Journal of Asian Earth Sciences 20 (2002) 867±877

www.elsevier.com/locate/jseaes

Nitrogen and noble gas isotopes in ma®c and ultrama®c inclusions in the alkali basalts from Kutch and Reunion Ð implications for their mantle sources Ratan K. Mohapatra 1, S.V.S. Murty* Physical Research Laboratory, Navrangpura, Ahmedabad 380009, India Received 6 September 2000; revised 25 July 2001; accepted 23 August 2001

Abstract Olivine phenocrysts and ultrama®c xenoliths from the alkali basalts of Kutch and Reunion have been analyzed simultaneously for nitrogen and light noble gases. The samples have low amounts of neon (22 Ne ˆ 1:4±5:2 £ 10210 cc STP g21 ) and argon (36 Ar ˆ 0:7±12:6 £ 10210 cc STP g21 ). The variations in 20Ne/ 22Ne and 21Ne/ 22Ne ratios indicate the presence of two trapped components: air and mantle (OIB and MORB), consistent with the range of 40Ar/ 36Ar ratios observed (2166±9268 in the xenoliths, ,550 in the olivine phenocryst from Kutch, and 350 in the Reunion sample). The noble gas data indicate an OIB type mantle source for the phenocrysts from Kutch and Reunion, but a MORB type mantle source for the Kutch xenoliths, and are consistent with the plume origin of Deccan. Nitrogen concentrations in these samples vary from 1 to 2.5 ppm. Its isotopic composition, which varies from 215 to 115½ in the stepped temperature data, is consistent with the presence of at least two components mantle (d15 N , 215½) and recycled nitrogen (d15 N , 115½), often in the same sample. The simultaneous presence of the mantle and recycled components in these samples, which is also consistent with the accompanying argon isotopic data, requires a millimeter-scale heterogeneity that can be explained by the presence of recycled materials in the form of a global enriched region at the lithosphere/asthenosphere boundary beneath Kutch and Reunion. q 2002 Elsevier Science Ltd. All rights reserved. Keywords: Nitrogen; Noble gases; Ultrama®c inclusions; Alkali basalt; Mantle sources; Kutch; Reunion

1. Introduction The alkali basalts in Kutch, Western India (Fig. 1) are attributed to the activity of the Reunion hotspot at ,65 Ma (Morgan, 1981; Mahoney, 1988; Venkatesan et al., 1993). Although they lie outside the Deccan Traps volcanic province, these rocks have been shown to be genetically related to the Deccan Traps, and have been studied by various workers with a hope to obtain geochemical clues for the origin of the Deccan Traps and the temporal evolution of the Reunion hotspot (e.g. Mahoney, 1988; Baxter, 1990; White et al., 1990; White and McKenzie, 1995). Samples from Reunion have been extensively studied for noble gases (Kaneoka et al., 1986; Staudacher et al., 1990; Graham et al., 1990; Moreira and Allegre, 1998) wherein * Corresponding author. Tel.: 191-79-630219; fax: 191-79-6301502. E-mail addresses: [email protected] (R.K. Mohapatra), [email protected] (S.V.S. Murty). 1 Present address: Abt. Kosmochemie, Max-Planck-Institut fuÈr Chemie, Becher Weg 27, D 55128 Mainz, Germany.

distinct plume signatures (3 He=4 He . 12 times air ratio) have been observed. But, to date there is just one nitrogen datum that exists in the literature for the Reunion basalts (Mohapatra and Murty, 2000b). Similarly, noble gas data for the Deccan Traps basalts are limited to a few measurements of helium and its isotopic ratio ( 3He/ 4He) by Basu et al. (1993), which suggest a plume type mantle source for the Deccan and its associated alkali basalts. But the alkali basalts in Kutch and their ultrama®c xenoliths have so far not been studied for nitrogen and noble gases. Nitrogen is an important geochemical tracer in the Earth's mantle where, in its most common form (N2), it is chemically inert like the noble gases (Marty and Humbert, 1997). It has two stable isotopes with masses 14 and 15. The isotopic composition, as de®ned below: d15 N …½† ˆ   15 N=14 N

Sample

2



15

 .   15 N=14 N N=14 N £ 1000 Air

Air

provides clues for identifying the parent materials for the

1367-9120/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved. PII: S 1367-912 0(01)00070-0

868

R.K. Mohapatra, S.V.S. Murty / Journal of Asian Earth Sciences 20 (2002) 867±877

Fig. 1. (a) Geographic locations of Kutch and the Reunion Islands (inset), and a geological map of Reunion showing the sample locations (modi®ed after Staudacher et al., 1990). (b) Geological map of Western India showing the location of the sampling sites in Kutch (modi®ed after Biswas, 1970).

accretion of the Earth, and for understanding its evolution (Javoy, 1998; Mohapatra, 1998; Tolstikhin and Marty, 1998). Together with argon, nitrogen provides a means to quantitatively assess the recycled component in mantlederived materials (Marty and Humbert, 1997; Mohapatra and Murty, 1998; Sano et al., 1998; Nishio et al., 1999;

Marty and Zimmermann, 1999; Mohapatra and Murty, 2000a, b). The noble gas isotopic systematics of mantlederived materials has been instrumental in understanding the degassing history of the Earth (Allegre et al., 1986/87; Tolstikhin and Marty, 1998), and the different geochemical regimes of the mantle (e.g. degassed and undegassed

R.K. Mohapatra, S.V.S. Murty / Journal of Asian Earth Sciences 20 (2002) 867±877 Table 1 List of the Kutch and Reunion samples used in the present study Samples a

Weight (g)

Remarks

Kutch (1) B835.Cp

0.339

(2) B8511.Cp (3) D872.Ol (4) D874.Cp (5) D875.Cp (6) VTH.Ol

0.505 0.996 0.319 0.388 1.144

Grains 2±3 mm in size; handpicked from disaggregated xenoliths from the Bhuj Hills (1 and 2, at the center of the Bhuj town) and Dhrubia Hill (3±5, near Nakhsatrana). Original ®eld sample names: (1) BN83-5.Cpx; (2) KB85-11.Cpx; (3) DN87-2.Ol; (4) DN87-4.Cpx; (5) DN87-5.Cpx

Reunion islands (7) RE.Ol

1.231

(8) RE.DU b

1.320

a b

Phenocryst grains (2±3 mm in size) from an alkali basalt sample from a plug at the Vithon village, near Bhuj Phenocryst grains (Re 78, Kaneoka et al., 1986) from a 2.0±0.4 Ma old oceanite from Piton des Neiges volcano (Fig. 1a), Reunion Island (Indian Ocean), obtained from Prof. Ichiro Kaneoka, Tokyo University A `Recent' coarse grained (2±3 mm) dunite (olivine 1 spinel), believed to be a cumulate (Re496-16, Kaneoka et al., 1986), from Piton Chisny, Reunion Islands (Indian Ocean)

Cp, Cpx ˆ clino pyroxene; DU ˆ dunite; Ol ˆ olivine. Mohapatra and Murty (2000b).

mantle, see Farley and Neroda, 1998, for a recent review). Nitrogen isotopic studies of the mantle are mostly concentrated on diamonds (Javoy et al., 1984; Boyd et al., 1987 and 1992; Boyd and Pillinger, 1994; Cartigny et al., 1997, 1998a, b; Mohapatra and Murty, 2001) and oceanic basalts (Marty and Humbert, 1997; Sano et al., 1998; Marty and Zimmermann, 1999; Mohapatra and Murty, 2000b). Mantle-derived ma®c and ultrama®c inclusions (cognates, ma®c phenocrysts and xenoliths), however, have not received much attention. A recent study of mantle xenoliths from San Carlos (Mohapatra and Murty, 2000a) has shown the importance of these samples in understanding the geochemistry of the sub-continental mantle (SCM). We have studied a few ma®c and ultrama®c inclusions from the basalts of Kutch and Reunion for their nitrogen and light (Ne and Ar) noble gas isotopic compositions with an aim to understand the mantle sources for these rocks. In this paper, we present the data obtained so far and discuss their implications for the geochemical evolution of the Kutch alkali basalts, and their relation to the Reunion hotspot.

2. Samples Table 1 compiles the details of samples used in the

869

present study. They include hand-picked (under a binocular microscope) mineral grains of clinopyroxenes (e.g. B835.Cp) and olivines (e.g. D872.Ol) from disaggregated xenoliths, along with phenocrystic olivine (e.g. VTH.Ol and RE.Ol) from host basalts. We also include sample RE.DU, a dunite nodule from Reunion, from the literature (Mohapatra and Murty, 2000b). The dunite nodule and the phenocrysts are a result of early crystallization of the magma that formed their host basalts, and would have trapped nitrogen and noble gases from the mantle sources that gave rise to their host magmas. Mantle xenoliths, on the other hand, are foreign to the magma and are not geochemically related to the host rocks. Although they may not convey anything about the source of their host rocks, mantle xenoliths provide insights into the nitrogen and noble gas signatures of SCM (Mohapatra and Murty, 2000a). Alkali basalts at Kutch (2382 0 ±23826 0 N; 69813 0 ± 69845 0 E) occur as intrusive plugs into the Mesozoic sandstones and contain numerous ultrama®c (spinel lhezolite) xenoliths of mantle origin (De, 1964; Pande, 1988; Krishnamurthy et al., 1988). However, due to improper preservation resulting from quarrying activities, it is very dif®cult to obtain samples of the required quantity and quality. We collected a few samples from the PRL's collection (Pande, 1988), and from a ®eld trip to Bhuj and nearby areas, for the present study. Fig. 1b is a geological map of Western India showing the locations of the alkali basalt plugs from which the present samples have been collected. The Reunion sample (RE.Ol), which is from Piton des Neiges, has been kindly provided for the study by Prof. Ichiro Kaneoka of the Tokyo University, Japan. Details of the geological history and noble gas data are given in Kaneoka et al. (1986). 3. Experimental techniques Several hundred milligrams (Table 1) of the mineral separates, thoroughly cleaned in ethanol and acetone (often in an ultrasonic bath) and wrapped in aluminum foil, were loaded into the sample tree of the gas extraction system. Prior to analysis, the samples were degassed in a vacuum at a temperature of ,1508C (using infra red lamps), to remove the sur®cial contamination (adsorbed atmospheric gases) resulting from sample preparation. To further remove the traces of atmospheric contamination, they were combusted in 2 Torr of oxygen (generated inline from a copper-oxide ®nger) at 4008C. Subsequently, gases were extracted from the samples by radio frequency heating at different temperatures, and analyzed for their nitrogen and noble gas isotopes using a VG 1200 noble gas mass spectrometer. Details of the mass spectrometric procedures are given in Mohapatra and Murty (2000a). Blanks, interspersed between the samples, were measured at each temperature in a fashion similar to the sample measurements. Blank amounts increase with extraction temperature but are generally ,10% of the signal for all the gases. Compositionally, the

870

R.K. Mohapatra, S.V.S. Murty / Journal of Asian Earth Sciences 20 (2002) 867±877

Table 2 Neon data for the Kutch and Reunion samples (Errors in the concentrations are ^10%; errors in the isotopic ratios (numbers in the parentheses) represent 95% CL) Samples

Temp. (8C)

22

D872.Ol

900 1800

0.3 4.9 5.2

10.40 (0.05) 9.88 (0.02) 9.92 (0.02)

0.034 (0.002) 0.031 (0.001) 0.031 (0.001)

900 1450

1.8 2.2 4.0

10.30 (0.32) 10.41 (0.04) 10.36 (0.17)

0.037 (0.002) 0.037 (0.004) 0.037 (0.003)

800 1200 1600

0.4 0.3 1.7 2.4

10.47 (0.03) 10.03 (0.06) 9.93 (0.04) 10.03 (0.04)

0.034 (0.001) 0.031 (0.001) 0.029 (0.001) 0.030 (0.001)

800 1200 1600

0.5 0.4 3.4 4.3

9.77 (0.17) 9.82 (0.11) 10.09 (0.05) 10.03 (0.07)

0.032 (0.001) 0.029 (0.003) 0.029 (0.001) 0.029 (0.001)

900 1800

0.2 1.2 1.4

9.77 (0.04) 9.98 (0.04) 9.94 (0.040

0.030 (0.001) 0.030 (0.001) 0.030 (0.001)

Total D875.Cp Total VTH.Ol Total RE.Ol Total RE.DU Total

Ne (10 210 cc STP g 21)

20

Ne/ 22Ne

21

Ne/ 22Ne

Table 3 Nitrogen and argon data for the Kutch and Reunion samples (Errors in the concentrations are ^10%; errors in isotopic ratios (numbers in the parentheses) represent 95% CL) Samples

Temp. (8C)

N (ppm)

d 15N (½)

36

B835.Cp

1400 1500

0.9 0.4 1.3

6.17 (1.36) 214.64 (0.98) 20.12 (1.24)

12.6 , 0.07 a 12.6

2166 (10) ± 2166 (10)

900 1500

0.7 1.8 2.5

3.64 (1.58) 12.34 (0.42) 9.82 (0.76)

, 0.04 a 6.8 6.8

± 4318 (26) 4318 (26)

900 1800

0.3 1.0 1.3

11.64 (0.92) 8.71 (0.41) 9.43 (0.54)

0.4 0.6 1.0

376 (2) 7300 (26) 4358 (2)

600 900 1500

0.04 0.26 0.8 1.1

3.52 (0.79) 12.94 (0.48) 14.29 (0.48) 13.53 (0.49)

900 1500

0.2 0.9 1.1

24.51 (0.64) 7.52 (0.52) 5.08 (0.54)

1.0 7.2 8.2

419 (10) 10 511 (65) 9268 (58)

800 1200 1600

0.8 1.0 0.1 1.9

6.13 (0.66) 5.08 (0.94) 6.24 (0.60) 5.58 (0.81)

6.2 3.2 1.4 10.8

317 (60) 376 (7) 1903 (31) 544 (10)

800 1200 1600

0.2 0.4 0.5 1.1

7.83 (0.65) 12.56 (0.27) 9.82 (0.77) 10.43 (0.57)

, 0.05 a , 0.06 a 0.7 0.7

900 1800

0.5 0.9 1.4

215.34 (0.40) 7.15 (0.36) 21.04 (0.38)

Total B8511.Cp Total D872.Ol Total D874.Cp Total D875.Cp Total VTH.Ol Total RE.Ol Total RE.DU b Total a b

Blank level concentrations. From Mohapatra and Murty (2000b).

Ar (10 210 cc STP g 21)

, 0.03 a , 0.04 a , 0.07 a Blank level a

0.2 1.0 1.2

40

Ar/ 36Ar

± ± ± ±

± ± 350 (1) 350 (1) 620 (3) 3043 (13) 2611 (11)

R.K. Mohapatra, S.V.S. Murty / Journal of Asian Earth Sciences 20 (2002) 867±877

871

in the compilations. We have also used the N and Ar data for another Reunion sample (RE.DU) from the literature (Mohapatra and Murty, 2000b), which are relevant to the present discussion. 4.1. Noble gases

Fig. 2. Plot of 20Ne/ 22Ne against 21Ne/ 22Ne for the total neon data (in Table 2) for the Kutch and Reunion samples. Also shown are the point air and the trends for OIBs (Honda et al., 1991; Valbracht et al., 1997) and MORBs (Sarda et al., 1988). The Reunion samples and the olivine phenocryst from Kutch (VTH.Ol) follow the air±OIB mixing line, while the Kutch xenoliths follow the air±MORB mixing line.

noble gas blanks are close to atmospheric within measurement errors. d 15N of the blanks are slightly positive, lying in the range 3 ^ 2½. A typical 18008C blank gave 22 Ne ˆ 3 £ 10211 , 36 Ar ˆ 1:4 £ 10211 (in cc STP units) and about 200 pg of nitrogen. Air standards were run to assess the sensitivity and the instrumental mass discrimination. The data reported here have been corrected for blanks and mass discrimination and the associated errors have been propagated. The nitrogen isotopic data have been additionally corrected for carbon monoxide (CO) interference as detailed in Murty (1997). This was carried out by routine measurement of mass 30 along with the nitrogen peaks (masses 28 and 29). The excess at mass 30 was assigned to a contribution from CO and the corresponding contributions at masses 28 and 29 were corrected. Correction to d 15N due to CO was always #1½. The neon data have been corrected for interferences from 40Ar 21 (at 20Ne) and 44 22 CO21 2 (at Ne) by measuring the backgrounds at masses 40 and 44 routinely in the neon measurements. The concentrations of 40Ar and CO2 in the mass spectrometer are kept to a minimum during Ne analysis by keeping the mass spectrometer volume open to a stainless steel mesh ®nger maintained at the liquid nitrogen temperature. From the signals of 40 and 44 the contributions at the neon masses were estimated and corrected. Such corrections were always at ,1% level. 4. Results and discussion Tables 2 and 3 show the stepped temperature noble gas and nitrogen data from the present study. Gas amounts in the combustion step are at blank levels (22 Ne , 1£ 10211 cc STP g21 , 36 Ar , 2 £ 10212 cc STP g21 , N , 50 pg with d15 N ˆ 5 ^ 3½† and have not been included

The concentrations of the noble gases in these samples are in general small. In some samples, only argon and neon are in amounts measurable for isotopic composition. Neon has a concentration (for 22Ne) of 4±5.2 (in 10 210 cc STP g 21 units) in the xenoliths, 2.4 (in 10 210 cc STP g 21 units) in the olivine phenocryst from Kutch (henceforth VTH.Ol), and varies from 1.4 to 4.3 (in 10 210 cc STP g 21 units) in the Reunion samples (Table 2). The isotopic ratios of 20 Ne/ 22Ne and 21Ne/ 22Ne in the temperature steps vary from near atmospheric (9.8 and 0.029, respectively) to slightly higher values (of 10.47 and 0.037, respectively) in these samples. On a neon three-isotope plot (Fig. 2), the present data point to two neon components Ð air and mantle (MORB and OIB), a feature that is consistent with their geochemical evolution (Ozima, 1994; Farley and Neroda, 1998). Also shown in Fig. 2 are the air±OIB (Honda et al., 1991; Valbracht et al., 1997) and air± MORB mixing trends (Sarda et al., 1988). Neon data in the Kutch xenoliths follow the air±MORB mixing trend, suggesting the presence of a MORB type component in these samples. The Reunion samples and VTH.Ol, which are cogenetic with the host basalts, follow the air±OIB mixing trend, indicating the presence of an OIB-type neon in these samples. The data plotted in Fig. 2 include only the totals from Table 2. The stepped temperature data (within errors) are also consistent with the above component structure, but create a cluster of points, and hence have not been used in the plot. Argon in the present data varies in the concentration of 36 Ar from 1 to 12.6 (in 10 210 cc STP g 21 units) in the xenoliths, 10.8 (in 10 210 cc STP g 21 units) in VTH.Ol, and 0.7± 1.2 (in 10 210 cc STP g 21 units) in the Reunion samples (Table 3). The 38Ar/ 36Ar ratios have large errors, but are essentially air-like (not reported in Table 3). 40Ar/ 36Ar varies in the temperature fractions from 376 to 10 511 in the xenoliths, 317 to 1903 in VTH.Ol, and from 350 to 3043 in the Reunion samples. The measured 40Ar/ 36Ar ratios in the Kutch samples should re¯ect their trapped gas signatures (from the mantle source), since the K content is too small (,35 ppm; Pande, 1988) to produce signi®cant in situ 40Ar p from the decay of 40K over the past 65 Ma since crystallization. Similarly, with the low crystallization age (#2 Ma) of the Reunion samples, it is not possible to produce significant 40Ar p by in situ radioactive decay of 40K. The 40 Ar/ 36Ar signature of 10 511 measured in the xenoliths can be explained by a two-component mixture of a mantle component similar to MORB (40 Ar=36 Ar $ 42 000; Burnard et al., 1997) and atmospheric argon (40 Ar=36 Ar ˆ 295:5; Ozima and Podosek, 1983). The 40Ar/ 36Ar signatures of

872

R.K. Mohapatra, S.V.S. Murty / Journal of Asian Earth Sciences 20 (2002) 867±877

Fig. 3. Plot of d 15N against the percentage release of nitrogen, showing the typical stepped temperature release patterns for d 15N. Nitrogen in these samples requires the presence of at least two components with d15 N , 215 and , 1 15½ often in the same sample (refer to text for discussion).

the Reunion samples and VTH.Ol are consistent with a mixture of an OIB type argon (40 Ar=36 Ar , 6000; Farley and Neroda, 1998) with atmospheric argon. The present noble gas data for the Reunion samples are consistent with those obtained from the earlier studies (Kaneoka et al., 1986; Staudacher et al., 1990; Moreira and Allegre, 1998), which also suggest a plume type mantle source for the Reunion basalts. The neon and argon data for the Kutch xenoliths and VTH.Ol, although they constitute the only data on the Kutch alkali basalts, are interesting in light of the observation of different mantle components (MORB and OIB type) in the mantle xenoliths and the phenocrysts (and hence in the host alkali basalts). We discuss the implications of these observations later in the paper. 4.2. Nitrogen The concentrations of nitrogen in the Kutch samples vary from 1.1 to 2.5 ppm in the xenoliths, and 1.9 ppm in VTH.Ol (Table 3). In the Reunion samples, it varies from 1.1 to 1.4 ppm. The isotopic composition of nitrogen released in the temperature steps varies over wide limits (215 to 115½), in different samples. At least two nitrogen components, one having d15 N , 215½ and the other with d15 N , 115½, are needed to explain the observed spread. A similar spread in d 15N has also been observed in mantle xenoliths from San Carlos (Mohapatra and Murty, 2000a), oceanic basalts (Mohapatra and Murty, 2000b) and diamonds (Boyd and Pillinger, 1994; Mohapatra and Murty, 2001). The light nitrogen isotopic composition (d 15N ranging from 225 to 25½) in mantle-derived samples has been shown to be of mantle origin (Javoy et al., 1984; Boyd et al., 1987; Javoy and Pineau, 1991; Boyd and Pillinger, 1994; Javoy, 1998; Cartigny et al., 1997, 1998a, b; Marty and Humbert, 1997; Sano et al., 1998; Mohapatra and Murty, 1998, 2000a, b; Marty and Zimmermann, 1999; Nishio et al., 1999) and has been regarded as

characterising pristine mantle nitrogen. Heavy nitrogen isotopic signatures in these materials can be explained either by contamination from organic nitrogen (Rau et al., 1987) or by contribution from recycled nitrogen (Haendel et al., 1986; Bebout and Fogel, 1992). Any low temperature sur®cial component in the present samples cannot survive the combustion at 4008C routinely done on each of them prior to the main gas extraction steps by pyrolysis (Murty, 1997; Mohapatra and Murty, 2000a, b). This rules out the possibility of an organic origin for the heavy nitrogen in the present samples. Fig. 3 is a plot of typical step temperature release patterns for the isotopic composition of nitrogen for the present samples. The heavy nitrogen in the present samples is released at temperatures .900 and up to 18008C, which is quite high for the survival of any kind of sur®cial contamination. A similar observation has also been made in the study of oceanic basalts (Mohapatra and Murty, 2000b) and mantle xenoliths (Mohapatra and Murty, 2000a), in which nitrogen with a d 15N as heavy as 19½ have been observed. The high release temperatures (#the melting temperatures of the host phase) together with the accompanying 40Ar/ 36Ar (often much above the atmospheric value) show that the heavy nitrogen component has been sampled from the mantle by these samples, and is of a recycled nature. Thus, a mixture of pristine mantle and recycled nitrogen can explain the whole spectrum of nitrogen data from the present study. This may appear to be inconsistent with the noble gas data, which require a two-component mixture of the mantle and atmospheric gases, but can be understood by the following reasoning. The presence of atmospheric gases will have a d 15N value (0½) that lies between the two extreme values 215 and 115½ and therefore will be indistinguishable by the nitrogen isotopic systematics. Similarly the recycled gases, which are similar to the atmospheric gases in their noble gas isotopic compositions (Staudacher and Allegre, 1988), are not distinguished by the noble gas isotopic systematics (Mohapatra and Murty, 1998, 2000a, b). Below we further investigate the component structure of these samples by using nitrogen and argon systematics together. 4.3. Nitrogen and argon isotopic systematics Mohapatra and Murty (2000a, b) have clearly shown that the nitrogen and argon isotopic compositions of mantlederived materials (ultrama®c xenoliths and MORBs) can be explained by contributions from three components: the MORB mantle (MORB: d15 N ˆ 215½ , 40 Ar=36 Ar ˆ 42 000), materials recycled into the mantle at the subduction zones (R: d15 N ˆ 119½, 40 Ar=36 Ar ˆ 375) and air-saturated water (ASW: d15 N ˆ 0½, 40 Ar=36 Ar ˆ 295:5). Although there are indications for a zoning in the 40 Ar/ 36Ar ratio between the MORB source mantle (upper mantle: 40 Ar=36 Ar $ 40 000) and the OIB source mantle (lower mantle: 40 Ar=36 Ar , 6000) (e.g. Farley and Neroda,

R.K. Mohapatra, S.V.S. Murty / Journal of Asian Earth Sciences 20 (2002) 867±877

873

Fig. 4. (a) Plot of d 15N against 40Ar/ 36Ar for data given in Table 2. Also included are the OIB data for HAW.DU from Mohapatra and Murty (2000b)and (the ®eld, shaded box) for the Iceland samples deduced from Marty and Dauphas (2000). The end-members: MORB mantle (MORB), OIB mantle (OIB), recycled materials (R) and air-saturated water (ASW) and the binary mixing lines (Langmuir et al., 1978) between them are based on the parameters given in Table 4 d 15N value for OIB source has been taken as 13½ (the question mark is to indicate the uncertainity in the OIB value). Tick marks on the MORB±R mixing curve represent mixing with different fractions (f ˆ 0:250, 0.50 and 0.75) of R. Tick marks on the MORB±ASW mixing curve represent mixing with different fractions (f ˆ 0:01 and 0.50) of ASW. (b) Same plot as (a), except that the d 15N value for the OIB source has been considered to be 215½ (the arrow and the question mark are to indicate the uncertainty in the OIB value). Tick marks on the OIB±R mixing curve represent mixing with different fractions (f ˆ 0:25, 0.50 and 0.75) of R. Tick marks on the and OIB±ASW mixing curve represent mixing with different fractions (f ˆ 0:01 and 0.50) of ASW. The R-values (ratio of 36Ar/ 14N in two components) for the mixing lines: MORB±R and OIB±R ˆ 2:7, MORB±ASW and OIB±ASW ˆ 0:002, R±ASW ˆ 0:0007, are applicable for both (a) and (b).

1998 for a recent review), available nitrogen data on oceanic basalts are inadequate at present to attest to such a zoning in the d 15N signatures of these two mantle reservoirs. To date, there are only three measurements that address the d 15N of the lower mantle, which is believed to be the source of the OIBs. Cartigny et al. (1997), from their measurement on a diamond sample from China, infer a d 15N value of 225½ for the lower mantle; Dauphas and Marty (1999), from a study of cabonatites, suggest a value of 13½, while Mohapatra and Murty (2000b) have observed a d 15N value of 215½ in a dunite sample from Reunion and about 23½ in a sample from Hawaii. The value of 225½ reported by Cartigny et al. (1997) is based on a single measurement on diamonds, but is consistent with the model of a predominant enstatite chondritic component in the Earth, which is based on results from oxygen isotopes (e.g. Javoy 1998), noble gases (Harper and Jacobsen, 1996) and major and trace element geochemistry (especially the refractory lithophile elements, e.g. WaÈnke (1981)). The positive d 15N (13½) value proposed by Dauphas and Marty (1999) could be due to the contributions from recycled materials, as the authors suggest. This is consistent with recent observations, based on carbon and oxygen isotopes, of recycled component in the carbonatites from India (Ray et al., 1999), as well as the presence of more than one nitrogen component in the apatites from carbonatites (Murty et al., 2001). If we adopt the d 15N value of 13½ for the lower mantle, it would mean that the whole lower mantle contains a signi®cant proportion of recycled materials, which is not consistent with other geophysical and geochemical observations (e.g. Zindler and

Hart, 1986; Dickin, 1995). Therefore, it would be imprudent to accept the d 15N values observed in carbonatites as directly representative of the lower mantle. On the contrary, the value of 215½ is based on measurements on oceanic basalts from the Paci®c and Indian Oceans (Mohapatra and Murty, 2000b). Such a value is in line with the value predicted by the enstatite chondritic parentage for the Earth (which is a conclusion based on the results from three different lines of arguments: oxygen isotopes, noble gases and major and trace elements), and is also consistent with the nitrogen isotopic data on diamonds (Javoy et al., 1984; Boyd et al., 1987; Boyd and Pillinger, 1994; Cartigny et al., 1997; Cartigny et al., 1998a, b; Mohapatra and Murty, 2001); oceanic basalts (Marty and Humbert, 1997; Sano et al., 1998; Nishio et al., 1999; Marty and Zimmermann, 1999; Mohapatra and Murty, 2000b), and mantle xenoliths (Mohapatra and Murty, 2000a). The available nitrogen data on the OIBs and other mantle-derived materials of lower mantle origin are too few at present to characterize the lower mantle d 15N signature with con®dence. Considering the primitive nature of the lower mantle (as deduced from the noble gases, e.g. Farley and Neroda (1998); solid element geochemistry, e.g. Dickin (1995); and the models about the origin and evolution of the Earth, e.g. Javoy (1998)), however, the lower mantle d 15N should be lighter than that (215½) of the upper mantle. The present nitrogen isotopic data as well as those of the Iceland basalts (Marty and Dauphas, 2000) are more consistent with a lower mantle d 15N value of 215½ as illustrated below (Fig. 4a and b). Fig. 4a and b are plots of d 15N against 40Ar/ 36Ar for both

874

R.K. Mohapatra, S.V.S. Murty / Journal of Asian Earth Sciences 20 (2002) 867±877

Table 4 Nitrogen and argon systematics of the end members End members

36

Ar/ 14N (10 27 units)

Air ASW R MORB OIB

202 a 420 c 0.3 d 0.8 d 0.8 i

d 15N (½)

40

Ar/ 36Ar

0b 0b 119 e 215 g 215 i 13 k

295.5 a 295.5 a 375 f $ 42 000 h 6000 j 6000 j

a

Ozima and Podosek (1983). Air is used as the standard for reporting nitrogen isotopic compositions. c Marty (1995). d After Mohapatra and Murty (2000a). e After Mohapatra and Murty (2000b). f Staudacher and Allegre (1988). g Mohapatra and Murty (1998). h Burnard et al. (1997). i The N±Ar signatures of the OIB mantle have been assumed to be similar to that for the MORB mantle (refer to text for discussion). j Farley and Neroda (1998). k Dauphas and Marty (1999). b

the step temperature data and totals (in Table 3) for the Kutch and Reunion samples. Also included are the data for a dunite sample from Hawaii (HAW.DU) from Mohapatra and Murty (2000b), and for samples from Iceland (the ®eld represented by the grey box), deduced from the data of Marty and Dauphas (2000). The end members: MORB mantle (MORB), OIB mantle (OIB), recycled (R) and air-saturated water (ASW), along with the binary mixing curves (Langmuir et al., 1978) between them are based on the parameters given in Table 4. Because of the scarcity of nitrogen data on the OIB source mantle, we have assumed its 36Ar/ 14N signature to be similar to that of the MORB source mantle. Similarly, taking into account the discrepancy in d 15N of the OIB source mantle (brought out in the preceding paragraph) we have used two plots (as per the suggestion of one of the reviewers), using the different literature d 15N values: 13 (Fig. 4a) and 215½ (Fig. 4b) for the OIB source in order to demonstrate that the choice of d15 N # 215½ for the OIB source can better explain the data. As can be seen in Fig. 4a, a d 15N value of 13½ for the OIB source mantle cannot explain the light signatures observed in RE.DU, HAW.DU and the majority of Icelandic samples. A d 15N of 215½ or lighter (Fig. 4b), however, explains all the OIB data except for the 18008C step of HAW.DU (Mohapatra and Murty, 2000b), which can be explained either by the MORB end member (which would mean limited mixing with the MORB source mantle) or a recycled end member with 40Ar/ 36Ar much higher (.10 000) than the present R. Clearly one would require more data on the Hawaiian samples to resolve this, and more data on the OIBs to better constrain the signatures of the OIB source mantle. The present uncertainties in the d 15N of the OIB source mantle are shown in Fig. 4a and b by the `?' mark. Although all (except RE.DU Ð 9008C) the data points fall within the area enclosed by MORB, R and ASW, the

data for VTH.Ol, the Reunion samples, HAW.DU (except the 18008C) and Iceland show consistently lower 40Ar/ 36Ar within the mixing ®eld of OIB±R±ASW (Fig. 4b). This suggests that while all these samples have contributions in their argon and nitrogen from the R and ASW components, they seem to have different pristine mantle components. The pristine mantle component in these samples can be MORB type (as in the xenoliths) or OIB type (as in VTH.Ol, Reunion samples, HAW.DU and Iceland samples), an observation consistent with the inferences drawn from the noble gas data (Tables 2 and 3; Kaneoka et al., 1986; Kaneoka and Takaoka, 1980; Harrison et al., 1999). The N±Ar systematics depicted in Fig. 4a and b show that nitrogen and noble gases in the mantle-derived samples form a complimentary pair of tracers. While nitrogen is an ef®cient tracer of the mantle and recycled components, noble gases are ef®cient in resolving the mantle and atmospheric components (air contamination). The stepped temperature release patterns of d 15N indicate that these components often occur as unequilibrated components (as in RE.Ol) that are released at different temperatures (Fig. 3). Olivine from the San Carlos xenoliths also show a similar feature (Mohapatra and Murty, 2000a). Below 9008C, the released nitrogen is close to atmospheric in its d 15N while at higher temperatures, it has a heavier d 15N signature re¯ecting the R component. The pristine mantle component, distinct in its d 15N, is released either at the melting step (from the lattice impurities, B835.Cp) or at the low temperature (from ¯uid inclusions, RE.DU) steps depending upon the trapping sites (Mohapatra and Murty, 2000a, b). The release patterns also show that in some samples, it is not possible to identify all these components (e.g. VTH.Ol), as the components have equilibrated with each other. To summarize, the present nitrogen and light noble gas data indicate the presence of a MORB type component in the Kutch xenoliths and an OIB type component in the olivine phenocryst from Kutch (VTH.Ol) and in the samples from Reunion. These observations provide constraints on the nature of their mantle sources. 4.4. Geochemical evolution The stepped temperature nitrogen and argon data reveal the presence of three, often unequilibrated, components (ASW, R and pristine mantle) in the present samples, which is consistent with the recent nitrogen and argon data on oceanic basalts (Mohapatra and Murty, 2000b) and mantle xenoliths (Mohapatra and Murty, 2000a). This has useful implications for the geochemical evolution of these samples. The ASW component is released during low temperature (#9008C) extractions. Such a component is ubiquitously observed in the noble gas signatures of mantle-derived materials, and has been explained by contamination from atmospheric gases. It is commonly explained by limited interaction of these samples with

R.K. Mohapatra, S.V.S. Murty / Journal of Asian Earth Sciences 20 (2002) 867±877

surface waters during or after their emplacement on the surface of the Earth (e.g. Farley and Neroda, 1998; Mohapatra and Murty, 2000a). However, Ballentine and Barford (2000) have recently shown that an atmospheric component, which survives the usual preheating of these samples at 1508C, could also be from modern air incorporated during sample preparation. Such a possibility is doubtful in the present case since, in addition to the usual preheating, the samples have been routinely combusted at 4008C prior to the main gas extraction by pyrolysis, which ef®ciently removes any sur®cial component (Murty, 1997; Mohapatra and Murty, 2000a). Below the veil of atmospheric contamination, the samples show the simultaneous presence of recycled and pristine mantle components. These samples are expected to have a mantle component trapped during their formation within the mantle. The step temperature release of nitrogen (Fig. 3) and the accompanying non-atmospheric 40Ar/ 36Ar ratios suggest that the recycled component was acquired by these samples during their mantle residence. It is important to consider the source of the recycled component in the mantle and how it was acquired by these samples, in order to understand their geochemical evolution and overall mantle geochemistry. The presence of an unequilibrated recycled component along with the pristine mantle component has been observed in mantle xenoliths from the SCM (Mohapatra and Murty, 2000a; Dunai and Baur, 1995) and in the samples from Island Arc settings (Matsumoto et al., 2001). This is consistent with observation from the solid element geochemistry (e.g. Hawkesworth et al., 1984; Zindler and Jagoutz, 1988), and is commonly explained by the presence of metasomatic ¯uids (from subduction zones) in the SCM. However, the simultaneous presence of the pristine mantle and recycled components in the basalts (as shown by the N±Ar signatures of VTH.Ol and the Reunion samples), which produces a small-scale (,1 mm) heterogeneity in these samples, is very dif®cult to explain by the conventional models for their geochemical evolution. Basalts form by partial melting of large portions (several cubic kilometers) of the mantle, a process, which would smoothen any small-scale heterogeneity in their mantle source. Therefore an unequilibrated recycled component in these basalts could not have been inherited from their mantle source. On the other hand, it is possible to introduce the recycled component late in the magmatic history so that it does not have a chance to completely mix with the ambient mantle component. This can be explained if we have a global shallow horizon in the mantle that contains the recycled gases (Mohapatra and Murty, 1998, 2000b). Such a zone develops in the mantle from ¯uids (CO2 ±H2O rich) that are expelled from the subducting slabs. Because of their much lower density (as compared to surrounding mantle) these ¯uids migrate upward in the mantle until their vertical motion is arrested by the overlying lithosphere, and they spread laterally to form a shallow global zone that is enriched in the recycled materials (Anderson, 1994; Mohapatra and Murty, 1998, 2000b). Thus the model of a global shallow enriched mantle

875

(EM) does not require the actual presence of a subduction zone near the places (e.g. Kutch or Reunion) where the samples have been emplaced at the Earth's surface. Being a buoyant and plastic zone, the EM remains as a passive layer at the lithosphere/asthenosphere (L/A) boundary and is not affected by the mantle convection, thus preserving its identity in the mantle. As it is present as a global layer at the L/A boundary, EM interacts with every magma that leaves the mantle. Interaction of the magma with EM is for a short duration, which does not always allow complete equilibration of the pristine mantle and recycled components. The presence of a recycled component has recently been observed in mid-oceanic ridge basalts from the Atlantic (Sarda et al., 1999), Paci®c (Cousens, 1996; Mohapatra and Murty, 2000b) and the Indian Oceans (Mohapatra and Murty, 2000b), and is well explained by the EM. Because of its mobile behaviour, EM migrates as metasomatic ¯uids into the SCM through fractures (Haggerty, 1987) at the base of the continental lithosphere. Since the SCM is a rigid body, the recycled component carried in the invading ¯uids does not often mix with the ambient pristine mantle component. Therefore mantle xenoliths from the SCM contain both the pristine mantle and recycled components, as observed in the present samples and in the mantle xenoliths of San Carlos (Mohapatra and Murty, 2000a). The present data, which show the ubiquitous presence of recycled nitrogen and noble gases in mantle-derived materials, has interesting implications for the solid element (Pb, Sr, Nd, etc.) geochemistry that is used to model their mantle sources. Although the concept of mantle enrichment discussed above concerns nitrogen and noble gases only, recent correlation (Sarda et al., 1999) of the isotopic systematics of argon (a noble gas) with that of the solid elements (Pb, Nd and Sr) suggests that varying contributions of recycled component in these elements cannot be ignored. Because of the ubiquitous presence of the recycled component in mantle-derived materials (acquired probably late in their geochemical evolution), their measured geochemical signatures cannot be directly used to derive the geochemical signatures of their pristine mantle sources. 4.5. Nature of mantle sources for Kutch and Reunion VTH.Ol and RE.Ol are phenocrysts while RE.DU is a cognate nodule, all of which are the result of early crystallization of their parent magmas and would have trapped gases from the magma. Therefore, their nitrogen and noble gas isotopic compositions can be considered similar to that of the magma that formed their host basalts (Kutch alkali basalts and Reunion basalts). The present noble gas data for the alkali basalts of Kutch, although very few at the moment, and Reunion suggest an OIB type mantle source for these rocks, and are consistent with a plume origin for these rocks which has also been inferred from various other geochemical and geophysical observations (Mahoney, 1988; White and McKenzie, 1995). For example, helium

876

R.K. Mohapatra, S.V.S. Murty / Journal of Asian Earth Sciences 20 (2002) 867±877

and neon isotopic compositions of the Reunion basalts have been interpreted as indicators for a plume type mantle source for Reunion (Kaneoka et al., 1986; Graham et al., 1990; Staudacher et al., 1990; Moreira and Allegre, 1998). Similarly, observations of plume type 3He/ 4He for the Deccan and its associated alkali basalts also suggest a plume type mantle source for these rocks (Basu et al., 1993). The mantle xenoliths from Kutch contain a MORB type noble gas and nitrogen component. Thermo-barometric data place their source in the SCM (lithosphere) at ,70 km depth (Krishnamurthy et al., 1988). The solid element geochemistry of these xenoliths also suggests a MORB type mantle source (Krishnamurthy et al., 1988; Mahoney, 1988; Pande, 1988) and requires the presence of a MORB type lithosphere beneath Kutch. A similar conclusion has also been drawn for other xenolith-bearing localities in the world (Zindler and Jagoutz, 1988; Dunai and Baur, 1995; Mohapatra and Murty, 2000a). A MORB-type reservoir in the lithosphere beneath continents is a remnant of the continental crust extraction process from the upper mantle early (,3 Ga ago) in the Earth's history (Zindler and Jagoutz, 1988; Haggerty, 1987). The mantle xenoliths of Kutch could be samples from such a reservoir, and have been carried in the host alkali basalt (from an OIB type mantle source) while the latter was passing through the lithosphere. 5. Summary We have studied mantle-derived ma®c and ultrama®c inclusions from Kutch and the Reunion Islands. Our nitrogen and light noble gas data show the presence of an OIB-type mantle component in the Kutch alkali basalts and the Reunion basalts. This is consistent with the model that the Kutch alkali basalts and the Deccan Traps volcanism have resulted from the activity of the Reunion hotspot ,65 My ago. The mantle xenoliths found in the Kutch alkali basalts, however, contain a dominant MORB type mantle component and require the presence of a MORB type lithosphere beneath Kutch. The present data shows the ubiquitous presence of variable proportions of recycled nitrogen and argon, consistent with the presence of a global shallow zone at the lithosphere/asthenosphere boundary that was enriched in ¯uids derived from subducting slabs. This also suggests that variable contributions from the recycled component cannot be ignored in the solid incompatible elements (Pb, Nd, Sr, etc.) which in turn shows that the measured geochemical signatures of these samples cannot be directly related to their mantle source composition. Acknowledgements We are thankful to Prof. Ichiro Kaneoka of the Tokyo University for providing the Reunion samples. We are grateful to Dr Kanchan Pande of PRL for providing most

of the Kutch samples from his collection, his help in a ®eld trip to Bhuj in 1994, and his keen interest in the present study. Constructive and critical reviews from two anonymous reviewers and suggestions from the editor, Kevin Burke have been very helpful in improving the manuscript. We dedicate this paper to the victims of the recent killer earthquake in Kutch.

References Allegre, C.J., Staudacher, Th., Sarda, P., 1986/87. Rare gas systematics: formation of the atmosphere, evolution and structure of the Earth's mantle. Earth Planet. Sci. Lett. 81, 127±150. Anderson, D.L., 1994. Lithosphere, asthenosphere, and perisphere. Rev. Geophys. 33, 125±149. Ballentine, C.J., Barford, D.N., 2000. The origin of air-like noble gases in MORB and OIB. Earth Planet. Sci. Lett. 180, 39±48. Basu, A., Renne, P.R., DasGupta, D.K., Teichmann, F., Poreda, R.J., 1993. Early and late alkali igneous pulses and a high- 3He plume origin for the Deccan ¯ood basalts. Science 261, 902. Baxter, A.N., 1990. Major and trace element variations in basalts from LEG115. Proc. Ocean Drilling Prog. 115, 11±21 Scienti®c Results. Bebout, G.E., Fogel, M.L., 1992. Nitrogen-isotopic composition of metasedimentary rocks in Catalina Schist, California: implications for metamorphic devolatization history. Geochim. Cosmochim. Acta 56, 2139± 2149. Biswas, S.K., 1970. A note on the geology of Kutch. Quart. J. Geol. Min. Melt. Soc. India 43, 223±235. Boyd, S.R., Pillinger, C.T., 1994. A preliminary study of 15N/ 14N in octahedral growth form diamonds. Chem. Geol. 116, 43±59. Boyd, S.R., Mattey, D.P., Pillinger, C.T., Milledge, H.J., Mendelssohn, M., Seal, M., 1987. Multiple growth events during diamond genesis: an integrated study of carbon and nitrogen isotopes and nitrogen aggregation state in coated stones. Earth Planet. Sci. Lett. 86, 341±353. Boyd, S.R., Pillinger, C.T., Milledge, H.J., Mendelssohn, M., Seal, M., 1992. C and N isotopic composition and infrared absorption spectra of coated diamonds: evidence for the regional uniformity of CO2 ± H2O rich ¯uids in lithospheric mantle. Earth Planet. Sci. Lett. 109, 633±644. Burnard, P.G., Graham, D., Turner, G., 1997. Vesicle-speci®c noble gas analyses of popping rock: implications for primordial noble gases in the Earth. Science 276, 568±571. Cartigny, P., Boyd, S.R., Harris, J.W., Javoy, M., 1997. Diamonds and the isotopic composition of mantle N. Terra Nova 9, 175±179. Cartigny, P., Harris, J.W., Javoy, M., 1998a. Eclogitic diamond formation at Jwaneng: no room for a recycled component. Science 280, 1421±1424. Cartigny, P., Harris, J.W., Phillips, D., Girard, M., Javoy, M., 1998b. Subduction-related diamonds? Ð The evidence for a mantle-derived origin from coupled d 13C±d 15N determinations. Chem. Geol. 147, 147±159. Cousens, B.L., 1996. Enriched upper mantle sources for Basaltic rocks from diverse environments in the northern Paci®c ocean: the generation of oceanic alkaline vs. tholeiitic basalts. In: Hart, S., Basu, A. (Eds.). Earth Processes: Reading the Isotopic Code. Monograph 95. Am. Geophys. Union, pp. 207±231. Dauphas, N., Marty, B., 1999. Heavy nitrogen in carbonatites of the Kola pennisula: a possible signal of the deep mantle. Science 286, 2488±2490. De, A., 1964. Iron titanium oxides and silicate minerals of the alkali olivine basalts, tholeiites, and acidic rocks of the Deccan Trap series and their signi®cance. Int. Geol. Cong. Rept., 22nd session, New Delhi, Part III, pp. 126±138. Dickin, A.P., 1995. Radiogenic Isotope Geology. Cambridge University Press, Cambridge. Dunai, T.J., Baur, H., 1995. Helium, neon, and argon systematics of the

R.K. Mohapatra, S.V.S. Murty / Journal of Asian Earth Sciences 20 (2002) 867±877 European sub-continental mantle: implications for its geochemical evolution. Geochim. Cosmochim. Acta 59, 2767±2783. Farley, K.A., Neroda, E., 1998. Noble gases in the Earth's mantle. Ann. Rev. Earth Planet. Sci. 26, 189±218. Graham, D., Lupton, J., Albarede, F., Condomines, M., 1990. Extreme temporal homogeneity of helium isotopes at Piton de la Fournaise, Reunion Islands. Nature 347, 545±548. Haendel, D., Muhle, K., Nitzsche, H., Stiehl, G., Wand, U., 1986. Isotopic variations of the ®xed N in metamorphic rocks. Geochim. Cosmochim. Acta 50, 749±758. Haggerty, S.E., 1987. Diamond genesis in a multiply Ð constrained model. Nature 320, 34±38. Harper Jr., C.L., Jacobsen, S.B., 1996. Noble gases and Earth accretion. Science 273, 1814±1818. Harrison, D., Burnard, G., Turner, G., 1999. Noble gas behaviour and composition in the mantle: constraints from the Iceland plume. Earth Planet. Sci. Lett. 171, 199±207. Hawkesworth, C.J., Rogers, N.W., van Calsteren, P.W.C., Menzies, M.A., 1984. Mantle enrichment processes. Nature 311, 331±335. Honda, M., McDougall, I., Patterson, D., Doulgeris, A., Clauge, D., 1991. Possible solar noble gas component in Hawaiian basalts. Nature 349, 149±151. Javoy, M., 1998. The birth of the Earth's atmosphere: the behavior and fate of its major elements. Chem. Geol. 147, 11±26. Javoy, M., Pineau, F., 1991. The volatiles record of a popping rock from the mid-Atlantic ridge at 148N: chemical and isotopic composition of gas trapped in the vesicles. Earth Planet. Sci. Lett. 107, 598±611. Javoy, M., Pineau, F., Demaiffe, D., 1984. Nitrogen and carbon isotopic composition of the Mbuji Mayi (Zaire). Earth Planet. Sci. Lett. 68, 399± 412. Kaneoka, I., Takaoka, N., 1980. Rare gas isotopes in Hawaiian ultrama®c nodules and volcanic rocks: constraints on genetic relationships. Science 208, 1366±1368. Kaneoka, I., Takaoka, N., Upton, B.G.J., 1986. Noble gas systematics in basalts and dunite nodule from Reunion and Grand Comore islands, Indian Ocean. Chem. Geol. (Isot. Geo. Sci.) 59, 35±42. Krishnamurthy, P., Pande, K., Gopalan, K., Mcdougall, J.D., 1988. Upper mantle xenoliths in alkali basalts related to Deccan Trap volcanism. J. Geol. Soc. India 10, 53±67. Langmuir, C.L., Vocke Jr., R.D., Hanson, G.N., 1978. A general mixing equation with applications to Iceland basalts. Earth Planet. Sci. Lett. 37, 380±392. Mahoney, J.J., 1988. Deccan Traps. In: Macdougall, J.D. (Ed.). Continental Flood Basalts. Kluwer, Dordrecht, pp. 151±194. Marty, B., 1995. Nitrogen content of the mantle inferred from N±Ar correlation in oceanic basalts. Nature 377, 326±329. Marty, B., Humbert, F., 1997. Nitrogen and argon isotopes in oceanic basalts. Earth Planet. Sci. Lett. 152, 101±112. Marty, B., Zimmermann, L., 1999. Volatiles (He, C, N, Ar) in mid-oceanic ridge basalts: assessment of shallow-level fractionation and characterization of source composition. Geochim. Cosmochim. Acta 63, 3619±3633. Marty, B., Dauphas, N., 2000. Nitrogen isotope systematics of the mantle and the fate of organic matter through time. Goldschmidt Conference 2000. J. Conf. Abstr. 5 (2), 672±673. Matsumoto, T., Chen, Y., Matsuda, J., 2001. Concomitant occurrence of primordial and recycled noble gases in the Earth's mantle. Earth Planet. Sci. Lett. 185, 35±47. Mohapatra, R.K., 1998. Study of the abundance and isotopic composition of nitrogen in Earth's mantle. PhD Thesis, M.S. University of Baroda, India. Mohapatra, R.K., Murty, S.V.S., 1998. Nitrogen in the Earth's mantle: inferences from a comprehensive study of the mantle-derived materials. Origin of the Earth and Moon Conference, Monterey, California, A4008.

877

Mohapatra, R.K., Murty, S.V.S., 2000a. Search for mantle N in the ultrama®c xenoliths from San Carlos, Arizona. Chem. Geol. 164, 305±320. Mohapatra, R.K., Murty, S.V.S., 2000b. Origin of air-like noble gases in oceanic basalts. Geophys. Res. Lett. 27, 1583±1586. Mohapatra, R.K., Murty, S.V.S., 2001. Nitrogen components in diamonds, (in preparation). Morgan, W.J., 1981. Hotspot tracks and the opening of the Atlantic and Indian Oceans. In: Emiliani, C. (Ed.). The Sea, vol. 7. Wiley Interscience, New York, pp. 443±475. Moreira, M., Allegre, C.J., 1998. Helium±neon systematics and the structure of the mantle. Chem. Geol. 147, 53±59. Murty, S.V.S., 1997. Noble gases and nitrogen in Muong Nong Tektites. Meteor. Planet. Sci. 32, 687±691. Murty, S.V.S., Basu, S., Kumar, A., 2001. Nitrogen components in carbonatites: clues to the magma source. Symposium on Carbonatites and Associated Alkaline Rocks, Chennai, India, pp. 18±19. Nishio, Y., Ishii, T., Gamo, T., Sano, Y., 1999. Volatile element isotopic systematics of the Rodrigues Triple Junction Indian Ocean MORB. Earth Planet. Sci. Lett. 170, 241±253. Ozima, M., 1994. Noble gas state in the mantle. Rev. Geophys. 32, 405± 426. Ozima, M., Podosek, F.A., 1983. Noble Gas Geochemistry. Cambridge University Press, Cambridge. Pande, K., 1988. Isotopic and chemical studies of ultrama®c nodules in alkali basalts from Kutch, Western India. PhD Thesis, Gujarat University, India. Rau, G.H., Arthur, M.A., Dean, W.E., 1987. 15N/ 14N variations in Cretaceous Atlantic sequences: implications for past changes in marine nitrogen biogeochemistry. Earth Planet. Sci. Lett. 82, 269±279. Ray, J.S., Ramesh, R., Pande, K., 1999. Carbon isotopes in Kerguelen Plume derived carbonatites: evidence for recycled inorganic carbon in carbonatites. Earth Planet. Sci. Lett. 170, 205±214. Sano, Y., Takahata, N., Nishio, Y., Marty, B., 1998. Nitrogen recycling in subduction zones. Geophys. Res. Lett. 25, 2289±2292. Sarda, P., Staudacher, T., Allegre, C.J., 1988. Neon isotopes in submarine basalts. Earth Planet. Sci. Lett. 91, 73±88. Sarda, P., Moreira, M., Staudacher, T., 1999. Argon±lead isotopic correlation in the mid-Atlantic Ridge Basalts. Science 283, 666±668. Staudacher, T., Allegre, C.J., 1988. Recycling of oceanic crust and sediments: the noble gas subduction barrier. Earth Planet. Sci. Lett. 89, 173±183. Staudacher, T., Sarda, P., Allegre, C.J., 1990. Noble gas systematics of Reunion Island, Indian Ocean. Chem. Geol. 89, 1±17. Tolstikhin, I.N., Marty, B., 1998. The evolution of terrestrial volatiles: a view from helium, neon and argon isotope modeling. Chem. Geol. 147, 27±52. Valbracht, P., Staudacher, T., Malahoff, A., Moreira, M., Allegre, C.J., 1997. Noble gas systematics of deep rift zone glasses from Loihi Sea Mount, Hawaii. Earth Planet. Sci. Lett. 150, 399±441. Venkatesan, T.R., Pande, K., Gopalan, K., 1993. Did Deccan volcanism pre-date the Cretaceous±Tertiary transition? Earth Planet. Sci. Lett. 119, 181. WaÈnke, H., 1981. Composition of terrestrial planets. Phil. Trans. R. Soc. Lond. A 303, 287±302. White, W.M., Cheatham, M.M., Duncan, R.A., 1990. Isotope geochemistry of leg 115 basalts and inferences on the history of the Reunion mantle plume. Proc. Ocean Drilling Prog. 115, 53±61 Scienti®c Results. White, R.S., McKenzie, D., 1995. Mantle plumes and ¯ood basalts. J. Geophys. Res. 100, 17543±17585. Zindler, A., Hart, S.R., 1986. Chemical geodynamics. Ann. Rev. Earth Planet. Sci. 14, 493±571. Zindler, A., Jagoutz, E., 1988. Mantle cryptology. Geochim. Cosmochim. Acta 52, 321±350.