75 0.22 2.96 0.276 0.83 0.13 0.22 0.82 0.05 0.11 0.21 0.08 0.12 0.06 152-918D-108R-1 (Piece 2B, 54 58 cm) 108R1-4 1197 11 50.50 0.99 13.94 13.22 0.25 7.54 11.43 1.84 0.14 0.06 2.04 0.504 0.16 0.05 0.11 0.33 0.05 0.08 0.19 0.05 0.02 0.03 108R1-5 1186 7 51.39 1.05 13.99 13.69 0.27 6.93 11.75 2.07 0.13 0.09 2.07 0.474 0.14 0.08 0.14 0.43 0.01 0.05 0.10 0.06 0.01 0.03 108R1-6 1177 7 51.54 1.13 13.54 13.23 0.25 6.53 11.26 1.97 0.12 0.07 2.17 0.468 0.29 0.07 0.10 0.27 0.03 0.10 0.14 0.08 0.02 0.02 108R1-7 1168 6 51.71 1.32 13.40 14.19 0.24 6.28 10.80 2.20 0.16 0.09 2.20 0.441 0.25 0.09 0.11 0.23 0.04 0.14 0.22 0.11 0.03 0.02 108R1-9 1158 6 50.69 1.38 12.85 15.04 0.28 5.82 10.44 2.34 0.18 0.13 2.34 0.408 0.24 0.09 0.15 0.32 0.04 0.06 0.14 0.06 0.02 0.04 108R1-10 1148 6 50.68 1.82 12.40 15.56 0.28 5.07 9.67 2.31 0.19 0.16 2.31 0.367 0.29 0.12 0.31 0.16 0.05 0.13 0.07 0.09 0.03 0.04 108R1-11 1136 6 49.61 1.85 11.85 18.68 0.30 5.17 9.72 2.27 0.24 0.14 2.47 0.330 0.28 0.07 0.07 0.28 0.04 0.08 0.26 0.08 0.03 0.04 108R1-12 1130 4 49.38 2.36 11.53 19.19 0.30 4.64 9.54 2.21 0.23 0.18 2.41 0.301 0.44 0.04 0.23 0.54 0.03 0.20 0.13 0.05 0.01 0.05 108R1-8 1111 3 48.73 3.11 10.63 21.84 0.31 3.21 9.18 2.22 0.33 0.37 2.32 0.208 0.21 0.13 0.33 0.33 0.02 0.70 0.10 0.16 0.01 0.02 108R1-21 1103 4 48.50 3.77 10.21 20.82 0.34 3.53 9.44 1.77 0.26 0.30 1.97 0.232 0.44 0.17 0.25 0.99 0.03 0.25 0.18 0.15 0.04 0.04 163-989B-10R-7 (Piece 5A, 55 59 cm) 989B-8 1186 10 48.48 1.11 13.95 13.57 0.21 8.04 11.06 2.24 0.13 0.16 2.24 0.514 0.83 0.06 0.16 1.02 0.04 0.34 0.27 0.17 0.03 0.03 989B-7 1182 10 48.73 1.12 14.14 13.39 0.17 7.38 11.12 2.28 0.12 0.16 2.28 0.496 0.38 0.08 0.17 0.39 0.06 0.12 0.18 0.10 0.03 0.02 989B-4 1172 7 49.08 1.21 14.07 13.77 0.17 6.76 11.44 2.41 0.14 0.18 2.41 0.467 0.49 0.03 0.09 0.27 0.02 0.10 0.16 0.12 0.02 0.04 989B-10 1162 8 49.01 1.33 13.42 14.25 0.21 6.41 11.54 2.27 0.11 0.18 2.27 0.445 0.40 0.04 0.10 0.44 0.07 0.07 0.20 0.09 0.02 0.04 989B-9 1158 5 49.27 1.51 13.05 14.61 0.16 5.99 11.14 2.43 0.15 0.21 2.43 0.422 0.78 0.08 0.07 0.26 0.05 0.12 0.23 0.10 0.02 0.05 989B-1 1150 6 51.04 1.61 13.21 14.89 0.24 5.58 10.65 2.46 0.29 0.20 2.53 0.400 0.26 0.08 0.15 0.22 0.06 0.09 0.22 0.06 0.02 0.05 989B-5 1125 7 50.54 2.33 12.33 17.22 0.30 4.58 9.50 2.69 0.21 0.30 2.69 0.322 0.52 0.15 0.24 0.59 0.06 0.14 0.21 0.10 0.03 0.03 989B-3 1101 5 50.15 3.58 11.21 19.27 0.31 3.63 8.69 2.36 0.42 0.45 2.46 0.251 0.25 0.23 0.50 0.39 0.06 0.91 0.25 0.11 0.03 0.05 P. THY, C.E. LESHER, J.D. MAYFIELD Notes: Average of individual analyses and associated one standard deviations (SD) of oxides and Mg#. Temp = the experimental temperature in C. N = the number of analyses used for calculating each average. FeO* = total iron as FeO; Cr 2 O 3 is below detection limit; Mg# = Mg/(Mg+Fe) calculated with all iron as FeO. Corr. = corrected. Na 2 O values are based on the estimated losses from Table 2.

LOW-PRESSURE MELTING STUDIES AND DACITE ORIGIN Table 4. Composition of experimental olivine. Run # Temp ( C) N SiO 2 FeO MnO MgO CaO NiO Fo SD (mol%) SD K D SiO 2 FeO MnO MgO CaO NiO 152-917A-27R-2 (Piece 4, 38 43 cm) 27R2-7 1168 5 38.22 25.00 0.20 38.02 0.28 0.13 73.1 0.6 0.303 0.24 0.80 0.05 0.56 0.04 0.03 27R2-9 1158 4 37.14 26.74 0.24 36.28 0.31 0.13 70.7 0.4 0.329 0.30 0.46 0.02 0.56 0.03 0.04 27R2-10 1148 5 37.62 29.18 0.27 34.87 0.31 0.10 68.1 0.5 0.330 0.41 0.37 0.03 0.70 0.03 0.06 27R2-11 1136 5 37.76 30.88 0.29 32.51 0.32 0.17 65.2 0.6 0.347 0.33 0.43 0.05 0.98 0.05 0.05 27R2-12 1130 5 36.18 33.64 0.33 30.89 0.40 0.11 62.1 0.5 0.338 0.41 0.71 0.06 0.35 0.03 0.05 27R2-13 1120 5 35.88 34.87 0.39 28.94 0.41 0.13 59.7 0.7 0.332 0.29 0.75 0.03 1.08 0.03 0.09 27R2-8 1111 5 36.03 37.06 0.35 29.15 0.42 0.15 58.4 0.5 0.245 0.22 0.42 0.04 0.31 0.05 0.06 27R2-19 1101 4 34.25 37.97 0.41 27.08 0.42 0.11 56.0 0.5 0.343 0.22 0.45 0.05 0.27 0.03 0.02 152-918B-108R-1 (Piece 2B, 54 58 cm) 108R1-5 1186 8 39.13 20.97 0.34 41.38 0.37 0.19 77.9 1.4 0.293 0.33 1.29 0.03 1.02 0.05 0.06 108R1-6 1177 4 38.59 21.51 0.36 40.44 0.39 0.14 77.0 0.3 0.300 0.14 0.37 0.03 0.09 0.08 0.06 108R1-7 1168 8 37.34 25.10 0.37 37.15 0.37 0.12 72.5 0.5 0.341 0.40 0.54 0.11 0.54 0.08 0.06 108R1-9 1158 4 37.60 27.46 0.45 35.64 0.44 0.16 69.8 0.6 0.342 0.58 0.30 0.04 0.66 0.03 0.03 108R1-10 1148 5 36.32 29.71 0.42 32.40 0.46 0.17 66.0 0.5 0.342 0.19 0.85 0.02 0.25 0.07 0.05 108R1-11 1136 4 37.11 29.78 0.48 32.22 0.42 0.21 65.9 0.4 0.295 0.13 0.73 0.04 0.59 0.05 0.11 108R1-12 1130 4 35.42 34.52 0.55 29.81 0.48 0.16 60.6 0.8 0.323 0.24 0.81 0.07 0.54 0.06 0.10 108R1-9 1111 4 35.33 40.46 0.57 25.19 0.54 0.12 52.6 0.2 0.274 0.25 0.45 0.07 0.07 0.06 0.03 108R1-21 1103 4 35.02 43.74 0.58 23.17 0.60 0.15 51.4 0.9 0.370 0.59 0.74 0.07 0.22 0.04 0.04 163-989B-10R-7 (Piece 5B, 55 59 cm) 989B-7 1186 6 37.26 21.66 0.33 40.06 0.37 0.17 76.7 0.9 0.341 0.84 0.76 0.05 0.61 0.04 0.08 989B-4 1172 7 37.43 21.00 0.25 41.28 0.34 0.09 77.8 1.6 0.286 0.28 1.41 0.06 1.06 0.07 0.05 989B-10 1162 6 37.05 23.53 0.34 37.27 0.93 0.04 73.8 1.8 0.326 0.54 2.44 0.11 3.40 1.19 0.04 989B-9 1158 3 36.87 26.36 0.37 35.64 0.47 0.10 70.7 1.1 0.349 1.01 0.88 0.02 1.45 0.05 0.02 989B-5 1125 7 35.55 37.05 0.50 28.54 0.50 0.02 57.8 1.1 0.397 0.45 1.11 0.08 0.55 0.07 0.04 989B-3 1101 6 35.00 39.96 0.48 24.87 0.65 0.08 52.6 1.0 0.349 0.37 1.08 0.06 0.49 0.09 0.09 Notes: Averages of individual analyses and associated one standard deviations (SD) of oxides and Mg#. Temp = the experimental temperature in C. N = number of analyses. K D is calculated as discussed in text. 1142 ± 6 C and plagioclase and olivine at 1173 ± 5 C. Section 152-918D-108R-1 melts in the order of low-ca pyroxene at 1153 ± 5 C, augite at 1182 ± 5 C, and olivine and plagioclase at 1192 ± 6 C. Section 163-989B-10R-7 melts in the order of low-ca pyroxene at 1113 ± 12 C, augite at 1167 ± 5 C, plagioclase at 1177 ± 5 C, and olivine at 1184 ± 2 C. Fe-Ti oxide minerals were not detected in any of the melting experiments. There is a remarkably good correlation between melting temperature, phase proportions, and compositional variables of the experimental liquids. The variation of melting temperature and liquid fraction (Fig. 1) allows a minimum estimate of the solidus temperature by extrapolation to zero melt fraction and gives solidus temperatures of ~1110 C for the most magnesian-rich lavas (Sections 152-917A- 11R-4 and 152-917A-86R-7) and ~1050 C for the most magnesianpoor lava (Section 152-917A-27R-2). Phase Compositions and Liquid Lines of Descent The compositional liquid lines of descent for the investigated samples (Table 3) (Thy et al., 1998) are represented in Figure 2 as a function of MgO content. The main variation is caused by cosaturation of olivine, plagioclase, and augite as reflected in the decreasing Al 2 O 3 and CaO with decreasing MgO. Olivine is on the liquidus for all samples and present throughout the melting intervals. The composition of olivine correlates with melting temperature and liquid Mg/(Mg + Fe 2+ ) ratio (Table 4). The distribution coefficient for the moles of MgO and FeO between olivine and liquid (K D FeO/MgO [ol/liq] = [FeO/MgO] ol /[FeO/MgO] liq ) is 0.32 ± 0.03 for 49 determinations (Fig. 3A) and is consistent with previous determinations (Roeder and Emslie, 1970; Ulmer, 1989). These results indicate that olivine and liquid are well equilibrated for most experiments. Plagioclase composition (Table 5) shows wide and irregular variations with melting variables and the composition of the coexisting liquid. The relationship between plagioclase and liquid composition can be expressed in terms of the exchange of Na and Ca (K D Na/Ca [pl/ liq] = [Na/Ca] pl /[Na/Ca] liq ) (Fig. 4A). The calculated K D values range between 0.6 and 1.6 and show a general increase with decreasing temperature (and liquid fraction) (Fig. 4B). The results indicate that the K D for the basaltic andesite (Section 152-917A-27R-2) is lower than for any of the other samples. These values are largely in accord with previous experimental results (Grove et al., 1982; Mahood and Baker, 1986; Tormey et al., 1987; Ussler and Glazner, 1989). The covariation between the An content (mol%) of plagioclase and Fo content (mol%) of olivine shows systematic differences between the low-na 2 O (Sections 152-918D-108R-1 and 163-989B- 10R-7) and the high-na 2 O basaltic samples (Sections 152-917A- 11R-4 and 152-917A-86R-7) as a steepening in the slopes of the Fo- An covariation for the former group (Fig. 5), likely to be related to the depletion rates of Na 2 O of the coexisting liquids. The errors associated with plagioclase compositions are especially large for the lowtemperature runs because of incomplete equilibration between plagioclase and liquid. The pyroxenes form two compositional groups, a high-ca group and a low-ca group. The high-ca group are salites to augites, while the low-ca group are magnesian to intermediate pigeonites (Poldervaart and Hess, 1951; Deer et al., 1978). Both augite and pigeonite show increasing ferrosalite content with decreasing melting temperature (Table 6). The exchange of Fe and Mg between pyroxene and liquid can be calculated as for olivine, except that all iron in the liquid is calculated as Fe 2+. The calculated K D FeO/MgO (aug/liq) vary between 0.22 ± 0.02 (Sections 152-917A-11R-4 and 152-917A-86R-7; n = 15) and 0.26 ± 0.02 (Sections 152-918D-108R-1 and 163-989B-10R-7; n = 8) (Fig. 3B) and are consistent with other experimental results as well as natural pyroxene-liquid pairs (Grove and Bryan, 1983; Perfit and Fornari, 1983; Toplis and Carroll, 1995). The pigeonites indicate a K D FeO/MgO (pig/liq) of 0.23 ± 0.01 (Section 152-917A-27R-2; n = 5), 0.25 (Section 152-918D-108R-1; n = 2), and 0.23 (Section 163-99

100 Run # Temp ( C) N SiO 2 TiO 2 Al 2 0 3 FeO MgO Table 5. Composition of experimental plagioclase. CaO Na 2 O K 2 O An (mol%) SD SiO 2 TiO 2 Al 2 O 3 FeO MgO CaO Na 2 O K 2 O SD P. THY, C.E. LESHER, J.D. MAYFIELD 152-917A-27R-2 (Piece 4, 38 43 cm) 27R2-7 1168 8 53.81 0.06 29.46 1.10 0.17 12.26 4.26 0.17 60.8 5.6 1.53 0.03 0.95 0.11 0.04 1.03 0.64 0.05 27R2-9 1158 6 54.29 0.08 28.79 1.20 0.26 11.57 4.49 0.16 58.2 4.9 1.07 0.03 0.98 0.10 0.05 1.05 0.50 0.03 27R2-10 1148 4 54.06 0.13 28.06 1.27 0.35 11.48 4.40 0.21 58.3 5.6 1.50 0.07 0.81 0.08 0.08 1.24 0.54 0.04 27R2-11 1136 5 54.15 0.11 28.72 1.29 0.17 11.59 4.50 0.21 58.0 2.7 0.69 0.05 0.72 0.07 0.03 0.56 0.28 0.05 27R2-12 1130 8 55.61 0.09 27.74 1.42 0.26 10.97 4.85 0.22 54.8 2.2 0.64 0.04 0.43 0.19 0.06 0.49 0.23 0.02 27R2-13 1120 5 55.17 0.08 27.92 1.43 0.16 10.94 4.86 0.27 54.6 3.3 0.70 0.04 0.51 0.15 0.03 0.63 0.35 0.06 27R2-8 1111 10 55.81 0.11 28.37 1.41 0.17 10.96 4.95 0.31 54.0 2.4 0.89 0.05 0.68 0.15 0.03 0.60 0.24 0.03 27R2-19 1101 16 55.42 0.11 27.35 1.27 0.18 10.47 5.07 0.32 52.3 2.4 0.71 0.04 0.77 0.17 0.07 0.39 0.35 0.04 152-918D-108R-1 (Piece 2B, 54 58 cm) 108R1-5 1186 6 50.43 0.08 31.94 1.18 0.32 15.26 2.58 0.03 76.4 1.2 0.15 0.06 1.35 0.15 0.12 0.23 0.12 0.01 108R1-6 1177 11 49.36 0.07 31.24 1.11 0.24 15.10 2.74 0.06 75.0 1.6 0.85 0.04 0.49 0.08 0.03 0.76 0.13 0.02 108R1-7 1168 9 50.77 0.06 28.77 1.35 0.33 13.97 3.30 0.04 69.9 3.8 1.27 0.03 0.85 0.23 0.10 0.73 0.42 0.01 108R1-9 1158 5 50.70 0.07 31.29 1.13 0.21 14.95 3.04 0.04 72.9 2.6 0.73 0.05 0.75 0.05 0.06 0.53 0.29 0.01 108R1-10 1148 8 50.85 0.06 29.89 1.04 0.21 13.60 3.41 0.04 68.6 3.9 1.43 0.01 1.15 0.16 0.02 0.79 0.42 0.01 108R1-11 1136 5 50.64 0.06 30.67 1.23 0.18 14.16 3.25 0.06 70.4 6.4 1.68 0.03 1.05 0.23 0.07 1.28 0.73 0.02 108R1-12 1130 6 51.51 0.08 30.79 1.14 0.14 13.88 3.57 0.07 68.0 4.8 1.39 0.03 1.13 0.06 0.04 0.99 0.55 0.03 108R1-8 1111 5 53.69 0.06 29.11 1.21 0.11 12.64 3.94 0.08 63.6 7.9 0.92 0.04 0.72 0.16 0.03 0.44 0.41 0.01 108R1-21 1103 6 51.61 0.08 29.91 1.00 0.17 13.91 3.75 0.08 66.9 6.3 2.08 0.29 1.10 0.25 0.11 1.46 0.71 0.04 163-989B-10R-7 (Piece 5B, 55 59 cm) 989B-4 1172 6 49.48 0.05 30.98 1.14 0.25 14.90 3.26 0.03 71.5 4.8 1.49 0.03 0.84 0.24 0.03 0.97 0.56 0.02 989B-10 1162 7 49.92 0.07 30.68 1.33 0.26 14.37 3.44 0.03 69.7 3.5 1.03 0.03 0.47 0.19 0.02 0.70 0.42 0.02 989B-9 1158 7 50.45 0.05 30.59 1.42 0.38 14.45 3.32 0.04 70.5 2.6 0.69 0.02 0.52 0.19 0.19 0.67 0.26 0.02 989B-5 1125 6 52.05 0.05 29.88 1.51 0.21 13.68 3.82 0.04 66.3 3.5 1.01 0.03 0.66 0.20 0.05 0.79 0.38 0.02 989B-3 1101 9 53.01 0.07 30.19 1.44 0.18 12.94 4.20 0.06 62.8 3.9 1.27 0.04 0.78 0.15 0.05 0.70 0.49 0.04 Notes: Averages of individual analyses and associated one standard deviations (SD) of oxides and An. Temp = the experimental temperature in C. N = number of analyses.

Table 6. Composition of experimental pyroxenes. SD Run # Temp ( C) N SiO 2 TiO 2 Al 2 O 3 FeO MnO MgO CaO Na 2 O Cr 2 O 3 En (mol%) Fs (mol%) Wo (mol%) Mg# SD K D SiO 2 TiO 2 Al 2 O 3 FeO MnO MgO CaO Na 2 O Cr 2 O 3 101 152-917A-27R-2 (Piece 4, 38 43 cm) 27R2-11 1136 5 52.91 0.28 1.19 18.45 0.32 23.70 3.53 0.06 64.8 28.3 6.9 69.6 1.6 0.248 0.44 0.06 0.33 1.56 0.14 0.61 1.20 0.03 27R2-12 1130 3 52.71 0.35 1.21 19.12 0.29 23.55 3.68 0.05 63.8 29.1 7.2 68.7 0.9 0.220 0.33 0.05 0.03 0.91 0.02 0.29 0.75 0.01 27R2-13 1120 3 52.44 0.28 0.94 21.12 0.45 22.23 3.72 0.07 60.5 32.2 7.3 65.2 1.5 0.227 0.19 0.08 0.20 0.95 0.19 0.52 0.47 0.04 27R2-8 1111 3 53.41 0.33 0.91 21.81 0.39 21.61 3.89 0.09 59.0 33.4 7.6 63.8 0.6 0.227 0.27 0.11 0.31 0.25 0.04 0.35 0.25 0.04 27R2-19 1101 12 51.83 0.42 0.89 22.01 0.36 19.88 4.61 0.09 55.9 34.8 9.3 60.8 2.9 0.238 0.66 0.09 0.17 0.53 0.04 0.93 0.45 0.04 152-918D-108R-1 (Piece 2B, 54 58 cm) 108R1-6 1177 16 51.82 0.30 1.88 8.92 0.21 16.58 18.97 0.20 0.17 47.1 14.2 38.7 76.8 2.4 0.266 0.67 0.05 0.15 2.52 0.03 0.56 0.44 0.04 0.05 108R1-7 1168 6 50.71 0.35 2.47 10.64 0.28 17.50 17.47 0.16 0.19 48.6 16.6 34.9 74.6 3.6 0.269 0.92 0.08 0.70 1.53 0.03 1.88 2.13 0.02 0.06 108R1-9 1158 6 52.30 0.35 1.68 11.50 0.30 16.57 17.67 0.15 0.13 46.4 18.1 35.6 72.0 2.6 0.269 0.39 0.04 0.09 1.14 0.06 0.62 0.80 0.03 0.07 108R1-10 1148 9 50.95 0.60 2.72 12.65 0.33 17.35 14.05 0.17 0.19 50.2 20.6 29.2 71.0 1.0 0.237 0.82 0.24 0.62 0.96 0.03 0.96 1.79 0.04 0.03 108R1-10 1148 3 52.12 0.31 1.15 18.19 0.46 21.55 5.47 0.08 0.09 60.4 28.6 11.0 67.9 0.6 0.275 0.62 0.04 0.15 0.38 0.01 0.16 0.11 0.01 0.01 108R1-11 1136 10 50.66 0.45 1.85 13.24 0.30 15.15 17.40 0.21 0.10 43.2 21.2 35.6 67.1 4.3 0.242 0.30 0.15 0.30 1.92 0.05 0.82 1.11 0.04 0.05 108R1-12 1130 4 51.10 0.50 1.70 14.88 0.32 14.54 16.85 0.20 0.09 41.5 23.9 34.6 63.5 4.3 0.248 1.21 0.12 0.12 2.18 0.07 0.77 0.91 0.06 0.08 108R1-12 1130 5 52.83 0.30 1.02 19.41 0.52 21.13 5.88 0.06 0.05 58.3 30.1 11.7 66.0 0.8 0.222 0.46 0.04 0.23 0.83 0.05 0.57 1.09 0.03 0.03 108R1-8 1111 7 51.60 0.63 1.76 16.92 0.33 14.01 15.57 0.21 0.16 40.4 27.4 32.3 59.6 4.7 0.178 0.77 0.21 0.54 2.14 0.06 1.10 1.23 0.06 0.09 108R1-8 1111 4 52.88 0.32 0.89 23.25 0.54 18.91 5.32 0.06 52.9 36.5 10.7 59.2 1.7 0.181 0.4 50.04 0.07 1.29 0.04 0.46 0.79 0.01 108R1-21 1103 5 50.67 0.59 2.17 16.40 0.34 14.74 15.77 0.16 0.14 41.8 26.1 32.1 61.6 7.4 0.189 0.22 0.51 3.68 0.10 1.40 2.16 0.06 0.10 108R1-21 1103 3 52.47 0.31 0.91 21.49 0.42 20.90 4.76 0.06 57.5 33.2 9.4 63.4 0.7 0.174 0.08 0.02 0.08 0.60 0.08 0.34 0.09 0.03 163-989B-10R-7 (Piece 5B, 55 59 cm) 989B-10 1162 7 50.60 0.42 2.24 10.96 0.27 17.54 17.24 0.19 0.12 48.6 17.0 34.4 74.1 5.1 0.281 0.91 0.14 0.30 2.02 0.04 1.65 1.52 0.06 0.07 989B-9 1158 8 50.53 0.51 3.04 10.93 0.27 17.31 17.35 0.21 0.01 48.2 17.1 34.7 73.8 3.5 0.259 1.07 0.15 1.02 1.22 0.05 1.34 1.18 0.04 0.03 989B-5 1125 5 51.05 0.48 2.59 11.20 0.23 16.33 18.26 0.21 0.06 45.6 17.8 36.6 72.2 4.7 0.183 0.86 0.10 0.74 2.03 0.06 1.20 1.28 0.05 0.04 989B-3 1101 3 51.32 0.53 2.73 11.74 0.28 15.77 18.34 0.21 0.05 44.4 18.5 37.1 70.2 3.6 0.140 1.59 0.21 0.42 3.56 0.08 1.77 1.35 0.04 0.03 989B-3 1101 3 51.82 0.65 1.63 21.55 0.44 17.50 7.23 0.17 50.3 34.8 14.9 59.1 2.4 0.232 0.44 0.39 0.97 1.06 0.08 1.20 1.25 0.22 Notes: Averages of individual analyses and associated one standard deviations (SD) of oxides and Mg#. Temp = the experimental temperature in o C. N = number of analyses. Mg# = Mg/(Mg+Fe) calculated on an atomic basis. En = enstatite, Fs = ferrosilite, Wo = wollastonite; all on a molecular basis. K D is calculated as discussed in text. LOW-PRESSURE MELTING STUDIES AND DACITE ORIGIN

P. THY, C.E. LESHER, J.D. MAYFIELD 989B-10R-7, n = 1). However, the very low K D for one of the pigeonite and liquid pairs (Section 152-918D-108R-1) suggests disequilibrium (Table 6; Fig. 3B). Phase Equilibria Figure 1. Melting temperature as a function of liquid proportion. The first appearance of pyroxene is indicated (aug = augite; lpx = low-ca pyroxene). The lines are best linear fits to the data. The experimentally determined phase relations are portrayed graphically in terms of the normative components quartz (q), plagioclase (pl), olivine (ol), and diopside (di) in Figure 6. Figure 6A shows the multiply-saturated liquids and coexisting pyroxenes for projections from pl to the plane of ol-di-q. Figure 6B shows the same data set as projected from di to the plane ol-pl-q. In contrast to olivine and plagioclase that plot near their ideal normative compositions (not shown), the experimental augites and pigeonites plot significantly away from their ideal normative compositions and form an array between normative hypersthene (hy) and a point off normative diopside (Fig. 6B). Augite becomes increasingly subcalcic in the sequence Sections 152-917A-11R-4, 152-917A-86R-7, 163-989B-10R-7, and 152-918D-108R-1. The experimental liquids for Section 152-918D-108R-1 and 163-989B-10R-7 show a systematic increase in normative quartz with decreasing temperature and move away from a fairly constant volume defined by coexisting olivine, plagioclase, and augite. The displace- Figure 2. Compositions of the experimental liquids as a function of MgO content. All analyses have been calculated to 100% with iron distributed between Fe 2 O 3 and FeO using the equation of Kilinc et al. (1983). The curves are best polynomial fits to the data. 102

LOW-PRESSURE MELTING STUDIES AND DACITE ORIGIN A B Figure 3. The molecular ratios FeO/MgO of mafic minerals as a function of the FeO/MgO ratio of the liquid. A. Olivine and coexisting liquid with iron of the liquid distributed according to Kilinc et al. (1983). B. Pyroxenes and coexisting liquid with iron calculated as total FeO (FeO*). A B Figure 4. Plagioclase liquid relations. A. The albite (Ab)/anorthite (An) ratio plotted against the Na/Ca ratio of the coexisting liquid. B. The calculated K D Na/Ca (pl/liq) as a function of melting temperature. See text for calculations. ment of the multiply-saturated cotectics for Sections 163-989B-10R7 and 152-918D-108R-1 toward lower normative plagioclase (Fig. 6) are principally caused by the variation in composition of the experimental augite not fully accounted for in projecting from normative diopside. However, the differences are small and the near linear variation in the liquid trends in these projections indicate that the cotectic proportions of olivine, plagioclase, and augite are nearly constant. Sections 152-917A-11R-4, 152-917A-86R-7, and 163-989B-10R-7 show early crystallization of olivine and/or plagioclase (cotectic proportion: 29% ol and 71% pl), followed by augite. The olivine, plagioclase, and augite cotectic corresponds to 7% olivine, 52% plagioclase, and 41% augite. Of these three sections, only Section 163-989B-10R-7 reaches low-ca pyroxene saturation. The liquid line of descent for the basaltic andesite sample (Section 152-917A-27R-4) is controlled by olivine, plagioclase, and pigeonite. This multiply-saturated relation is a reaction boundary along which olivine reacts with the liquid to form pigeonite (Grove et al., 1982, 1983; Grove and Juster, 1989; Juster et al., 1989). The result is that the liquid compositions lie outside the triangle defined by equilibrium plagioclase, pigeonite, and olivine. The results for Section 152-917A-27R-4 define a point on the pigeonite-saturated reaction curve. By extrapolating pigeonite-olivine-plagioclase relations to in- 103

P. THY, C.E. LESHER, J.D. MAYFIELD tersect with the olivine-plagioclase-augite cotectic as defined by Sections 152-917A-11R-4 and 152-917A-86R-7, we can estimate an approximate location of the pseudoinvariant point in Figure 6B, where olivine + liquid plagioclase + augite + pigeonite. Section 152-918D-108R-1 shows an extended interval of olivine, plagioclase, augite, and pigeonite saturation, but defines the lowvariance pigeonite-saturated conditions at a relatively higher temperature (1153 C) and lower silica saturation compared to typical MORB compositions (Fig. 6B) (Juster et al., 1989; Grove and Juster, 1989). Section 163-989B-10R-7 is saturated in pigeonite at the lowest melting temperature (1113 C) and slightly higher normative quartz composition than 152-918D-108R-1. Sections 152-917A- 86R-7 and 152-917A-11R-4 were not saturated in low-ca pyroxenes, and low-ca saturation likely will be below the lowest temperature melted (1111 1120 C) and the highest normative quartz of any of the investigated sections. The present results extend the stability of low-ca pyroxenes to lower normative quartz than obtained by Juster et al. (1989) and Grove and Juster (1989) for basalts and basaltic andesite lavas (Fig. 6). MAGMATIC EVOLUTION OF THE EAST GREENLAND RIFTED MARGIN Phase Equilibria Constraints Figure 5. Forsterite content (Fo mol%) of olivine as a function of anorthite content (An mol%) of coexisting plagioclase. The lines are best linear fits to the data. The analyzed flows sampled during Legs 152 and 163 (Larsen, Saunders, Clift, et al., 1994; Duncan, Larsen, Allan, et al., 1996; Larsen et al., 1998; Larsen et al., this volume) have been recast into the same normative components using the same calculation procedure as for illustrating the experimental phase equilibria. The results are shown in Figure 7 and highlight several important features. As was observed by Thy et al. (1998), the primitive flows from Hole 917A in the Lower and Upper Series plot away from the experimentally determined multiply-saturated cotectic. This is consistent with the phenocryst assemblages found in these lavas (Larsen, Saunders, Clift, et al., 1994; Demant, 1998). Thy et al. (1998) used inferred shifts in the locations of the olivine-plagioclase cotectic to suggest A B Figure 6. Normative projections of the olivine (ol), plagioclase (pl), diopside (di), quartz (q) basalt tetrahedron on the triangular diagrams (A) ol-pl-q and (B) oldi-q. Shown are liquids multiply-saturated in olivine, plagioclase, and pyroxene together with coexisting pyroxenes. The normative components have been calculated using a CIPW molecular norm with iron distributed as for Figure 2. The pyroxenes have been plotted assuming that all iron occurs as FeO. The vertical lines marked 108R1 and 10R7 (stippled) and 27R2 (full drawn) are schematic illustration of pigeonite saturation (marked by + ) for the three melted sections, respectively 152-918D-108R-1, 163-989B-10R-7, and 152-917A-27R-4. Section 152-917A-86R-7 is not saturated in pigeonite at any temperature melted. The low-ca pyroxene saturated liquids marked by are from Juster et al. (1989) and Grove and Juster (1989) and range in composition from basaltic to andesitic. Other abbreviations used are hy = hypersthene, pig = pigeonite, and aug = augite. 104

LOW-PRESSURE MELTING STUDIES AND DACITE ORIGIN that some differentiated Lower Series lavas were modified by crustal contamination and concluded that excessive olivine fractionation caused by suppressed plagioclase crystallization at elevated partial water pressures could explain some of the compositional features of the Lower Series lavas. In general, however, both the Lower and Upper Series were controlled by near-anhydrous crystallization conditions. In contrast, lavas of the Middle Series show large variations in the normative projections and plot systematically below the projected olivine-plagioclase-augite cotectic in Figure 7B. This position implies fractionating phase assemblages of olivine or low-ca pyroxene and plagioclase. Such fractionating assemblages would not lead to the large variation in SiO 2 content shown by the Middle Series lavas. Recovery at Site 988 consisted of differentiated olivine, plagioclase, and augite phyric lava from only one fresh flow. The analyzed samples from this flow cluster near the multiply-saturated experimental cotectic, consistent with the evolved nature of the flow and its phenocryst assemblage. In these respects, the Site 988 lava is similar to the onshore plateau lavas in the Scoresby Sund area (Larsen et al., 1989; Fram and Lesher, 1997). The flows analyzed from Sites 989 and 990 are compositionally very similar to each other and narrowly cluster along the multiply-saturated cotectic, consistent with their aphyric to highly olivine, plagioclase, and augite phyric nature. The multiply-saturated liquid line of descent for the Site 918 sample is slightly offset from the trend determined for the basaltic Site 917 samples by lower normative diopside and plagioclase (Fig. 6). The flows recovered from the oceanic succession at Site 918 have compositions that deviate from their experimentally determined liquid line of descent (compare Figs. 6 and 7). The majority of the Site 918 lavas are aphyric and contain microphenocrysts of olivine and plagioclase and, occasionally, augite phenocrysts (Larsen, Saunders, Clift, et al., 1994). Therefore, these are likely to have been controlled by olivine and plagioclase crystallization and thus plot away from the olivine, plagioclase, augite cotectic toward lower normative diopside (Fig. 7). Lesher et al. (Chap. 12, this volume) discuss important kinetic controls on the nucleation of augite that further help to explain the discrepancies between the petrographic observations and the expectations based on the equilibrium phase relations for the basalts of the oceanic series. Low-Pressure Fractional Crystallization Modeling Quantitative modeling of fractional crystallization is presented in Figure 8. The modeling is accomplished by incrementally removing the calculated solid fractionate in equilibrium with the residual liquid. The calculation of the solid fractionate is based on the experimentally determined mineral compositions and their proportions from this study, Thy et al. (1996, 1998), and Toplis and Carroll (1995). The first model illustrates the effect of fractionation on the estimated primary liquid in equilibrium with mantle olivine (Fo 91 ) for Section 152-917A-86R-7 and uses the crystallization order experimentally determined (Thy et al., 1998). The second model shows the results for the section from Hole 989B using the crystallization order from Table 2. We can draw several conclusions from this modeling. Most of the compositional variation within groups of flows is accounted for by <15% fractionation. The exceptions are a few TiO 2 -rich flows from Site 917 that suggest extensive olivine, plagiocase, and augite removal approximating 50% 60% fractionation. The flows from Site 988, as well as a small subset from Site 918, also suggest derivation by extensive fractionation and may require a more Ti-rich parental composition than used in the calculations (Fig. 8). Origin and Evolution of the Middle Series Magmas The Middle Series at Site 917 deviates from the experimental liquid lines of descent (Fig. 7), as well as the modeling of compositional effects of fractional crystallization (Fig. 8). This is especially true for FeO and TiO 2 that do not show increasing concentrations with decreasing MgO as predicted for typical tholeiitic evolution trends. An alternative fractionation model involving early low-ti magnetite saturation (~1180 1200 C) could account for an early decrease in FeO and only modest variation in TiO 2. However, early saturation of Fe- Ti oxides is neither supported by the present study nor predicted based on Fe-Ti oxide solubility in basaltic systems (e.g., Toplis and Carroll, 1995). Furthermore, magnetite and ilmenite are conspicuously absent in the lavas of the Middle Series (Larsen, Saunders, Clift, et al., 1994). A B Figure 7. Normative projections of the olivine (ol), plagioclase (pl), diopside (di), quartz (q) basalt tetrahedron on the triangular diagrams (A) ol-pl-q and (B) oldi-q. Shown are the analyzed flows sampled during Leg 152 (Sites 917 and 918) and Leg 163 (Sites 988, 989, and 990), excluding analyses with >5% loss on ignition. Calculation, projection, and abbreviations are as for Figure 6. The solid lines are the pyroxene, olivine, and plagioclase cotectics taken from Figure 6. 105

P. THY, C.E. LESHER, J.D. MAYFIELD Figure 8. Oxide variation diagrams with MgO as abscissa for all available analyses of flows from Leg 152 and Leg 163. All in wt% oxides calculated anhydrous and iron distributed between FeO and Fe 2 O 3 (not shown) following Kilinc et al. (1983). Analyses with >5% loss on ignition has been excluded. Model calculation of fractional crystallization is shown and calculated as described in the text for Sections 163-989B-10R-7 and 152-917A-86R-7 (Table 2). The latter is extended to the inferred Primary Melt to show the early olivine fractionation (Thy et al., 1998). The model calculation for 152-917A-86R-7 is ticked by intervals of 5% fractionation. The appearances of Fe-Ti oxides in the crystallization models are at 75% fractionation for 152-917A-86R7 and 65% for 163-989B-10R- 7. The curve marked olivine illustrates the effect of suppression of plagioclase (and augite). The marked bends of the model curves are for the appearances of plagioclase, augite, and Fe-Ti oxide minerals, except for the latter, as experimentally determined. A consequence is that the major element variation observed for the Middle Series can neither be accounted for by the phenocryst assemblages nor by the experimental phase equilibria. The flows of the Middle Series are aphyric to plagioclase phyric basalts with embayed plagioclase and rare augite phenocrysts. The dacites and silicic tuffs are generally restricted to the lower and uppermost parts of the Middle Series and contain abundant embayed and sieved textured plagioclase phenocrysts and minor amounts of augite phenocrysts. Olivine and quartz are only observed as rare xenocrysts. The occurrence of augite phenocrysts, although not common, is inconsistent with the low-pressure phase relations shown in Figure 7, and it is possible that these augites are also xenocrystic. Alternatively, Thy et al. (1998) proposed that differentiation of Lower Series basaltic magmas was influenced by the addition of water, possibly during the early stages of crustal contamination. The effect of elevated water content is to reduce the stability of plagioclase relative to olivine and augite. This could explain the occurrence of pyroxene phenocrysts and sieve textured plagioclase phenocrysts. The Leg 152 Shipboard Scientific Party concluded that flows and tuffs of the Middle Series were contaminated by significant amounts of crustal material and suggested a process of magma mixing between basaltic and silicic components (Larsen, Saunders, Clift, et al., 1994). Fitton et al. (1998a, 1998b) argued that the flows of the Lower Series at Site 917 were contaminated by granulite facies leucogneiss, while the Middle Series was contaminated by amphibolite facies leucogneiss. This temporal change in crustal contaminant is interpreted as caused by a shallowing of magma chambers during development of the Lower and Middle Series succession. The modeling of Fitton et al. (1998a) implies relatively high extents of fractionation (50% 70%) and/or assimilation to account for the dacitic flows of the 106