TY - JOUR
AU - Hawkins, James W.
AB - Abstract We present the results of an experimental study on a refractory back-arc basin glass composition 123 95–1 recovered close to the intersection of the North Western Lau Spreading Centre and the Peggy Ridge in the Central Lau Basin. We used both the inverse and forward experimental approaches to determine that a picrite composition of ∼16 wt % MgO was parental to the refractory back-arc basalt composition and that this picrite was in equilibrium with a residual lherzolite assemblage at ∼2.1 GPa, ∼1460°C. Our experimental results suggest that this primary picrite represents neither an aggregate of small melt fractions collected over a depth interval in a melting column nor a small melt fraction. Instead, the primary picrite composition represents a significant melt fraction (15% melting) of a lherzolite source, and its composition is closely modelled by equilibrium batch melting. Our preferred model of magma genesis involves three-dimensional diapiric flow during which significant solid–melt re-equilibration occurs. Melts are trapped within an ascending diapir until melting reaches a moderate to high fraction (15–25%), at which point the primary picrite magmas segregate from the diapir. Introduction The Lau Basin (SW Pacific, Fig. 1) is postulated to have formed, in common with many other back-arc basins (Taylor & Karner, 1983), as a result of new ocean crust formation during lithospheric extension in response to the upwelling and partial melting of asthenospheric mantle behind island arcs (Crawford et al., 1981; Hawkins et al., 1984). Recent geophysical and geologic studies suggest that the Lau Basin has a complex history. The initial period (starting 6–6.5 Ma) was marked by protracted amagmatic back-arc extension in association with diffuse arc-related volcanism. This was followed by the development, at around 5–5.5 Ma, in the central Lau Basin of a propagating spreading ridge extending southward from the Peggy Ridge (Hawkins, 1995b). This spreading ridge was followed by a second propagating ridge, forming the present-day Central Lau Spreading Centre (CLSC) at ∼1.5 Ma (Hawkins, 1994, 1995a, 1995b; Hawkins et al., 1994). ‘GLORIA’ imaging and, more recently, bathymetry-corrected, magnetic data of the Lau Basin seafloor have defined the configuration and location of currently active spreading centres in the Lau Basin (Tiffin, 1993; Taylor et al., 1996; see Fig. 1). In the central and southern Lau Basin two spreading centres have been identified: the Central and Eastern Lau Spreading Centres (CLSC and ELSC, respectively). The well-studied Valu Fa Ridge (VF) forms the southern part of the ELSC. In the northern Lau Basin two new spreading centres have been identified: (1) the North Western Lau Spreading Centre (NWLSC), connected to the CLSC via the Peggy Ridge (PR), which is itself an extensional transform zone (Taylor et al., 1996); (2) the King's Triple Junction [KTJ; Mangatolu Triple Junction (MTJ) of Hawkins (1995a, 1995b)], which is a ridge–ridge–ridge feature identified in the North Eastern Lau Basin close to the termination of the north Tonga Ridge. Fig. 1. Open in new tabDownload slide Tectonic interpretation of the Lau Basin after Taylor et al. (1996) showing the location of sample TWD 123 95–1 (crossed circle). Spreading segments, strike-slip faults, propagation boundaries, pseudofaults and the trench axis are indicated by bold lines, lines with double arrows, ticked lines, dashed lines and lines with teeth, respectively. Islands are irregular shapes, filled black, and surrounded by the 1500 m contour (fine line). Ocean Drilling Program (ODP) Leg 135 drill sites located by open circles. Filled diamonds with numbers are stations 25 and 24 (see text, Falloon et al., 1987). NWLSC, North Western Lau Spreading Centre; ECVZ, east Cikobia volcanic zone; CLSC, Central Lau Spreading Centre; ELSC, Eastern Lau Spreading Centre; VF, Valu Fa Ridge; KTJ, King's Triple Junction; PR, Peggy Ridge; ETZ, extensional transform zone. Sampling of lavas from these modern-day spreading ridges has revealed that there is a remarkable diversity in major element, minor element and isotopic chemistry of erupted magmas (Hawkins, 1976, 1994, 1995a, 1995b; Hawkins & Melchior, 1985; Falloon et al., 1992; Ewart et al., 1994; Pearce et al., 1995). The VF spreading centre, located close (<50 km) to the present-day Tofua Volcanic Arc (TVA), has erupted magmas of predominantlybasaltic andesite to andesite composition similar to the BABB magma type defined by Sinton & Fryer (1987). Spreading centres located well away from the TVA (CLSC, PR) have erupted mainly olivine tholeiite to tholeiite magma compositions similar to those currently erupted at major mid-ocean ridges. Lavas sampled from the KTJ range from basalt to andesite, with the majority of these lavas showing mid-ocean ridge basalt (MORB)-like chemistry, with a small subset having compositions similar to the BABB magma type (Falloon et al., 1992; Hawkins, 1995b; J. W. Hawkins, unpublished data, 1986). In some of the above tectonic–magmatic associations there is evidence for the presence of very refractory primitive magmas. In the case of spreading ridges located close to the active arc volcanic front, rocks have been sampled off-axis which contain disequilibrium mineral assemblages as a result of mixing of an evolved andesitic to dacitic magma with a very refractory, olivine-bearing magma or magmas (Falloon et al., 1992; Kamenetsky et al., 1997). These refractory end-member magmas have crystallized magnesian olivines (up to Fo94), chromite (cr-number up to 0.87) and magnesian endiopside (up to mg-number 94) Kamenetsky et al., 1997). The melt inclusion study of Kamenetsky et al. (1997) suggests the generation and subsequent mixing of very refractory high-Ca and low-Ca boninite magmas associated with the southward propagation of the VF into shallow hydrated sub-arc lithosphere. Refractory high-Ca boninite lavas have also been sampled off-axis to the KTJ at the northern termination of the north Tonga Ridge (stations 24 and 25, Falloon et al., 1987, 1989; Fig. 1). At locations well away from the arc volcanic front, the most magnesian end-member sampled to date is an olivine tholeiite glass in equilibrium with olivine microphenocrysts (Fo86) and with significantly lower TiO2 and higher CaO/Na2O and CaO/Al2O3 values than normal MORB. This second refractory magma type is represented by sample 123 95-1, which was recovered by dredging close to the intersection of the PR and the NWLSC (16°23.6′S, 177°28.3′W) in 999–1322 m water depth (Hawkins, 1976; Fig. 1). This basalt glass has very low TiO2 (0.4 wt %, Table 1) and a high CaO/Na2O value (13.5). It also has low La/Yb with light rare earth element (LREE) concentrations at ∼1 × chondritic abundances and 6 × chondritic for the heavy rare earth elements (HREE) (Gill, 1976; Fig. 2). These characteristics imply a source region and/or melt extraction process in which the refractory olivine tholeiite has captured the characteristics of a residue from prior melt extraction within the garnet lherzolite field. Furthermore, this depletion process was more extensive (higher CaO/Na2O value, more depleted REE pattern) than similar depletion required for N-MORB sources (Figs 2 and 3). Figure 3 demonstrates that sample 123 95–1 has significantly lower abundances of large ion lithophile elements (LILE) and REE compared with the rest of the magma types so far identified in the Lau Basin, and that subduction-related enrichment in LILE and LREE has been relatively minor. Table 1: Selected Lau Basin basalt analyses and comparison with primitive MORB . 1 . 2 . 3 . 4 . 5 . 6 . 7 . 8 . 9 . 10 . 11 . 12 . . Peggy . Peggy . NWLSC . CLSC M- . ELCS 123 . ELSC . VF . KTJ . KTJ . KTJ . off-MORB . MORB . . Ridge . Ridge . PPTU 19–1 2231-1 . 74-1 . ODP . off-axis . RNDB . M-2218-9 . axis . Atlantic . Pacific . . TWD 123 . St-31 . . . . 836B-7R2, . SO35-128 . 19-1 . . boninite . DSDP-3- . ODP Leg . . 95-1 . 11-3 . . . . 58-62 . . . . St-25 7–18 18-7-1 . 148 896A- 27R-1, 124-130, pc.15 . Sio2 48.21 50.22 49.17 50.35 50.24 48.08 53.38 50.89 49.94 54.72 49.70 49.03 TiO2 0.40 0.65 0.82 1.20 0.76 0.72 0.50 0.94 1.13 0.45 0.72 0.64 Al2O3 16.01 15.33 17.15 14.97 15.80 16.21 13.44 15.93 15.66 10.90 16.40 16.07 FeO 9.10 9.57 8.72 10.48 9.13 8.13 8.45 8.53 9.48 8.65 7.90 9.12 MnO 0.16 0.18 0.18 0.21 0.14 0.23 0.17 0.17 0.17 0.12 0.12 MgO 11.10 8.87 9.51 7.84 7.39 9.07 8.20 7.52 8.44 12.97 10.10 9.43 CaO 13.90 13.40 12.75 12.44 13.37 13.78 11.98 13.10 12.43 9.65 13.10 13.66 Na2O 1.00 1.96 2.11 2.42 2.10 2.33 1.25 2.14 2.60 1.52 2.00 1.62 K2O 0.03 0.03 0.05 0.11 0.01 0.51 0.21 0.09 0.71 0.01 0.02 P2O5 0.04 0.06 0.09 0.14 0.09 0.11 0.06 0.26 0.03 . 1 . 2 . 3 . 4 . 5 . 6 . 7 . 8 . 9 . 10 . 11 . 12 . . Peggy . Peggy . NWLSC . CLSC M- . ELCS 123 . ELSC . VF . KTJ . KTJ . KTJ . off-MORB . MORB . . Ridge . Ridge . PPTU 19–1 2231-1 . 74-1 . ODP . off-axis . RNDB . M-2218-9 . axis . Atlantic . Pacific . . TWD 123 . St-31 . . . . 836B-7R2, . SO35-128 . 19-1 . . boninite . DSDP-3- . ODP Leg . . 95-1 . 11-3 . . . . 58-62 . . . . St-25 7–18 18-7-1 . 148 896A- 27R-1, 124-130, pc.15 . Sio2 48.21 50.22 49.17 50.35 50.24 48.08 53.38 50.89 49.94 54.72 49.70 49.03 TiO2 0.40 0.65 0.82 1.20 0.76 0.72 0.50 0.94 1.13 0.45 0.72 0.64 Al2O3 16.01 15.33 17.15 14.97 15.80 16.21 13.44 15.93 15.66 10.90 16.40 16.07 FeO 9.10 9.57 8.72 10.48 9.13 8.13 8.45 8.53 9.48 8.65 7.90 9.12 MnO 0.16 0.18 0.18 0.21 0.14 0.23 0.17 0.17 0.17 0.12 0.12 MgO 11.10 8.87 9.51 7.84 7.39 9.07 8.20 7.52 8.44 12.97 10.10 9.43 CaO 13.90 13.40 12.75 12.44 13.37 13.78 11.98 13.10 12.43 9.65 13.10 13.66 Na2O 1.00 1.96 2.11 2.42 2.10 2.33 1.25 2.14 2.60 1.52 2.00 1.62 K2O 0.03 0.03 0.05 0.11 0.01 0.51 0.21 0.09 0.71 0.01 0.02 P2O5 0.04 0.06 0.09 0.14 0.09 0.11 0.06 0.26 0.03 NWLSC, North Western Lau Spreading Centre; CLSC, Central Lau Spreading Centre; ELSC, Eastern Lau Spreading Centre; VF, Valu Fa Ridge; KTJ, King's Triple Junction. Data sources: 1, Hawkins (1976); 2 and 10, Falloon et al. (1987); 3, 5, 6 and 8, Hawkins (1995a); 7, Sunkel (1990); 4 and 9, Falloon et al. (1992); 11, Frey et al. (1974); 12, McNeill & Danyushevsky (1996). All analyses have been normalized to 100 wt % anhydrous. Open in new tab Table 1: Selected Lau Basin basalt analyses and comparison with primitive MORB . 1 . 2 . 3 . 4 . 5 . 6 . 7 . 8 . 9 . 10 . 11 . 12 . . Peggy . Peggy . NWLSC . CLSC M- . ELCS 123 . ELSC . VF . KTJ . KTJ . KTJ . off-MORB . MORB . . Ridge . Ridge . PPTU 19–1 2231-1 . 74-1 . ODP . off-axis . RNDB . M-2218-9 . axis . Atlantic . Pacific . . TWD 123 . St-31 . . . . 836B-7R2, . SO35-128 . 19-1 . . boninite . DSDP-3- . ODP Leg . . 95-1 . 11-3 . . . . 58-62 . . . . St-25 7–18 18-7-1 . 148 896A- 27R-1, 124-130, pc.15 . Sio2 48.21 50.22 49.17 50.35 50.24 48.08 53.38 50.89 49.94 54.72 49.70 49.03 TiO2 0.40 0.65 0.82 1.20 0.76 0.72 0.50 0.94 1.13 0.45 0.72 0.64 Al2O3 16.01 15.33 17.15 14.97 15.80 16.21 13.44 15.93 15.66 10.90 16.40 16.07 FeO 9.10 9.57 8.72 10.48 9.13 8.13 8.45 8.53 9.48 8.65 7.90 9.12 MnO 0.16 0.18 0.18 0.21 0.14 0.23 0.17 0.17 0.17 0.12 0.12 MgO 11.10 8.87 9.51 7.84 7.39 9.07 8.20 7.52 8.44 12.97 10.10 9.43 CaO 13.90 13.40 12.75 12.44 13.37 13.78 11.98 13.10 12.43 9.65 13.10 13.66 Na2O 1.00 1.96 2.11 2.42 2.10 2.33 1.25 2.14 2.60 1.52 2.00 1.62 K2O 0.03 0.03 0.05 0.11 0.01 0.51 0.21 0.09 0.71 0.01 0.02 P2O5 0.04 0.06 0.09 0.14 0.09 0.11 0.06 0.26 0.03 . 1 . 2 . 3 . 4 . 5 . 6 . 7 . 8 . 9 . 10 . 11 . 12 . . Peggy . Peggy . NWLSC . CLSC M- . ELCS 123 . ELSC . VF . KTJ . KTJ . KTJ . off-MORB . MORB . . Ridge . Ridge . PPTU 19–1 2231-1 . 74-1 . ODP . off-axis . RNDB . M-2218-9 . axis . Atlantic . Pacific . . TWD 123 . St-31 . . . . 836B-7R2, . SO35-128 . 19-1 . . boninite . DSDP-3- . ODP Leg . . 95-1 . 11-3 . . . . 58-62 . . . . St-25 7–18 18-7-1 . 148 896A- 27R-1, 124-130, pc.15 . Sio2 48.21 50.22 49.17 50.35 50.24 48.08 53.38 50.89 49.94 54.72 49.70 49.03 TiO2 0.40 0.65 0.82 1.20 0.76 0.72 0.50 0.94 1.13 0.45 0.72 0.64 Al2O3 16.01 15.33 17.15 14.97 15.80 16.21 13.44 15.93 15.66 10.90 16.40 16.07 FeO 9.10 9.57 8.72 10.48 9.13 8.13 8.45 8.53 9.48 8.65 7.90 9.12 MnO 0.16 0.18 0.18 0.21 0.14 0.23 0.17 0.17 0.17 0.12 0.12 MgO 11.10 8.87 9.51 7.84 7.39 9.07 8.20 7.52 8.44 12.97 10.10 9.43 CaO 13.90 13.40 12.75 12.44 13.37 13.78 11.98 13.10 12.43 9.65 13.10 13.66 Na2O 1.00 1.96 2.11 2.42 2.10 2.33 1.25 2.14 2.60 1.52 2.00 1.62 K2O 0.03 0.03 0.05 0.11 0.01 0.51 0.21 0.09 0.71 0.01 0.02 P2O5 0.04 0.06 0.09 0.14 0.09 0.11 0.06 0.26 0.03 NWLSC, North Western Lau Spreading Centre; CLSC, Central Lau Spreading Centre; ELSC, Eastern Lau Spreading Centre; VF, Valu Fa Ridge; KTJ, King's Triple Junction. Data sources: 1, Hawkins (1976); 2 and 10, Falloon et al. (1987); 3, 5, 6 and 8, Hawkins (1995a); 7, Sunkel (1990); 4 and 9, Falloon et al. (1992); 11, Frey et al. (1974); 12, McNeill & Danyushevsky (1996). All analyses have been normalized to 100 wt % anhydrous. Open in new tab Fig. 2. Open in new tabDownload slide Chondrite-normalized REE pattern of sample 123 95–1 compared with primitive MORB. Data sources as follows: 95–1 (TWD 123 95-1) from Gill (1976); 3-18, southern Atlantic [Deep Sea Drilling Project (DSDP) 3-18-7-1] from Frey et al. (1974); 896A, Pacific (ODP Leg 148 site 896A-27R-1, 124-130, pc. 15) from A. W. McNeill et al. (unpublished data, 1997). Chondrite normalizing values from Sun & McDonough (1989). Fig. 3. Open in new tabDownload slide N-MORB normalized spider diagram for sample 123 95–1 compared with representative Lau Basin magma types. N-MORB normalizing values from Sun & McDonough (1989). Data sources as follows: 95–1 (Gill, 1976; Hawkins & Melchior, 1985); N-MORB, sample M-2231-3, CLSC (Falloon et al., 1992); BABB, sample M-2212-2, KTJ (Falloon et al., 1992); ARC, sample ODP Leg 135 site 839B-25R-1, 27-32, unit 3 (matrix) (Ewart et al., 1994); E-MORB, sample Dr. 120-1-1, KTJ (T. J. Falloon et al., unpublished data, 1996). Sample 123 95–1 is of interest not only because of its primitive and refractory nature but also because its location on the PR suggests that it may be representative of the first magmas erupted as the Lau Basin opened (Hawkins, 1995b). Thus the petrogenesis of sample 123 95–1 will have important implications for our understanding of back-arc basin initiation and subsequent evolution. In this paper we present an experimental study of the refractory composition 123 95–1 utilizing both the inverse and forward experimental approaches [Basaltic Volcanism Study Project (BSVP), 1981; see below] to: determine whether the composition 123 95–1 could have been in equilibrium with an upper-mantle wall-rock residue; find the pressure and temperature at which 123 95–1 or its more primitive parent (in the case that 123 95-1is not a primary magma) was last in equilibrium with mantle wall-rock, i.e. the point of magma segregation or of melt aggregation (see below); constrain the nature of the mantle residue at the point of magma segregation or aggregation and the degree of partial melting of appropriate mantle sources; place petrogenetic and geodynamic constraints on magma genesis in the Lau Basin. Rationale of the Experimental Approach The Basaltic Volcanism Study Project (BSVP, 1981) suggested that primary basaltic liquids could be successfully used as probes into planetary interiors and that both the experimental inverse and forward approaches could successfully yield valuable information on the nature of the mantle source regions of basaltic magmatism. The inverse experimental approach takes the basaltic liquid selected as primitive or primary and infers the nature of the mantle source regions from the nature of liquidus phases of the basalt. The forward experimental approach makes assumptions on the nature of the mantle source composition and attempts to directly determine the compositions of melts from this source and compare them with primary basaltic magmatism. Both approaches are based on two very important assumptions. The first assumption is that during magmatic processes such as crystallization and partial melting, the major element compositions of crystals and liquids will closely resemble equilibrium assemblages. Or, in terms of chemical thermodynamics, the system chosen for study has minimized its overall Gibbs free energy. We believe this assumption is justified on the basis that (1) the timescales involved in magmatic processes are of the order of thousands of years (Rubin & Macdougall, 1988, 1990) and (2) at magmatic temperatures, the rates of chemical equilibration observed in laboratory experiments are very rapid. The second very important assumption is that mantle melts leave their source regions as liquids, not as some mixture of matrix and liquid. This assumption is critical in evaluating experimental data that seek to demonstrate equilibrium with mantle mineralogies. More complex models, in which rapidly moving magma entrains xenoliths, xenocrysts or residual phases and even acts as a density filter on such materials, can be suggested and are potentially important for both extreme magma types such as peridotitic komatiite or volatile-rich magmas. The inverse experimental approach The inverse experimental approach is based on the concept and interpretation of experimentally determined multiple saturation points on the high-pressure liquidi of basalt magmas. A multiple saturation point is a P, T point on the liquidus where two or more phases coexist. The liquidus phases (assuming no reaction relationships) indicate the residual source mineralogy for a simple melting model (BSVP, 1981). This P, T condition has also been called the ‘depth of magma segregation’ at which the magma finally leaves its wall-rock environment (Green & Ringwood, 1967). The nature of the residual mantle phases at the depth of magma segregation will be critical in determining the major element and compatible trace element characteristics of any primary mantle-derived magma. The inverse approach thus seeks to test whether there are conditions at which natural magmas are in equilibrium with mantle mineralogies. The concept of multiple saturation is commonly used to test whether a basalt composition of interest is a primary magma from peridotitic mantle. If the composition of interest is primary then it should be possible to find multiple saturation on the liquidus at which olivine + orthopyroxene ± clinopyroxene ± spinel ± garnet ± plagioclase (assuming no reaction relationships) are liquidus phases. Experimental studies are designed to find such P, T conditions for the chosen magma. In most published experimental studies the point of multiple saturation is at the interpolated intersection of two liquidus fields, olivine as the lower-pressure phase and orthopyroxene as the liquidus phase at higher pressures. However, the absence of multiple saturation (in those minerals appropriate for a mantle residue) is taken to indicate that either (1) the composition of interest is unlikely to be a primary magma and has suffered some degree of modification following primary magma segregation (i.e. crystal fractionation, magma mixing, assimilation, wall-rock reaction) or (2) a reaction relationship exists between crystals and liquid at the P, T of magma segregation. Only a more detailed exploration of compositional space close to the composition of interest can distinguish between these two possibilities and this is where the forward experimental approach (see below) can be very useful in interpreting the significance of the presence or absence of multiple saturation on the liquidi of basaltic magmas. The inverse experimental approach is incapable of constraining the modal proportions of crystals vs melt at the depth of magma segregation because if we have an equilibrium system we are free to change the proportions of equilibrium phases without changing their compositions (BSVP, 1981). The advantage of the inverse experimental approach, however, lies in the fact that the equilibrium conditions of multiple saturation (depth of magma segregation) so determined can be applied as a constraint on a range of models of mantle melting. For example, a primary magma multiply saturated in olivine and orthopyroxene at 2 GPa and 1450°C tells us that the composition of interest is in equilibrium with a harzburgite wall-rock environment but the melt itself could represent: a fractional melt of ≪ 1% finally losing contact with the surrounding matrix in a model of magma genesis involving porous flow in a melting column (McKenzie & Bickle, 1988); a large melt fraction of >20%, which itself could represent a batch melt in a diapiric model of magma genesis (Green et al., 1979; Green & Falloon, 1998); an aggregated fractional melt composition (Eggins, 1992) in a harzburgite melt column; or a magma that has finally lost contact with its wall-rock environment after a significant period of wall-rock reaction (Kelemen et al., 1992). The inverse approach is more difficult to apply if primary magma compositions are determined in the final stages by porous flow and equilibration within a monomineralic (olivine-only) channelway. Nevertheless, olivine addition and subtraction vectors point toward compositions previously at olivine and orthopyroxene saturation and reaction vectors must be obtained from the experimental studies to address such reactive, porous flow models. The forward experimental approach If a basaltic magma of interest is a primary magma produced by partial melting of a mantle peridotite composition, then it should be possible to produce similar compositions by direct partial melting studies on suitable peridotite compositions. The success of the forward experimental approach depends on the suitability of the chosen source composition and on the experimental technique chosen for determining partial melt compositions (Falloon et al., 1996, 1997). The forward experimental approach can be undertaken in two different ways: (1) partial melt products can be determined from a suitably chosen peridotite starting composition and compared directly with the natural magmas; (2) a basaltic composition of interest can be ‘reacted’ with a potential source peridotite composition in a sandwich or mixed reaction experiment (Falloon & Green, 1987; Falloon et al., 1988), forcing the composition of interest to reach equilibrium with a residual peridotite mineralogy at the chosen P, T conditions. If the basaltic composition is indeed close to a primary melt for these P, T conditions then there should be little compositional shift between the original basaltic composition and the final equilibrated composition. An advantage of the sandwich or mixed reaction experiment is that it can clarify any reaction relationships between liquids and the phases entering the liquid. As discussed above, the inverse technique may not give definitive answers if reaction relationships exist. The advantage of the forward approach is that it can provide information on the modal abundances of equilibrium residual phases, and the degree of partial melting of the likely source composition for a model that involves simple melting of peridotite (either batch or fractional). In this experimental study we have used both the inverse and forward experimental approaches to investigate the depth of magma segregation and the nature of the primary magma for the refractory Lau Basin composition 123 95-1. Experimental Technique All 1 atm furnace and piston-cylinder experiments were performed in the High Pressure Laboratory in the School of Earth Sciences at the University of Tasmania. Starting compositions (Table 2) for the experimental study are all sintered oxide mixes of the composition of 123 95–1 (95-1, Table 2) and progressively more olivine-enriched compositions (95-2, -3, -4, Table 2). Sintered oxide mixes of peridotite compositions were also used (TQ-40, MM-3, Table 2). Compositions 95-5, -6, -7, -8 (Table 2) were prepared by the addition of orthopyroxene (either pure synthetic clinoenstatite or natural minerals, Table 2) to the olivine-enriched compositions. Starting compositions were prepared from analytical grade oxides, ground under acetone, pelletized and sintered overnight at 950°C. Fayalite was then added to the mixes. The mixes were stored in an oven at 110°C. Starting materials were either crystalline or glass as indicated in Table 2. The compositions of our mixes as determined by electron microprobe analysis are presented in Table 2. Table 2 demonstrates that the prepared composition of mix 95–1 differs slightly from the modelled natural glass 123 95–1 in having slightly less FeO and slightly higher CaO and Na2O contents. These differences arose because of Fe loss to the Pt crucible on fusion (to prepare a glass starting composition from the sintered oxide mix), and alkali contamination of the batch of silica used in the preparation of the 95–1 mix. It is sufficient to note that the olivine-enriched compositions 95-2, 95–3 and 95–4 do not suffer from the high Na2O content and closely model 123 95–1 plus olivine. These differences in the 95–1 mix will have little effect on the phase relations other than to slightly raise the liquidus temperature, resulting in a slightly more Mg-rich liquidus olivine and to cause a less calcic plagioclase to crystallize than would otherwise crystallize from the natural composition. The sintered oxide mix 95–2 was used in the atmospheric pressure experiments (Table 3), as this composition was not affected by the slight compositional problems outlined above. For the 1 atm experiments ∼20 mg of sample was placed in spec-pure Fe capsules, which were sealed in evacuated silica tubes and suspended in a vertical, Pt-wound, atmospheric pressure, quenching furnace. Use of crystalline starting material for low-pressure runs is preferable as glass starting material may inhibit nucleation of plagioclase. In the case of peridotite reaction experiments (Table 3) a layer of sintered oxide basalt mix (95-1 or 95-3) was placed above a layer of peridotite mix (TQ-40 or MM-3, Table 3). Table 2: Starting compositions used in the experimental study Mix . . SiO2 . TiO2 . Al2O3 . FeO . MnO . MgO . CaO . Na2O . K2O . Cr2O3 . 95-1 48.21 0.40 16.01 9.10 0.16 11.10 13.90 1.00 0.03 0.10 EMPA 47.90 0.42 16.52 8.49 0.11 10.67 14.30 1.40 0.07 0.11 95-2 47.39 0.36 14.28 9.17 0.14 15.18 12.40 0.89 0.03 0.16 EMPA 47.40 0.35 13.83 9.42 0.07 15.17 12.61 0.97 0.05 0.13 95-3 47.18 0.35 13.83 9.19 0.14 16.25 12.01 0.86 0.03 0.16 EMPA 46.98 0.37 13.5 9.38 0.10 16.25 12.17 0.98 0.05 0.21 95-4 46.92 0.33 13.26 9.21 0.13 17.63 11.51 0.83 0.02 0.16 EMPA 47.65 0.35 13.16 8.59 0.09 17.64 11.37 0.96 0.05 0.15 95-5 47.27 0.35 13.70 9.19 0.14 16.41 11.90 0.85 0.03 0.16 95-6 47.31 0.35 13.69 9.10 0.14 16.49 11.89 0.85 0.03 0.16 95-7 47.35 0.35 13.63 9.14 0.14 16.53 11.83 0.85 0.03 0.16 EMPA 47.96 0.38 13.10 9.04 0.09 16.52 11.75 0.93 0.05 0.18 95-8 47.38 0.34 13.61 9.04 0.14 16.63 11.82 0.85 0.03 0.16 Opx (WSS1) 57.61 0.16 1.07 6.20 0.14 33.90 0.59 0.00 0.00 0.32 Opx (2539) 56.46 0.00 0.90 9.40 0.00 32.68 0.55 0.00 0.00 0.00 TQ-40 47.51 0.13 5.35 7.51 0.13 32.81 4.97 0.3 0.03 0.75 MM-3 45.50 0.11 3.98 7.18 0.13 38.3 3.57 0.31 — 0.68 Mix . . SiO2 . TiO2 . Al2O3 . FeO . MnO . MgO . CaO . Na2O . K2O . Cr2O3 . 95-1 48.21 0.40 16.01 9.10 0.16 11.10 13.90 1.00 0.03 0.10 EMPA 47.90 0.42 16.52 8.49 0.11 10.67 14.30 1.40 0.07 0.11 95-2 47.39 0.36 14.28 9.17 0.14 15.18 12.40 0.89 0.03 0.16 EMPA 47.40 0.35 13.83 9.42 0.07 15.17 12.61 0.97 0.05 0.13 95-3 47.18 0.35 13.83 9.19 0.14 16.25 12.01 0.86 0.03 0.16 EMPA 46.98 0.37 13.5 9.38 0.10 16.25 12.17 0.98 0.05 0.21 95-4 46.92 0.33 13.26 9.21 0.13 17.63 11.51 0.83 0.02 0.16 EMPA 47.65 0.35 13.16 8.59 0.09 17.64 11.37 0.96 0.05 0.15 95-5 47.27 0.35 13.70 9.19 0.14 16.41 11.90 0.85 0.03 0.16 95-6 47.31 0.35 13.69 9.10 0.14 16.49 11.89 0.85 0.03 0.16 95-7 47.35 0.35 13.63 9.14 0.14 16.53 11.83 0.85 0.03 0.16 EMPA 47.96 0.38 13.10 9.04 0.09 16.52 11.75 0.93 0.05 0.18 95-8 47.38 0.34 13.61 9.04 0.14 16.63 11.82 0.85 0.03 0.16 Opx (WSS1) 57.61 0.16 1.07 6.20 0.14 33.90 0.59 0.00 0.00 0.32 Opx (2539) 56.46 0.00 0.90 9.40 0.00 32.68 0.55 0.00 0.00 0.00 TQ-40 47.51 0.13 5.35 7.51 0.13 32.81 4.97 0.3 0.03 0.75 MM-3 45.50 0.11 3.98 7.18 0.13 38.3 3.57 0.31 — 0.68 Starting compositions: 95-1, Lau Basin glass 123 95–1 (Gill, 1976; Hawkins, 1976); 95-2, 89 wt % 95–1 + 11 wt % olivine (Fo90); 95-3, 87 wt % 95–1 + 13 wt % olivine (Fo90); 95-4, 83 wt % 95–1 + 17 wt % olivine (Fo90); 95-5, 99 wt % 95–3 + 1 wt % orthopyroxene (2539); 95-6, 99 wt % 95–3 + 1 wt % clinoenstatite; 95-7, 98 wt % 95–3 + 2 wt % orthopyroxene (WSS1); 95-8, 98 wt % 95–3 + 2 wt % clinoenstatite. EMPA, electron microprobe analysis of glass (normalized to 100 wt %) in above-liquidus experiments (Table 3). Open in new tab Table 2: Starting compositions used in the experimental study Mix . . SiO2 . TiO2 . Al2O3 . FeO . MnO . MgO . CaO . Na2O . K2O . Cr2O3 . 95-1 48.21 0.40 16.01 9.10 0.16 11.10 13.90 1.00 0.03 0.10 EMPA 47.90 0.42 16.52 8.49 0.11 10.67 14.30 1.40 0.07 0.11 95-2 47.39 0.36 14.28 9.17 0.14 15.18 12.40 0.89 0.03 0.16 EMPA 47.40 0.35 13.83 9.42 0.07 15.17 12.61 0.97 0.05 0.13 95-3 47.18 0.35 13.83 9.19 0.14 16.25 12.01 0.86 0.03 0.16 EMPA 46.98 0.37 13.5 9.38 0.10 16.25 12.17 0.98 0.05 0.21 95-4 46.92 0.33 13.26 9.21 0.13 17.63 11.51 0.83 0.02 0.16 EMPA 47.65 0.35 13.16 8.59 0.09 17.64 11.37 0.96 0.05 0.15 95-5 47.27 0.35 13.70 9.19 0.14 16.41 11.90 0.85 0.03 0.16 95-6 47.31 0.35 13.69 9.10 0.14 16.49 11.89 0.85 0.03 0.16 95-7 47.35 0.35 13.63 9.14 0.14 16.53 11.83 0.85 0.03 0.16 EMPA 47.96 0.38 13.10 9.04 0.09 16.52 11.75 0.93 0.05 0.18 95-8 47.38 0.34 13.61 9.04 0.14 16.63 11.82 0.85 0.03 0.16 Opx (WSS1) 57.61 0.16 1.07 6.20 0.14 33.90 0.59 0.00 0.00 0.32 Opx (2539) 56.46 0.00 0.90 9.40 0.00 32.68 0.55 0.00 0.00 0.00 TQ-40 47.51 0.13 5.35 7.51 0.13 32.81 4.97 0.3 0.03 0.75 MM-3 45.50 0.11 3.98 7.18 0.13 38.3 3.57 0.31 — 0.68 Mix . . SiO2 . TiO2 . Al2O3 . FeO . MnO . MgO . CaO . Na2O . K2O . Cr2O3 . 95-1 48.21 0.40 16.01 9.10 0.16 11.10 13.90 1.00 0.03 0.10 EMPA 47.90 0.42 16.52 8.49 0.11 10.67 14.30 1.40 0.07 0.11 95-2 47.39 0.36 14.28 9.17 0.14 15.18 12.40 0.89 0.03 0.16 EMPA 47.40 0.35 13.83 9.42 0.07 15.17 12.61 0.97 0.05 0.13 95-3 47.18 0.35 13.83 9.19 0.14 16.25 12.01 0.86 0.03 0.16 EMPA 46.98 0.37 13.5 9.38 0.10 16.25 12.17 0.98 0.05 0.21 95-4 46.92 0.33 13.26 9.21 0.13 17.63 11.51 0.83 0.02 0.16 EMPA 47.65 0.35 13.16 8.59 0.09 17.64 11.37 0.96 0.05 0.15 95-5 47.27 0.35 13.70 9.19 0.14 16.41 11.90 0.85 0.03 0.16 95-6 47.31 0.35 13.69 9.10 0.14 16.49 11.89 0.85 0.03 0.16 95-7 47.35 0.35 13.63 9.14 0.14 16.53 11.83 0.85 0.03 0.16 EMPA 47.96 0.38 13.10 9.04 0.09 16.52 11.75 0.93 0.05 0.18 95-8 47.38 0.34 13.61 9.04 0.14 16.63 11.82 0.85 0.03 0.16 Opx (WSS1) 57.61 0.16 1.07 6.20 0.14 33.90 0.59 0.00 0.00 0.32 Opx (2539) 56.46 0.00 0.90 9.40 0.00 32.68 0.55 0.00 0.00 0.00 TQ-40 47.51 0.13 5.35 7.51 0.13 32.81 4.97 0.3 0.03 0.75 MM-3 45.50 0.11 3.98 7.18 0.13 38.3 3.57 0.31 — 0.68 Starting compositions: 95-1, Lau Basin glass 123 95–1 (Gill, 1976; Hawkins, 1976); 95-2, 89 wt % 95–1 + 11 wt % olivine (Fo90); 95-3, 87 wt % 95–1 + 13 wt % olivine (Fo90); 95-4, 83 wt % 95–1 + 17 wt % olivine (Fo90); 95-5, 99 wt % 95–3 + 1 wt % orthopyroxene (2539); 95-6, 99 wt % 95–3 + 1 wt % clinoenstatite; 95-7, 98 wt % 95–3 + 2 wt % orthopyroxene (WSS1); 95-8, 98 wt % 95–3 + 2 wt % clinoenstatite. EMPA, electron microprobe analysis of glass (normalized to 100 wt %) in above-liquidus experiments (Table 3). Open in new tab Both Pt/Pt90Rh10 and W75Re25/W97Re3 thermocouples (runs T-2611 and higher) were used in the course of this experimental study and temperatures are accurate to within ±15°C. Talc–Pyrex or NaCl–Pyrex assemblies with graphite heaters were used for piston-cylinder experiments. All assemblies were stored in an oven at 110°C. Pressures are accurate to within ±0.1 GPa. Graphite, graphite sealed in platinum, platinum and iron were used as capsule materials (Table 3) during the experimental study. Graphite capsules were used for runs at 1 GPa and above to avoid the problems of Fe loss or gain. Sealed Pt capsules, which were immersed in fine Fe powder and heated at 900–1000°C for 4–6 h in an attempt to minimize Fe loss, were used for the 0.2 GPa and some 0.5 GPa runs in preference to the graphite capsules, because at pressures below 0.5 GPa use of graphite capsules results in f(O2) well below the quartz–fayalite–magnetite buffer, causing some FeO to be reduced to metallic Fe. Near-liquidus runs at 0.5 GPa in graphite capsules differed from those in Pt capsules by the presence of chrome spinel in runs in Pt capsules and its absence in graphite runs. Despite the attempt to presaturate the Pt capsules, some iron was lost to the Pt capsules during the runs, resulting in slightly more Mg-rich compositions than those obtained in graphite. At the end of each experiment selected pieces of run products, in the case of crystallization experiments, were mounted and polished for the purpose of electron microprobe analysis and scanning electron micrographs (using a Phillips 505 scanning electron microscope, operating conditions 20 kV and spot size 100 nm), whereas in the case of peridotite reaction experiments (Table 3) the entire experimental charge was mounted and sectioned longitudinally before polishing. Crystalline phases were analysed by either energy-dispersive (EDS) microanalysis using a JEOL JX-50A electron probe microanalyser (operating conditions 15 kV and 5 nA, calibrated using pure Cu, formerly housed in the Central Science Laboratory, University of Tasmania) or wavelength-dispersive (WDS) microanalysis using a Cameca SX50 microprobe currently housed in the Central Science Laboratory, University of Tasmania (operating conditions 15 kV and 20 nA). All glass analyses presented in Table 4 were obtained by WDS microanalysis using the international glass standard VG-2 (Jarosewich et al., 1979). Table 3: Experimental run data Run no. . Pressure (kbar) . Temperature (°C) . Time (min) . Capsule . Material . Mix . Phase assemblage . 123 95–1 crystallization experiments T-44 5 1260 45 C gl 95-1 L T-39 5 1240 45 C gl 95-1 Ol + L T-91 5 1240 40 Pt gl 95-1 Ol + Sp + L T-42 5 1220 45 C gl 95-1 Ol + Plg + L T-43 5 1210 45 C gl 95-1 Ol + Plg + L1 T-40 5 1200 45 C gl 95-1 Ol + Plg + L T-59 10 1300 45 C gl 95-1 L T-58 10 1290 45 C gl 95-1 Ol + L T-54 10 1280 45 C gl 95-1 Ol + L T-92 10 1280 45 Pt gl 95-1 Ol + Sp + L T-55 10 1270 45 C gl 95-1 Ol + L T-45 10 1260 45 C gl 95-1 Ol + Cpx1 + L T-48 10 1240 45 C gl 95-1 Ol + Cpx3 + Plg + L1 T-79 12 1310 45 C gl 95-1 L T-80 12 1290 45 C gl 95-1 Ol + Cpx1 + L T-68 15 1360 45 C gl 95-1 L T-69 15 1340 45 C gl 95-1 L T-66 15 1320 45 C gl 95-1 Cpx1 + L T-62 15 1300 45 C gl 95-1 Cpx3 + L1 Olivine addition crystallization experiments Series 1 (11 wt % olivine) AT-17 0 1360 15 Fe cryst 95-2 L AT-13 0 1340 15 Fe cryst 95-2 Ol + L AT-11 0 1300 60 Fe cryst 95-2 Ol + L AT-15 0 1230 120 Fe cryst 95-2 Ol + Plg + L AT-14 0 1210 120 Fe cryst 95-2 Ol + Plg + L AT-10 0 1200 120 Fe cryst 95-2 Ol + Plg + L1 AT-12 0 1190 120 Fe cryst 95-2 Ol + Plg + Cpx5 + L1 AT-9 0 1160 60 Fe cryst 95-2 Ol + Plg + Cpx3 + L1 T-188 2 1320 30 Pt gl 95-2 L T-189 2 1280 30 Pt gl 95-2 Ol + Sp + L T-190 2 1240 40 Pt gl 95-2 Ol + Sp + L T-376 2 1200 120 Fe cryst 95-2 Ol + Plg + L T-200 2 1180 45 Pt gl 95-2 Ol + Plg1 + Sp1 + L T-209 2 1160 50 Pt gl 95-2 Ol1 + Plg1 + Cpx3 + L1 T-369 5 1360 30 C cryst 95-2 L T-179 5 1340 45 C gl 95-2 Ol + L T-224 5 1330 30 Pt gl 95-2 Ol + Sp1 + L T-193 5 1300 30 Pt gl 95-2 Ol + Sp + L T-195 5 1260 45 Pt gl 95-2 Ol + Sp + L T-375 5 1230 120 Fe cryst 95-2 Ol + L T-196 5 1220 45 Pt gl 95-2 Ol + Sp1 + L T-219 5 1210 45 Pt gl 95-2 Ol + Plg + L T-197 5 1200 45 Pt gl 95-2 Ol + Plg + Cpx3 + L T-181 10 1380 45 C gl 95-2 L T-176 10 1360 45 C gl 95-2 Ol + L T-205 10 1260 45 C gl 95-2 Ol + Cpx1 + L T-169 15 1400 45 C gl 95-2 L T-182 15 1380 45 C gl 95-2 Ol + L T-171 15 1360 45 C gl 95-2 Ol + L T-220 15 1320 45 C gl 95-2 Ol + Cpx3 + L1 T-204 18 1400 40 C gl 95-2 Ol + Cpx2 + L T-221 20 1460 40 C gl 95-2 L T-213 20 1440 40 C gl 95-2 Cpx2 + L Series 2 (13 wt % olivine) T-352 18 1420 60 C cryst 95-3 Ol + L1 T-343 18 1400 60 C cryst 95-3 Ol + Cpx2 + L T-353 20 1460 60 C cryst 95-3 Ol + L T-364 20 1450 60 C cryst 95-3 Ol + L T-344 20 1440 60 C cryst 95-3 Ol? + Cpx2 + L1 T-414 22 1480 30 C cryst 95-3 L T-356 22 1460 60 C cryst 95-3 Cpx2 + L Series 3 (17 wt % olivine) T-531 25 1540 30 C cryst 95-4 L T-537 25 1520 30 C cryst 95-4 Ol + L T-543 25 1515 40 C cryst 95-4 Ol + Cpx3 + L T-542 25 1510 40 C cryst 95-4 Ol + Cpx3 + L1 T-540 25 1500 40 C cryst 95-4 Ol + Cpx3 + L1 T-530 25 1500 60 C cryst 95-4 Ol + Cpx + L1 T-2662 26 1530 150 Pt/C cryst 95-4 Cpx3 + Ga + L1 T-2611 27 1540 60 Pt/C cryst 95-4 Cpx3 + Ga? + L T-2617 27 1500 60 Pt/C cryst 95-4 Ol + Cpx3 + Ga + L1 Olivine and orthopyroxene addition crystallization experiments Series 1 (1 wt % orthopyroxene and 13 wt % olivine) T-361 20 1450 60 C cryst 95-5 Ol + Opx + Cpx2 + L T-377 22 1460 30 C cryst 95-6 Cpx2 + L Series 2 (2 wt % orthopyroxene and 13 wt % olivine) T-524 21 1480 60 C cryst 95-7 Ol + L T-517 21 1460 60 C cryst 95-8 Ol + Cpx2 + L T-518 21 1460 60 C cryst 95-8 Ol + Opx + Cpx2 + L T-525 23 1500 60 C cryst 95-7 L T-528 23 1490 60 C cryst 95-7 Cpx2 + L T-529 23 1490 60 C cryst 95-7 Cpx4 + L T-523 23 1480 60 C cryst 95-8 Cpx4 + L Peridotite reaction experiments Series A T-4151 21 1460 1452 C cryst 95-3(19)/TQ-40 Ol + Opx + Cpx3 + L T-4152 21 1460 1452 C cryst 95-3(27)/MM-3 Ol + Opx + Cpx3 + L Series B T-4013 13 1325 1440 C cryst 95-1(25)/TQ-40 Ol + Opx + Cpx5 + Sp + L T-3996 18 1410 2880 C cryst 95-3(27)/TQ-40 Ol + Opx + Cpx3 + L T-3997 22 1480 1530 C cryst 95-3(28)/TQ-40 Ol + Opx + Cpx3 + L T-3998 25 1530 1530 C cryst 95-3(25)/TQ-40 Ol + Opx + L Run no. . Pressure (kbar) . Temperature (°C) . Time (min) . Capsule . Material . Mix . Phase assemblage . 123 95–1 crystallization experiments T-44 5 1260 45 C gl 95-1 L T-39 5 1240 45 C gl 95-1 Ol + L T-91 5 1240 40 Pt gl 95-1 Ol + Sp + L T-42 5 1220 45 C gl 95-1 Ol + Plg + L T-43 5 1210 45 C gl 95-1 Ol + Plg + L1 T-40 5 1200 45 C gl 95-1 Ol + Plg + L T-59 10 1300 45 C gl 95-1 L T-58 10 1290 45 C gl 95-1 Ol + L T-54 10 1280 45 C gl 95-1 Ol + L T-92 10 1280 45 Pt gl 95-1 Ol + Sp + L T-55 10 1270 45 C gl 95-1 Ol + L T-45 10 1260 45 C gl 95-1 Ol + Cpx1 + L T-48 10 1240 45 C gl 95-1 Ol + Cpx3 + Plg + L1 T-79 12 1310 45 C gl 95-1 L T-80 12 1290 45 C gl 95-1 Ol + Cpx1 + L T-68 15 1360 45 C gl 95-1 L T-69 15 1340 45 C gl 95-1 L T-66 15 1320 45 C gl 95-1 Cpx1 + L T-62 15 1300 45 C gl 95-1 Cpx3 + L1 Olivine addition crystallization experiments Series 1 (11 wt % olivine) AT-17 0 1360 15 Fe cryst 95-2 L AT-13 0 1340 15 Fe cryst 95-2 Ol + L AT-11 0 1300 60 Fe cryst 95-2 Ol + L AT-15 0 1230 120 Fe cryst 95-2 Ol + Plg + L AT-14 0 1210 120 Fe cryst 95-2 Ol + Plg + L AT-10 0 1200 120 Fe cryst 95-2 Ol + Plg + L1 AT-12 0 1190 120 Fe cryst 95-2 Ol + Plg + Cpx5 + L1 AT-9 0 1160 60 Fe cryst 95-2 Ol + Plg + Cpx3 + L1 T-188 2 1320 30 Pt gl 95-2 L T-189 2 1280 30 Pt gl 95-2 Ol + Sp + L T-190 2 1240 40 Pt gl 95-2 Ol + Sp + L T-376 2 1200 120 Fe cryst 95-2 Ol + Plg + L T-200 2 1180 45 Pt gl 95-2 Ol + Plg1 + Sp1 + L T-209 2 1160 50 Pt gl 95-2 Ol1 + Plg1 + Cpx3 + L1 T-369 5 1360 30 C cryst 95-2 L T-179 5 1340 45 C gl 95-2 Ol + L T-224 5 1330 30 Pt gl 95-2 Ol + Sp1 + L T-193 5 1300 30 Pt gl 95-2 Ol + Sp + L T-195 5 1260 45 Pt gl 95-2 Ol + Sp + L T-375 5 1230 120 Fe cryst 95-2 Ol + L T-196 5 1220 45 Pt gl 95-2 Ol + Sp1 + L T-219 5 1210 45 Pt gl 95-2 Ol + Plg + L T-197 5 1200 45 Pt gl 95-2 Ol + Plg + Cpx3 + L T-181 10 1380 45 C gl 95-2 L T-176 10 1360 45 C gl 95-2 Ol + L T-205 10 1260 45 C gl 95-2 Ol + Cpx1 + L T-169 15 1400 45 C gl 95-2 L T-182 15 1380 45 C gl 95-2 Ol + L T-171 15 1360 45 C gl 95-2 Ol + L T-220 15 1320 45 C gl 95-2 Ol + Cpx3 + L1 T-204 18 1400 40 C gl 95-2 Ol + Cpx2 + L T-221 20 1460 40 C gl 95-2 L T-213 20 1440 40 C gl 95-2 Cpx2 + L Series 2 (13 wt % olivine) T-352 18 1420 60 C cryst 95-3 Ol + L1 T-343 18 1400 60 C cryst 95-3 Ol + Cpx2 + L T-353 20 1460 60 C cryst 95-3 Ol + L T-364 20 1450 60 C cryst 95-3 Ol + L T-344 20 1440 60 C cryst 95-3 Ol? + Cpx2 + L1 T-414 22 1480 30 C cryst 95-3 L T-356 22 1460 60 C cryst 95-3 Cpx2 + L Series 3 (17 wt % olivine) T-531 25 1540 30 C cryst 95-4 L T-537 25 1520 30 C cryst 95-4 Ol + L T-543 25 1515 40 C cryst 95-4 Ol + Cpx3 + L T-542 25 1510 40 C cryst 95-4 Ol + Cpx3 + L1 T-540 25 1500 40 C cryst 95-4 Ol + Cpx3 + L1 T-530 25 1500 60 C cryst 95-4 Ol + Cpx + L1 T-2662 26 1530 150 Pt/C cryst 95-4 Cpx3 + Ga + L1 T-2611 27 1540 60 Pt/C cryst 95-4 Cpx3 + Ga? + L T-2617 27 1500 60 Pt/C cryst 95-4 Ol + Cpx3 + Ga + L1 Olivine and orthopyroxene addition crystallization experiments Series 1 (1 wt % orthopyroxene and 13 wt % olivine) T-361 20 1450 60 C cryst 95-5 Ol + Opx + Cpx2 + L T-377 22 1460 30 C cryst 95-6 Cpx2 + L Series 2 (2 wt % orthopyroxene and 13 wt % olivine) T-524 21 1480 60 C cryst 95-7 Ol + L T-517 21 1460 60 C cryst 95-8 Ol + Cpx2 + L T-518 21 1460 60 C cryst 95-8 Ol + Opx + Cpx2 + L T-525 23 1500 60 C cryst 95-7 L T-528 23 1490 60 C cryst 95-7 Cpx2 + L T-529 23 1490 60 C cryst 95-7 Cpx4 + L T-523 23 1480 60 C cryst 95-8 Cpx4 + L Peridotite reaction experiments Series A T-4151 21 1460 1452 C cryst 95-3(19)/TQ-40 Ol + Opx + Cpx3 + L T-4152 21 1460 1452 C cryst 95-3(27)/MM-3 Ol + Opx + Cpx3 + L Series B T-4013 13 1325 1440 C cryst 95-1(25)/TQ-40 Ol + Opx + Cpx5 + Sp + L T-3996 18 1410 2880 C cryst 95-3(27)/TQ-40 Ol + Opx + Cpx3 + L T-3997 22 1480 1530 C cryst 95-3(28)/TQ-40 Ol + Opx + Cpx3 + L T-3998 25 1530 1530 C cryst 95-3(25)/TQ-40 Ol + Opx + L Ol, olivine (1olivine too fine-grained to analyse); Cpx, clinopyroxene (1sector zoning, 2zoning to subcalcic pyroxene, 3unzoned, 4zoning to relict ‘seed’ pyroxene, 5zoned); Opx, orthopyroxene; Sp, spinel (1spinel too fine-grained to analyse); Plg, plagioclase (1plagioclase too fine-grained to analyse); L, glass (1highly crystalline charge or strongly quench modified glass, no reliable analysis possible);?, presence of phase inferred but not found during EMPA of selected pieces of run product; cryst, sintered oxide starting material; gl, glass starting material; C, graphite; Pt, platinum; Fe, iron. Mix compositions are given in Table 2. Numbers in parentheses next to the mix composition for peridotite reaction experiments refer to the percentage basalt mix in the bulk composition. Open in new tab Table 3: Experimental run data Run no. . Pressure (kbar) . Temperature (°C) . Time (min) . Capsule . Material . Mix . Phase assemblage . 123 95–1 crystallization experiments T-44 5 1260 45 C gl 95-1 L T-39 5 1240 45 C gl 95-1 Ol + L T-91 5 1240 40 Pt gl 95-1 Ol + Sp + L T-42 5 1220 45 C gl 95-1 Ol + Plg + L T-43 5 1210 45 C gl 95-1 Ol + Plg + L1 T-40 5 1200 45 C gl 95-1 Ol + Plg + L T-59 10 1300 45 C gl 95-1 L T-58 10 1290 45 C gl 95-1 Ol + L T-54 10 1280 45 C gl 95-1 Ol + L T-92 10 1280 45 Pt gl 95-1 Ol + Sp + L T-55 10 1270 45 C gl 95-1 Ol + L T-45 10 1260 45 C gl 95-1 Ol + Cpx1 + L T-48 10 1240 45 C gl 95-1 Ol + Cpx3 + Plg + L1 T-79 12 1310 45 C gl 95-1 L T-80 12 1290 45 C gl 95-1 Ol + Cpx1 + L T-68 15 1360 45 C gl 95-1 L T-69 15 1340 45 C gl 95-1 L T-66 15 1320 45 C gl 95-1 Cpx1 + L T-62 15 1300 45 C gl 95-1 Cpx3 + L1 Olivine addition crystallization experiments Series 1 (11 wt % olivine) AT-17 0 1360 15 Fe cryst 95-2 L AT-13 0 1340 15 Fe cryst 95-2 Ol + L AT-11 0 1300 60 Fe cryst 95-2 Ol + L AT-15 0 1230 120 Fe cryst 95-2 Ol + Plg + L AT-14 0 1210 120 Fe cryst 95-2 Ol + Plg + L AT-10 0 1200 120 Fe cryst 95-2 Ol + Plg + L1 AT-12 0 1190 120 Fe cryst 95-2 Ol + Plg + Cpx5 + L1 AT-9 0 1160 60 Fe cryst 95-2 Ol + Plg + Cpx3 + L1 T-188 2 1320 30 Pt gl 95-2 L T-189 2 1280 30 Pt gl 95-2 Ol + Sp + L T-190 2 1240 40 Pt gl 95-2 Ol + Sp + L T-376 2 1200 120 Fe cryst 95-2 Ol + Plg + L T-200 2 1180 45 Pt gl 95-2 Ol + Plg1 + Sp1 + L T-209 2 1160 50 Pt gl 95-2 Ol1 + Plg1 + Cpx3 + L1 T-369 5 1360 30 C cryst 95-2 L T-179 5 1340 45 C gl 95-2 Ol + L T-224 5 1330 30 Pt gl 95-2 Ol + Sp1 + L T-193 5 1300 30 Pt gl 95-2 Ol + Sp + L T-195 5 1260 45 Pt gl 95-2 Ol + Sp + L T-375 5 1230 120 Fe cryst 95-2 Ol + L T-196 5 1220 45 Pt gl 95-2 Ol + Sp1 + L T-219 5 1210 45 Pt gl 95-2 Ol + Plg + L T-197 5 1200 45 Pt gl 95-2 Ol + Plg + Cpx3 + L T-181 10 1380 45 C gl 95-2 L T-176 10 1360 45 C gl 95-2 Ol + L T-205 10 1260 45 C gl 95-2 Ol + Cpx1 + L T-169 15 1400 45 C gl 95-2 L T-182 15 1380 45 C gl 95-2 Ol + L T-171 15 1360 45 C gl 95-2 Ol + L T-220 15 1320 45 C gl 95-2 Ol + Cpx3 + L1 T-204 18 1400 40 C gl 95-2 Ol + Cpx2 + L T-221 20 1460 40 C gl 95-2 L T-213 20 1440 40 C gl 95-2 Cpx2 + L Series 2 (13 wt % olivine) T-352 18 1420 60 C cryst 95-3 Ol + L1 T-343 18 1400 60 C cryst 95-3 Ol + Cpx2 + L T-353 20 1460 60 C cryst 95-3 Ol + L T-364 20 1450 60 C cryst 95-3 Ol + L T-344 20 1440 60 C cryst 95-3 Ol? + Cpx2 + L1 T-414 22 1480 30 C cryst 95-3 L T-356 22 1460 60 C cryst 95-3 Cpx2 + L Series 3 (17 wt % olivine) T-531 25 1540 30 C cryst 95-4 L T-537 25 1520 30 C cryst 95-4 Ol + L T-543 25 1515 40 C cryst 95-4 Ol + Cpx3 + L T-542 25 1510 40 C cryst 95-4 Ol + Cpx3 + L1 T-540 25 1500 40 C cryst 95-4 Ol + Cpx3 + L1 T-530 25 1500 60 C cryst 95-4 Ol + Cpx + L1 T-2662 26 1530 150 Pt/C cryst 95-4 Cpx3 + Ga + L1 T-2611 27 1540 60 Pt/C cryst 95-4 Cpx3 + Ga? + L T-2617 27 1500 60 Pt/C cryst 95-4 Ol + Cpx3 + Ga + L1 Olivine and orthopyroxene addition crystallization experiments Series 1 (1 wt % orthopyroxene and 13 wt % olivine) T-361 20 1450 60 C cryst 95-5 Ol + Opx + Cpx2 + L T-377 22 1460 30 C cryst 95-6 Cpx2 + L Series 2 (2 wt % orthopyroxene and 13 wt % olivine) T-524 21 1480 60 C cryst 95-7 Ol + L T-517 21 1460 60 C cryst 95-8 Ol + Cpx2 + L T-518 21 1460 60 C cryst 95-8 Ol + Opx + Cpx2 + L T-525 23 1500 60 C cryst 95-7 L T-528 23 1490 60 C cryst 95-7 Cpx2 + L T-529 23 1490 60 C cryst 95-7 Cpx4 + L T-523 23 1480 60 C cryst 95-8 Cpx4 + L Peridotite reaction experiments Series A T-4151 21 1460 1452 C cryst 95-3(19)/TQ-40 Ol + Opx + Cpx3 + L T-4152 21 1460 1452 C cryst 95-3(27)/MM-3 Ol + Opx + Cpx3 + L Series B T-4013 13 1325 1440 C cryst 95-1(25)/TQ-40 Ol + Opx + Cpx5 + Sp + L T-3996 18 1410 2880 C cryst 95-3(27)/TQ-40 Ol + Opx + Cpx3 + L T-3997 22 1480 1530 C cryst 95-3(28)/TQ-40 Ol + Opx + Cpx3 + L T-3998 25 1530 1530 C cryst 95-3(25)/TQ-40 Ol + Opx + L Run no. . Pressure (kbar) . Temperature (°C) . Time (min) . Capsule . Material . Mix . Phase assemblage . 123 95–1 crystallization experiments T-44 5 1260 45 C gl 95-1 L T-39 5 1240 45 C gl 95-1 Ol + L T-91 5 1240 40 Pt gl 95-1 Ol + Sp + L T-42 5 1220 45 C gl 95-1 Ol + Plg + L T-43 5 1210 45 C gl 95-1 Ol + Plg + L1 T-40 5 1200 45 C gl 95-1 Ol + Plg + L T-59 10 1300 45 C gl 95-1 L T-58 10 1290 45 C gl 95-1 Ol + L T-54 10 1280 45 C gl 95-1 Ol + L T-92 10 1280 45 Pt gl 95-1 Ol + Sp + L T-55 10 1270 45 C gl 95-1 Ol + L T-45 10 1260 45 C gl 95-1 Ol + Cpx1 + L T-48 10 1240 45 C gl 95-1 Ol + Cpx3 + Plg + L1 T-79 12 1310 45 C gl 95-1 L T-80 12 1290 45 C gl 95-1 Ol + Cpx1 + L T-68 15 1360 45 C gl 95-1 L T-69 15 1340 45 C gl 95-1 L T-66 15 1320 45 C gl 95-1 Cpx1 + L T-62 15 1300 45 C gl 95-1 Cpx3 + L1 Olivine addition crystallization experiments Series 1 (11 wt % olivine) AT-17 0 1360 15 Fe cryst 95-2 L AT-13 0 1340 15 Fe cryst 95-2 Ol + L AT-11 0 1300 60 Fe cryst 95-2 Ol + L AT-15 0 1230 120 Fe cryst 95-2 Ol + Plg + L AT-14 0 1210 120 Fe cryst 95-2 Ol + Plg + L AT-10 0 1200 120 Fe cryst 95-2 Ol + Plg + L1 AT-12 0 1190 120 Fe cryst 95-2 Ol + Plg + Cpx5 + L1 AT-9 0 1160 60 Fe cryst 95-2 Ol + Plg + Cpx3 + L1 T-188 2 1320 30 Pt gl 95-2 L T-189 2 1280 30 Pt gl 95-2 Ol + Sp + L T-190 2 1240 40 Pt gl 95-2 Ol + Sp + L T-376 2 1200 120 Fe cryst 95-2 Ol + Plg + L T-200 2 1180 45 Pt gl 95-2 Ol + Plg1 + Sp1 + L T-209 2 1160 50 Pt gl 95-2 Ol1 + Plg1 + Cpx3 + L1 T-369 5 1360 30 C cryst 95-2 L T-179 5 1340 45 C gl 95-2 Ol + L T-224 5 1330 30 Pt gl 95-2 Ol + Sp1 + L T-193 5 1300 30 Pt gl 95-2 Ol + Sp + L T-195 5 1260 45 Pt gl 95-2 Ol + Sp + L T-375 5 1230 120 Fe cryst 95-2 Ol + L T-196 5 1220 45 Pt gl 95-2 Ol + Sp1 + L T-219 5 1210 45 Pt gl 95-2 Ol + Plg + L T-197 5 1200 45 Pt gl 95-2 Ol + Plg + Cpx3 + L T-181 10 1380 45 C gl 95-2 L T-176 10 1360 45 C gl 95-2 Ol + L T-205 10 1260 45 C gl 95-2 Ol + Cpx1 + L T-169 15 1400 45 C gl 95-2 L T-182 15 1380 45 C gl 95-2 Ol + L T-171 15 1360 45 C gl 95-2 Ol + L T-220 15 1320 45 C gl 95-2 Ol + Cpx3 + L1 T-204 18 1400 40 C gl 95-2 Ol + Cpx2 + L T-221 20 1460 40 C gl 95-2 L T-213 20 1440 40 C gl 95-2 Cpx2 + L Series 2 (13 wt % olivine) T-352 18 1420 60 C cryst 95-3 Ol + L1 T-343 18 1400 60 C cryst 95-3 Ol + Cpx2 + L T-353 20 1460 60 C cryst 95-3 Ol + L T-364 20 1450 60 C cryst 95-3 Ol + L T-344 20 1440 60 C cryst 95-3 Ol? + Cpx2 + L1 T-414 22 1480 30 C cryst 95-3 L T-356 22 1460 60 C cryst 95-3 Cpx2 + L Series 3 (17 wt % olivine) T-531 25 1540 30 C cryst 95-4 L T-537 25 1520 30 C cryst 95-4 Ol + L T-543 25 1515 40 C cryst 95-4 Ol + Cpx3 + L T-542 25 1510 40 C cryst 95-4 Ol + Cpx3 + L1 T-540 25 1500 40 C cryst 95-4 Ol + Cpx3 + L1 T-530 25 1500 60 C cryst 95-4 Ol + Cpx + L1 T-2662 26 1530 150 Pt/C cryst 95-4 Cpx3 + Ga + L1 T-2611 27 1540 60 Pt/C cryst 95-4 Cpx3 + Ga? + L T-2617 27 1500 60 Pt/C cryst 95-4 Ol + Cpx3 + Ga + L1 Olivine and orthopyroxene addition crystallization experiments Series 1 (1 wt % orthopyroxene and 13 wt % olivine) T-361 20 1450 60 C cryst 95-5 Ol + Opx + Cpx2 + L T-377 22 1460 30 C cryst 95-6 Cpx2 + L Series 2 (2 wt % orthopyroxene and 13 wt % olivine) T-524 21 1480 60 C cryst 95-7 Ol + L T-517 21 1460 60 C cryst 95-8 Ol + Cpx2 + L T-518 21 1460 60 C cryst 95-8 Ol + Opx + Cpx2 + L T-525 23 1500 60 C cryst 95-7 L T-528 23 1490 60 C cryst 95-7 Cpx2 + L T-529 23 1490 60 C cryst 95-7 Cpx4 + L T-523 23 1480 60 C cryst 95-8 Cpx4 + L Peridotite reaction experiments Series A T-4151 21 1460 1452 C cryst 95-3(19)/TQ-40 Ol + Opx + Cpx3 + L T-4152 21 1460 1452 C cryst 95-3(27)/MM-3 Ol + Opx + Cpx3 + L Series B T-4013 13 1325 1440 C cryst 95-1(25)/TQ-40 Ol + Opx + Cpx5 + Sp + L T-3996 18 1410 2880 C cryst 95-3(27)/TQ-40 Ol + Opx + Cpx3 + L T-3997 22 1480 1530 C cryst 95-3(28)/TQ-40 Ol + Opx + Cpx3 + L T-3998 25 1530 1530 C cryst 95-3(25)/TQ-40 Ol + Opx + L Ol, olivine (1olivine too fine-grained to analyse); Cpx, clinopyroxene (1sector zoning, 2zoning to subcalcic pyroxene, 3unzoned, 4zoning to relict ‘seed’ pyroxene, 5zoned); Opx, orthopyroxene; Sp, spinel (1spinel too fine-grained to analyse); Plg, plagioclase (1plagioclase too fine-grained to analyse); L, glass (1highly crystalline charge or strongly quench modified glass, no reliable analysis possible);?, presence of phase inferred but not found during EMPA of selected pieces of run product; cryst, sintered oxide starting material; gl, glass starting material; C, graphite; Pt, platinum; Fe, iron. Mix compositions are given in Table 2. Numbers in parentheses next to the mix composition for peridotite reaction experiments refer to the percentage basalt mix in the bulk composition. Open in new tab Run times for near-liquidus crystallization experiments were generally short (15–60 min), especially in those runs using Pt or Fe capsules to minimize Fe loss or gain. Run times for peridotite reaction experiments are longer (24–48 h) and are based on run times that in our experience produce a close approach to equilibrium assemblages (Falloon & Green, 1988; Falloon et al., in preparation). In general, our run products that have not been affected by Fe loss or gain or quench modification of liquids show good mass balance and equilibrium Kd values for olivine–liquid Fe–Mg exchange. Also, mineral phases are unzoned even in our shortest near-liquidus runs. However, in many cases microprobe analyses of clinopyroxene displayed significant composition variation (Table 3). This compositional variation is attributed to the following factors: (1) sector zoning in the clinopyroxene; (2) in near-liquidus experiments at P >1.8 GPa and T >1400°C, the presence of metastable sub-calcic cores or relict seed crystals (Fig. 4) in the case of addition experiments (see Table 3); (3) overlap with other phases, especially glass and coexisting orthopyroxene (in the case of peridotite reaction experiments) because of the small crystal size or habit of the clinopyroxene (1–2 μm thin, tabular plates). The presence of metastable sub-calcic cores in some of the high-pressure and -temperature near-liquidus runs indicates that these experiments have not fully come to equilibrium; however, as discussed below, the presence of these metastable sub-calcic cores is significant in assessing how close the olivine-enriched 95–1 compositions are to multiple saturation with mantle peridotite. Experimental Results Our experimental run data are presented in Table 3. Representative analyses of the run products are presented in Table 4. The experimental results are also summarized in Figs 5–8. Inverse experimental approach The 123 95–1 crystallization experiments The phase relationships of mix 95–1 from atmospheric pressure to 1.5 GPa are presented in Fig. 5, which also incorporates phase equilibrium results from the more olivine-enriched composition 95–2 at atmospheric pressure and 0.2, 0.5 and 1 GPa where the temperatures overlap with experiments performed on the 95–1 composition and the MgO content of the quenched glass is <11.1 wt %. Because olivine is the liquidus phase for both compositions at these pressures, both 95–1 and 95–2 can be used to establish the phase relationships of 123 95-1. The agreement between the 95–1 and 95–2 experiments indicates that the differences in Na2O and FeO between glass 95–1 and the natural composition 123 95–1 have not produced significantly different phase relations. At atmospheric pressure, olivine crystallizes alone to 1230°C, where it is joined by plagioclase. Olivine compositions change systematically, becoming less magnesian with decreasing temperature, Fo88 at 1340°C, Fo84 at 1230°C and Fo70 at 1190°C. Modelling using both SILMIN (Ghiorso et al., 1983) and PETROLOG (a program designed by L. V. Danyushevsky et al. for modelling crystallization processes) predicts that olivine (Fo86) is the liquidus phase for the 123 95–1 natural composition at a temperature of ∼1260°C. Thus we infer that the olivine microphenocrysts (Fo86) present in the natural rock are in equilibrium with the composition 123 95–1 and that any prior history of crystal fractionation involved olivine only. Plagioclase (An91) crystallizes at 1230°C and becomes less calcic with decreasing temperature (An79 at 1160°C). Clinopyroxene (mg-number 82.6) crystallizes at 1190°C. Olivine remains the liquidus phase to ∼1.3 GPa, where it is replaced by clinopyroxene. Clinopyroxene replaces plagioclase as the second crystallizing phase at 0.8–1.0 GPa (Fig. 6). Multiple saturation in olivine + orthopyroxene ± clinopyroxene, which is expected for equilibrium with a mantle residue mineralogy, does not occur near the liquidus for the 95–1 composition. This indicates that 95–1 is not a primary, mantle-derived magma; it underwent some olivine fractionation before eruption. We have therefore investigated a number of olivine-enriched (11–17 wt %) compositions (95-2, 95–3 and 95-4, Table 2) to determine whether a more picritic composition could be in equilibrium with a mantle residue assemblage. Table 4: Experimental run products Run no. . Pressure (kbar) . Temperature (°C) . Phase . sio2 . Tio2 . Al2o3 . FeO . MnO . MgO . CaO . Na2O . K2O . Cr2O3 . 123 95–1 crystallization experiments T-39 5 1240 Ol 40.39 11.62 47.08 0.57 L 47.47 0.40 16.58 8.75 0.10 10.05 15.01 1.43 0.09 0.12 T-91 5 1240 Ol 40.10 11.70 47.20 0.56 0.16 Sp 35.80 13.90 18.50 0.34 31.50 L 48.33 0.42 16.74 8.11 0.09 10.74 14.13 1.31 0.05 0.09 T-42 5 1220 Ol 40.00 11.80 47.20 0.62 Plg 47.50 33.60 0.26 0.28 16.80 1.56 L 48.67 0.47 16.35 8.04 0.11 9.23 15.36 1.57 0.09 0.12 T-43 5 1210 Ol 40.00 14.00 44.90 0.67 Plg 48.10 33.00 17.30 1.59 T-40 5 1200 Ol 39.70 0.66 15.20 43.60 0.88 Plg 48.10 31.30 0.78 0.83 17.40 1.63 L 48.83 0.54 15.21 9.57 0.14 8.34 15.54 1.58 0.10 0.15 T-58 10 1290 Ol 40.40 11.50 47.80 0.41 L 47.98 0.43 16.93 8.46 0.11 9.64 14.78 1.44 0.08 0.14 T-54 10 1280 Ol 40.30 11.52 47.54 0.48 L 47.74 0.43 16.70 8.80 0.13 9.97 14.58 1.44 0.08 0.11 T-92 10 1280 Ol 40.20 11.70 47.70 0.45 Sp 41.10 12.80 18.80 0.36 26.90 L 48.32 0.41 16.67 8.06 0.11 10.82 14.14 1.33 0.04 0.11 T-55 10 1270 Ol 40.20 11.80 47.60 0.36 L 47.83 0.42 16.60 8.74 0.12 10.24 14.41 1.42 0.07 0.15 T-45 10 1260 Ol 40.00 12.20 47.30 0.55 Cpx A 52.43 0.18 4.82 5.42 20.86 15.56 0.16 0.37 Cpx B 49.2 0.4 9.98 3.99 16.11 19.59 0.29 0.37 L 48.08 0.41 17.34 8.59 0.13 9.78 13.98 1.55 0.08 0.07 T-48 10 1240 Ol 38.60 0.96 18.90 40.90 0.63 Plg 48.90 32.00 0.35 0.53 16.10 2.14 Cpx 50.21 0.47 8.92 6.55 15.43 17.46 0.55 0.28 T-80 12 1290 Ol 40.10 12.00 47.30 0.53 0.16 Cpx A 52.25 0.19 5.25 5.29 19.55 17.12 0.25 0.17 Cpx B 49.48 0.33 9.39 4.03 15.96 19.98 0.30 0.36 L 47.67 0.42 16.83 9.19 0.09 10.27 13.85 1.49 0.07 0.11 T-66 15 1320 Cpx A 52.03 0.25 6.07 4.72 20.05 16.21 0.34 0.26 Cpx B 49.24 0.33 10.34 4.32 16.77 18.17 0.42 0.30 L 47.05 0.42 18.01 8.55 0.12 10.48 13.85 1.36 0.08 0.09 T-62 15 1300 Cpx 48.69 0.28 12.04 5.03 15.18 18.10 0.39 Olivine addition crystallization experiments Series 1 (11 wt % olivine) AT-13 0 1340 Ol 40.69 11.45 0.11 48.11 0.49 L 46.92 0.38 14.21 10.35 0.06 14.06 12.80 0.97 0.05 0.20 AT-11 0 1300 Ol 40.24 12.19 0.11 47.38 0.52 L 47.22 0.39 15.06 10.38 0.07 12.14 13.48 1.09 0.06 0.12 AT-15 0 1230 Ol 39.42 15.12 44.87 0.59 Plg 45.60 34.00 0.35 0.57 18.40 1.02 L 48.32 0.44 15.83 10.27 0.03 9.09 14.64 1.18 0.06 0.14 AT-14 0 1210 Ol 39.90 15.00 44.60 0.52 Plg 46.10 33.30 0.51 0.65 18.30 1.11 L 49.18 0.48 14.60 10.52 0.10 8.42 15.25 1.24 0.06 0.15 AT-10 0 1200 Ol 39.09 17.24 43.14 0.54 Plg 45.30 34.70 0.84 1.67 16.20 1.19 AT-12 0 1190 Ol 38.3 22.96 0.07 37.99 0.68 Plg 45.90 31.80 1.41 1.25 17.10 1.43 Cpx 53.10 4.33 6.32 16.80 19.50 AT-9 0 1160 Ol 37.63 26.79 35.15 0.44 Plg 46.94 34.52 1.01 0.25 15.06 2.21 Cpx 52.53 0.39 2.67 9.68 15.62 18.77 0.22 T-189 2 1280 Ol 40.84 8.69 49.61 0.37 Sp 0.53 0.28 29.31 13.41 17.66 0.28 38.53 L 48.24 0.36 14.77 8.64 0.11 13.66 13.11 0.92 0.06 0.14 T-190 2 1240 Ol 40.53 9.40 49.13 0.30 Sp 0.54 0.29 29.44 12.53 16.80 0.28 40.12 L 49.13 0.41 15.94 7.51 0.08 11.68 14.11 0.98 0.05 0.11 T-376 2 1200 Ol 39.20 16.30 44.10 0.43 Plg 47.20 32.70 0.68 18.30 1.08 L 48.50 0.47 15.00 11.18 0.10 8.25 15.04 1.23 0.07 0.17 T-200 2 1180 Ol 39.19 14.38 0.30 45.62 0.50 T-209 2 1160 Cpx 51.47 5.74 5.51 16.50 20.27 T-179 5 1340 Ol 40.80 9.42 49.60 0.27 L 47.51 0.38 13.96 9.43 0.16 15.20 12.30 0.83 0.04 0.19 T-224 5 1330 Ol 41.08 8.03 50.17 0.38 L 48.30 0.37 14.21 7.85 0.10 15.42 12.60 0.91 0.04 0.20 T-193 5 1300 Ol 40.21 10.35 0.18 48.58 0.43 Sp 30.20 14.50 19.40 0.49 35.40 L 48.62 0.40 15.06 8.28 0.10 12.80 13.58 0.95 0.05 0.16 T-195 5 1260 Ol 40.60 9.84 49.20 0.31 Sp 32.00 15.00 18.60 0.59 33.80 L 49.12 0.40 15.76 7.64 0.09 11.45 14.33 1.04 0.06 0.11 T-375 5 1230 Ol 39.70 13.80 46.10 0.43 L 47.76 0.39 15.85 10.45 0.05 10.01 14.22 1.07 0.06 0.14 T-196 5 1220 Ol 40.33 10.47 48.20 0.51 L 49.50 0.40 16.26 8.02 0.13 9.88 14.60 1.08 0.05 0.08 T-219 5 1210 Ol 39.92 13.66 45.72 0.46 Plag 47.30 32.70 0.63 0.89 17.20 1.58 L 49.07 0.47 15.91 8.92 0.11 8.89 15.38 1.04 0.05 0.16 T-197 5 1200 Ol 39.60 14.90 45.20 0.41 Plag 48.30 30.90 0.80 0.91 17.70 1.37 Cpx 52.40 4.08 6.59 19.20 17.40 0.94 L 50.24 0.56 15.23 9.23 0.17 9.43 13.86 1.14 0.08 0.06 T-176 10 1360 Ol 40.80 9.96 49.20 L 47.69 0.39 14.76 9.32 0.14 14.54 12.10 0.85 0.06 0.16 T-205 10 1260 Ol 40.00 12.70 46.90 0.36 Cpx 51.60 7.03 4.50 18.60 17.80 0.63 L 48.58 0.40 16.19 8.99 0.11 9.94 14.54 1.03 0.05 0.18 T-182 15 1380 Ol 40.60 10.70 48.60 0.17 L 47.82 0.38 14.59 9.44 0.12 14.02 12.56 0.84 0.05 0.18 T-171 15 1360 Ol 40.40 11.10 47.90 0.48 L 47.90 0.36 14.94 9.43 0.10 13.14 12.99 0.92 0.05 0.17 T-220 15 1320 Ol 40.25 12.21 47.02 0.38 Cpx 50.81 0.2 7.83 5.4 19.92 15.21 0.20 0.33 T-204 18 1400 Ol 40.10 11.50 48.20 0.29 Cpx 53.10 7.10 5.48 23.30 12.20 0.46 L 46.94 0.41 14.96 10.19 0.11 13.84 12.38 0.97 0.06 0.14 T-213 20 1440 Cpx 51.90 7.51 5.18 23.10 11.30 0.31 L 47.30 0.40 14.35 9.99 0.10 14.53 12.22 0.91 0.05 0.14 Series 2 (13 wt % olivine) T-352 18 1420 Ol 41.1 6.84 51.65 0.37 T-343 18 1400 Ol 40.41 11.20 0.15 47.59 0.38 Cpx 50.31 0.18 7.83 5.04 21.08 14.68 0.33 0.45 L 46.83 0.47 15.67 9.90 0.11 12.84 12.84 1.19 0.06 0.09 T-353 20 1460 Ol 40.79 8.44 50.25 0.4 L 48.15 0.35 13.96 8.01 0.13 15.78 12.42 0.98 0.05 0.17 T-364 20 1450 Ol 40.64 8.27 0.2 50.51 0.33 L 48.60 0.39 13.34 8.38 0.10 16.30 11.79 0.87 0.05 0.19 T-344 20 1440 Cpx 51.16 0.17 8.66 5.01 20.80 13.62 0.15 0.29 T-356 22 1460 Cpx 50.77 0.16 9.47 5.12 20.36 13.17 0.50 0.31 L 46.65 0.40 15.07 10.00 0.12 14.33 12.11 1.13 0.06 0.13 Series 3 (17 wt % olivine) T-537 25 1520 Ol 41.72 8.74 50.04 0.27 L 47.37 0.33 13.35 9.03 0.12 16.85 11.84 0.90 0.06 0.15 T-543 25 1515 Ol 41.28 9.66 48.71 0.35 Cpx 53.38 6.75 4.71 24.26 10.13 0.35 0.43 L 46.52 0.40 14.11 9.89 0.13 15.89 11.85 1.02 0.06 0.13 T-542 25 1510 Ol 40.70 10.17 48.76 0.37 Cpx 52.01 0.11 7.56 5.18 23.20 11.24 0.30 0.37 T-540 25 1500 Ol 40.60 10.01 48.77 0.44 Cpx 51.63 0.10 7.43 5.37 23.68 11.07 0.34 0.31 T-530 25 1500 Ol 39.98 11.26 47.98 0.51 Cpx 50.82 0.12 8.62 5.66 21.59 12.39 0.46 0.34 T-2662 26 1530 Cpx 50.26 0.13 10.27 6.71 18.62 13.42 0.73 Ga 42.89 0.16 22.57 7.96 19.04 7.29 0.15 T-2611 27 1540 Cpx 53.67 5.96 4.76 25.07 9.87 0.32 0.35 L 47.39 0.36 13.41 9.13 0.09 16.59 11.89 0.93 0.06 0.16 T-2617 27 1500 Ol 39.34 0.26 14.91 44.14 0.44 Cpx 50.76 0.22 9.51 6.01 18.20 14.77 0.53 Ga 42.14 0.19 23.60 8.25 19.35 6.21 0.27 Olivine and orthopyroxene addition crystallization experiments Series 1 (1 wt % orthopyroxene and 13 wt % olivine) T-361 20 1450 Ol 39.85 10.27 49.41 0.38 Opx 53.62 0.11 6.03 6.07 30.83 2.70 0.19 Cpx 52.44 0.11 5.68 5.82 26.47 9.14 0.22 L 48.02 0.35 14.41 9.16 0.07 14.59 12.27 0.95 0.06 0.11 T-377 22 1460 Cpx 50.49 0.17 10.09 5.22 22.27 11.03 0.33 0.29 L 46.04 0.48 15.81 10.57 0.15 13.69 11.90 1.17 0.10 0.10 Series 2 (2 wt % orthopyroxene and 13 wt % olivine) T-524 21 1480 Ol 41.43 8.96 49.39 0.23 L 48.00 0.37 13.53 9.05 0.09 16.07 11.72 0.91 0.04 0.20 T-517 21 1460 Ol 40.92 0.23 9.78 49.16 0.45 Cpx 52.53 0.07 6.21 5.39 24.14 10.94 0.27 0.37 L 47.54 0.41 14.17 9.34 0.11 15.07 12.16 1.01 0.06 0.13 T-518 21 1460 Ol 40.51 9.27 49.53 0.38 Opx 55.67 0.05 3.03 5.66 32.46 2.71 0.10 0.23 Cpx 54.28 0.06 3.85 5.56 28.49 7.05 0.17 0.33 L 48.0 0.37 13.62 9.21 0.10 15.43 12.06 1.03 0.05 0.15 T-528 23 1490 Cpx 52.20 0.08 6.03 5.45 25.06 10.33 0.37 0.36 L 47.40 0.40 13.85 9.57 0.10 15.57 11.89 1.00 0.06 0.15 T-529 23 1490 Cpx 52.99 0.06 5.98 5.24 26.35 8.65 0.23 0.40 L 47.49 0.38 13.90 9.20 0.12 15.84 11.86 0.98 0.07 0.16 T-523 23 1480 Cpx 52.67 0.07 6.62 4.92 24.97 10.03 0.27 0.33 L 47.15 0.38 14.38 9.45 0.12 15.51 11.81 1.03 0.06 0.11 Peridotite reaction experiments Series A T-4151 21 1460 Ol 41.01 9.23 0.17 48.65 0.35 Opx 54.63 0.05 4.93 5.45 0.11 31.03 2.70 0.10 1.01 Cpx 53.04 0.03 5.65 4.81 0.17 24.11 10.71 0.32 1.16 L 47.18 0.55 13.84 9.46 0.16 15.15 12.08 1.02 0.15 0.41 T-4152 21 1460 Ol 41.48 8.39 0.15 49.59 0.37 Opx 54.42 0.06 5.26 4.98 0.10 31.22 2.71 0.10 1.15 Cpx 52.83 0.06 5.42 4.34 0.14 23.59 12.09 0.35 1.19 L 47.24 0.54 13.81 8.41 0.14 16.20 12.23 0.94 0.06 0.43 Series B T-4013 13 1325 Ol 40.10 12.67 46.71 0.17 Opx 52.82 7.62 7.05 29.52 2.43 0.56 Cpx 49.95 0.28 9.30 4.50 17.34 17.86 0.17 0.43 Sp 0.70 59.05 10.05 19.59 0.08 10.53 L 48.45 0.61 18.56 8.69 0.15 9.26 11.50 2.47 0.26 0.04 T-3996 18 1410 Ol 40.52 9.84 48.49 0.23 Opx 53.59 6.08 5.83 30.74 2.50 1.07 Cpx 51.88 7.06 4.51 20.92 14.13 0.20 1.30 L 47.18 0.54 14.64 8.73 0.16 14.99 12.51 0.85 0.13 0.29 T-3997 22 1480 Ol 40.81 8.61 49.59 0.24 Opx 54.25 5.06 5.19 31.71 2.51 1.17 Cpx 53.50 5.06 4.56 24.64 10.72 0.15 1.22 L 47.31 0.41 12.36 8.85 0.18 17.26 12.29 0.79 0.08 0.48 T-3998 25 1530 Ol 41.37 7.42 51.07 0.15 Opx 56.06 3.22 4.39 33.77 1.71 0.85 L 48.14 0.33 10.81 8.57 0.15 20.36 10.36 0.63 0.06 0.59 Average standard deviations Olivine 0.16(9) 0.12(7) 0.06(8) 0.14(9) 0.03(2) 0.03(2) Pyroxenes 0.3(2) 0.06(9) 0.3(2) 0.15(9) 0.04(4) 0.4(2) 0.3(3) 0.03(3) 0.3(3) Spinel 0.8(6) 0.16(8) 0.3(1) 0.05(3) 0.9(6) Plagioclase 0.3(2) 0.2(1) 0.06(5) 0.1(2) 0.2(2) 0.11(6) Glass 0.15(9) 0.03(2) 0.07(5) 0.04(2) 0.04(2) 0.10(7) 0.08(2) 0.03(3) 0.01(1) 0.03(2) Run no. . Pressure (kbar) . Temperature (°C) . Phase . sio2 . Tio2 . Al2o3 . FeO . MnO . MgO . CaO . Na2O . K2O . Cr2O3 . 123 95–1 crystallization experiments T-39 5 1240 Ol 40.39 11.62 47.08 0.57 L 47.47 0.40 16.58 8.75 0.10 10.05 15.01 1.43 0.09 0.12 T-91 5 1240 Ol 40.10 11.70 47.20 0.56 0.16 Sp 35.80 13.90 18.50 0.34 31.50 L 48.33 0.42 16.74 8.11 0.09 10.74 14.13 1.31 0.05 0.09 T-42 5 1220 Ol 40.00 11.80 47.20 0.62 Plg 47.50 33.60 0.26 0.28 16.80 1.56 L 48.67 0.47 16.35 8.04 0.11 9.23 15.36 1.57 0.09 0.12 T-43 5 1210 Ol 40.00 14.00 44.90 0.67 Plg 48.10 33.00 17.30 1.59 T-40 5 1200 Ol 39.70 0.66 15.20 43.60 0.88 Plg 48.10 31.30 0.78 0.83 17.40 1.63 L 48.83 0.54 15.21 9.57 0.14 8.34 15.54 1.58 0.10 0.15 T-58 10 1290 Ol 40.40 11.50 47.80 0.41 L 47.98 0.43 16.93 8.46 0.11 9.64 14.78 1.44 0.08 0.14 T-54 10 1280 Ol 40.30 11.52 47.54 0.48 L 47.74 0.43 16.70 8.80 0.13 9.97 14.58 1.44 0.08 0.11 T-92 10 1280 Ol 40.20 11.70 47.70 0.45 Sp 41.10 12.80 18.80 0.36 26.90 L 48.32 0.41 16.67 8.06 0.11 10.82 14.14 1.33 0.04 0.11 T-55 10 1270 Ol 40.20 11.80 47.60 0.36 L 47.83 0.42 16.60 8.74 0.12 10.24 14.41 1.42 0.07 0.15 T-45 10 1260 Ol 40.00 12.20 47.30 0.55 Cpx A 52.43 0.18 4.82 5.42 20.86 15.56 0.16 0.37 Cpx B 49.2 0.4 9.98 3.99 16.11 19.59 0.29 0.37 L 48.08 0.41 17.34 8.59 0.13 9.78 13.98 1.55 0.08 0.07 T-48 10 1240 Ol 38.60 0.96 18.90 40.90 0.63 Plg 48.90 32.00 0.35 0.53 16.10 2.14 Cpx 50.21 0.47 8.92 6.55 15.43 17.46 0.55 0.28 T-80 12 1290 Ol 40.10 12.00 47.30 0.53 0.16 Cpx A 52.25 0.19 5.25 5.29 19.55 17.12 0.25 0.17 Cpx B 49.48 0.33 9.39 4.03 15.96 19.98 0.30 0.36 L 47.67 0.42 16.83 9.19 0.09 10.27 13.85 1.49 0.07 0.11 T-66 15 1320 Cpx A 52.03 0.25 6.07 4.72 20.05 16.21 0.34 0.26 Cpx B 49.24 0.33 10.34 4.32 16.77 18.17 0.42 0.30 L 47.05 0.42 18.01 8.55 0.12 10.48 13.85 1.36 0.08 0.09 T-62 15 1300 Cpx 48.69 0.28 12.04 5.03 15.18 18.10 0.39 Olivine addition crystallization experiments Series 1 (11 wt % olivine) AT-13 0 1340 Ol 40.69 11.45 0.11 48.11 0.49 L 46.92 0.38 14.21 10.35 0.06 14.06 12.80 0.97 0.05 0.20 AT-11 0 1300 Ol 40.24 12.19 0.11 47.38 0.52 L 47.22 0.39 15.06 10.38 0.07 12.14 13.48 1.09 0.06 0.12 AT-15 0 1230 Ol 39.42 15.12 44.87 0.59 Plg 45.60 34.00 0.35 0.57 18.40 1.02 L 48.32 0.44 15.83 10.27 0.03 9.09 14.64 1.18 0.06 0.14 AT-14 0 1210 Ol 39.90 15.00 44.60 0.52 Plg 46.10 33.30 0.51 0.65 18.30 1.11 L 49.18 0.48 14.60 10.52 0.10 8.42 15.25 1.24 0.06 0.15 AT-10 0 1200 Ol 39.09 17.24 43.14 0.54 Plg 45.30 34.70 0.84 1.67 16.20 1.19 AT-12 0 1190 Ol 38.3 22.96 0.07 37.99 0.68 Plg 45.90 31.80 1.41 1.25 17.10 1.43 Cpx 53.10 4.33 6.32 16.80 19.50 AT-9 0 1160 Ol 37.63 26.79 35.15 0.44 Plg 46.94 34.52 1.01 0.25 15.06 2.21 Cpx 52.53 0.39 2.67 9.68 15.62 18.77 0.22 T-189 2 1280 Ol 40.84 8.69 49.61 0.37 Sp 0.53 0.28 29.31 13.41 17.66 0.28 38.53 L 48.24 0.36 14.77 8.64 0.11 13.66 13.11 0.92 0.06 0.14 T-190 2 1240 Ol 40.53 9.40 49.13 0.30 Sp 0.54 0.29 29.44 12.53 16.80 0.28 40.12 L 49.13 0.41 15.94 7.51 0.08 11.68 14.11 0.98 0.05 0.11 T-376 2 1200 Ol 39.20 16.30 44.10 0.43 Plg 47.20 32.70 0.68 18.30 1.08 L 48.50 0.47 15.00 11.18 0.10 8.25 15.04 1.23 0.07 0.17 T-200 2 1180 Ol 39.19 14.38 0.30 45.62 0.50 T-209 2 1160 Cpx 51.47 5.74 5.51 16.50 20.27 T-179 5 1340 Ol 40.80 9.42 49.60 0.27 L 47.51 0.38 13.96 9.43 0.16 15.20 12.30 0.83 0.04 0.19 T-224 5 1330 Ol 41.08 8.03 50.17 0.38 L 48.30 0.37 14.21 7.85 0.10 15.42 12.60 0.91 0.04 0.20 T-193 5 1300 Ol 40.21 10.35 0.18 48.58 0.43 Sp 30.20 14.50 19.40 0.49 35.40 L 48.62 0.40 15.06 8.28 0.10 12.80 13.58 0.95 0.05 0.16 T-195 5 1260 Ol 40.60 9.84 49.20 0.31 Sp 32.00 15.00 18.60 0.59 33.80 L 49.12 0.40 15.76 7.64 0.09 11.45 14.33 1.04 0.06 0.11 T-375 5 1230 Ol 39.70 13.80 46.10 0.43 L 47.76 0.39 15.85 10.45 0.05 10.01 14.22 1.07 0.06 0.14 T-196 5 1220 Ol 40.33 10.47 48.20 0.51 L 49.50 0.40 16.26 8.02 0.13 9.88 14.60 1.08 0.05 0.08 T-219 5 1210 Ol 39.92 13.66 45.72 0.46 Plag 47.30 32.70 0.63 0.89 17.20 1.58 L 49.07 0.47 15.91 8.92 0.11 8.89 15.38 1.04 0.05 0.16 T-197 5 1200 Ol 39.60 14.90 45.20 0.41 Plag 48.30 30.90 0.80 0.91 17.70 1.37 Cpx 52.40 4.08 6.59 19.20 17.40 0.94 L 50.24 0.56 15.23 9.23 0.17 9.43 13.86 1.14 0.08 0.06 T-176 10 1360 Ol 40.80 9.96 49.20 L 47.69 0.39 14.76 9.32 0.14 14.54 12.10 0.85 0.06 0.16 T-205 10 1260 Ol 40.00 12.70 46.90 0.36 Cpx 51.60 7.03 4.50 18.60 17.80 0.63 L 48.58 0.40 16.19 8.99 0.11 9.94 14.54 1.03 0.05 0.18 T-182 15 1380 Ol 40.60 10.70 48.60 0.17 L 47.82 0.38 14.59 9.44 0.12 14.02 12.56 0.84 0.05 0.18 T-171 15 1360 Ol 40.40 11.10 47.90 0.48 L 47.90 0.36 14.94 9.43 0.10 13.14 12.99 0.92 0.05 0.17 T-220 15 1320 Ol 40.25 12.21 47.02 0.38 Cpx 50.81 0.2 7.83 5.4 19.92 15.21 0.20 0.33 T-204 18 1400 Ol 40.10 11.50 48.20 0.29 Cpx 53.10 7.10 5.48 23.30 12.20 0.46 L 46.94 0.41 14.96 10.19 0.11 13.84 12.38 0.97 0.06 0.14 T-213 20 1440 Cpx 51.90 7.51 5.18 23.10 11.30 0.31 L 47.30 0.40 14.35 9.99 0.10 14.53 12.22 0.91 0.05 0.14 Series 2 (13 wt % olivine) T-352 18 1420 Ol 41.1 6.84 51.65 0.37 T-343 18 1400 Ol 40.41 11.20 0.15 47.59 0.38 Cpx 50.31 0.18 7.83 5.04 21.08 14.68 0.33 0.45 L 46.83 0.47 15.67 9.90 0.11 12.84 12.84 1.19 0.06 0.09 T-353 20 1460 Ol 40.79 8.44 50.25 0.4 L 48.15 0.35 13.96 8.01 0.13 15.78 12.42 0.98 0.05 0.17 T-364 20 1450 Ol 40.64 8.27 0.2 50.51 0.33 L 48.60 0.39 13.34 8.38 0.10 16.30 11.79 0.87 0.05 0.19 T-344 20 1440 Cpx 51.16 0.17 8.66 5.01 20.80 13.62 0.15 0.29 T-356 22 1460 Cpx 50.77 0.16 9.47 5.12 20.36 13.17 0.50 0.31 L 46.65 0.40 15.07 10.00 0.12 14.33 12.11 1.13 0.06 0.13 Series 3 (17 wt % olivine) T-537 25 1520 Ol 41.72 8.74 50.04 0.27 L 47.37 0.33 13.35 9.03 0.12 16.85 11.84 0.90 0.06 0.15 T-543 25 1515 Ol 41.28 9.66 48.71 0.35 Cpx 53.38 6.75 4.71 24.26 10.13 0.35 0.43 L 46.52 0.40 14.11 9.89 0.13 15.89 11.85 1.02 0.06 0.13 T-542 25 1510 Ol 40.70 10.17 48.76 0.37 Cpx 52.01 0.11 7.56 5.18 23.20 11.24 0.30 0.37 T-540 25 1500 Ol 40.60 10.01 48.77 0.44 Cpx 51.63 0.10 7.43 5.37 23.68 11.07 0.34 0.31 T-530 25 1500 Ol 39.98 11.26 47.98 0.51 Cpx 50.82 0.12 8.62 5.66 21.59 12.39 0.46 0.34 T-2662 26 1530 Cpx 50.26 0.13 10.27 6.71 18.62 13.42 0.73 Ga 42.89 0.16 22.57 7.96 19.04 7.29 0.15 T-2611 27 1540 Cpx 53.67 5.96 4.76 25.07 9.87 0.32 0.35 L 47.39 0.36 13.41 9.13 0.09 16.59 11.89 0.93 0.06 0.16 T-2617 27 1500 Ol 39.34 0.26 14.91 44.14 0.44 Cpx 50.76 0.22 9.51 6.01 18.20 14.77 0.53 Ga 42.14 0.19 23.60 8.25 19.35 6.21 0.27 Olivine and orthopyroxene addition crystallization experiments Series 1 (1 wt % orthopyroxene and 13 wt % olivine) T-361 20 1450 Ol 39.85 10.27 49.41 0.38 Opx 53.62 0.11 6.03 6.07 30.83 2.70 0.19 Cpx 52.44 0.11 5.68 5.82 26.47 9.14 0.22 L 48.02 0.35 14.41 9.16 0.07 14.59 12.27 0.95 0.06 0.11 T-377 22 1460 Cpx 50.49 0.17 10.09 5.22 22.27 11.03 0.33 0.29 L 46.04 0.48 15.81 10.57 0.15 13.69 11.90 1.17 0.10 0.10 Series 2 (2 wt % orthopyroxene and 13 wt % olivine) T-524 21 1480 Ol 41.43 8.96 49.39 0.23 L 48.00 0.37 13.53 9.05 0.09 16.07 11.72 0.91 0.04 0.20 T-517 21 1460 Ol 40.92 0.23 9.78 49.16 0.45 Cpx 52.53 0.07 6.21 5.39 24.14 10.94 0.27 0.37 L 47.54 0.41 14.17 9.34 0.11 15.07 12.16 1.01 0.06 0.13 T-518 21 1460 Ol 40.51 9.27 49.53 0.38 Opx 55.67 0.05 3.03 5.66 32.46 2.71 0.10 0.23 Cpx 54.28 0.06 3.85 5.56 28.49 7.05 0.17 0.33 L 48.0 0.37 13.62 9.21 0.10 15.43 12.06 1.03 0.05 0.15 T-528 23 1490 Cpx 52.20 0.08 6.03 5.45 25.06 10.33 0.37 0.36 L 47.40 0.40 13.85 9.57 0.10 15.57 11.89 1.00 0.06 0.15 T-529 23 1490 Cpx 52.99 0.06 5.98 5.24 26.35 8.65 0.23 0.40 L 47.49 0.38 13.90 9.20 0.12 15.84 11.86 0.98 0.07 0.16 T-523 23 1480 Cpx 52.67 0.07 6.62 4.92 24.97 10.03 0.27 0.33 L 47.15 0.38 14.38 9.45 0.12 15.51 11.81 1.03 0.06 0.11 Peridotite reaction experiments Series A T-4151 21 1460 Ol 41.01 9.23 0.17 48.65 0.35 Opx 54.63 0.05 4.93 5.45 0.11 31.03 2.70 0.10 1.01 Cpx 53.04 0.03 5.65 4.81 0.17 24.11 10.71 0.32 1.16 L 47.18 0.55 13.84 9.46 0.16 15.15 12.08 1.02 0.15 0.41 T-4152 21 1460 Ol 41.48 8.39 0.15 49.59 0.37 Opx 54.42 0.06 5.26 4.98 0.10 31.22 2.71 0.10 1.15 Cpx 52.83 0.06 5.42 4.34 0.14 23.59 12.09 0.35 1.19 L 47.24 0.54 13.81 8.41 0.14 16.20 12.23 0.94 0.06 0.43 Series B T-4013 13 1325 Ol 40.10 12.67 46.71 0.17 Opx 52.82 7.62 7.05 29.52 2.43 0.56 Cpx 49.95 0.28 9.30 4.50 17.34 17.86 0.17 0.43 Sp 0.70 59.05 10.05 19.59 0.08 10.53 L 48.45 0.61 18.56 8.69 0.15 9.26 11.50 2.47 0.26 0.04 T-3996 18 1410 Ol 40.52 9.84 48.49 0.23 Opx 53.59 6.08 5.83 30.74 2.50 1.07 Cpx 51.88 7.06 4.51 20.92 14.13 0.20 1.30 L 47.18 0.54 14.64 8.73 0.16 14.99 12.51 0.85 0.13 0.29 T-3997 22 1480 Ol 40.81 8.61 49.59 0.24 Opx 54.25 5.06 5.19 31.71 2.51 1.17 Cpx 53.50 5.06 4.56 24.64 10.72 0.15 1.22 L 47.31 0.41 12.36 8.85 0.18 17.26 12.29 0.79 0.08 0.48 T-3998 25 1530 Ol 41.37 7.42 51.07 0.15 Opx 56.06 3.22 4.39 33.77 1.71 0.85 L 48.14 0.33 10.81 8.57 0.15 20.36 10.36 0.63 0.06 0.59 Average standard deviations Olivine 0.16(9) 0.12(7) 0.06(8) 0.14(9) 0.03(2) 0.03(2) Pyroxenes 0.3(2) 0.06(9) 0.3(2) 0.15(9) 0.04(4) 0.4(2) 0.3(3) 0.03(3) 0.3(3) Spinel 0.8(6) 0.16(8) 0.3(1) 0.05(3) 0.9(6) Plagioclase 0.3(2) 0.2(1) 0.06(5) 0.1(2) 0.2(2) 0.11(6) Glass 0.15(9) 0.03(2) 0.07(5) 0.04(2) 0.04(2) 0.10(7) 0.08(2) 0.03(3) 0.01(1) 0.03(2) Ol, olivine; Sp, spinel; Plg, plagioclase; Cpx, clinopyroxene (A, B refer to coexisting compositions in sector zoned clinopyroxene); Opx, orthopyroxene; L, glass. All analyses have been normalized to 100 wt %, average standard deviations are in wt % and numbers in parentheses are 1σ in terms of the last unit cited; e.g. 0.15(9) represents 0.15 ± 0.09. Open in new tab Table 4: Experimental run products Run no. . Pressure (kbar) . Temperature (°C) . Phase . sio2 . Tio2 . Al2o3 . FeO . MnO . MgO . CaO . Na2O . K2O . Cr2O3 . 123 95–1 crystallization experiments T-39 5 1240 Ol 40.39 11.62 47.08 0.57 L 47.47 0.40 16.58 8.75 0.10 10.05 15.01 1.43 0.09 0.12 T-91 5 1240 Ol 40.10 11.70 47.20 0.56 0.16 Sp 35.80 13.90 18.50 0.34 31.50 L 48.33 0.42 16.74 8.11 0.09 10.74 14.13 1.31 0.05 0.09 T-42 5 1220 Ol 40.00 11.80 47.20 0.62 Plg 47.50 33.60 0.26 0.28 16.80 1.56 L 48.67 0.47 16.35 8.04 0.11 9.23 15.36 1.57 0.09 0.12 T-43 5 1210 Ol 40.00 14.00 44.90 0.67 Plg 48.10 33.00 17.30 1.59 T-40 5 1200 Ol 39.70 0.66 15.20 43.60 0.88 Plg 48.10 31.30 0.78 0.83 17.40 1.63 L 48.83 0.54 15.21 9.57 0.14 8.34 15.54 1.58 0.10 0.15 T-58 10 1290 Ol 40.40 11.50 47.80 0.41 L 47.98 0.43 16.93 8.46 0.11 9.64 14.78 1.44 0.08 0.14 T-54 10 1280 Ol 40.30 11.52 47.54 0.48 L 47.74 0.43 16.70 8.80 0.13 9.97 14.58 1.44 0.08 0.11 T-92 10 1280 Ol 40.20 11.70 47.70 0.45 Sp 41.10 12.80 18.80 0.36 26.90 L 48.32 0.41 16.67 8.06 0.11 10.82 14.14 1.33 0.04 0.11 T-55 10 1270 Ol 40.20 11.80 47.60 0.36 L 47.83 0.42 16.60 8.74 0.12 10.24 14.41 1.42 0.07 0.15 T-45 10 1260 Ol 40.00 12.20 47.30 0.55 Cpx A 52.43 0.18 4.82 5.42 20.86 15.56 0.16 0.37 Cpx B 49.2 0.4 9.98 3.99 16.11 19.59 0.29 0.37 L 48.08 0.41 17.34 8.59 0.13 9.78 13.98 1.55 0.08 0.07 T-48 10 1240 Ol 38.60 0.96 18.90 40.90 0.63 Plg 48.90 32.00 0.35 0.53 16.10 2.14 Cpx 50.21 0.47 8.92 6.55 15.43 17.46 0.55 0.28 T-80 12 1290 Ol 40.10 12.00 47.30 0.53 0.16 Cpx A 52.25 0.19 5.25 5.29 19.55 17.12 0.25 0.17 Cpx B 49.48 0.33 9.39 4.03 15.96 19.98 0.30 0.36 L 47.67 0.42 16.83 9.19 0.09 10.27 13.85 1.49 0.07 0.11 T-66 15 1320 Cpx A 52.03 0.25 6.07 4.72 20.05 16.21 0.34 0.26 Cpx B 49.24 0.33 10.34 4.32 16.77 18.17 0.42 0.30 L 47.05 0.42 18.01 8.55 0.12 10.48 13.85 1.36 0.08 0.09 T-62 15 1300 Cpx 48.69 0.28 12.04 5.03 15.18 18.10 0.39 Olivine addition crystallization experiments Series 1 (11 wt % olivine) AT-13 0 1340 Ol 40.69 11.45 0.11 48.11 0.49 L 46.92 0.38 14.21 10.35 0.06 14.06 12.80 0.97 0.05 0.20 AT-11 0 1300 Ol 40.24 12.19 0.11 47.38 0.52 L 47.22 0.39 15.06 10.38 0.07 12.14 13.48 1.09 0.06 0.12 AT-15 0 1230 Ol 39.42 15.12 44.87 0.59 Plg 45.60 34.00 0.35 0.57 18.40 1.02 L 48.32 0.44 15.83 10.27 0.03 9.09 14.64 1.18 0.06 0.14 AT-14 0 1210 Ol 39.90 15.00 44.60 0.52 Plg 46.10 33.30 0.51 0.65 18.30 1.11 L 49.18 0.48 14.60 10.52 0.10 8.42 15.25 1.24 0.06 0.15 AT-10 0 1200 Ol 39.09 17.24 43.14 0.54 Plg 45.30 34.70 0.84 1.67 16.20 1.19 AT-12 0 1190 Ol 38.3 22.96 0.07 37.99 0.68 Plg 45.90 31.80 1.41 1.25 17.10 1.43 Cpx 53.10 4.33 6.32 16.80 19.50 AT-9 0 1160 Ol 37.63 26.79 35.15 0.44 Plg 46.94 34.52 1.01 0.25 15.06 2.21 Cpx 52.53 0.39 2.67 9.68 15.62 18.77 0.22 T-189 2 1280 Ol 40.84 8.69 49.61 0.37 Sp 0.53 0.28 29.31 13.41 17.66 0.28 38.53 L 48.24 0.36 14.77 8.64 0.11 13.66 13.11 0.92 0.06 0.14 T-190 2 1240 Ol 40.53 9.40 49.13 0.30 Sp 0.54 0.29 29.44 12.53 16.80 0.28 40.12 L 49.13 0.41 15.94 7.51 0.08 11.68 14.11 0.98 0.05 0.11 T-376 2 1200 Ol 39.20 16.30 44.10 0.43 Plg 47.20 32.70 0.68 18.30 1.08 L 48.50 0.47 15.00 11.18 0.10 8.25 15.04 1.23 0.07 0.17 T-200 2 1180 Ol 39.19 14.38 0.30 45.62 0.50 T-209 2 1160 Cpx 51.47 5.74 5.51 16.50 20.27 T-179 5 1340 Ol 40.80 9.42 49.60 0.27 L 47.51 0.38 13.96 9.43 0.16 15.20 12.30 0.83 0.04 0.19 T-224 5 1330 Ol 41.08 8.03 50.17 0.38 L 48.30 0.37 14.21 7.85 0.10 15.42 12.60 0.91 0.04 0.20 T-193 5 1300 Ol 40.21 10.35 0.18 48.58 0.43 Sp 30.20 14.50 19.40 0.49 35.40 L 48.62 0.40 15.06 8.28 0.10 12.80 13.58 0.95 0.05 0.16 T-195 5 1260 Ol 40.60 9.84 49.20 0.31 Sp 32.00 15.00 18.60 0.59 33.80 L 49.12 0.40 15.76 7.64 0.09 11.45 14.33 1.04 0.06 0.11 T-375 5 1230 Ol 39.70 13.80 46.10 0.43 L 47.76 0.39 15.85 10.45 0.05 10.01 14.22 1.07 0.06 0.14 T-196 5 1220 Ol 40.33 10.47 48.20 0.51 L 49.50 0.40 16.26 8.02 0.13 9.88 14.60 1.08 0.05 0.08 T-219 5 1210 Ol 39.92 13.66 45.72 0.46 Plag 47.30 32.70 0.63 0.89 17.20 1.58 L 49.07 0.47 15.91 8.92 0.11 8.89 15.38 1.04 0.05 0.16 T-197 5 1200 Ol 39.60 14.90 45.20 0.41 Plag 48.30 30.90 0.80 0.91 17.70 1.37 Cpx 52.40 4.08 6.59 19.20 17.40 0.94 L 50.24 0.56 15.23 9.23 0.17 9.43 13.86 1.14 0.08 0.06 T-176 10 1360 Ol 40.80 9.96 49.20 L 47.69 0.39 14.76 9.32 0.14 14.54 12.10 0.85 0.06 0.16 T-205 10 1260 Ol 40.00 12.70 46.90 0.36 Cpx 51.60 7.03 4.50 18.60 17.80 0.63 L 48.58 0.40 16.19 8.99 0.11 9.94 14.54 1.03 0.05 0.18 T-182 15 1380 Ol 40.60 10.70 48.60 0.17 L 47.82 0.38 14.59 9.44 0.12 14.02 12.56 0.84 0.05 0.18 T-171 15 1360 Ol 40.40 11.10 47.90 0.48 L 47.90 0.36 14.94 9.43 0.10 13.14 12.99 0.92 0.05 0.17 T-220 15 1320 Ol 40.25 12.21 47.02 0.38 Cpx 50.81 0.2 7.83 5.4 19.92 15.21 0.20 0.33 T-204 18 1400 Ol 40.10 11.50 48.20 0.29 Cpx 53.10 7.10 5.48 23.30 12.20 0.46 L 46.94 0.41 14.96 10.19 0.11 13.84 12.38 0.97 0.06 0.14 T-213 20 1440 Cpx 51.90 7.51 5.18 23.10 11.30 0.31 L 47.30 0.40 14.35 9.99 0.10 14.53 12.22 0.91 0.05 0.14 Series 2 (13 wt % olivine) T-352 18 1420 Ol 41.1 6.84 51.65 0.37 T-343 18 1400 Ol 40.41 11.20 0.15 47.59 0.38 Cpx 50.31 0.18 7.83 5.04 21.08 14.68 0.33 0.45 L 46.83 0.47 15.67 9.90 0.11 12.84 12.84 1.19 0.06 0.09 T-353 20 1460 Ol 40.79 8.44 50.25 0.4 L 48.15 0.35 13.96 8.01 0.13 15.78 12.42 0.98 0.05 0.17 T-364 20 1450 Ol 40.64 8.27 0.2 50.51 0.33 L 48.60 0.39 13.34 8.38 0.10 16.30 11.79 0.87 0.05 0.19 T-344 20 1440 Cpx 51.16 0.17 8.66 5.01 20.80 13.62 0.15 0.29 T-356 22 1460 Cpx 50.77 0.16 9.47 5.12 20.36 13.17 0.50 0.31 L 46.65 0.40 15.07 10.00 0.12 14.33 12.11 1.13 0.06 0.13 Series 3 (17 wt % olivine) T-537 25 1520 Ol 41.72 8.74 50.04 0.27 L 47.37 0.33 13.35 9.03 0.12 16.85 11.84 0.90 0.06 0.15 T-543 25 1515 Ol 41.28 9.66 48.71 0.35 Cpx 53.38 6.75 4.71 24.26 10.13 0.35 0.43 L 46.52 0.40 14.11 9.89 0.13 15.89 11.85 1.02 0.06 0.13 T-542 25 1510 Ol 40.70 10.17 48.76 0.37 Cpx 52.01 0.11 7.56 5.18 23.20 11.24 0.30 0.37 T-540 25 1500 Ol 40.60 10.01 48.77 0.44 Cpx 51.63 0.10 7.43 5.37 23.68 11.07 0.34 0.31 T-530 25 1500 Ol 39.98 11.26 47.98 0.51 Cpx 50.82 0.12 8.62 5.66 21.59 12.39 0.46 0.34 T-2662 26 1530 Cpx 50.26 0.13 10.27 6.71 18.62 13.42 0.73 Ga 42.89 0.16 22.57 7.96 19.04 7.29 0.15 T-2611 27 1540 Cpx 53.67 5.96 4.76 25.07 9.87 0.32 0.35 L 47.39 0.36 13.41 9.13 0.09 16.59 11.89 0.93 0.06 0.16 T-2617 27 1500 Ol 39.34 0.26 14.91 44.14 0.44 Cpx 50.76 0.22 9.51 6.01 18.20 14.77 0.53 Ga 42.14 0.19 23.60 8.25 19.35 6.21 0.27 Olivine and orthopyroxene addition crystallization experiments Series 1 (1 wt % orthopyroxene and 13 wt % olivine) T-361 20 1450 Ol 39.85 10.27 49.41 0.38 Opx 53.62 0.11 6.03 6.07 30.83 2.70 0.19 Cpx 52.44 0.11 5.68 5.82 26.47 9.14 0.22 L 48.02 0.35 14.41 9.16 0.07 14.59 12.27 0.95 0.06 0.11 T-377 22 1460 Cpx 50.49 0.17 10.09 5.22 22.27 11.03 0.33 0.29 L 46.04 0.48 15.81 10.57 0.15 13.69 11.90 1.17 0.10 0.10 Series 2 (2 wt % orthopyroxene and 13 wt % olivine) T-524 21 1480 Ol 41.43 8.96 49.39 0.23 L 48.00 0.37 13.53 9.05 0.09 16.07 11.72 0.91 0.04 0.20 T-517 21 1460 Ol 40.92 0.23 9.78 49.16 0.45 Cpx 52.53 0.07 6.21 5.39 24.14 10.94 0.27 0.37 L 47.54 0.41 14.17 9.34 0.11 15.07 12.16 1.01 0.06 0.13 T-518 21 1460 Ol 40.51 9.27 49.53 0.38 Opx 55.67 0.05 3.03 5.66 32.46 2.71 0.10 0.23 Cpx 54.28 0.06 3.85 5.56 28.49 7.05 0.17 0.33 L 48.0 0.37 13.62 9.21 0.10 15.43 12.06 1.03 0.05 0.15 T-528 23 1490 Cpx 52.20 0.08 6.03 5.45 25.06 10.33 0.37 0.36 L 47.40 0.40 13.85 9.57 0.10 15.57 11.89 1.00 0.06 0.15 T-529 23 1490 Cpx 52.99 0.06 5.98 5.24 26.35 8.65 0.23 0.40 L 47.49 0.38 13.90 9.20 0.12 15.84 11.86 0.98 0.07 0.16 T-523 23 1480 Cpx 52.67 0.07 6.62 4.92 24.97 10.03 0.27 0.33 L 47.15 0.38 14.38 9.45 0.12 15.51 11.81 1.03 0.06 0.11 Peridotite reaction experiments Series A T-4151 21 1460 Ol 41.01 9.23 0.17 48.65 0.35 Opx 54.63 0.05 4.93 5.45 0.11 31.03 2.70 0.10 1.01 Cpx 53.04 0.03 5.65 4.81 0.17 24.11 10.71 0.32 1.16 L 47.18 0.55 13.84 9.46 0.16 15.15 12.08 1.02 0.15 0.41 T-4152 21 1460 Ol 41.48 8.39 0.15 49.59 0.37 Opx 54.42 0.06 5.26 4.98 0.10 31.22 2.71 0.10 1.15 Cpx 52.83 0.06 5.42 4.34 0.14 23.59 12.09 0.35 1.19 L 47.24 0.54 13.81 8.41 0.14 16.20 12.23 0.94 0.06 0.43 Series B T-4013 13 1325 Ol 40.10 12.67 46.71 0.17 Opx 52.82 7.62 7.05 29.52 2.43 0.56 Cpx 49.95 0.28 9.30 4.50 17.34 17.86 0.17 0.43 Sp 0.70 59.05 10.05 19.59 0.08 10.53 L 48.45 0.61 18.56 8.69 0.15 9.26 11.50 2.47 0.26 0.04 T-3996 18 1410 Ol 40.52 9.84 48.49 0.23 Opx 53.59 6.08 5.83 30.74 2.50 1.07 Cpx 51.88 7.06 4.51 20.92 14.13 0.20 1.30 L 47.18 0.54 14.64 8.73 0.16 14.99 12.51 0.85 0.13 0.29 T-3997 22 1480 Ol 40.81 8.61 49.59 0.24 Opx 54.25 5.06 5.19 31.71 2.51 1.17 Cpx 53.50 5.06 4.56 24.64 10.72 0.15 1.22 L 47.31 0.41 12.36 8.85 0.18 17.26 12.29 0.79 0.08 0.48 T-3998 25 1530 Ol 41.37 7.42 51.07 0.15 Opx 56.06 3.22 4.39 33.77 1.71 0.85 L 48.14 0.33 10.81 8.57 0.15 20.36 10.36 0.63 0.06 0.59 Average standard deviations Olivine 0.16(9) 0.12(7) 0.06(8) 0.14(9) 0.03(2) 0.03(2) Pyroxenes 0.3(2) 0.06(9) 0.3(2) 0.15(9) 0.04(4) 0.4(2) 0.3(3) 0.03(3) 0.3(3) Spinel 0.8(6) 0.16(8) 0.3(1) 0.05(3) 0.9(6) Plagioclase 0.3(2) 0.2(1) 0.06(5) 0.1(2) 0.2(2) 0.11(6) Glass 0.15(9) 0.03(2) 0.07(5) 0.04(2) 0.04(2) 0.10(7) 0.08(2) 0.03(3) 0.01(1) 0.03(2) Run no. . Pressure (kbar) . Temperature (°C) . Phase . sio2 . Tio2 . Al2o3 . FeO . MnO . MgO . CaO . Na2O . K2O . Cr2O3 . 123 95–1 crystallization experiments T-39 5 1240 Ol 40.39 11.62 47.08 0.57 L 47.47 0.40 16.58 8.75 0.10 10.05 15.01 1.43 0.09 0.12 T-91 5 1240 Ol 40.10 11.70 47.20 0.56 0.16 Sp 35.80 13.90 18.50 0.34 31.50 L 48.33 0.42 16.74 8.11 0.09 10.74 14.13 1.31 0.05 0.09 T-42 5 1220 Ol 40.00 11.80 47.20 0.62 Plg 47.50 33.60 0.26 0.28 16.80 1.56 L 48.67 0.47 16.35 8.04 0.11 9.23 15.36 1.57 0.09 0.12 T-43 5 1210 Ol 40.00 14.00 44.90 0.67 Plg 48.10 33.00 17.30 1.59 T-40 5 1200 Ol 39.70 0.66 15.20 43.60 0.88 Plg 48.10 31.30 0.78 0.83 17.40 1.63 L 48.83 0.54 15.21 9.57 0.14 8.34 15.54 1.58 0.10 0.15 T-58 10 1290 Ol 40.40 11.50 47.80 0.41 L 47.98 0.43 16.93 8.46 0.11 9.64 14.78 1.44 0.08 0.14 T-54 10 1280 Ol 40.30 11.52 47.54 0.48 L 47.74 0.43 16.70 8.80 0.13 9.97 14.58 1.44 0.08 0.11 T-92 10 1280 Ol 40.20 11.70 47.70 0.45 Sp 41.10 12.80 18.80 0.36 26.90 L 48.32 0.41 16.67 8.06 0.11 10.82 14.14 1.33 0.04 0.11 T-55 10 1270 Ol 40.20 11.80 47.60 0.36 L 47.83 0.42 16.60 8.74 0.12 10.24 14.41 1.42 0.07 0.15 T-45 10 1260 Ol 40.00 12.20 47.30 0.55 Cpx A 52.43 0.18 4.82 5.42 20.86 15.56 0.16 0.37 Cpx B 49.2 0.4 9.98 3.99 16.11 19.59 0.29 0.37 L 48.08 0.41 17.34 8.59 0.13 9.78 13.98 1.55 0.08 0.07 T-48 10 1240 Ol 38.60 0.96 18.90 40.90 0.63 Plg 48.90 32.00 0.35 0.53 16.10 2.14 Cpx 50.21 0.47 8.92 6.55 15.43 17.46 0.55 0.28 T-80 12 1290 Ol 40.10 12.00 47.30 0.53 0.16 Cpx A 52.25 0.19 5.25 5.29 19.55 17.12 0.25 0.17 Cpx B 49.48 0.33 9.39 4.03 15.96 19.98 0.30 0.36 L 47.67 0.42 16.83 9.19 0.09 10.27 13.85 1.49 0.07 0.11 T-66 15 1320 Cpx A 52.03 0.25 6.07 4.72 20.05 16.21 0.34 0.26 Cpx B 49.24 0.33 10.34 4.32 16.77 18.17 0.42 0.30 L 47.05 0.42 18.01 8.55 0.12 10.48 13.85 1.36 0.08 0.09 T-62 15 1300 Cpx 48.69 0.28 12.04 5.03 15.18 18.10 0.39 Olivine addition crystallization experiments Series 1 (11 wt % olivine) AT-13 0 1340 Ol 40.69 11.45 0.11 48.11 0.49 L 46.92 0.38 14.21 10.35 0.06 14.06 12.80 0.97 0.05 0.20 AT-11 0 1300 Ol 40.24 12.19 0.11 47.38 0.52 L 47.22 0.39 15.06 10.38 0.07 12.14 13.48 1.09 0.06 0.12 AT-15 0 1230 Ol 39.42 15.12 44.87 0.59 Plg 45.60 34.00 0.35 0.57 18.40 1.02 L 48.32 0.44 15.83 10.27 0.03 9.09 14.64 1.18 0.06 0.14 AT-14 0 1210 Ol 39.90 15.00 44.60 0.52 Plg 46.10 33.30 0.51 0.65 18.30 1.11 L 49.18 0.48 14.60 10.52 0.10 8.42 15.25 1.24 0.06 0.15 AT-10 0 1200 Ol 39.09 17.24 43.14 0.54 Plg 45.30 34.70 0.84 1.67 16.20 1.19 AT-12 0 1190 Ol 38.3 22.96 0.07 37.99 0.68 Plg 45.90 31.80 1.41 1.25 17.10 1.43 Cpx 53.10 4.33 6.32 16.80 19.50 AT-9 0 1160 Ol 37.63 26.79 35.15 0.44 Plg 46.94 34.52 1.01 0.25 15.06 2.21 Cpx 52.53 0.39 2.67 9.68 15.62 18.77 0.22 T-189 2 1280 Ol 40.84 8.69 49.61 0.37 Sp 0.53 0.28 29.31 13.41 17.66 0.28 38.53 L 48.24 0.36 14.77 8.64 0.11 13.66 13.11 0.92 0.06 0.14 T-190 2 1240 Ol 40.53 9.40 49.13 0.30 Sp 0.54 0.29 29.44 12.53 16.80 0.28 40.12 L 49.13 0.41 15.94 7.51 0.08 11.68 14.11 0.98 0.05 0.11 T-376 2 1200 Ol 39.20 16.30 44.10 0.43 Plg 47.20 32.70 0.68 18.30 1.08 L 48.50 0.47 15.00 11.18 0.10 8.25 15.04 1.23 0.07 0.17 T-200 2 1180 Ol 39.19 14.38 0.30 45.62 0.50 T-209 2 1160 Cpx 51.47 5.74 5.51 16.50 20.27 T-179 5 1340 Ol 40.80 9.42 49.60 0.27 L 47.51 0.38 13.96 9.43 0.16 15.20 12.30 0.83 0.04 0.19 T-224 5 1330 Ol 41.08 8.03 50.17 0.38 L 48.30 0.37 14.21 7.85 0.10 15.42 12.60 0.91 0.04 0.20 T-193 5 1300 Ol 40.21 10.35 0.18 48.58 0.43 Sp 30.20 14.50 19.40 0.49 35.40 L 48.62 0.40 15.06 8.28 0.10 12.80 13.58 0.95 0.05 0.16 T-195 5 1260 Ol 40.60 9.84 49.20 0.31 Sp 32.00 15.00 18.60 0.59 33.80 L 49.12 0.40 15.76 7.64 0.09 11.45 14.33 1.04 0.06 0.11 T-375 5 1230 Ol 39.70 13.80 46.10 0.43 L 47.76 0.39 15.85 10.45 0.05 10.01 14.22 1.07 0.06 0.14 T-196 5 1220 Ol 40.33 10.47 48.20 0.51 L 49.50 0.40 16.26 8.02 0.13 9.88 14.60 1.08 0.05 0.08 T-219 5 1210 Ol 39.92 13.66 45.72 0.46 Plag 47.30 32.70 0.63 0.89 17.20 1.58 L 49.07 0.47 15.91 8.92 0.11 8.89 15.38 1.04 0.05 0.16 T-197 5 1200 Ol 39.60 14.90 45.20 0.41 Plag 48.30 30.90 0.80 0.91 17.70 1.37 Cpx 52.40 4.08 6.59 19.20 17.40 0.94 L 50.24 0.56 15.23 9.23 0.17 9.43 13.86 1.14 0.08 0.06 T-176 10 1360 Ol 40.80 9.96 49.20 L 47.69 0.39 14.76 9.32 0.14 14.54 12.10 0.85 0.06 0.16 T-205 10 1260 Ol 40.00 12.70 46.90 0.36 Cpx 51.60 7.03 4.50 18.60 17.80 0.63 L 48.58 0.40 16.19 8.99 0.11 9.94 14.54 1.03 0.05 0.18 T-182 15 1380 Ol 40.60 10.70 48.60 0.17 L 47.82 0.38 14.59 9.44 0.12 14.02 12.56 0.84 0.05 0.18 T-171 15 1360 Ol 40.40 11.10 47.90 0.48 L 47.90 0.36 14.94 9.43 0.10 13.14 12.99 0.92 0.05 0.17 T-220 15 1320 Ol 40.25 12.21 47.02 0.38 Cpx 50.81 0.2 7.83 5.4 19.92 15.21 0.20 0.33 T-204 18 1400 Ol 40.10 11.50 48.20 0.29 Cpx 53.10 7.10 5.48 23.30 12.20 0.46 L 46.94 0.41 14.96 10.19 0.11 13.84 12.38 0.97 0.06 0.14 T-213 20 1440 Cpx 51.90 7.51 5.18 23.10 11.30 0.31 L 47.30 0.40 14.35 9.99 0.10 14.53 12.22 0.91 0.05 0.14 Series 2 (13 wt % olivine) T-352 18 1420 Ol 41.1 6.84 51.65 0.37 T-343 18 1400 Ol 40.41 11.20 0.15 47.59 0.38 Cpx 50.31 0.18 7.83 5.04 21.08 14.68 0.33 0.45 L 46.83 0.47 15.67 9.90 0.11 12.84 12.84 1.19 0.06 0.09 T-353 20 1460 Ol 40.79 8.44 50.25 0.4 L 48.15 0.35 13.96 8.01 0.13 15.78 12.42 0.98 0.05 0.17 T-364 20 1450 Ol 40.64 8.27 0.2 50.51 0.33 L 48.60 0.39 13.34 8.38 0.10 16.30 11.79 0.87 0.05 0.19 T-344 20 1440 Cpx 51.16 0.17 8.66 5.01 20.80 13.62 0.15 0.29 T-356 22 1460 Cpx 50.77 0.16 9.47 5.12 20.36 13.17 0.50 0.31 L 46.65 0.40 15.07 10.00 0.12 14.33 12.11 1.13 0.06 0.13 Series 3 (17 wt % olivine) T-537 25 1520 Ol 41.72 8.74 50.04 0.27 L 47.37 0.33 13.35 9.03 0.12 16.85 11.84 0.90 0.06 0.15 T-543 25 1515 Ol 41.28 9.66 48.71 0.35 Cpx 53.38 6.75 4.71 24.26 10.13 0.35 0.43 L 46.52 0.40 14.11 9.89 0.13 15.89 11.85 1.02 0.06 0.13 T-542 25 1510 Ol 40.70 10.17 48.76 0.37 Cpx 52.01 0.11 7.56 5.18 23.20 11.24 0.30 0.37 T-540 25 1500 Ol 40.60 10.01 48.77 0.44 Cpx 51.63 0.10 7.43 5.37 23.68 11.07 0.34 0.31 T-530 25 1500 Ol 39.98 11.26 47.98 0.51 Cpx 50.82 0.12 8.62 5.66 21.59 12.39 0.46 0.34 T-2662 26 1530 Cpx 50.26 0.13 10.27 6.71 18.62 13.42 0.73 Ga 42.89 0.16 22.57 7.96 19.04 7.29 0.15 T-2611 27 1540 Cpx 53.67 5.96 4.76 25.07 9.87 0.32 0.35 L 47.39 0.36 13.41 9.13 0.09 16.59 11.89 0.93 0.06 0.16 T-2617 27 1500 Ol 39.34 0.26 14.91 44.14 0.44 Cpx 50.76 0.22 9.51 6.01 18.20 14.77 0.53 Ga 42.14 0.19 23.60 8.25 19.35 6.21 0.27 Olivine and orthopyroxene addition crystallization experiments Series 1 (1 wt % orthopyroxene and 13 wt % olivine) T-361 20 1450 Ol 39.85 10.27 49.41 0.38 Opx 53.62 0.11 6.03 6.07 30.83 2.70 0.19 Cpx 52.44 0.11 5.68 5.82 26.47 9.14 0.22 L 48.02 0.35 14.41 9.16 0.07 14.59 12.27 0.95 0.06 0.11 T-377 22 1460 Cpx 50.49 0.17 10.09 5.22 22.27 11.03 0.33 0.29 L 46.04 0.48 15.81 10.57 0.15 13.69 11.90 1.17 0.10 0.10 Series 2 (2 wt % orthopyroxene and 13 wt % olivine) T-524 21 1480 Ol 41.43 8.96 49.39 0.23 L 48.00 0.37 13.53 9.05 0.09 16.07 11.72 0.91 0.04 0.20 T-517 21 1460 Ol 40.92 0.23 9.78 49.16 0.45 Cpx 52.53 0.07 6.21 5.39 24.14 10.94 0.27 0.37 L 47.54 0.41 14.17 9.34 0.11 15.07 12.16 1.01 0.06 0.13 T-518 21 1460 Ol 40.51 9.27 49.53 0.38 Opx 55.67 0.05 3.03 5.66 32.46 2.71 0.10 0.23 Cpx 54.28 0.06 3.85 5.56 28.49 7.05 0.17 0.33 L 48.0 0.37 13.62 9.21 0.10 15.43 12.06 1.03 0.05 0.15 T-528 23 1490 Cpx 52.20 0.08 6.03 5.45 25.06 10.33 0.37 0.36 L 47.40 0.40 13.85 9.57 0.10 15.57 11.89 1.00 0.06 0.15 T-529 23 1490 Cpx 52.99 0.06 5.98 5.24 26.35 8.65 0.23 0.40 L 47.49 0.38 13.90 9.20 0.12 15.84 11.86 0.98 0.07 0.16 T-523 23 1480 Cpx 52.67 0.07 6.62 4.92 24.97 10.03 0.27 0.33 L 47.15 0.38 14.38 9.45 0.12 15.51 11.81 1.03 0.06 0.11 Peridotite reaction experiments Series A T-4151 21 1460 Ol 41.01 9.23 0.17 48.65 0.35 Opx 54.63 0.05 4.93 5.45 0.11 31.03 2.70 0.10 1.01 Cpx 53.04 0.03 5.65 4.81 0.17 24.11 10.71 0.32 1.16 L 47.18 0.55 13.84 9.46 0.16 15.15 12.08 1.02 0.15 0.41 T-4152 21 1460 Ol 41.48 8.39 0.15 49.59 0.37 Opx 54.42 0.06 5.26 4.98 0.10 31.22 2.71 0.10 1.15 Cpx 52.83 0.06 5.42 4.34 0.14 23.59 12.09 0.35 1.19 L 47.24 0.54 13.81 8.41 0.14 16.20 12.23 0.94 0.06 0.43 Series B T-4013 13 1325 Ol 40.10 12.67 46.71 0.17 Opx 52.82 7.62 7.05 29.52 2.43 0.56 Cpx 49.95 0.28 9.30 4.50 17.34 17.86 0.17 0.43 Sp 0.70 59.05 10.05 19.59 0.08 10.53 L 48.45 0.61 18.56 8.69 0.15 9.26 11.50 2.47 0.26 0.04 T-3996 18 1410 Ol 40.52 9.84 48.49 0.23 Opx 53.59 6.08 5.83 30.74 2.50 1.07 Cpx 51.88 7.06 4.51 20.92 14.13 0.20 1.30 L 47.18 0.54 14.64 8.73 0.16 14.99 12.51 0.85 0.13 0.29 T-3997 22 1480 Ol 40.81 8.61 49.59 0.24 Opx 54.25 5.06 5.19 31.71 2.51 1.17 Cpx 53.50 5.06 4.56 24.64 10.72 0.15 1.22 L 47.31 0.41 12.36 8.85 0.18 17.26 12.29 0.79 0.08 0.48 T-3998 25 1530 Ol 41.37 7.42 51.07 0.15 Opx 56.06 3.22 4.39 33.77 1.71 0.85 L 48.14 0.33 10.81 8.57 0.15 20.36 10.36 0.63 0.06 0.59 Average standard deviations Olivine 0.16(9) 0.12(7) 0.06(8) 0.14(9) 0.03(2) 0.03(2) Pyroxenes 0.3(2) 0.06(9) 0.3(2) 0.15(9) 0.04(4) 0.4(2) 0.3(3) 0.03(3) 0.3(3) Spinel 0.8(6) 0.16(8) 0.3(1) 0.05(3) 0.9(6) Plagioclase 0.3(2) 0.2(1) 0.06(5) 0.1(2) 0.2(2) 0.11(6) Glass 0.15(9) 0.03(2) 0.07(5) 0.04(2) 0.04(2) 0.10(7) 0.08(2) 0.03(3) 0.01(1) 0.03(2) Ol, olivine; Sp, spinel; Plg, plagioclase; Cpx, clinopyroxene (A, B refer to coexisting compositions in sector zoned clinopyroxene); Opx, orthopyroxene; L, glass. All analyses have been normalized to 100 wt %, average standard deviations are in wt % and numbers in parentheses are 1σ in terms of the last unit cited; e.g. 0.15(9) represents 0.15 ± 0.09. Open in new tab Olivine addition crystallization experiments Series 1 (11 wt % olivine). The phase relations of the 95–2 composition (15 wt % MgO) from 1 to 2 GPa are presented in Fig. 6. Olivine is the liquidus phase to ∼1.8 GPa, where it is replaced by clinopyroxene. As for mix composition 95-1, there exists no liquidus or near-liquidus field of orthopyroxene although the clinopyroxene at the liquidus at 2 GPa is more sub-calcic than the liquidus clinopyroxene for 95–1 at 1.5 GPa and 1.2 GPa. Series 2 (13 wt % olivine). The phase relations of the 95–3 composition (16 wt % MgO) from 1.8 to 2.2 GPa are presented in Fig. 7. Olivine is the liquidus phase at 1.8 and 2.0 GPa and clinopyroxene is close to the liquidus at 2.2 GPa. We infer that clinopyroxene replaces olivine as the liquidus phase at ∼2.2 GPa. As with 95–1 and 95–2 compositions there exists no liquidus or near-liquidus field of orthopyroxene. Fig. 4. Open in new tabDownload slide Back-scattered electron image of run T-361 (Tables 3 and 4) showing initial growth of subcalcic pyroxene (arrow 2) on seed orthopyroxene (arrow 1), and subsequent development and growth of an equilibrium clinopyroxene (arrow 3) and orthopyroxene pair (arrow 4). Series 3 (17 wt % olivine). The phase relations of the 95–4 composition (18 wt % MgO) from 2.5 to 2.7 GPa arepresented in Fig. 8. Olivine is the liquidus phase at 2.5 GPa and garnet is inferred to be the liquidus phase at 2.6 and 2.7 GPa. The presence of garnet as a near-liquidus phase close to the inferred inflexion point on the liquidus surface (∼2.55 GPa) indicates that 95–4 is not a suitable composition for a primary magma parental, via olivine fractionation, to glass composition 123 95-1. This is because the REE pattern of 123 95–1 (Fig. 2) is not consistent with equilibrium with residual garnet. Therefore, there is no need to investigate more olivine-enriched mix compositions in an attempt to find a multiple saturation point in olivine + orthopyroxene ± clinopyroxene. Fig. 5. Open in new tabDownload slide Pressure (in this and subsequent figures, given in kilobars; to convert to GPa, divide by 10) vs temperature plot presenting the experimental results of crystallization experiments on mix 95–1 and inferred phase boundaries and phase fields. Filled circles, above liquidus; crosses, olivine; open circles, clinopyroxene; squares, plagioclase. Bold text indicates the inferred P and T of the change in liquidus phase from olivine to clinopyroxene. Fig. 6. Open in new tabDownload slide Pressure and temperature plot presenting the experimental results of crystallization experiments on mix 95–2 and inferred phase boundaries and phase fields. Symbols as for Fig. 5. Bold text indicates the inferred P and T of the change in liquidus phase from olivine to clinopyroxene. Fig. 7. Open in new tabDownload slide Pressure and temperature plot presenting the experimental results of crystallization experiments on mix 95–3 and inferred phase boundaries and phase fields. Symbols as for Fig. 5. Bold text indicates the inferred P and T of the change in liquidus phase from olivine to clinopyroxene. Olivine and orthopyroxene addition crystallization experiments The results of the crystallization experiments on mixes 95–1 to 95–4 fail to demonstrate a near-liquidus field in olivine + orthopyroxene ± clinopyroxene. However, at pressures >1.8 GPa, near-liquidus clinopyroxene is distinctly sub-calcic (Table 4, e.g. T-344, T-213) or is zoned towards low-calcium cores (Table 3). This may indicate that the 95–1 + olivine-enriched compositions are very close to orthopyroxene saturation but orthopyroxene has failed to nucleate, possibly because of the presence of early formed metastable subcalcic pyroxene or reaction relationships in melting (see discussion above). Experimental studies of liquidus pyroxenes in lunar basalts have shown that experimental difficulties do exist in crystallization of orthopyroxene, and that metastable nucleation and growth of subcalcic (‘pigeonitic’) clinopyroxenes in place of the stable coexisting orthopyroxene–clinopyroxene pair may occur (Green, 1976; Green et al., 1975). To test this possibility we have added orthopyroxene seeds to mix 95–3 to test for orthopyroxene saturation in a similar manner to Green et al. (1975, 1979). Two series of experiments were conducted. In Series 1 experiments, 1 wt % oforthopyroxene was added as either bronzite (mix 95-5, Table 2) or Tem-Pres synthetic clinoenstatite (mix 95-6, Table 2). In Series 2 experiments 2 wt % of orthopyroxene was added as either orthopyroxene WSS1 (mix 95-7, Table 2) or Tem-Pres synthetic clinoenstatite (mix 95-8, Table 2). Fig. 8. Open in new tabDownload slide Pressure and temperature plot presenting the experimental results of crystallization experiments on mix 95–4 and inferred phase boundaries and phase fields. Symbols as for Fig. 5, except for large open circles, which indicate garnet. Bold text indicates the inferred P and T of the change in liquidus phase from olivine to clinopyroxene. Both series of orthopyroxene addition experiments were successful in producing melt compositions in equilibrium with a lherzolite assemblage (runs T-361 and T-518, Tables 3 and 4) at 2-2.1 GPa, 1450–1460°C. Figure 4 is a back-scattered electron image of pyroxene in run T-361 showing initial subcalcic pyroxene that has grown on seed orthopyroxene laths, and during the course of the experiment (1 h, Table 3) an orthopyroxene and clinopyroxene pair has grown, with both phases now in contact with coexisting melt. The high-Ca clinopyroxene and the coexisting orthopyroxene are inferred to be an equilibrium pair. More importantly, the melt compositions in T-361 and T-518 closely match olivine-enriched 123 95–1 compositions (intermediate between 95–2 and 95–3 compositions). The results of the orthopyroxene addition experiments demonstrate that olivine-enriched 123 95–1 parental compositions are in equilibrium with upper-mantle lherzolite at 2 GPa, 1450–1460°C. Forward experimental approach The inverse experimental approach in combination with orthopyroxene addition experiments has been successful in determining a multiple saturation point at ∼2–2.1 GPa, 1450–1460°C. How robust is this conclusion? There are two areas of concern. The first is the presence of zoning and relict seed orthopyroxene in runs T-361 and T-518, which indicates that the entire bulk composition has not come to full equilibrium. Second, it may be possible that more olivine-poor or -rich parental compositions could also be close to equilibrium with the mantle if we took the trouble to perform more seeding experiments. We therefore have used the forward experimental approach in the form of peridotite reaction experiments to (1) perform reversals on the orthopyroxene addition experiments (Series A) and (2) test whether other olivine-enriched parental compositions are close to being in equilibrium with mantle peridotite (Series B). Series A peridotite reaction experiments We performed two reaction experiments at 2.1 GPa, 1460°C using mix 95–3 and peridotite compositions TQ-40 (T-4151, Table 3) and MM-3 (T-4152, Table 3). Both reaction experiments resulted in glass compositions closely similar to the glass in T-518 (Table 4), i.e. the multiple-saturation point defined by the inverse experimental approach. However, the residual pyroxenes in the reaction experiments are slightly more aluminous and Cr rich, and the clinopyroxene is significantly more calcic than present in run T-518, indicating that run T-518 was not fully equilibrated, though it had approached an equilibrium multiply saturated assemblage. Series B peridotite reaction experiments The rationale of this series of experiments is outlined with reference to Fig. 9. In Fig. 9 the clinopyroxene, olivine and garnet (inferred) liquidus surfaces are shown on a P–T plot for all the olivine-enriched mix compositions studied (95-1 to 95-4) and compared with the solidus and clinopyroxene-out curves determined for peridotite composition TQ–40 (Falloon et al., 1988; Green & Falloon, 1998). Of particular importance is the liquidus surface for clinopyroxene, which has a positive P–T slope, and crosses the solidus for TQ-40 at ∼1.3 GPa, 1310°C and the cpx-out curve for TQ-40 at ∼2.5 GPa, 1525°C. Consequently, there is the potential for 123 95–1 and more olivine-enriched parental compositions to be in equilibrium with a lherzolite residue over this entire pressure and temperature range, and that the results of the inverse approach are non-unique. To test this possibility we have performed reaction experiments at each of the inferred inflexion points defined by the changeover from olivine to clinopyroxene liquidus phase for each composition studied. The inferred inflexion points are shown in Fig. 9 as well as Figs 5–8. We used 95–1 as a reactant at 1.3 GPa and 95–3 as a reactant at 1.8–2.5 GPa. Reaction experiments from 1.3 to 2.2 GPa resulted in lherzolite residues whereas the reaction experiment at 2.5 GPa resulted in a harzburgite residue. The compositions of glass in the reaction experiments are plotted in the molecular normative tetrahedron in Fig. 10a and b, and compared with the olivine-enriched parental compositions, glass compositions from runs T-518 and T-361, and glass compositions from the series A reaction experiments. Figure 10a and b demonstrates that the conclusions from the inverse approach experiments are unique because of the fact that olivine-enriched compositions define a vector (an olivine control line towards the Oliv apex) which cuts across the trend of compositions defined by the peridotite reaction experiments in the projection from diopside onto the base [Jd + CaTs + Lc]–Qz–Ol (Fig. 10b). The glass compositions from runs T-518, T-4151 and T-4152 plot on or close to this intersection. In the projection from olivine onto the face Qz–Di–[Jd + CaTs + Lc] (Fig. 10a), all olivine-enriched parental compositions plot at the same position and as with the projection from diopside, glass compositions from runs T-518, T-4151 and T-4152 closely match the position of 123 95–1 and more olivine-enriched compositions whereas glass compositions from reaction experiments at different P, T plot away from 123 95-1. Figure 10b shows that peridotite reaction experiments between 1.8 and 2.2 GPa appear to plot on or close to the olivine control line, and in particular the reaction experiment at 1.8 GPa (Tables 3 and 4) falls on the olivine control line in both the projection from diopside and olivine (Fig. 10a, b). However, the 1.8 and 2.2 GPa glass compositions have significantly different Al2O3 and CaO contents compared with the olivine-enriched compositions at a given MgO content. The series A peridotite reaction experiments provide the closest matching to the olivine-enriched parental compositions in terms of major elements. However, although the series A reaction experiments plot on the olivine control line in Fig. 10b, they plot slightly away towards higher normative Di contents in Fig. 10a. These small inconsistencies apparent in Fig. 10a and b reflect the ability to resolve small but important differences in major element compositions on the basis of normative chemistry. Fig. 9. Open in new tabDownload slide Summary pressure vs temperature plot comparing the clinopyroxene liquidus slope for 123 95–1 and olivine-enriched compositions with the solidus and clinopyroxene-out curves for Tinaquillo Lherzolite. Filled circle is the inferred point of magma segregation for the primary picrite magma parental to 123 95-1. Open squares indicate the positions of Series B peridotite reaction experiments. Continuous line, clinopyroxene liquidus for 123 95–1 and olivine-enriched compositions (95-2, -3, -4); dashed–dotted line, solidus for Tinquillo Lherzolite; long dashed line, clinopyroxene-out for Tinquillo Lherzolite; short dashed line, inferred garnet liquidus for 95–4 composition; dotted lines, inferred olivine liquidus for 123 95–1 and olivine-enriched compositions (95-2, -3, -4). Implications For Magma Genesis in Back-Arc Basins The experimental results from both the inverse and forward approach demonstrate that a picrite parental composition of ∼16 wt % MgO is in equilibrium with a residual lherzolite assemblage at a unique and well-constrained pressure of ∼2.1 GPa, and temperature of ∼1460°C. Although we have established the above conditions of origin for the parental picrite to 123 95-1, the degree of partial melting is model dependent, as the nature of our experimental approach does not directly give us this information. We are free to vary our modal abundances in an equilibrium assemblage and hence there is potentially a wide range of chemical compositions and thus in the degree of partial melting that can be applied to the parental picrite for 123 95-1. However, we can constrain these calculations by making an informed choice on potential peridotite source compositions, noting that the mantle-derived lherzolite samples form a coherent compositional trend (Green & Falloon, 1998). If we assume that naturally occurring mantle samples are representative of asthenospheric sources and residues (Green & Falloon, 1998), then it is possible to estimate the degree of partial melting via mass balance using the experimentally determined equilibrium phase assemblages and a suitable mantle sample. The suitability of natural mantle samples may be assessed in terms of their distinctive minor and trace element abundances and, in this case, mantle samples with strongly LREE-depleted patterns such as Tinaquillo Lherzolite or Lizard Lherzolite (Frey, 1970; Haskin & Frey, 1966). These compositions are lherzolitic rather than harzburgitic and are very similar in major element composition to fertile (MORB pyrolite) or enriched (Hawaiian pyrolite) model mantle compositions. However, in addition to being strongly depleted in LREE, these lherzolites have lower K2O, TiO2, P2O5 and Na2O contents than the above compositions. If we take the Tinaquillo Lherzolite (Green, 1963) as representative of a more refractory lherzolite composition then mass balance calculations using the equilibrium assemblage from run T-4152 give a very good fit (Table 5) and suggest that the parental picrite 123 95–1 represents ∼15 wt % melting of a lherzolite source approximating refractory Tinaquillo Lherzolite. The mass balance calculation (Table 5) also indicates that only a trace of clinopyroxene is left in the residue, which is essentially harzburgite. Fig. 10. Open in new tabDownload slide Molecular normative projection from olivine (a) onto the face Di–Qz–[Jd + CaTs + Lc] and from diopside (b) onto the base [Jd + CaTs + Lc]–Qz–Ol of the ‘basalt tetrahedron’ (Falloon & Green, 1988) displaying the experimental results of the orthopyroxene addition and peridotite reaction experiments. Open circle, composition 123 95-1; dotted arrow in (b) is an olivine addition vector; open square, series B peridotite reaction experiments with pressure in kilobars; filled circle, series A peridotite reaction experiments; inverted triangle, orthopyroxene addition experiments (T-361 and T-518; see Tables 3 and 4). Table 5: Mass balance calculation for run T-4152 and Tinaquillo Lherzolite Phase . Glass . Olivine . Orthopyroxene . Clinopyroxene . Tinaquillo Lherzolite . . . . . . Observed . Estimated . sio2 47.24 41.48 54.42 52.83 44.95 44.94 Tio2 0.54 0.06 0.06 0.08 0.10 Al2O3 13.81 5.26 5.42 3.22 3.30 FeO 8.41 8.39 4.98 4.34 7.66 7.55 MgO 16.20 49.59 31.22 23.59 40.03 40.06 CaO 12.23 0.37 2.71 12.09 2.99 2.98 Na2O 0.94 0.10 0.35 0.18 0.19 Cr2O3 0.43 1.15 1.19 0.45 0.34 Calculated proportions 0.15(1) 0.61(1) 0.23(1) 0.0003(1) Square sum of the residuals 0.0324 Phase . Glass . Olivine . Orthopyroxene . Clinopyroxene . Tinaquillo Lherzolite . . . . . . Observed . Estimated . sio2 47.24 41.48 54.42 52.83 44.95 44.94 Tio2 0.54 0.06 0.06 0.08 0.10 Al2O3 13.81 5.26 5.42 3.22 3.30 FeO 8.41 8.39 4.98 4.34 7.66 7.55 MgO 16.20 49.59 31.22 23.59 40.03 40.06 CaO 12.23 0.37 2.71 12.09 2.99 2.98 Na2O 0.94 0.10 0.35 0.18 0.19 Cr2O3 0.43 1.15 1.19 0.45 0.34 Calculated proportions 0.15(1) 0.61(1) 0.23(1) 0.0003(1) Square sum of the residuals 0.0324 For Tinaquillo Lherzolite, ‘observed’ is the target composition, and ‘estimated’ is the result of mixing the phases in run T-4152 (see text, and Tables 3 and 4). Open in new tab Table 5: Mass balance calculation for run T-4152 and Tinaquillo Lherzolite Phase . Glass . Olivine . Orthopyroxene . Clinopyroxene . Tinaquillo Lherzolite . . . . . . Observed . Estimated . sio2 47.24 41.48 54.42 52.83 44.95 44.94 Tio2 0.54 0.06 0.06 0.08 0.10 Al2O3 13.81 5.26 5.42 3.22 3.30 FeO 8.41 8.39 4.98 4.34 7.66 7.55 MgO 16.20 49.59 31.22 23.59 40.03 40.06 CaO 12.23 0.37 2.71 12.09 2.99 2.98 Na2O 0.94 0.10 0.35 0.18 0.19 Cr2O3 0.43 1.15 1.19 0.45 0.34 Calculated proportions 0.15(1) 0.61(1) 0.23(1) 0.0003(1) Square sum of the residuals 0.0324 Phase . Glass . Olivine . Orthopyroxene . Clinopyroxene . Tinaquillo Lherzolite . . . . . . Observed . Estimated . sio2 47.24 41.48 54.42 52.83 44.95 44.94 Tio2 0.54 0.06 0.06 0.08 0.10 Al2O3 13.81 5.26 5.42 3.22 3.30 FeO 8.41 8.39 4.98 4.34 7.66 7.55 MgO 16.20 49.59 31.22 23.59 40.03 40.06 CaO 12.23 0.37 2.71 12.09 2.99 2.98 Na2O 0.94 0.10 0.35 0.18 0.19 Cr2O3 0.43 1.15 1.19 0.45 0.34 Calculated proportions 0.15(1) 0.61(1) 0.23(1) 0.0003(1) Square sum of the residuals 0.0324 For Tinaquillo Lherzolite, ‘observed’ is the target composition, and ‘estimated’ is the result of mixing the phases in run T-4152 (see text, and Tables 3 and 4). Open in new tab A batch melting model for 123 95–1 parental picrite is one of diapirism of a high-temperature lherzolite such as Tinaquillo or Lizard Lherzolite, previously depleted by loss of a very small melt at T >1460°C. The diapir attains a melt fraction of ∼15% at depths of ∼70 km, T ∼1460°C, with melt separation and movement to a sub-volcanic magma accumulation region with some precipitation of olivine en route. However, most current models of magma genesis in mid-ocean ridge environments invoke the generation of small melt fractions over a depth interval (melting column), with the efficient removal and isolation of melt fractions before aggregation of these melt fractions at the top of the melting column or in a sub-axial magma chamber (McKenzie & Bickle, 1988; Langmuir et al., 1992; Shen & Forysth, 1995). Therefore, is it possible that the parental picrite composition for 123 95–1 is either a small melt fraction or an aggregate of small melt fractions? To answer these questions we need to know the nature of small melt fractions of upwelling mantle peridotite undergoing dynamic melting. In early studies, Green & Ringwood (1967) and Green (1970) inferred that low-degree melts from fertile mantle lherzolite were nepheline normative from <1 GPa to >3 GPa and further, that enriched sources were required for Hawaiian primitive magmas and depleted sources required for MORB primitive magmas. Thompson (1984, 1987) used a wt % CIPW normative projection [a Di–Hy–Ol–Ne–Qz plot; see fig. 4 of Thompson (1984) and fig. 6 of Thompson (1987)] and experimental data available at that time to estimate the positions of initial melt fractions for a fertile and a depleted mantle composition. Thompson (1984, 1987) estimated the normative composition of initial melt fractions by extrapolating isobaric melting trends, determined at higher melt fractions, towards likely lower initial melt fractions. He also concluded that initial melt fractions for fertile peridotite were all nepheline normative from 1 to ∼3.5 GPa, and ol–hy normative at pressures >3.5 GPa. MgO contents of initial melts for fertile mantle varied from ∼8 wt % MgO at 1 GPa to ∼25 wt % at 3.3 GPa. For more depleted peridotite, initial melts become ol–hy normative in this pressure range with MgO contents ranging from 9 to 18 wt % from 0.8 to 2 GPa. Thompson (1987) demonstrated, on the basis of the available data, that source fertility had a major control on the compositions of initial melt fractions, being strongly nepheline normative for fertile peridotite but ol–hy normative for more refractory peridotite. Recent experimental work by Walter & Presnall (1994) in the simple system NCMAS (Na2O–CaO–MgO–Al2O3–SiO2) and Falloon et al. (1997) at 1 GPa in more complex Fe-bearing systems has helped to further constrain the normative characteristics of initial melt fractions and the change in initial melt fractions with source fertility. In Fig. 11a and b we present the isobaric melting systematics for model peridotite compositions (MORB pyrolite and Tinaquillo Lherzolite) at 1 and 2 GPa, and the range of initial melt fractions possible from mantle peridotite based on the work of Falloon & Green (1987, 1988), Walter & Presnall (1994), Falloon et al. (1997, in preparation). At 1 GPa, plagioclase forms part of the sub-solidus phase assemblage for lherzolite and initial melts approach ‘eutectic melting’ in which the composition of the plagioclase is an important control on melt composition. The line An40–An100 in Fig. 11a and b is the locus of initial melt fractions for fertile to refractory lherzolite compositions at 1 GPa (Falloon et al., 1997). Figure 11a indicates that initial melt fractions of plagioclase lherzolite at 1 GPa define a smooth trend within the tetrahedron, illustrated by projections from olivine (Fig. 11a) and diopside (Fig. 11b). The initial or near-solidus melts have higher [(Na + K)/Ca] than the bulk composition but, so long as plagioclase is a residual or solidus phase, liquids lie on the line An40–An100. For more fertile compositions, liquids are nepheline normative. For more refractory lherzolite compositions (i.e. with sub-solidus plagioclase of >An65 composition, or with chrome-spinel lherzolite mineralogy) initial melts become more tholeiitic with normative olivine and hypersthene. The presence of clinopyroxene as a residual phase buffers the initial melts at approximately constant normative diopside contents (10 mol % in this projection; Fig. 11a) but if plagioclase is eliminated from the residue, then liquids move to higher normative diopside contents along olivine + orthopyroxene + clinopyroxene + chrome-spinel cotectics. Isobaric melting at 1 GPa for a MORB pyrolite composition is illustrated in Fig. 11. Melting begins at a point on the Ab40–An100 curve appropriate for its bulk composition. Plagioclase is eliminated from the residue at temperatures slightly above the solidus, and indeed the abundance of plagioclase strongly controls the proportion of melt phase at and just above the solidus. As melting progresses, melt compositions become progressively more diopside normative (Fig. 11a) and olivine normative (Fig. 11b). At the point of clinopyroxene elimination, partial melt compositions have reached a maximum in normative diopside content and now lie on an olivine + orthopyroxene control line (Fig. 11a). Melting will then proceed along an olivine + orthopyroxene cotectic until orthopyroxene is eliminated and melt compositions will then be constrained to lie on an olivine control line through the bulk composition. Melts in equilibrium with olivine will therefore define a straight line in the diopside projection (Fig. 11b) and plot on top of the bulk composition in the projection from olivine (Fig. 11a). Figure 11a and b also illustrates that the melting systematics established at 1 GPa are applicable to higher pressures, and we illustrate the isobaric melting of MORB pyrolite and the range of initial melts from mantle peridotite at 2 GPa. Initial melts at 20 kbar form a smooth trend (trend Jd–Di in Fig. 11b) from nepheline-normative Na2O rich melts from fertile or enriched compositions towards ol–hy normative melts from refractory compositions dependent again on bulk composition and the jadeite content of sub-solidus clinopyroxene. Although plagioclase (albite) is only a sub-solidus phase at 2 GPa in very high Na/Ca lherzolites, the Jd/Di ratio of sub-solidus pyroxene is an important control on melt compositions, reflecting Na–Ca partitioning between clinopyroxene and liquid at 2 GPa. Figure 11 also illustrates that the isobaric melting trend for MORB pyrolite closely resembles that at 10 kbar except that, in the projection from olivine (Fig. 11a), clinopyroxene is eliminated at melt compositions with higher normative diopside (Fig. 11a). Fig. 11. Open in new tabDownload slide Molecular normative projection from olivine (a) onto the face Di–Qz–[Jd + CaTs + Lc] and from diopside (b) onto the base [Jd + CaTs + Lc]–Qz–Ol of the ‘basalt tetrahedron’ (Falloon & Green, 1988), showing the phase relationships of peridotite melting at 10 and 20 kbar for MORB pyrolite (continuous lines) and Tinaquillo Lherzolite (dashed lines). Large open circle, primary picrite parental to 123 95-1; filled circle, glass 123 95–1 [note that in (a) 123 95–1 plots on top of the picrite parent]; arrow in (b), olivine control line connecting the picrite parent to 123 95-1; filled cross in (a), MORB pyrolite; open cross in (a), Tinaquillo Lherzolite; bold line marked ‘An40–An100’ is the locus of initial melts in equilibrium with a range of plagioclase contents (Walter & Presnall, 1994; Falloon et al., 1997); bold dashed line with arrow marked ‘Jd–Di’ is the estimated (based on Walter & Presnall, 1994) locus of initial melts in equilibrium with a range of clinopyroxene compositions with differing jadeite contents at 20 kbar; arrows in (b) represent melts in equilibrium with a dunite residue and as they define an olivine control line, they plot on top of MORB pyrolite and Tinaquillo Lherzolite in (a) [and are therefore not plotted in (a)]; open diamonds, field of Lau Basin glasses with >8 wt % MgO. Data sources: Hawkins & Melchior (1985); Davis et al. (1987); Falloon et al. (1987, 1992); Jenner et al. (1987); Boespflug et al. (1990); Frenzel et al. (1990); Sunkel (1990); Sinton et al. (1991); Hawkins & Allan (1994); Hawkins, (1995a, 1995b). Experimental data from: Falloon & Green (1988); Falloon et al. (1988, 1997, in preparation); Walter & Presnall (1994); T. J. Falloon (unpublished data, 1998). A very important feature of Fig. 11a is the illustration at 2 GPa of the locus of melt compositions at the solidus of lherzolite compositions that vary from enriched or fertile (with higher Jd/Di ratio in sub-solidus clinopyroxene) to refractory (limited by the Na-free system CMAS). All have olivine + orthopyroxene + clinopyroxene + spinel as the sub-solidus mineralogy. The melts are only slightly higher in normative diopside than those at 1 GPa. In models of fractional melting and instantaneous melt extraction they represent the instantaneous fractional melts available at 2 GPa over the range of temperatures and bulk compositions in a model melt column. We are now in a position to assess the composition of melts derived by fractional melting, or incremental melting, and melt aggregation over a melt column from 2 GPa (70 km) to 1 GPa (35 km) and, by extrapolation, over a larger depth range. In particular, we can assess whether the composition of 123 95–1 or related olivine-enriched picrites (shown by the filled or open circles in Fig. 11) could result from pooling of melt increments in a melt column with bulk composition varying from fertile at high pressures to refractory at low pressures. In both projections (Fig. 11a and b) any mixing contribution from deep, more fertile compositions is too olivine rich and too silica undersaturated in relation to the target composition (olivine control line through 123 95-1). Compositions overlapping 123 95–1 in the projection from diopside (Fig. 11b) could represent melt increments at ∼1–1.2 GPa and residue without plagioclase and with clinopyroxene either absent or about to disappear, i.e. harzburgitic residue. However, this is the incremental melt composition and for 123 95–1 to be a pooled melt composition then significant melt contributions from harzburgitic residue at P ∼0.5 GPa would be required. Although this is not precluded by the diopside projection (Fig. 11b), it is precluded by the olivine projection (Fig. 11a). In this projection, the high relative normative diopside of 123 95–1 (an olivine-enriched picrites) is incompatible with incremental melts from both higher-pressure fertile compositions (too low in diopside, too nepheline normative) and lower-pressure refractory (harzburgitic) compositions (too low in diopside, too quartz normative). There is no conceptual melt column, whether beginning well within the stability field for solidus garnet (P >3 GPa) or ending well within the low-pressure field for quartz-normative liquids (P ∼0.2–0.5 GPa), that can produce pooled melt compositions matching 123 95–1 or olivine-enriched picrites. The phase equilibria constraints on melting of lherzolite, whether ‘fertile’ or ‘refractory’, are considered to preclude the derivation of 123 95–1 glass by a process of fractional melting and melt aggregation. Although the phase equilibria constraints on major element compositions of melts preclude the dynamic melting or column melting models for upwelling asthenospheric mantle as appropriate for the refractory magma 123 95-1, we also need to examine models in which melts with incompatible minor and trace element compositions determined at deeper levels have major element compositions determined not by mixing of low-pressure melt increments bu by reaction and re-equilibration with lherzolite–harzburgite–dunite at low pressure. Th ‘wall-rock’ reaction concept is a vintage one, being proposed by Green & Ringwood (1967) to explain the diversity and disjunction between trace-element and isotopic signatures of magmas and their major element–phase equilibria characteristics. If models invoke re-equilibration with lherzolite at low pressure then, in the current context, we are simply changing the bulk composition, and liquids will lie on olivine + orthopyroxene + clinopyroxene ± plagioclase ± spinel cotectics at the postulated pressure of reaction and re-equilibration. The data presented here for 123 95–1 (and olivine-enriched picrites) constrain such reaction and re-equilibration to occur at ∼2.1 GPa, 1460°C if it is postulated that the residue or reacting wall-rock is lherzolite. Similarly, if the reacting wall-rock remains as harzburgite, then the only condition at which 123 95–1 (and olivine-related picrites) is in equilibrium with harzburgite is ∼2.1 GPa, T >1460°C. However, if it is postulated that reaction with harzburgite at lower pressure has eliminated orthopyroxene from the wall-rock so that 123 95–1 attains its observed composition within a dunite channel through harzburgite, then the precursor (pre-reaction) magma for 123 95–1 would lie on a vector from the quartz apex through 123 95–1 (or related picrite) composition in both projections. This is because the reaction between higher-pressure magma and harzburgite at lower pressure is an expansion of the olivine field at the expense of orthopyroxene. In the projection from diopside such precursor melts could lie on higher-pressure lherzolite residue trends and would be less hypersthene rich than 123 95-1, implying smaller degrees of melting or deeper levels for melt pooling but still requiring a refractory source in terms of trace and minor elements. In the projection from olivine, the precursor melt would lie towards even higher relative ‘Di’ content and such compositions are not compatible with melting of lherzolite, unless at pressures >2 GPa. More complex wall-rock reaction processes may be suggested in which the magma evolves by elimination of orthopyroxene from the wall-rock and precipitation of olivine and calcic clinopyroxene, buVered in mg-number by the wall-rock olivine. In the projection of Fig. 11, this reaction trend is a vector through 123 95–1 expressing both addition of quartz and precipitation of diopside components. The precursor liquid, before reaction to produce 123 95-1, would lie between the two vectors from 123 95–1 towards the diopside apex and away from the quartz apex. This postulated precursor liquid is too rich in normative diopside to be a melt of mantle lherzolite. In our view, the major element composition of 123 95-1, and the liquidus and near-liquidus phase relationships, which are a consequence of the major element composition, preclude the derivation of this magma by either column melting or dynamic melting models, or by models in which such precursor melts react with lherzolitic or harzburgitic wall-rock at low pressure. In the last type of models, the wall-rock reaction model may be, in terms of equilibrium, indistinguishable from the batch-melting model, with the debate being transferred to hypotheses of how the particular lherzolite composition undergoing equilibrium melt–residue reaction attained its bulk composition. In presenting our preferred model of diapirism, we draw attention to examples of refractory lherzolite comoccur positions emplaced to crustal levels as high-temperature diapirs or mantle slices, for which batch melting along a P, T upwelling path through 2.1 GPa, 1460°C would yield picritic magma matching closely the major, minor and trace element characteristics of a parental picrite to 123 95–1 containing ∼16 wt % MgO (123 95–1 + 13 wt % olivine). The examples of such peridotite compositions include Lizard Lherzolite and Tinaquillo Lherzolite. Our preferred model is one of mantle upwelling of a previously depleted [by loss of incipient (1–2%) melt fraction] lherzolite with melt retention within a diapiric or small plume-head configuration. The P, T path through 2.1 GPa, 1460°C and the lherzolite bulk composition resulted in ∼15% melting and melt segregation. The primitive picrite precipitated ∼13 wt % olivine between source depth and eruption and quenching to 123 95–1 glass. The residue was high-temperature, high-pressure harzburgite (∼73% olivine, 27% orthopyroxene, <1% clinopyroxene) (Table 5). The preferred model is one of three-dimensional upwelling of plume-like or diapiric structures along essentially adiabatic paths through the lithosphere–asthenosphere transition (Green & Falloon, 1998). Possible support for three-dimensional diapiric upwelling is given by evidence for ‘local’ trends (Klein & Langmuir, 1987; Niu & Batiza, 1993) shown by some of the Lau Basin basalts on an Na8 vs Fe8 plot (site 834, KTJ, ELSC, Fig. 12). Local trends appear to be characteristic of slow, but can also occur at intermediate–fast spreading ridges (Niu & Batiza, 1993). Niu & Batiza (1993, 1994) presented strong arguments that the local vectors are best explained by focused buoyant mantle upwelling melting processes in which significant solid–melt equilibration and melt retention can occur. Niu & Batiza (1993, 1994) therefore argued that diapirism is a significant process in the generation of ocean crust. However, Pearce et al. (1995) have noted that in the case of the Lau Basin spreading ridges, especially the ELSC, the local trend is correlated with a subduction zone signature and suggests that local vectors may also reflect more depleted mantle sources for spreading ridges close to the volcanic front, which should lead to a lower degree of partial melt generation. This source depletion may be counterbalanced by the presence of water from the slab-derived subduction component, which should increase the degree of partial melting (Pearce et al., 1995). Fig. 12. Open in new tabDownload slide Na2O[8] vs FeO[8] (after Klein & Langmuir, 1987) for Lau Basin glasses. Crossed squares, ODP Leg 135 Site 839; diagonally filled squares, ODP Leg 135 Site 838; open squares, ODP Leg 135 site 836; inverted open triangles, ODP Leg 135 site 834; half-filled diamonds, King's Triple Junction; open circles, Central Lau Spreading Centre; filled squares, Eastern Lau Spreading Centre; crossed circles, N-type Lau Basin glasses; filled circles, E-type Lau Basin glasses; open cross, sample 123 95-1. Irregular polygon encloses the global array for MORB from Klein & Langmuir (1987). Short dashed lines define local trends for Site 834, King's Triple Junction and Eastern Lau Spreading Centre data. Glass data sources as the legend to Fig. 11. In this study we have established the composition of a primary magma for a refractory end-member Lau Basin upwelcomposition. However, the Lau Basin covers a spectrum of compositions (Hawkins, 1995a, 1995b; Pearce et al., 1995), well illustrated by Figs 3 and 12, and it is therefore important to understand whether the petrogenetic conditions established for 123 95–1 are characteristic of the Lau Basin in general. As mentioned above, the local vectors displayed by the Lau Basin data in Fig. 12 suggest that three-dimensional diapiric upwelling proposed for the petrogenesis of 123 95–1 composition is a general feature of the Lau Basin. If this is true, we would expect relatively primitive Lau Basin compositions to show batch melt characteristics in their normative chemistry. In Fig. 11 we plot the compositions of Lau Basin glasses with >8 wt % MgO (n = 22) and compare them with the mantle melting cotectics presented in Fig. 11. Figure 11a demonstrates that all of the Lau Basin glass compositions have significant normative diopside contents. The plotted position of Lau Basin glasses with relatively low normative diopside contents suggests they could be generated by melting (up to ∼25%) of a fertile MORB pyrolite source leaving lherzolite to harzburgite residues. The plotted position of Lau Basin glasses with relatively high normative diopside contents in Fig. 11a suggests they could be generated by lower degrees of partial melting (∼10–15%) of the more refractory Tinaquillo Lherzolite, leaving lherzolite residues, or harzburgite residues if the refractory source had a slightly lower CaO/Al2O3 value. Pressures of magma segregation for all the Lau Basin glasses are >∼1.5 GPa and temperatures >∼1400°C, and imply that, in general, primary magmas parental to the relatively low-MgO Lau Basin magmas should be picritic (13–16 wt % MgO). Acknowledgements We acknowledge the technical assistance of Wieslav Jablonski, Keith Harris and Leonid Danyushevsky. Trevor Falloon acknowledges support from the Australian Research Council. This manuscript was improved by constructive reviews from Drs Sherm Bloomer and John Sinton. A. L. Jaques publishes with the permission of the Executive Director of AGSO. References Basaltic Volcanism Study Project (BVSP) . , Basaltic Volcanism on the Terrestrial Planets , 1981 New York Pergamon Press Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Boespflug B. , Dosso L. , Bougault H. , Joron J.-L. . von Stackelberg U. , von Rad U. . Trace element and isotopic (Sr, Nd) geochemistry of volcanic rocks from the Lau Basin , Geological Evolution and Hydrothermal Activity in the Lau and North Fiji Basins, Southwest Pacific Ocean (Results of SONNE Cruise SO–35) Geologische Jahrbuch , 1990 , vol. 92 (pg. 503 - 516 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Crawford A. J. , Beccaluva L. , Serri G. . Tectono–magmatic evolution of the west Philippine–Mariana region and the origin of boninites , Earth and Planetary Science Letters , 1981 , vol. 54 (pg. 346 - 356 ) Google Scholar Crossref Search ADS WorldCat Davis A. S. , Clague D. A. , Morton J. L. . Volcanic glass compositions from two spreading centers in Lau Basin , US Geological Survey Open–File Report , 1987 (pg. 87 - 61 ) Google Scholar OpenURL Placeholder Text WorldCat Eggins S. M. . Petrogenesis of Hawaiian tholeiites: 2. Aspects of dynamic melt segregation , Contributions to Mineralogy and Petrology , 1992 , vol. 110 (pg. 398 - 410 ) Google Scholar Crossref Search ADS WorldCat Ewart A. , Bryan W. B. , Chappell B. W. , Rudnick R. L. . Hawkins J. , Parson L. , Allan J. , Resig J. , Weaver P. . Regional geochemistry of the Lau–Tonga arc and backarc systems , Proceedings of the Ocean Drilling Program, Scientific Results 135 , 1994 College Station, TX Ocean Drilling Program (pg. 385 - 425 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Falloon T. J. , Green D. H. . Anhydrous partial melting of MORB pyrolite and other peridotite compositions at 10 kbar and implications for the origin of primitive MORB glasses , Mineralogy and Petrology , 1987 , vol. 37 (pg. 181 - 219 ) Google Scholar Crossref Search ADS WorldCat Falloon T. J. , Green D. H. . Anhydrous partial melting of peridotite from 8 to 35 kbar and petrogenesis of MORB , Journal of Petrology, Special Lithosphere Issue , 1988 (pg. 379 - 414 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Falloon T. J. , Green D. H. , Crawford A. J. . Dredged igneous rocks from the northern termination of the Tofua magmatic arc, Tonga and adjacent Lau Basin , Australian Journal of Earth Sciences , 1987 , vol. 34 (pg. 487 - 506 ) Google Scholar Crossref Search ADS WorldCat Falloon T. J. , Green D. H. , Hatton C. J. , Harris K. L. . Anhydrous partial melting of fertile and depleted peridotite from 2 to 30 kbar and application to basalt petrogenesis , Journal of Petrology , 1988 , vol. 29 (pg. 257 - 282 ) Google Scholar Crossref Search ADS WorldCat Falloon T. J. , Green D. H. , McCulloch M. T. . Crawford A. J. . Petrogenesis of high–Mg and associated lavas from the north Tonga trench , Boninites and Related Rocks , 1989 London Unwin Hyman (pg. 357 - 395 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Falloon T. J. , Malahoff A. , Zonenshain L. P. , Bogdanov Y. . Petrology and geochemistry of back–arc basin basalts from Lau Basin spreading ridges at 15°, 18° and 19°S , Mineralogy and Petrology , 1992 , vol. 47 (pg. 1 - 35 ) Google Scholar Crossref Search ADS WorldCat Falloon T. J. , Green D. H. , O'Neill H. St. C. , Ballhaus C. G. . Quest for low–degree mantle melts , Nature , 1996 , vol. 381 pg. 285 Google Scholar Crossref Search ADS WorldCat Falloon T. J. , Green D. H. , O'Neill H. St. C. , Hibberson W. O. . Experimental tests of low degree peridotite partial melt compositions: implications for the nature of anhydrous near–solidus peridotite melts at 1 GPa , Earth and Planetary Science Letters , 1997 , vol. 152 (pg. 149 - 162 ) Google Scholar Crossref Search ADS WorldCat Frenzel G. , Muhe R. , Stoffers P. . von Stackelberg U. , von Rad U. . Petrology of the volcanic rocks from the Lau Basin Southwest Pacific , Geological Evolution and Hydrothermal Activity in the Lau and North Fiji Basins, Southwest Pacific Ocean (Results of SONNE Cruise SO–35) Geologische Jahrbuch , 1990 , vol. 92 (pg. 395 - 479 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Frey F. A. . Rare earth abundances in Alpine ultramafic rocks , Physics of Earth and Planetary Interiors , 1970 , vol. 3 (pg. 323 - 330 ) Google Scholar Crossref Search ADS WorldCat Frey F. A. , Bryan W. B. , Thompson G. . Atlantic ocean floor: geochemistry and petrology of basalts from Legs 2 and 3 of the Deep–Sea Drilling Project , Journal of Geophysical Research , 1974 , vol. 79 (pg. 5507 - 5527 ) Google Scholar Crossref Search ADS WorldCat Ghiorso M. S. , Carmichael I. S. E. , Rivers M. L. , Sack R. O. . The Gibbs free energy of mixing of natural silicate liquids; an expanded regular solution approximation for the calculation of magmatic intensive variables , Contributions to Mineralogy and Petrology , 1983 , vol. 84 (pg. 107 - 145 ) Google Scholar Crossref Search ADS WorldCat Gill J. B. . Composition and age of Lau Basin and Ridge volcanics: implications for evolution of an interarc basin and remnant arc , Geological Society of America Bulletin , 1976 , vol. 87 (pg. 1384 - 1395 ) Google Scholar Crossref Search ADS WorldCat Green D. H. . Alumina content of enstatite in a Venezuelan high temperature peridotite , Geological Society of America Bulletin , 1963 , vol. 74 (pg. 1397 - 1402 ) Google Scholar Crossref Search ADS WorldCat Green D. H. . The origin of basaltic and nephelinitic magmas , Transactions Leicester Literary and Philosophical Society , 1970 , vol. 64 (pg. 28 - 54 ) Google Scholar OpenURL Placeholder Text WorldCat Green D. H. . Orthopyroxene in the lunar interior and constraints on early lunar differentiation , Lunar Science , 1976 , vol. 7 (pg. 336 - 338 ) Google Scholar OpenURL Placeholder Text WorldCat Green D. H. , Falloon T. J. . Jackson I. . Pyrolite: a Ringwood concept and its current expression , The Earth's Mantle: Structure, Composition and Evolution—the Ringwood Volume , 1998 Cambridge Cambridge University Press (pg. 311 - 378 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Green D. H. , Ringwood A. E. . The genesis of basaltic magmas , Contributions to Mineralogy and Petrology , 1967 , vol. 15 (pg. 103 - 190 ) Google Scholar Crossref Search ADS WorldCat Green D. H. , Ringwood A. E. , Hibberson W. O. , Ware N. G. . Experimental petrology of Apollo 17 mare basalts , Proceedings of the 6th Lunar Science Conference , 1975 Houston, TX Lunar Science Institute (pg. 871 - 893 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Green D. H. , Hibberson W. O. , Jaques A. L. . McElhinny M. W. . Petrogenesis of mid–ocean ridge basalts , The Earth: its Origin, Structure and Evolution , 1979 London Academic Press (pg. 265 - 290 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Haskin L. A. , Frey F. A. . Dispersed and not–so–rare earths , Science , 1966 , vol. 152 pg. 299 Google Scholar Crossref Search ADS PubMed WorldCat Hawkins J. W. . Petrology and geochemistry of basaltic rocks of the Lau Basin , Earth and Planetary Science Letters , 1976 , vol. 28 (pg. 283 - 296 ) Google Scholar Crossref Search ADS WorldCat Hawkins J. W. . Hawkins J. , Parson L. , Allan J. , Resig J. , Weaver P. . Petrologic synthesis: Lau Basin transect (Leg 135) , Proceedings of the Ocean Drilling Program, Scientific Results, 135 , 1994 College Station, TX Ocean Drilling Program (pg. 879 - 905 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hawkins J. W. . Taylor B. , Natland J. . Evolution of the Lau Basin—insights from ODP Leg 135 , Active Margins and Marginal Basins of the Western Pacific. Geophysical Monograph, American Geophysical Union , 1995 , vol. 88 (pg. 125 - 173 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hawkins J. W. . Taylor B. . The geology of the Lau Basin , Backarc Basins: Tectonics and Magmatism , 1995 New York Plenum (pg. 63 - 138 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hawkins J. W. , Allan J. F. . Hawkins J. , Parson L. , Allan J. , Resig J. , Weaver P. . Petrologic evolution of Lau Basin sites 834 through 839 , Proceedings of the Ocean Drilling Program, Scientific Results, 135 , 1994 College Station, TX Ocean Drilling Program (pg. 427 - 470 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hawkins J. W. , Bloomer S. H. , Evans C. A. , Melchoir J. T. . Evolution of intra–oceanic arc–trench systems , Tectonophysics , 1984 , vol. 102 (pg. 174 - 205 ) Google Scholar Crossref Search ADS WorldCat Hawkins J. W. , Melchoir J. T. . Petrology of Mariana Trough and Lau Basin basalts , Journal of Geophysical Research , 1985 , vol. 90 (pg. 11431 - 11468 ) Google Scholar Crossref Search ADS WorldCat Hawkins J. W. , Parson L. M. , Allan J. F. , Resig J. , Weaver P. . , Proceedings of the Ocean Drilling Program, Scientific Results, 135 , 1994 College Station, TX Ocean Drilling Program Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Jarosewich E. , Nelen J. A. , Norberg J. A. . Electron microprobe reference samples for mineral analyses , Smithsonian Contributions to Earth Sciences , 1979 , vol. 22 (pg. 68 - 72 ) Google Scholar OpenURL Placeholder Text WorldCat Jenner G. A. , Cawood P. A. , Rautenschlein M. , White W. M. . Composition of back–arc basin volcanics, Valu Fa Ridge, Lau Basin: evidence for a slab–derived component in their mantle source , Journal of Volcanology and Geothermal Research , 1987 , vol. 32 (pg. 209 - 222 ) Google Scholar Crossref Search ADS WorldCat Kamenetsky V. S. , Crawford A. J. , Eggins S. , Muhe R. . Phenocryst and melt inclusion chemistry of near–axis seamounts, Valu Fa Ridge, Lau Basin: insight into mantle wedge melting and the addition of subduction components , Earth and Planetary Science Letters , 1997 , vol. 151 (pg. 205 - 223 ) Google Scholar Crossref Search ADS WorldCat Kelemen P. B. , Dick H. J. B. , Quick J. E. . Formation of harzburgite by pervasive melt/rock reaction in the upper mantle , Nature , 1992 , vol. 358 (pg. 635 - 641 ) Google Scholar Crossref Search ADS WorldCat Klein E. M. , Langmuir C. H. . Global correlations of ocean ridge basalt chemistry with axial depth and crustal thickness , Journal of Geophysical Research , 1987 , vol. 92 (pg. 8089 - 8115 ) Google Scholar Crossref Search ADS WorldCat Langmuir C. H. , Klein E. M. , Plank T. . Phipps Morgan J. , Blackman D. K. , Sinton J. M. . Petrological systematics of mid–ocean ridge basalts: constraints on melt generation beneath ocean ridges , Mantle Flow and Melt Generation at Mid–ocean Ridges. Geophysical Monograph, American Geophysical Union , 1992 , vol. 71 (pg. 183 - 280 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC McKenzie D. , Bickle M. J. . The volume and composition of melt generated by extension of the lithosphere , Journal of Petrology , 1988 , vol. 29 (pg. 625 - 679 ) Google Scholar Crossref Search ADS WorldCat McNeill A. W. , Danyushevsky L. V. . Alt J. C. , Kinshita H. , Stokking L. B. , Michael P. J. . Composition and crystallization temperatures of primary melts from Hole 896A basalts: evidence from melt inclusions studies , Proceedings of the Ocean Drilling Program, Scientific Results, 148 , 1996 College Station, TX Ocean Drilling Program (pg. 21 - 35 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Niu Y. , Batiza R. . Chemical variation trends at fast and slow spreading mid–ocean ridges , Journal of Geophysical Research , 1993 , vol. 98 (pg. 7887 - 7902 ) Google Scholar Crossref Search ADS WorldCat Niu Y. , Batiza R. . Magmatic processes at the Mid–Atlantic Ridge ∼26°S , Journal of Geophysical Research , 1994 , vol. 99 (pg. 19719 - 19740 ) Google Scholar Crossref Search ADS WorldCat Pearce J. A. , Ernewein M. , Bloomer S. H. , Parson L. M. , Murton B. J. , Johnson L. E. . Smellie J. L. . Geochemistry of Lau Basin volcanic rocks: influence of ridge segmentation and arc proximity , Volcanism Associated with Extension at Consuming Plate Margins. Geological Society, London, Special Publication , 1995 , vol. 81 (pg. 53 - 75 ) Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Rubin K. H. , Macdougall J. D. . 226Ra excesses in mid–ocean ridge basalts and mantle melting , Nature , 1988 , vol. 335 (pg. 158 - 161 ) Google Scholar Crossref Search ADS WorldCat Rubin K. H. , Macdougall J. D. . Dating of neovolcanic MORB using (226Ra/230Th) disequilibrium , Earth and Planetary Science Letters , 1990 , vol. 101 (pg. 313 - 322 ) Google Scholar Crossref Search ADS WorldCat Shen Y. , Forysth D. W. . Geochemical constraints on initial and final depth of melting beneath mid–ocean ridges , Journal of Geophysical Research , 1995 , vol. 100 (pg. 2211 - 2237 ) Google Scholar Crossref Search ADS WorldCat Sinton J. M. , Fryer P. . Mariana Trough lavas from 18°N: implications for the origin of back arc basin basalts , Journal of Geophysical Research , 1987 , vol. 92 (pg. 12782 - 12802 ) Google Scholar Crossref Search ADS WorldCat Sinton J. M. , Price R. C. , Johnston K. T. M. , Staudigel H. , Zindler A. . Kroenke L. W. , Eade J. V. . Petrology and geochemistry of submarine lavas from the Lau and North Fiji back–arc basins , Basin Formation Ridge Crest Processes, and Metallogenesis in the North Fiji Basin. Circum–Pacific Council for Energy and Mineral Resources, Earth Science Series, 12 , 1991 Tulsa, OK American Association of Petroleum Geologists (pg. 155 - 177 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Sun S.-S. , McDonough W. F. . Saunders A. D. , Norry M. J. . Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes , Magmatism in the Ocean Basins. Geological Society, London, Special Publication , 1989 , vol. 42 (pg. 313 - 345 ) Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Sunkel G. . Origin of petrological and geochemical variations of Lau Basin lavas (SW Pacific) , Marine Mining , 1990 , vol. 9 (pg. 205 - 234 ) Google Scholar OpenURL Placeholder Text WorldCat Taylor B. , Karner G. D. . On the evolution of marginal basins , Reviews of Geophysics , 1983 , vol. 21 (pg. 1727 - 1741 ) Google Scholar Crossref Search ADS WorldCat Taylor B. , Zellmer K. , Martinez F. , Goodliffe A. . Sea–floor spreading in the Lau back–arc basin , Earth and Planetary Science Letters , 1996 , vol. 144 (pg. 35 - 40 ) Google Scholar Crossref Search ADS WorldCat Thompson R. N. . Dispatches from the basalt front, I. Experiments , Proceedings of the Geological Association , 1984 , vol. 95 (pg. 249 - 262 ) Google Scholar Crossref Search ADS WorldCat Thompson R. N. . Phase–equilibria constraints on the genesis and magmatic evolution of oceanic basalts , Earth–Science Reviews , 1987 , vol. 24 (pg. 161 - 210 ) Google Scholar Crossref Search ADS WorldCat Tiffin D. L. . Tectonic and structural features of the Pacific/Indo–Australian plate boundary in the North Fiji–Lau Basin regions, southwest Pacific , Geo–Marine Letters , 1993 , vol. 13 (pg. 126 - 131 ) Google Scholar Crossref Search ADS WorldCat Walter M. J. , Presnall D. C. . Melting behavior of simplified lherzolite in the system CaO–MgO–Al2O3–SiO2–Na2O from 7 to 35 kbar , Journal of Petrology , 1994 , vol. 35 (pg. 329 - 360 ) Google Scholar Crossref Search ADS WorldCat © Oxford University Press 1999 © Oxford University Press 1999
TI - Refractory Magmas in Back-Arc Basin Settings—Experimental Constraints on the Petrogenesis of a Lau Basin Example
JF - Journal of Petrology
DO - 10.1093/petroj/40.2.255
DA - 1999-02-01
UR - https://www.deepdyve.com/lp/oxford-university-press/refractory-magmas-in-back-arc-basin-settings-experimental-constraints-0CbqwYUCU0
SP - 255
EP - 277
VL - 40
IS - 2
DP - DeepDyve
ER -