Journal of Petrology

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (30) Request Permissions Google Scholar Articles by BAKER, J. A. Articles by MATTEY, D. P. Search for Related Content GeoRef GeoRef Citation Social Bookmarking          What's this?

Journal of Petrology Volume 41 Number 12 Pages 1805-1820 2000
© Oxford University Press 2000

Resolving Crustal and Mantle Contributions to Continental Flood Volcanism, Yemen; Constraints from Mineral Oxygen Isotope Data

J. A. BAKER^1,*, C. G. MACPHERSON¹, M. A. MENZIES¹, M. F. THIRLWALL¹, M. AL-KADASI² and D. P. MATTEY¹

¹DEPARTMENT OF GEOLOGY, ROYAL HOLLOWAY UNIVERSITY OF LONDON, EGHAM HILL, EGHAM TW20 0EX, UK
²DEPARTMENT OF GEOLOGY, UNIVERSITY OF SANA’A, SANA’A, YEMEN

Received January 28, 1999; Revised typescript accepted May 2, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION SAMPLES ANALYTICAL TECHNIQUES RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Oxygen isotope ratios determined by laser fluorination analysis on olivine, clinopyroxene and plagioclase separated from 31 Oligocene flood basalts and rhyolites from Yemen display small but significant variations (5·1–6·2 for olivine; 5·5–6·9 for clinopyroxene; 5·9–6·9 for plagioclase). The range in ¹⁸O values exceeds: (1) the analytical reproducibility of the technique (±0·15; 2 SD); (2) the range expected for minerals that would have crystallized from uncontaminated oceanic basalts or primary magmas in equilibrium with mantle peridotite; (3) the range in melt values and equilibrium phenocryst compositions that could be produced by fractional crystallization of these magmas. Samples with the highest ¹⁸O values exhibit increases in ⁸⁷Sr/⁸⁶Sr ratio, decreases in ¹⁴³Nd/¹⁴⁴Nd ratio, and increasing Pb isotopic heterogeneity. Samples with the lowest ¹⁸O values have radiogenic isotope ratios that approach those inferred for the Afar plume. The oxygen isotope data provide unequivocal evidence that assimilation of heterogeneous lower and upper Pan-African crust was the primary control on isotopic variation in this continental flood basalt province. Moreover, new radiogenic and oxygen isotope data for Pan-African crustal samples from Yemen have appropriate crustal isotopic compositions to generate the observed isotopic variations in the volcanic rocks. A near-primary high-MgO basalt with low Nd and extreme Pb isotope ratios contains strongly zoned clinopyroxene crystals that range from green cores through to greenish brown, brownish green and dark brown or black rims. Handpicked crystals of each colour type display the following correlated range in isotope ratios: ⁸⁷Sr/⁸⁶Sr = 0·7036–0·7049; ¹⁴³Nd/¹⁴⁴Nd = 0·5129–0·5127; ²⁰⁶Pb/²⁰⁴Pb = 18·6–17·9; ¹⁸O = 5·67–6·86. The Sr–Nd–Pb–O isotope variations are attributed to rapid assimilation of 25% Pan-African continental crust by hot mafic magma during clinopyroxene crystallization. Contamination in this flood basalt province varied from combined assimilation and fractional crystallization to rapid assimilation of crust by hot mafic magmas with little fractionation. Laser fluorination oxygen isotope analysis of mineral separates allows small differences in ¹⁸O to be correlated with radiogenic isotope data and is a powerful tool for evaluating the relative roles of enriched lithospheric mantle and continental crust in suites of continental flood basalts.

KEY WORDS: oxygen isotopes; phenocrysts; flood volcanism; crustal contamination; Yemen

	INTRODUCTION

TOP ABSTRACT INTRODUCTION SAMPLES ANALYTICAL TECHNIQUES RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Continental flood volcanism has in the past produced massive (>1 x 10⁶ km³) outpourings of largely basaltic magmas in only a few millions of years (Coffin & Eldholm, 1993). The large size of these volcanic episodes, coupled with the rapid rate at which they are produced, has led to the inference that they are directly related to mantle geodynamics, i.e. deep-seated mantle plumes (e.g. White & McKenzie, 1989). Moreover, the intense volcanic activity associated with continental flood volcanism may have affected the atmosphere and hydrosphere leading to abrupt mass extinction events (e.g. Officer & Drake, 1983).

Continental flood basalts exhibit a wide range in trace element and isotopic compositions and, whereas some flood basalts clearly have chemical and isotopic signatures akin to oceanic basalts, many flood basalts have signatures that are more typical of continental crust (e.g. Ferrar; Hergt et al., 1989). Even examples of flood volcanism that have isotope ratios that fall entirely within the range of oceanic basalts display substantial isotopic variability that must represent multiple contributions to volcanism [e.g. Yemen; Fig. 1 (Baker et al., 1996b)]. There is general agreement that the enriched isotope ratios (e.g. low Nd) and lithospheric trace element signatures (e.g. high Rb, K, Ba, Pb and La) of many flood basalts, which distinguishes them from oceanic basalts, must reflect a contribution from the continental lithosphere traversed by these magmas. To date, those who study flood basalt provinces have failed to reach any consensus on the general origin of this lithospheric signature, although it needs to be stressed that the source of this lithospheric signature need not be the same in each province. However, recent studies have reached directly conflicting conclusions on the major control on isotopic variability in many individual examples of continental flood volcanism, e.g. Columbia River Basalt Group (Carlson et al., 1981; Carlson, 1984; Brandon et al., 1993; Hooper & Hawkesworth, 1993), Ethiopian–Yemeni Traps (Chazot & Bertrand, 1993; Deniel et al., 1994; Baker et al., 1996b; Pik et al., 1999); Deccan Traps (Lightfoot et al., 1990a; Peng et al., 1994); Siberian Traps (Lightfoot et al., 1990b, 1993; Arndt et al., 1993; Wooden et al., 1993; Horan et al., 1995).

View larger version (39K): [in this window] [in a new window]   Fig. 1. Sr–Nd isotope variations in Oligocene flood volcanic rocks from Yemen compared with local lithospheric mantle (LM), plume (Afar plume) and depleted upper-mantle (MORB) sources, and global oceanic basalt compositions. Sources of data: MORB–Afar plume, Schilling et al. (1992) and Volker et al. (1993); Afar plume, Vidal et al. (1991) and Schilling et al. (1992); LM, Henjes-Kunst et al. (1990), Blusztajn et al. (1995) and Baker et al. (1998); Yemen flood volcanism (herein) and additional data from Baker (1996) and Baker et al. (1996b); global oceanic basalts (Hofmann, 1997).

 

Proponents of crustal contamination models argue that crustal contamination is inevitable during the storage and transfer of mantle-derived melts through the lithosphere and, moreover, that the lithospheric mantle (LM) is too cold and refractory and does not have the appropriate composition to generate large volumes of flood basalt (Arndt & Christensen, 1992; Menzies, 1992). However, others have argued that addition of small amounts of volatiles to the LM can depress solidus temperatures, allowing large amounts of melting in the presence of a thermal anomaly when coupled with lithospheric extension (Gallagher & Hawkesworth, 1992). Presumably, hydration and introduction of subducted sediment into the LM by ancient subduction zone processes produces a crustal-like signature in the LM. Recent studies have also suggested that plume-derived carbonatitic metasomatism of the LM could impart a similar signature (Hauri et al., 1993; Baker et al., 1998).

What remains beyond question is that only when the primary chemical and isotopic signature of flood volcanism has been identified is it possible to test rigorously and develop models for flood basalt petrogenesis. Using primary magma compositions to identify and temporal and spatial variability in mantle source composition(s) and the degree(s) and depth(s) of partial melting is a powerful tool for testing and developing geochemical and geodynamic models for flood basalt genesis (e.g. White & McKenzie, 1989, 1995; Ellam, 1992; Gallagher & Hawkesworth, 1992).

Identification of crustal contamination typically relies on identification of correlations between indices of fractionation and chemical or isotopic data. For example, correlations between MgO (negative) or SiO₂ (positive) contents with ⁸⁷Sr/⁸⁶Sr ratios are taken as evidence for progressive contamination of magmas while fractionation was taking place in lithospheric magma chambers. In principle, however, this approach has a number of limitations: (1) such correlations are not always observed when reconnaissance suites of samples are considered that do not represent a coherent magmatic system (e.g. Baker et al., 1996b); (2) similar correlations can be produced during fractionation of magmas within the LM, accompanied by a progressive addition of melts from enriched LM (Ellam & Cox, 1991); (3) contamination processes are thought to vary considerably from combined assimilation and fractional crystallization (AFC; DePaolo, 1981) through to assimilation of crust by hot mafic magmas with little or no concomitant fractionation (Thirlwall & Jones, 1983; Cox & Hawkesworth, 1985; Huppert & Sparks, 1985; Devey & Cox, 1987; Kerr et al., 1995); (4) MgO and SiO₂ are not always useful indicators of fractionation, e.g. only small changes in MgO are generated by fractionation of a plagioclase-dominated gabbroic assemblage.

The now well-defined and generally restricted oxygen isotope ratios of crustally uncontaminated oceanic basalts, and mantle peridotites (Ito et al., 1987; Mattey et al., 1994; Eiler et al., 1997), provide a powerful and unequivocal method to test the relative roles of crust and LM in the petrogenesis of continental flood basalts. Assimilation of crustal materials that typically, but not always, have elevated ¹⁸O relative to mantle-derived magmas should generate ¹⁸O enrichment in the contaminated basalts with predictable correlated changes in radiogenic isotope ratios. To date, oxygen isotope study of continental flood basalts has been hampered by the often altered nature of these rocks and, in some cases, their sparsely porphyritic nature. Susceptibility of the oxygen isotope system to post-eruptive alteration largely precludes the use of whole-rock oxygen isotope data in evaluating the relative roles of crust and mantle in flood basalts (e.g. Hergt et al., 1989) and perhaps volcanic rocks in general (e.g. Eiler et al., 1995, 1997).

Herein, we present laser fluorination oxygen isotope data on milligram-sized phenocrysts separated from flood basalts and rhyolites that were erupted between 31 and 26 Ma in Yemen, which form part of the Oligocene Ethiopian flood basalt province (Baker et al., 1996a). These rocks have radiogenic isotope ratios that extend out of the fields defined by local LM, mantle plume (Afar plume) and depleted upper mantle [DMM or mid-ocean ridge basalt (MORB) mantle] compositions, although the isotopic variation is less than that exhibited by oceanic basalts (Fig. 1). Baker et al. (1996b) concluded that correlations between MgO and SiO₂ and isotope data, the probable availability of appropriate crustal components, and unavailability of appropriate enriched LM components required much of the chemical and isotopic variability in the Yemen flood basalts to result from contamination of plume-derived magmas with regionally heterogeneous crustal components. Whereas the evidence presented by Baker et al. (1996b) supporting crustal contamination is largely circumstantial and equivocal, the mineral oxygen isotope data presented herein show substantial variability that exceeds, and extends to higher values than, that observed in oceanic basalts and mantle peridotites. Thus, the oxygen isotope data provide unequivocal evidence for crustal contamination in this flood basalt province and also confirm that contamination mechanisms were variable. We also present new radiogenic and oxygen isotope data for crustal basement samples from Yemen that, in some cases, have precisely the compositions required to generate some of the mixing arrays exhibited by the volcanic rocks in Sr–Nd–Pb–O isotopic space.

	SAMPLES

TOP ABSTRACT INTRODUCTION SAMPLES ANALYTICAL TECHNIQUES RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
The samples that form the basis of this study are flood volcanic rocks that were erupted from 31 to 29 Ma in Yemen (Baker et al., 1996a) and form part of the Ethiopian–Yemeni large igneous province. The volcanic rocks comprise a diverse variety of rock types, including a spectrum of basaltic compositions (basanite, trachybasalt, basalt, basaltic andesite and basaltic trachyandesite) and also rhyolitic rocks (Baker, 1996). Mafic samples are variably olivine + clinopyroxene phyric and more evolved mafic samples may also be plagioclase ± Fe–Ti oxide phyric. Rhyolitic samples contain phenocrysts of sodic clinopyroxene + alkali feldspar + Fe–Ti oxide ± amphibole ± quartz.

The volcanic rocks can be divided into two suites on the basis of field relationships, chemistry and isotope data. Rocks erupted through the western part of the province display limited and distinct isotopic variations compared with those erupted through the eastern part of the province (Baker, 1996; Baker et al., 1996b; see the ‘Origin of isotopic variations’ section, below) and, as such, this is used to separate the samples into two groups in this paper. Rhyolitic pyroclastic rocks were also erupted from the western part of the province at sites now marked by the presence of granite plutons, which are unroofed caldera centres exhumed in the last 25 my.

	ANALYTICAL TECHNIQUES

TOP ABSTRACT INTRODUCTION SAMPLES ANALYTICAL TECHNIQUES RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Oxygen isotope analyses
Approximately 1–2 mg of clean olivine, clinopyroxene, feldspar (where no mafic silicates were available for analysis) or Fe–Ti oxide phenocrysts were handpicked from coarsely crushed whole-rock volcanic samples to avoid inclusion-rich and altered material. One mafic flood basalt (JB172) contains strongly colour-zoned clinopyroxene crystals that vary from apple-green cores, through slightly brownish green, greenish brown and pale brown zones to black or dark brown crystal rims. The large size of these crystals (1–3 mm) allowed the O–Sr–Nd isotope ratios of the differently coloured zones in this sample to be determined, along with the Pb isotope ratios of the core material. Bulk separates of each colour fraction of the clinopyroxene from JB172 were prepared and homogenized, and these were then split into portions for radiogenic isotope analysis and replicate oxygen isotope analyses. A number of mineral phases (amphibole, biotite, quartz, garnet) were separated from two Pan-African crustal samples for oxygen isotope analysis and, in one case, a fine-grained chip of a whole-rock sample (plagioclase + amphibole) was also analysed.

Oxygen isotope analyses were carried out using the laser fluorination technique of Mattey & Macpherson (1993) at Royal Holloway University of London. Mineral separates were heated with a Nd–YAG laser in the presence of ClF₃. To optimize oxygen yield, laser power and the quantity of ClF₃ used during the fluorination were varied according to the phase being analysed. Feldspar, which is transparent to the Nd–YAG laser, was mixed with a mafic silicate phase of known oxygen isotope composition (San Carlos mantle olivine; SCOL) to assist fluorination reactions. Using a simple mixing equation involving the relative weight proportions of feldspar to olivine in the reaction mixture, their respective oxygen concentrations, the oxygen isotope ratio of the olivine and the oxygen isotope ratio of the mixture, the isotopic composition of the feldspar could be devolved. After clean-up with KBr, the liberated O₂ was converted to CO₂ by reaction with hot graphite. CO₂ was analysed on a VG PRISM mass spectrometer. Oxygen isotope ratios are reported as the per mil deviation from Vienna Standard Mean Ocean Water in the standard notation.

Analytical precision can be judged from the mean values for in-house olivine (SCOL) and garnet (Beni Bousera peridotite massif; GP143) standards during the period of analysis: SCOL = 4·90 ± 0·15 (n = 8; 2 SD); GP143 = 7·23 ± 0·20 (n = 19; 2 SD). During the period of analysis, and in the 2 years before this, the following values were determined on these two in-house standards: SCOL = 4·86 ± 0·16 (n = 245; 2 SD); GP143 = 7·21 ± 0·20 (n = 106; 2 SD). Replicate analyses of standards or unknowns on the same day typically show enhanced reproducibility (<±0·1; 2 SD) compared with these long-term averages. As such, reproducibility of ¹⁸O values for analysed mafic silicates with fluorination yields >95% can be taken to be better than ±0·15 (2 SD). Yields for all unknowns and standards were 100 ± 2%.

Feldspar oxygen isotope analyses are inherently less precise than those of mafic phases because of the need to mix samples with a mafic silicate to induce fluorination reactions. However, the reliability of the feldspar analyses can be tested by replicate analyses of NBS28 quartz samples by the same technique, fluxing with the SCOL standard, which yielded a mean value of 9·57 ± 0·36 (n = 17; 2 SD; D. Lowry, unpublished data, 1996); this compares favourably with the recommended value for this NBS standard, i.e. 9.66. NBS-30 biotite standard analysed in this laboratory gives ¹⁸O = 5·03 ± 0·16 (2 SD; n = 80) compared with the recommended value of 5·1.

Sr–Nd–Pb isotope analyses
Details of the analytical techniques used to obtain Sr–Nd–Pb isotope ratios for host basaltic samples (MgO > 3 wt %), from which minerals were separated for oxygen isotope analyses, have been presented by Baker et al. (1996b). Sr–Nd–Pb isotope ratios were determined on whole-rock powders and are reported here as initial ratios; these isotopic data, along with measured isotope ratios and chemical data for the same samples, were presented by Baker et al. (1996b). Sr–Nd–Pb isotope ratios for the rhyolitic samples were obtained on acid-leached mineral separates [anorthoclase–sanidine (Sr–Pb), clinopyroxene (Sr–Nd) and amphibole (Sr–Nd)]. The feldspar Pb isotope data can be considered to be initial data given the low U/Pb and Th/Pb ratio of feldspar. However, all the remaining radiogenic isotope data presented herein for volcanic samples have been age-corrected using ⁴⁰Ar/³⁹Ar ages (31–26 Ma) from Baker et al. (1996a) and trace element data from Baker (1996) and Baker et al. (1996b), including Rb–Sr isotope dilution data determined on the rhyolitic mineral separates.

Sr–Nd–Pb isotope data on the zoned clinopyroxene phenocrysts from mafic flood basalt JB172 were determined on bulk separates (0·02–0·1 g) on splits of similarly coloured parts of the zoned phenocrysts as used for oxygen isotope analysis. Clinopyroxene was leached in hot HCl and repeatedly rinsed in deionized water before dissolution, conventional chemical separation and subsequent mass spectrometric analysis as described by Baker et al. (1996b). Only sufficient material from the cores was available to perform Pb isotope analyses. To assess isotopic homogeneity between the same-coloured splits used for oxygen and radiogenic isotope analyses, the fluoride residues of two differently coloured parts of the zoned clinopyroxene crystal were recovered after oxygen isotope analysis and analysed for Sr–Nd isotopes. The measured Sr–Nd isotope ratios were within analytical uncertainty of those determined on the bulk separates of the similarly coloured parts of the zoned clinopyroxene phenocrysts, implying that the colour fractions used for radiogenic and oxygen isotope analysis are isotopically homogeneous on the scale of the analytical procedures used in this study.

New Sr–Nd–Pb isotope ratios have also been determined on five crustal basement samples (whole-rock powders) that were collected from the eastern (two samples) and western (three samples) parts of Yemen.

External precision and reproducibility of Sr–Nd isotope ratios are better than ±0·000018 and ±0·000013 (n > 50; 2 SD), and ratios are reported relative to values of 0·710250 for SRM987 and 0·511424 for an in-house laboratory Nd standard. This ¹⁴³Nd/¹⁴⁴Nd ratio corresponds to 0·511860 and 0·512638 for the international standards La Jolla and BCR-1, respectively. External precision or reproducibility of SRM981 shows that the reproducibility of sample Pb isotope ratios is ±0·010, ±0·012 and ±0·030 (2 SD) for ²⁰⁶Pb/²⁰⁴Pb, ²⁰⁷Pb/²⁰⁴Pb and ²⁰⁸Pb/²⁰⁴Pb, respectively.

	RESULTS

TOP ABSTRACT INTRODUCTION SAMPLES ANALYTICAL TECHNIQUES RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Oxygen isotope analyses of 66 mineral separates from 31 volcanic samples are presented in Tables 1 and 2. Table 1 lists mineral oxygen isotope data, host volcanic rock MgO contents and initial Sr–Nd–Pb isotope ratios, along with five radiogenic and limited mineral oxygen isotope analyses of Pan-African crustal basement samples from Yemen. Table 2 lists O–Sr–Nd–Pb isotope data for differently coloured parts of the strongly zoned clinopyroxene phenocrysts from mafic flood basalt (JB172), along with oxygen isotope data for olivine phenocrysts from the same rock.

View this table: [in this window] [in a new window]   Table 1: O–Sr–Nd–Pb isotope data for mineral phenocrysts (O) and host flood volcanic rocks (Sr–Nd–Pb), and crustal basement (Sr–Nd–Pb–O) samples

 

View this table: [in this window] [in a new window]   Table 2: O–Sr–Nd–Pb isotope data for strongly zoned clinopyroxene phenocrysts and O isotope data for olivine phenocrysts from a mafic flood basalt (JB172)

 

Oxygen isotope data
Olivine ¹⁸O values vary from 5·1 to 6·2 and a number of samples have values (5·1–5·3) that overlap the range defined by olivine separated from (1) a wide variety of mantle peridotite xenoliths, largely from the subcontinental LM (5·18 ± 0·28; n = 76; Mattey et al., 1994), and (2) MORB and ocean island basalts (OIB) (for MORB, 5·16 ± 0·18, n = 6; for OIB, 5·17 ± 0·49, n = 62; Eiler et al., 1997) (Fig. 2).

View larger version (35K): [in this window] [in a new window]   Fig. 2. Raw 18O mineral data vs host rock MgO contents. The fields for olivine and clinopyroxene separated from mantle peridotites (1 SD; Mattey et al. (1994) are also shown. The diamonds show the hypothetical effects of fractional crystallization on melt and equilibrium clinopyroxene 18O values using the modelling described in Table 3.

 

View this table: [in this window] [in a new window]   Table 3: Modelling the hypothetical effects of fractional crystallization on evolving melt and equilibrium clinopyroxene phenocryst oxygen isotope compositions

 
Clinopyroxene ¹⁸O values also vary widely (5·5–6·9) and, again, a number of samples from the lower end of this range have ¹⁸O values that closely overlap the range defined by clinopyroxenes from mantle peridotites (5·57 ± 0·32; n = 57; Mattey et al., 1994). Four clinopyroxene separates from rhyolitic volcanic rocks have restricted ¹⁸O values of 5·9–6·1, which are actually lower than those observed in some of the basaltic samples. The differently coloured parts of the zoned clinopyroxene crystals separated from mafic flood basalt JB172 have highly variable ¹⁸O values (5·67–6·86) and this sample is discussed in more detail in the next section.

Plagioclase ¹⁸O values range from 5·9 to 6·9. As a result of the scarcity of plagioclase-facies mantle rocks, no mantle feldspar ¹⁸O values are available. However, it is notable, given the relatively small oxygen isotopic fractionation between melt and plagioclase at magmatic temperatures (_melt-plag = -0·1 to -0·3; Kalamarides, 1986; Kyser, 1986, 1990), that the lower feldspar ¹⁸O values (6·0 ± 0·1) would be in equilibrium with melts that have ¹⁸O 5·7–5·8 (Table 1). Such melt ¹⁸O values are characteristic of fresh MORB glasses (5·71 ± 0·34, Ito et al., 1987; 5·81 ± 0·10, Macpherson, 1995).

Oxygen isotope fractionations observed between different mineral phases separated from the same rock are: _ol-cpx = -0·20 to -0·35; _ol-plag = -0·86; _mt-plag = -1·86. These mineral–mineral fractionations are consistent with those measured in studies of phenocrysts from other volcanic rocks (Anderson et al., 1971; Singer et al., 1992; Macpherson, 1995; Eiler et al., 1997; Thirlwall et al., 1997; Macpherson et al., 1998). These fractionations are also broadly consistent with the mineral phases being in equilibrium at magmatic temperatures (1200–1000°C; Chiba et al., 1989; Zheng, 1993).

Figure 2 shows raw mineral ¹⁸O values plotted vs host rock MgO contents. A very general trend to higher mineral ¹⁸O values, of each mineral type, with decreasing host rock MgO may be evident. However, more importantly, there is much scatter in these possible trends and samples with any MgO contents may have elevated mineral ¹⁸O values; for example, olivine and clinopyroxene from JB172, clinopyroxene from JB82 and olivine from JB282.

In summary, ranges in ¹⁸O values for each phase (1) exceed the maximum analytical reproducibility of the technique (±0·15) and (2) extend to higher values than those recorded in either minerals (olivine and clinopyroxene) separated from crustally uncontaminated mantle-derived oceanic basalts (Eiler et al., 1997) and mantle peridotites (Mattey et al., 1994) or mineral compositions (plagioclase) that can be calculated to have been in equilibrium with MORB melts. The lowest ¹⁸O values obtained for each mineral phase overlap values for these phases in oceanic basalts, mantle peridotite xenoliths, or values that would characterize minerals that crystallized at basaltic temperatures from melts in equilibrium with mantle peridotites. These observations are independent of corrections to the raw mineral ¹⁸O values described below (in ‘Correlations between mineral oxygen and host rock radiogenic data’) to convert them to values equivalent to clinopyroxene or melt compositions so that all the data can be collectively compared with whole-rock Sr–Nd–Pb isotope ratios.

Oxygen and radiogenic isotope variations in a mafic flood basalt (JB172)
Sample JB172 is a magnesian basalt containing 10·95 wt % MgO (mg-number 0·68). Its whole-rock powder is characterized by high ⁸⁷Sr/⁸⁶Sr (0·70467) and low ¹⁴³Nd/¹⁴⁴Nd (0·51272) ratios, and the lowest ²⁰⁶Pb/²⁰⁴Pb ratio (17·9) and almost the highest 7/4 (+14·3) and 8/4 (+101·9) values of the flood volcanic samples from Yemen [ notation from Hart (1984)].

From core to rim the ¹⁸O values of the clinopyroxene increase systematically from 5·7 to 6·9 (Table 2; Fig. 3), and ¹⁸O correlates with changes in ⁸⁷Sr/⁸⁶Sr and ¹⁴³Nd/¹⁴⁴Nd ratios (Fig. 3). The cores of the crystals have low ⁸⁷Sr/⁸⁶Sr (0·70356) and high ¹⁴³Nd/¹⁴⁴Nd (0·51290) ratios, which are amongst the most depleted compositions measured for the western Yemen flood basalts, and similar to those inferred for the Afar plume (Vidal et al., 1991; Schilling et al., 1992; Baker et al., 1996b, 1997). The rims have high ⁸⁷Sr/⁸⁶Sr (0·70487) and low ¹⁴³Nd/¹⁴⁴Nd (0·51273) ratios that are similar to values measured on the whole-rock powder, where the analysed Sr and Nd would largely have been derived from the groundmass, which is evidently in equilibrium with the clinopyroxene rims. Pb isotope ratios of the green clinopyroxene cores are completely different from the measured whole-rock ratios (Table 2, footnote), falling close to the inferred Afar plume composition (Figs 4 and 5), and are similar to other whole-rock samples with low ⁸⁷Sr/⁸⁶Sr and high ¹⁴³Nd/¹⁴⁴Nd ratios.

View larger version (19K): [in this window] [in a new window]   Fig. 3. Sr–Nd–O isotope variations in strongly zoned clinopyroxene phenocrysts separated from a mafic flood basalt (JB172). Also annotated in these figures are data for two flows immediately above and below JB172 (JB171 and JB166) and the inferred fields for the Afar plume. Mixing arrays are bulk mixing curves between components with: (1) 87Sr/86Sr = 0·7035, 143Nd/144Nd = 0·5129, Sr = 476 ppm, Nd = 32 ppm, 18O = 5·6 and (2) 87Sr/86Sr = 0·7086, 143Nd/144Nd = 0·5120, Sr = 476 ppm, Nd = 26 ppm, 18O = 10·6. Component (1) is inferred to be that of a primary magma produced from the Afar plume before mixing with the high-18O component (2), and uses the isotope ratios inferred for the Afar plume (same references as for Fig. 1) and trace element concentrations measured for JB172 in the modelling calculations (Baker et al., 1996b). Component (2) is inferred to be continental crust and uses radiogenic isotopic ratios from an average of the eastern crustal basement samples from Yemen. The whole-rock oxygen isotope ratio for this sample was inferred to be similar to an analysis of a similar sample of Archaean gneiss from Sudan (Davidson & Wilson, 1989). Mineral oxygen isotope ratios determined herein on Yemen crust suggest that non-modal melting of Yemen gneiss F27 (i.e. biotite being preferentially melted) would also have resulted in addition of crustal melts with 18O = 10–11 to the basaltic magmas.

 

View larger version (20K): [in this window] [in a new window]   Fig. 4. 18O values (relative to clinopyroxene values) vs host rock Sr–Nd–Pb isotope ratios. Field for the Afar plume taken from references cited for Fig. 1 and assuming ‘normal’ mantle clinopyroxene 18O values (Mattey et al., (1994). Symbols in this figure reflect geographical groupings of samples not mineral type: , Yemen flood basalts erupted through the western part of the volcanic province; , rhyolites erupted through the western part of the volcanic province; , flood basalts erupted through the eastern part of the volcanic province. 8/4 values for analysed crustal lithologies (Table 1) are illustrated in the Pb–O isotope plot. Crustal 8/4 values from herein (Table 1), G. Chazot & J. A. Baker (unpublished data, 1998) and Brueckner et al. (1995)).

 

View larger version (25K): [in this window] [in a new window]   Fig. 5. Sr–Nd–Pb isotope variations in Oligocene flood volcanic rocks from Yemen compared with local lithospheric mantle, plume (Afar plume) and depleted upper-mantle (MORB) and continental crust compositions. Same symbols as for Fig. 4; •, analyses of continental crust. Sources of data: same as for Fig. 1, but with additional data for continental crust from herein (Table 1), G. Chazot & J. A. Baker (unpublished data, 1998) and Brueckner et al. (1995).

 

Olivine phenocrysts, which on petrographic grounds crystallized after the green clinopyroxene cores but before the clinopyroxene rims, have ¹⁸O 6·1. With our technique it is impossible to assess if there is any zonal heterogeneity with respect to Sr–Nd–Pb–O isotope ratios in the olivine crystals, but olivine ¹⁸O values of 6·1 would be in equilibrium with clinopyroxenes with ¹⁸O 6·4–6·5. This is intermediate between the values determined for the end-member core and rim zones of the clinopyroxene and in agreement with the petrographic evidence for the order of crystallization of these phases.

The correlated changes in Sr–Nd–O isotope ratios from core to rim in clinopyroxene crystals from JB172 suggest that a magnesian, near-primary magma with depleted Sr–Nd isotope ratios and Pb isotope ratios approaching those of the Afar plume, and with mantle-like ¹⁸O values, progressively, and rapidly, acquired more enriched Sr Nd isotope ratios and higher ¹⁸O during clinopyroxene crystallization.

Interestingly, two mafic flows intercalated with the flow from which JB172 was sampled are offset to slightly higher ⁸⁷Sr/⁸⁶Sr and lower ¹⁴³Nd/¹⁴⁴Nd ratios than the Sr–O and Nd–O mixing arrays defined by the JB172 analyses (Fig. 3). In these rocks (JB166 and JB171) the radiogenic isotope ratios were determined on whole-rock powders and the oxygen isotope ratios were measured on large clinopyroxene phenocrysts. It is therefore possible that these two samples, like JB172, also record some isotopic disequilibrium between phenocryst phases and groundmass, i.e. the clinopyroxenes are isotopically zoned. Unfortunately, the smaller size of the clinopyroxene phenocrysts and lack of colour zoning compared with the JB172 clinopyroxenes precluded further investigation of this hypothesis, in the absence of in situ analytical techniques.

Correlations between mineral oxygen and host rock radiogenic data
The different phenocryst phases present in the volcanic rocks precluded analysis of a common mineral phase in every sample. To compare the oxygen isotope ratios from samples with different analysed phases we have chosen to correct our mineral data to values that are equivalent to clinopyroxene values. We have taken this approach as: (1) clinopyroxene is the phase we have the most data for; (2) conversion to melt ¹⁸O values is precluded by a lack of knowledge of mineral–melt fractionation factors and adequate thermometric data for the Yemen flood basalts; (3) although melt ¹⁸O values are likely to increase during fractionation of olivine + clinopyroxene ± plagioclase ± Fe–Ti oxides, this is offset by increasing _melt-cpx fractionation at lower melt temperatures (e.g. Kalamarides, 1986; Fig. 2; Table 3).

The following mineral fractionation factors were used to convert the raw olivine, plagioclase and Fe–Ti oxide oxygen isotope data into equilibrium clinopyroxene isotope data: _cpx-ol = +0·4; _cpx-plag = -0·35; _cpx-mt = +1·6. These fractionation factors are consistent with those measured in basaltic and andesitic rocks (Anderson et al., 1971; Singer et al., 1992; Macpherson, 1995; Thirlwall et al., 1997; Macpherson et al., 1998) and also the fractionations predicted from experimental work and theoretical considerations between these phases at magmatic temperatures (1000–1200°C; Chiba et al., 1989; Zheng, 1993).

Actual or calculated equilibrium clinopyroxene ¹⁸O values, with JB172 represented by the analysis of crystal rims, are plotted vs host rock Sr–Nd–Pb isotope ratios (Fig. 4). Although there is considerable scatter in these plots, ¹⁸O values increase with: (1) increasing ⁸⁷Sr/⁸⁶Sr ratios and also with little change in ⁸⁷Sr/⁸⁶Sr; (2) decreasing ¹⁴³Nd/¹⁴⁴Nd ratios; (3) high 8/4 values (such as JB172), and also with little change in 8/4 (such as JB282), and possibly at low 8/4 values (such as JB259) (i.e. increasing Pb isotopic heterogeneity). When the higher ¹⁸O samples are considered, it is notable that the samples erupted through the western part of the volcanic province display more limited Nd isotopic variations and trend towards different Pb isotopic compositions compared with those erupted through the eastern part of the volcanic province.

Samples with the lowest ¹⁸O values tend to approach isotope ratios of ⁸⁷Sr/⁸⁶Sr = 0·7035, ¹⁴³Nd/¹⁴⁴Nd = 0·5129, ²⁰⁶Pb/²⁰⁴Pb = 19·0 and 8/4 40, which are similar to those inferred for the Afar plume (Fig. 4).

Clinopyroxene separates from the rhyolite samples fall on the same trends as the basaltic samples erupted through the western part of the volcanic province, but are displaced to higher ⁸⁷Sr/⁸⁶Sr ratios at the same ¹⁸O values compared with most of these samples. Like all the basaltic samples erupted through the western part of the volcanic province, the rhyolites exhibit rather limited radiogenic isotope variations compared with basalts erupted through the eastern part of the volcanic province.

Crustal Sr–Nd–Pb–O isotope data
Sr–Nd–Pb isotope ratios for five crustal basement samples display substantial variations (Table 1; Fig. 5): ⁸⁷Sr/⁸⁶Sr = 0·704–0·710; ¹⁴³Nd/¹⁴⁴Nd = 0·51286–0·5119; ²⁰⁶Pb/²⁰⁴Pb = 17·3–19·9. There is a clear distinction between the basement underlying the western and eastern parts of the volcanic province. Samples from the west have limited isotopic variability and are akin to juvenile, Late Proterozoic, Pan-African basement from elsewhere in the Arabian shield (e.g. Duyvermann et al., 1982; McGuire & Stern, 1993; Brueckner et al., 1995). Samples from basement underlying the eastern part of the volcanic province display more extreme isotope ratios and are examples of the Late Archaean crust that has been recently identified in this part of Yemen (Windley et al., 1996).

Oxygen isotope data for two crustal samples from these two crustal provinces also exhibit marked differences. A sample from the western part of the volcanic province has relatively low ¹⁸O, comparable with values of the flood basalts. However, an Archaean(?) gneiss from the eastern part of the province has much higher ¹⁸O (9·6–13·3) for all its mineral phases compared with the flood basalts. These values are almost identical to whole-rock ¹⁸O values reported by Davidson & Wilson (1989) for similar Archaean gneisses from Sudan.

	DISCUSSION

TOP ABSTRACT INTRODUCTION SAMPLES ANALYTICAL TECHNIQUES RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Origin of isotopic variations
Effects of magma fractionation
Few of the Yemen flood basalts and clearly none of the rhyolites are primary magmas. Petrographic observations along with qualitative and quantitative modelling of fractional crystallization processes (Baker, 1996) indicate that the main fractionating phases in the basaltic rocks (MgO > 7 wt %) were olivine + clinopyroxene with minor amounts of plagioclase and Fe–Ti oxides also joining the fractionating assemblage at lower MgO contents (MgO = 7–3 wt %). If the rhyolites are considered the product of fractionation from basaltic parents then their very low TiO₂, P₂O₅, Sr, Ba and V contents require further extensive fractionation of feldspar, Fe–Ti oxides and apatite (Chazot & Bertrand, 1995; Baker, 1996).

Modelling the effects of fractional crystallization on melt and equilibrium mineral oxygen isotope compositions is non-trivial, given the relatively poor constraints on mineral–melt fractionation factors and melt temperatures, and the fact that mineral–melt fractionation factors are temperature dependent and so will change in an evolving magma. However, in Table 3 we crudely illustrate the effects of fractional crystallization for three crystallization steps with changing crystallizing assemblages and mineral–melt fractionation factors in each step (MgO = 13·5–8·5, 8·5–4·0, 4·0–0·5 wt %). The modelling uses a starting magma with ¹⁸O = 5·70 and calculates an average mineral–melt fractionation factor over the crystallization interval using the relative proportions of fractionating phases obtained from least-squares mixing modelling of major and trace element variations (Chazot & Bertrand, 1993; Baker, 1996). Mineral–melt fractionation factors are calculated from the equations of Kalamarides (1986) assuming temperature varies as shown in Table 3. It is then finally necessary to recalculate equilibrium clinopyroxene compositions that would have been in equilibrium with the evolving melts as _melt-cpx increases with decreasing MgO (i.e. temperature).

These simple calculations show that fractionation of olivine + clinopyroxene and subsequent minor amounts of plagioclase + Fe–Ti oxides within the basaltic spectrum of rock types (MgO = 13·5–4·0 wt %) produces only small increases in melt ¹⁸O (<0·3) and that the increasing _melt-cpx fractionation factor with decreasing temperature (or MgO) means that an increase of only 0·1 in clinopyroxene ¹⁸O values is likely to result from fractional crystallization (Table 3; Fig. 2). Production of the rhyolitic rocks from basaltic parents requires extensive fractional crystallization (85%) and the significant amounts of fractionating Fe–Ti oxides result in a modest increase in melt ¹⁸O values (0·4–0·6). Clinopyroxene in equilibrium with such rhyolitic melts produced by protracted differentiation of basaltic parents would have ¹⁸O 5·9, similar to the lowest values measured in rhyolites in this study.

These calculations show that fractionation of the observed mineral phases in the Yemen flood basalts cannot generate the range in mineral ¹⁸O values observed in these rocks (>1). However, the marginally elevated ¹⁸O values measured on clinopyroxenes from the rhyolitic rocks can be largely accounted for by extensive fractional crystallization from basaltic parents. It is vital to note that relatively large ¹⁸O variations observed in individual mineral phases from samples with a limited range in MgO contents preclude fractional crystallization control on much of the oxygen isotopic variability in this dataset.

Mantle heterogeneity or crustal contamination?
Volcanic rocks from Yemen have radiogenic isotope ratios that exhibit considerable variation, but still fall within the field of oceanic basalts (Figs 1 and 5), and these correlations correlate with mineral oxygen isotope data (Figs 3 and 4). However, oxygen isotope variations (1·5) in minerals separated from the flood basalts exceed that observed in mantle peridotites (or minerals that would have crystallized from melts in equilibrium with mantle peridotites), including a large number of peridotites from the continental LM (Mattey et al., 1994). Oceanic basalts with a wide range in radiogenic isotope ratios exhibit limited oxygen isotope variations when analyses of mineral phases from these rocks are considered (Eiler et al., 1997). The previous discussion has shown that the mineral oxygen isotope variations are too large to be the product of fractional crystallization. We conclude that assimilation of continental crust, with elevated ¹⁸O relative to mantle-derived magmas, rather than a contribution from enriched LM or subcontinental mantle is responsible for the isotopic and also much of the incompatible trace element heterogeneity of the flood volcanic rocks from Yemen.

The oxygen isotope data confirm previous assertions that crustal contamination was an important process in this flood volcanic province (Baker et al., 1996b) and cast further doubt on the frequently inferred role of an enriched LM contribution to flood volcanism at the Afro-Arabian triple junction (Hart et al., 1989; Vidal et al., 1991; Chazot & Bertrand, 1993; Deniel et al., 1994).

Crustal provinciality from isotopic data
Sr–Nd–Pb–O isotope variations require the involvement of at least three crustal components (C1, C2 and C3) in the petrogenesis of the Yemen flood basalts and rhyolites (Figs 4 and 5).

Basalts erupted through the eastern part of the volcanic province are contaminated by a component with unradiogenic Pb isotope ratios and elevated 7/4 and 8/4 values, and high ⁸⁷Sr/⁸⁶Sr and low ¹⁴³Nd/¹⁴⁴Nd ratios (C1) (Fig. 5). This component has an isotopic composition remarkably similar to analyses of Late Archaean crustal gneisses from eastern Yemen presented herein (Fig. 5).

Basalts and rhyolites erupted through the western part of the volcanic province have been contaminated by at least two components. One component (C2) is difficult to identify on the basis of radiogenic isotopes as it has an isotopic composition relatively close to that of the uncontaminated basalts compared, for example, with C1. This component (C2) has Pb isotope ratios that fall close to the Northern Hemisphere Reference Line and the MORB–Afar plume array, with slightly unradiogenic Pb and slightly higher ⁸⁷Sr/⁸⁶Sr and lower ¹⁴³Nd/¹⁴⁴Nd ratios compared with the uncontaminated basalts. This composition is inferred to be Late Proterozoic Pan-African lower crust which is not old enough (or particularly chemically evolved enough) to have developed exotic isotope ratios compared with mantle-derived basalts. Analyses of lower-crustal granulite-facies gneisses from the Zabargad Island massif (Brueckner et al., 1995) and also such xenoliths entrained in Late Cenozoic volcanism throughout Saudi Arabia and Yemen (McGuire & Stern, 1993; G. Chazot & J. A. Baker, unpublished data, 1998) have precisely the required radiogenic isotope compositions to generate the observed mixing arrays in Sr–Nd–Pb isotopic space (Fig. 5). The second crustal component (C3) contaminating the basalts (and rhyolites) erupted through the western part of the volcanic province has higher ⁸⁷Sr/⁸⁶Sr ratios than, but similar ¹⁴³Nd/¹⁴⁴Nd ratios to, C2 and a distinctive Pb isotope signature marked by radiogenic ²⁰⁶Pb/²⁰⁴Pb isotope ratios with negative 8/4 values. This component is clearly distinct from the MORB–Afar plume mixing array in Pb isotope space and analyses of crustal basement from the western part of the volcanic province are a suitable crustal end-member.

Figure 6 illustrates the multi-element patterns of two relatively uncontaminated basalts with low ¹⁸O erupted through the western (JB281) and eastern (JB231II) parts of the volcanic province in Yemen. These samples are marked by relatively smooth multi-element patterns with a marked negative K anomaly. Figure 6 also illustrates examples of basalts contaminated by each of the inferred crustal contaminants. Samples contaminated by C1 and C2 show smaller K anomalies and enrichments in Ba and Pb relative to Rb, Th and U, which is consistent with these samples having assimilated Rb–Th–U-depleted, granulite-facies, lower crust. Samples contaminated by C3 show enrichments in all the large ion lithophile elements (LILE—Rb, Ba, K and Pb), Th and U relative to Nb. These features are consistent with C3 being an upper-crustal component.

View larger version (26K): [in this window] [in a new window]   Fig. 6. Multi-element plot of uncontaminated (JB281 and JB231II) and contaminated (JB259, JB172 and JB282) flood basalts illustrating the different effects on trace element signatures produced by lower- and upper-crustal contamination. Normalizing values for primitive mantle are from Sun & McDonough (1989).

 

Scattered arrays of data in Figs 1, 4 and 5 are a natural consequence of crustal contamination with isotopically heterogeneous crustal components, variable distribution coefficients for Sr during assimilation and fractional crystallization, and differences in primary magma incompatible trace element abundances rendering different samples’ isotopic systems to have had different susceptibilities to a finite amount of contamination. Semi-quantitative modelling of contamination processes has already been presented by Baker et al. (1996b) and in Fig. 3. These calculations show that the Yemen flood basalts and rhyolites have variably assimilated 0–25% crustal material.

Mechanisms of contamination
Mafic magmas erupted through the eastern part of the volcanic province have, in some cases, assimilated large amounts (20–25%) of a Late Archaean silicic lower-crustal component (C1) with little concomitant fractionation, e.g. JB172 and JB129. Although assimilation clearly involved little accompanying fractionation, the occurrence of this lower-crustal component at the surface today (crustal samples F25 and F27) means it is equivocal as to whether or not this assimilation took place at lower-crustal levels.

Mafic magmas erupted through the western part of the volcanic province have also commonly assimilated large amounts of lower-crustal material (C2), e.g. JB282. More evolved samples either show evidence for no further contamination or have a signature of contamination by upper continental crust (C3). The relatively limited isotopic contrast between the primary isotopic composition of the flood basalts and the young, Late Proterozoic, crustal components (C2 and C3) in the western part of the volcanic province means that the radiogenic and oxygen isotope effects of contamination are less obvious in this suite of samples than in the samples erupted through the eastern part of the volcanic province. It also makes it more difficult to model the amounts of contamination. However, given the ubiquitous ‘lithospheric’ trace element signature of continental crust, regardless of its age, the trace element effects of contamination are similar in both suites of basalts (Fig. 6).

Thus, contamination processes in Yemen varied from early lower-crustal contamination of hot mafic magmas with little or no concomitant fractionation (r > 1) through to combined assimilation and fractional crystallization typically within upper-crustal magma chambers. The most extreme result of the decline in the rate of assimilation to fractional crystallization is probably represented by the rhyolites, which require extensive fractionation with only limited amounts of contamination. The switch from lower- to upper-crustal contamination reflects establishment of magmatic plumbing systems; early intrusion of magmas into the lower crust resulted in hot mafic magmas removing the fusible parts of the lower crust and subsequent establishment of upper-crustal magma chambers produced more evolved magmas with a variable upper-crustal overprint. Plagioclase fractionation in the most evolved and upper crustally contaminated samples is consistent with a switch in the depth of fractionation and assimilation from deep to shallow crustal levels.

Identification of crustal contamination taking place in the lower crust when hot mafic magmas are introduced into the lithosphere has long been postulated as being important in flood basalt provinces (Thirlwall & Jones, 1983; Cox & Hawkesworth, 1985; Huppert & Sparks, 1985; Devey & Cox, 1987; Kerr et al., 1995; Martinez et al., 1996). This has now been verified by oxygen isotope studies of flood basalts from both the Deccan Traps and Yemen (Peng et al., 1994; herein). Presumably, assimilation is facilitated by the limited heat input required to melt already hot lower crust. What is also important to note is that magmas contaminated at this stage also have near-primary (i.e. low) incompatible trace element contents rendering them highly susceptible to small amounts of contamination and acquisition of lithospheric-like trace element and isotopic signatures.

Origin of rhyolitic volcanism
Clinopyroxene separated from four rhyolitic rocks has ¹⁸O = 5·9–6·1. These values are lower than those observed in some of the more contaminated flood basalts, including samples contaminated by crustal rocks in the western part of the volcanic province through which the rhyolites were erupted. Moreover, as discussed above, the slightly elevated ¹⁸O values of the rhyolite clinopyroxenes may largely be accounted for by extensive crystal fractionation.

Oxygen isotope data for the rhyolite samples might seem to preclude an origin for the rhyolites by extensive crustal melting in this province, and be more consistent with a model for rhyolite petrogenesis involving protracted fractionation with relatively limited assimilation from flood basalt parent magmas (Chazot & Bertrand, 1995). However, some Pan-African crust clearly has low ¹⁸O, which requires examination of other chemical and isotopic data to determine the petrogenesis of the rhyolites. Moreover, currently available chemical and isotope data are not sufficient to test an alternative hypothesis that the rhyolites might have been generated by melting of underplated basaltic material related to the flood volcanic episode (e.g. Harris & Erlank, 1991). Either of the latter two hypotheses has important implications for melt generation in this volcanic province—large amounts of unerupted mafic material equivalent to several times the volume of erupted rhyolite must be present in the lithosphere. Given that the basalt:rhyolite ratio is 1:1 in Yemen and Ethiopia, the volume of (basaltic) melt produced and the rate at which it was produced is at least three times greater than that currently inferred from study of the preserved volcanic volumes.

The primary oxygen isotope signature
Baker et al. (1996b) noted that the primary trace element and isotopic signature of the Yemen flood basalts is characterized by LILE depletion relative to the high field strength elements and a depleted Sr–Nd isotopic signature that approaches that of HIMU or PREMA mantle (Figs 5 and 6). However, Pb isotope ratios are not as radiogenic as HIMU mantle proper. These features are similar to those exhibited by magmas produced by the Iceland plume and have been explained in terms of recent recycling of oceanic lithosphere, i.e. the Afar and Iceland mantle plumes are immature HIMU mantle plumes (Thirlwall et al., 1994; Baker et al., 1996b).

It is noteworthy that, to date, the least crustally contaminated flood basalt samples from Yemen do not share the particularly low ¹⁸O values that have been observed in some rocks from Hawaii, Iceland and some other HIMU ocean islands (Eiler et al., 1997). Low K/Nb and Ba/Nb ratios (<150 and <5, respectively) in Yemen flood basalts with the lowest ¹⁸O values preclude these rocks being substantially crustally contaminated (<<5%) and having primary ¹⁸O values significantly lower than their measured values. Moreover, LM xenoliths metasomatized by the Afar plume (Baker et al., 1998) also do not have low ¹⁸O values (Chazot et al., 1997).

The final point concerning the oxygen isotope signature of the least contaminated Yemen flood basalts is that basanites and basalts sensu stricto have similar oxygen (and radiogenic) isotope ratios. Major and trace element differences between the basanite and basalt rocks are readily explained in terms of depth and degree of partial melting, with the basanites being the product of smaller average degrees of partial melting generated at greater depths than the basalts within the Afar plume. As such, there is apparently no partial melting control on oxygen isotope ratios in the Yemen flood basalt suite—at least within the resolution of the laser fluorination technique.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION SAMPLES ANALYTICAL TECHNIQUES RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Oxygen isotope heterogeneity of minerals separated from Oligocene flood volcanic rocks from Yemen exceeds analytical reproducibility and also that observed in oceanic basalts and mantle peridotites. High ¹⁸O values are correlated with changes in radiogenic isotope ratios, such as increasing ⁸⁷Sr/⁸⁶Sr. Samples with high ¹⁸O values include primitive and evolved compositions. The oxygen isotope data place the following constraints on magma genesis during flood volcanism in Yemen:

1. assimilation of heterogeneous Pan-African crust is unequivocally confirmed to have been the major control on isotopic and much trace element variability in this flood volcanic province.
   
2. The assimilated crustal components include both upper and lower crust of variable age. Assimilation of lower crust was a particularly important process in the Yemen flood volcanic province.
   
3. Crustal contamination processes can be demonstrated to have varied from combined assimilation and fractional crystallization through to rapid assimilation of crust by hot mafic magmas with little or no concomitant fractionation.
   
4. Rhyolitic magmas are not characterized by excessively high ¹⁸O values and/or crustal radiogenic isotope ratios, which seems to preclude their origin by wholesale crustal anatexis. The ¹⁸O data suggest they are probably either the product of extensive assimilation with small amounts of crustal contamination from basaltic parents or the result of melting underplated flood basalt material.
   
5. Least contaminated samples with low ¹⁸O have radiogenic isotope ratios that approach those inferred for the Afar plume, and these samples have ¹⁸O values within the range of other crustally uncontaminated plume-related magmas, i.e. ocean island basalts.
   

Laser fluorination oxygen isotope analysis of mineral separates is a powerful tool for examining the relative roles of crust and mantle in suites of continental volcanic rocks. Future combined oxygen and osmium isotope studies, along with in situ mineral multi-collector–inductively coupled plasma mass spectrometry studies, should finally resolve the debate regarding the relative contributions of crust and LM to flood basalt magmas.

	ACKNOWLEDGEMENTS

 
Abdulkarim Al-Subbary is thanked for assistance with fieldwork in Yemen. Gerry Ingram helped with radiogenic isotope analyses. The British Council, Royal Society and the Industrial Association at RHUL supported this research. The radiogenic and stable isotope laboratories at RHUL are University of London Intercollegiate facilities. Constructive reviews by John Eiler, Richard Carlson and Chris Harris, which improved this manuscript, are gratefully acknowledged.

	FOOTNOTES

 
^*Corresponding author. Present address: Danish Lithosphere Centre, Øster Voldgade 10, L, 1350 Copenhagen K, Denmark. Telephone: +45 38 14 26 42. Fax: +45 33 11 08 78. e-mail: jab{at}dlc.ku.dk

	REFERENCES

TOP ABSTRACT INTRODUCTION SAMPLES ANALYTICAL TECHNIQUES RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Anderson, A. T., Clayton, R. N. & Mayeda, T. K. (1971). Oxygen isotope thermometry of mafic igneous rocks. Journal of Geology 79, 715–729.

Arndt, N. T. & Christensen, U. (1992). The role of lithospheric mantle in continental flood volcanism: thermal and geochemical constraints. Journal of Geophysical Research 97, 10967–10981.

Arndt, N. T., Czamanske, G. K., Wooden, J. L. & Fedorenko, V. A. (1993). Mantle and crustal contributions to continental flood volcanism. Tectonophysics 223, 39–52.

Baker, J. A. (1996). Stratigraphy, geochronology and geochemistry of Cenozoic volcanism in western Yemen. Ph.D. Thesis, University of London, 386 pp.

Baker, J. A., Snee, L. & Menzies, M. A. (1996a). A brief period of Oligocene flood volcanism in western Yemen: implications for the duration and rate of continental flood volcanism at the Afro-Arabian triple junction. Earth and Planetary Science Letters 138, 39–56.

Baker, J. A., Thirlwall, M. F. & Menzies, M. A. (1996b). Sr–Nd–Pb isotopic and trace element evidence for crustal contamination of plume-derived flood basalts: Oligocene flood volcanism in western Yemen. Geochimica et Cosmochimica Acta 60, 2559–2581.

Baker, J. A., Menzies, M. A., Thirlwall, M. F. & Macpherson, C. G. (1997). Petrogenesis of Quaternary intraplate volcanism, Sana’a, Yemen: implications for plume–lithosphere interaction and polybaric melt hybridization. Journal of Petrology 38, 1359–1380.

Baker, J. A., Chazot, G., Menzies, M. & Thirlwall, M. (1998). Metasomatism of the shallow mantle beneath Yemen by the Afar plume—implications for mantle plumes, flood volcanism, and intraplate volcanism. Geology 26, 431–434.[Abstract/Free Full Text]

Blusztajn J., Hart, S. R., Shimizu, N. & McGuire, A. V. (1995). Trace element and isotopic characteristics of spinel peridotite xenoliths from Saudi Arabia. Chemical Geology 123, 53–65.

Brandon, A. D., Hooper, P. R., Goles, G. G. & Lambert, R. S. J. (1993). Evaluating crustal contamination in continental basalts: the isotopic composition of the Picture Gorge Basalt of the Columbia River Basalt Group. Contributions to Mineralogy and Petrology 114, 452–464.[Web of Science]

Brueckner, H. K., Elhaddad, M. A., Hamelin, B., Hemming, S., Kröner, A., Reisberg, L. & Seyler, M. (1995). A Pan-African origin and uplift for gneisses and peridotites of Zabargad Island, Red Sea: a Nd, Sr, Pb and Os isotope study. Journal of Geophysical Research 100, 22283–22297.

Carlson, R. W. (1984). Isotopic constraints on Columbia River flood basalt genesis and the nature of the subcontinental mantle. Geochimica et Cosmochimica Acta 48, 2357–2372.[Web of Science]

Carlson, R. W., Lugmair, J. & Macdougall, J. D. (1981). Columbia River volcanism: the question of mantle heterogeneity or crustal contamination. Geochimica et Cosmochimica Acta 45, 2483–2499.

Chazot, G. & Bertrand, H. (1993). Mantle sources and magma–continental crust interaction during early Red Sea–Aden rifting in southern Yemen: elemental and Sr–Nd–Pb isotope evidence. Journal of Geophysical Research 98, 1819–1835.

Chazot, G. & Bertrand, H. (1995). Genesis of silicic magmas during Tertiary continental rifting in Yemen. Lithos 36, 69–83.

Chazot, G., Lowry, D., Menzies, M. & Mattey, D. (1997). Oxygen isotopic compositions of hydrous and anhydrous mantle peridotites. Geochimica et Cosmochimica Acta 61, 161–169.

Chiba, H., Chacko, T., Clayton, R. N. & Goldsmith, J. R. (1989). Oxygen isotopic fractionation involving diopside, magnetite, and calcite: application to geothermometry. Geochimica et Cosmochimica Acta 53, 2985–2995.

Coffin, M. & Eldholm, O. (1993). Scratching the surface: estimating the dimensions of large igneous provinces. Geology 21, 515–518.[Abstract/Free Full Text]

Cox, K. G. & Hawkesworth, C. J. (1985). Geochemical stratigraphy of the Deccan Traps at Mahabaleshwar, Western Ghats, India, with implications for open system magmatic processes. Journal of Petrology 26, 355–377.[Abstract/Free Full Text]

Davidson, J. P. & Wilson, I. R. (1989). Evolution of an alkali basalt–trachyte suite from Jebel Marra volcano, Sudan, through assimilation and fractional crystallization. Earth and Planetary Science Letters 95, 141–160.

Deniel, C., Vidal, P., Coulon, C., Vellutini, P.-J. & Piguet, P. (1994). Temporal evolution of mantle sources during continental rifting: the volcanism of Djibouti (Afar). Journal of Geophysical Research 99, 2853–2869.

DePaolo, D. J. (1981). Trace element and isotopic effects of combined wallrock assimilation and fractional crystallization. Earth and Planetary Science Letters 53, 189–202.[Web of Science]

Devey, C. W. & Cox, K. G. (1987). Relationships between crustal contamination and crystallization in continental flood basalt magmas with special reference to the Deccan Traps of the western Ghats. Earth and Planetary Science Letters 84, 59–68.

Duyvermann, H, J., Harris, N. B. W. & Hawkesworth, C. J. (1982). Crustal accretion in the Pan-African: Nd and Sr isotope evidence from the Arabian shield. Earth and Planetary Science Letters 59, 315–326.

Eiler, J. M., Farley, K., Valley, J. W., Stolper, E. M., Hauri, E. H. & Craig, H. (1995). Oxygen isotope evidence against bulk recycled sediment in the mantle sources of Pitcairn Island lavas. Nature 377, 138–141.

Eiler, J. M., Farley, K., Valley, J. W., Hauri, E. H., Craig, H., Hart, S. R. & Stolper, E. M. (1997). Oxygen isotope variations in ocean island basalt phenocrysts. Geochimica et Cosmochimica Acta 61, 2281–2293.

Ellam, R. M. (1992). Lithospheric thickness as a control on basalt geochemistry. Geology 20, 153–156.[Abstract/Free Full Text]

Ellam, R. M. & Cox, K. G. (1991). An interpretation of Karoo picrites in terms of interaction between asthenospheric magmas and the mantle lithosphere. Earth and Planetary Science Letters 105, 330–342.[Web of Science]

Gallagher, K. & Hawkesworth, C. J. (1992). Dehydration melting and the generation of continental flood basalts. Nature 358, 57–59.

Harris, C. & Erlank, A. J. (1991). The production of large-volume low ¹⁸O rhyolites during the rifting of Africa and Antarctica: the Lebombo Monocline, southern Africa. Geochimica et Cosmochimica Acta 56, 3561–3570.

Hart, S. R. (1984). A large-scale isotope anomaly in the southern hemisphere. Nature 309, 753–757.

Hart, W. K., Woldegabriel, G., Walter, R. C. & Mertzman, S. A. (1989). Basaltic volcanism in Ethiopia: constraints on continental rifting and mantle interactions. Journal of Geophysical Research 94, 7731–7748.

Hauri, E. H., Shimizu, N., Dieu, J. J. & Hart, S. R. (1993). Evidence for hotspot-related carbonatite metasomatism in the oceanic upper mantle. Nature 365, 221–227.

Henjes-Kunst, F., Altherr, R. & Baumann, A. (1990). Evolution and composition of the lithospheric mantle underneath the western Arabian peninsula: constraints from Sr–Nd isotope systematics of mantle xenoliths. Contributions to Mineralogy and Petrology 105, 460–472.

Hergt, J. M., Chappell, B. W., McCulloch, M. T., MacDougall, I. & Chivas, A. R. (1989). Geochemical and isotopic constraints on the origin of the Jurassic dolerites of Tasmania. Journal of Petrology 30, 841–883.[Abstract/Free Full Text]

Hofmann, A. W. (1997). Mantle geochemistry: the message from oceanic volcanism. Nature 385, 219–229.

Hooper, P. R. & Hawkesworth, C. J. (1993). Isotopic and geochemical constraints on the origin and evolution of the Columbia River Basalt. Journal of Petrology 34, 1203–1246.[Abstract/Free Full Text]

Horan, M. F., Walker, R. J., Fedorenko, V. A. & Czamanske, G. K. (1995). Osmium and neodymium isotopic constraints on the temporal and spatial evolution of Siberian flood basalt sources. Geochimica et Cosmochimica Acta 59, 5159–5168.

Huppert, H. E. & Sparks, R. S. J. (1985). Cooling and contamination of mafic and ultramafic magmas during ascent through continental crust. Earth and Planetary Science Letters 74, 371–386.[Web of Science]

Ito, E., White, W. M. & Gopel, C. (1987). The O, Sr, Nd and Pb isotope geochemistry of MORB. Chemical Geology 62, 157–176.[Web of Science]

Kalamarides, R. (1986). High-temperature oxygen isotope fractionation among the phases of the Kiglapait Intrusion, Labrador, Canada. Chemical Geology (Isotope Geoscience Section) 58, 303–310.

Kerr, A. C., Kempton, P. D. & Thompson, R. N. (1995). Crustal assimilation during turbulent magma ascent (ATA); new isotopic evidence from the Mull Tertiary lava succession, NW Scotland. Contributions to Mineralogy and Petrology 119, 142–154.[Web of Science]

Kyser, T. K. (1986). Stable isotope variations in the mantle. In: Valley, J. W., Taylor, H. P. & O’Neill, J. R. (eds) Stable Isotopes in High Temperature Geological Processes. Mineralogical Society of America, Reviews in Mineralogy 16, 227–271.

Kyser, T. K. (1990). Stable isotopes in the continental lithospheric mantle. In: Menzies, M. (ed.) The Continental Lithosphere. Oxford: Clarendon Press, pp. 127–156.

Lightfoot, P. C., Hawkesworth, C. J., Devey, C. W., Rogers, N. W. & Van Casteren, P. W. (1990a). Source and differentiation of Deccan Trap lavas: implications of geochemical and mineral chemical variations. Journal of Petrology 31, 1165–1200.[Abstract/Free Full Text]

Lightfoot, P. C., Naldrett, A. J., Gorbachev, N. S., Doherty, W. & Fedorenko, V. A. (1990b). Geochemistry of the Siberian Traps of the Noril’sk area, USSR, with implications for the relative contributions of crust and mantle to flood basalt magmatism. Contributions to Mineralogy and Petrology 104, 631–644.

Lightfoot, P. C., Hawkesworth, C. J., Hergt, J., Naldrett, A. J., Gorbachev, N. S., Fedorenko, V. A. & Doherty, W. (1993). Remobilisation of the continental lithosphere by a mantle plume: major-, trace element and Sr-, Nd-, and Pb-isotope evidence from picritic and tholeiitic lavas of the Noril’sk District, Siberian Traps, Russia. Contributions to Mineralogy and Petrology 114, 171–188.[Web of Science]

Macpherson, C. G. (1995). New analytical approaches to the oxygen and carbon stable isotope geochemistry of some subduction-related lavas. Ph.D. Thesis, University of London, 206 pp.

Macpherson, C. G., Gamble, J. A. & Mattey, D. P. (1998). Oxygen isotope geochemistry of lavas from an oceanic to continental arc transition, Kermadec–Hikurangi margin, SW Pacific. Earth and Planetary Science Letters 160, 609–621.

Martinez, I. A., Harris, C., le Roex, A. P. & Milner S. C. (1996). Oxygen isotope evidence for extensive crustal contamination in the Okenyenya igneous complex, Namibia. Geochimica et Cosmochimica Acta 60, 4497–4508.

Mattey, D. P. & Macpherson, C. G. (1993). High-precision oxygen isotope analysis of ferromagnesian minerals by laser fluorination. Chemical Geology (Isotope Geoscience Section) 105, 305–318.

Mattey, D. P., Lowry, D. & Macpherson, C. G. (1994). Oxygen isotope composition of mantle peridotites. Earth and Planetary Science Letters 128, 231–241.[Web of Science]

McGuire, A. V. & Stern, R. J. (1993). Granulite xenoliths from Saudi Arabia: the lower crust of the late Precambrian Arabian–Nubian shield. Contributions to Mineralogy and Petrology 114, 395–408.

Menzies, M. A. (1992). The lower lithosphere as a major source for continental flood basalts: a re-appraisal. In: Storey, B., Alabaster, T. & Pankhurst, R. J. (eds) Magmatism and the Causes of Continental Break-Up. Geological Society, London, Special Publications 68, 293–304.

Officer, C. B. & Drake, C. L. (1983). The Cretaceous–Tertiary transition. Science 219, 1383–1390.[Abstract/Free Full Text]

Peng, Z. X., Mahoney, J., Hooper, P., Harris, C. & Beane, J. (1994). A role for lower continental crust in flood basalt genesis? Isotopic and incompatible element study of the lower six formations of the western Deccan Traps. Geochimica et Cosmochimica Acta 58, 267–288.[Web of Science]

Pik, R., Deniel, C., Coulon, C., Yirghu, G. & Marty, B. (1999). Isotopic and trace element signatures of Ethiopian Flood Basalts: evidence for plume–lithosphere interactions. Geochimica et Cosmochimica Acta 63, 2263–2279.

Schilling, J.-G., Kingsley, R. H., Hanan, B. B. & McCully, B. L. (1992). Nd–Sr–Pb isotopic variations along the Gulf of Aden: evidence for mantle plume–continental lithosphere interaction. Journal of Geophysical Research 97, 10927–10966.

Singer, B. S., Brophy, J. G. & O’Neill, J. R. (1992). Oxygen isotope constraints on the petrogenesis of Aleutian arc magmas. Geology 20, 367–370.[Abstract/Free Full Text]

Sun, S.-s. & McDonough, W. F. (1989). Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In: Saunders, A. D. & Norry, M. J. (eds) Magmatism in the Ocean Basins. Geological Society, London, Special Publications 42, 313–345.

Taylor, H. P. & Sheppard, S. M. F. (1986). Igneous rocks: I. Processes of isotopic fractionation and isotope systematics. In: Valley, J. W., Taylor, H. P. & O’Neill, J. R. (eds) Stable Isotopes in High Temperature Processes. Mineralogical Society of America, Reviews in Mineralogy 16, 141–164.

Thirlwall, M. F. & Jones, N. W. (1983). Isotope geochemistry and contamination mechanics of Tertiary lavas from Skye, NW Scotland. In: Hawkesworth, C. J. & Norry, M. J. (eds) Continental Basalts and Mantle Xenoliths. Nantwich, UK: Shiva, pp. 186–208.

Thirlwall, M. F., Upton, B. G. J. & Jenkins, C. (1994). Interaction between continental lithosphere and the Iceland plume—Sr–Nd–Pb isotope chemistry of Tertiary basalts, NE Greenland. Journal of Petrology 35, 839–879.[Abstract/Free Full Text]

Thirlwall, M. F., Jenkins, C., Vroon, P. Z. & Mattey, D. P. (1997). Crustal interaction during construction of ocean islands: Pb–Sr–Nd–O isotope geochemistry of the shield basalts of Gran Canaria, Canary Islands. Chemical Geology (Isotope Geoscience Section) 52, 317–331.

Vidal, P., Deniel, C., Vellutini, P. J., Piguet, P., Coulon, C., Vincent, J. & Audin, J. (1991). Changes of mantle sources in the course of a rift evolution: the Afar case. Geophysical Research Letters 18, 1913–1916.[Web of Science]

Volker, F., McCulloch, M. T. & Altherr, R. (1993). Submarine basalts from the Red Sea: new Pb, Sr and Nd isotopic data. Geophysical Research Letters 20, 927–930.[Web of Science]

White, R. S. & McKenzie, D. P. (1989). Magmatism at rift zones: the generation of volcanic continental margins. Journal of Geophysical Research 94, 7685–7729.

White, R. S. & McKenzie, D. P. (1995). Mantle plumes and flood basalts. Journal of Geophysical Research 100, 17543–17585.

Windley, B. F., Whitehouse, M. J. & Ba-Bttat, M. A. O. (1996). Early Precambrian gneiss terranes and Pan-African island arcs in Yemen: crustal accretion of the eastern Arabian Shield. Geology 24, 131–134.[Abstract/Free Full Text]

Wooden, J. L., Czamanske, G. K., Fedorenko, V. A., Arndt, N. A., Chauvel, C., Bouse, R. M., King, B. W. & Siems, D. F. (1993). Isotopic and trace element constraints on mantle and crustal contributions to Siberian continental flood basalts, Noril’sk area, Siberia. Geochimica et Cosmochimica Acta 57, 3677–3704.

Zheng, Y. (1993). Calculation of oxygen fractionation in anhydrous silicate minerals. Geochimica et Cosmochimica Acta 57, 1079–1091.[Web of Science]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				L. Beccaluva, G. Bianchini, C. Natali, and F. Siena Continental Flood Basalts and Mantle Plumes: a Case Study of the Northern Ethiopian Plateau J. Petrology, July 1, 2009; 50(7): 1377 - 1403. [Abstract] [Full Text] [PDF]

				T. C. Feeley, M. A. Clynne, G. S. Winer, and W. C. Grice Oxygen Isotope Geochemistry of the Lassen Volcanic Center, California: Resolving Crustal and Mantle Contributions to Continental Arc Magmatism J. Petrology, May 1, 2008; 49(5): 971 - 997. [Abstract] [Full Text] [PDF]

				J. S. Beard Crystal-Melt Separation and the Development of Isotopic Heterogeneities in Hybrid Magmas J. Petrology, May 1, 2008; 49(5): 1027 - 1041. [Abstract] [Full Text] [PDF]

				M.-S. Krienitz, K. M. Haase, K. Mezger, and M. A. Shaikh-Mashail Magma Genesis and Mantle Dynamics at the Harrat Ash Shamah Volcanic Field (Southern Syria) J. Petrology, August 1, 2007; 48(8): 1513 - 1542. [Abstract] [Full Text] [PDF]

				T. Furman, J. Bryce, T. Rooney, B. Hanan, G. Yirgu, and D. Ayalew Heads and tails: 30 million years of the Afar plume Geological Society, London, Special Publications, January 1, 2006; 259(1): 95 - 119. [Abstract] [PDF]

				C. HARRIS, J. J. M. PRONOST, L. D. ASHWAL, and R. G. CAWTHORN Oxygen and Hydrogen Isotope Stratigraphy of the Rustenburg Layered Suite, Bushveld Complex: Constraints on Crustal Contamination J. Petrology, March 1, 2005; 46(3): 579 - 601. [Abstract] [Full Text] [PDF]

				B. KIEFFER, N. ARNDT, H. LAPIERRE, F. BASTIEN, D. BOSCH, A. PECHER, G. YIRGU, D. AYALEW, D. WEIS, D. A. JERRAM, et al. Flood and Shield Basalts from Ethiopia: Magmas from the African Superswell J. Petrology, April 1, 2004; 45(4): 793 - 834. [Abstract] [Full Text] [PDF]

				S. Pilet, J. Hernandez, F. Bussy, and P.J. Sylvester Short-term metasomatic control of Nb/Th ratios in the mantle sources of intraplate basalts Geology, February 1, 2004; 32(2): 113 - 116. [Abstract] [Full Text] [PDF]

				R. Halama, R. Halama, T. Wenzel, B. G. J. Upton, W. Siebel, and G. Markl A geochemical and Sr-Nd-O isotopic study of the Proterozoic Eriksfjord Basalts, Gardar Province, South Greenland: Reconstruction of an OIB signature in crustally contaminated rift-related basalts Mineralogical Magazine, October 1, 2003; 67(5): 831 - 853. [Abstract] [Full Text] [PDF]

				M. MARKS, T. VENNEMANN, W. SIEBEL, and G. MARKL Quantification of Magmatic and Hydrothermal Processes in a Peralkaline Syenite-Alkali Granite Complex Based on Textures, Phase Equilibria, and Stable and Radiogenic Isotopes J. Petrology, July 1, 2003; 44(7): 1247 - 1280. [Abstract] [Full Text] [PDF]

				D. Ayalew, D. Ayalew, and G. Yirgu Crustal contribution to the genesis of Ethiopian plateau rhyolitic ignimbrites: basalt and rhyolite geochemical provinciality Journal of the Geological Society, January 1, 2003; 160(1): 47 - 56. [Abstract] [Full Text] [PDF]

				J. M. Eiler and J. M. Eiler Oxygen Isotope Variations of Basaltic Lavas and Upper Mantle Rocks Reviews in Mineralogy and Geochemistry, January 1, 2001; 43(1): 319 - 364. [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (30) Request Permissions Google Scholar Articles by BAKER, J. A. Articles by MATTEY, D. P. Search for Related Content GeoRef GeoRef Citation Social Bookmarking          What's this?

Online ISSN 1460-2415 - Print ISSN 0022-3530

Copyright © 2010 Oxford University Press

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics