Dongre 2015

Available online at www.sciencedirect.
com
ScienceDirect
Geochimica et Cosmochimica Acta 166 (2015) 165–188
www.elsevier.com/locate/gca
Subduction-related origin of eclogite xenoliths from

the Wajrakarur kimberlite field, Eastern Dharwar
craton, Southern India: Constraints from
petrology and geochemistry
A.N. Dongre a,1, D.E. Jacob b,c,⇑, R.A. Stern d
a
Post Graduate Department of Geology, Institute of Science, Caves Road, Aurangabad 431004, India
b
Australian Research Council Centre of Excellence for Core to Crust Fluid Systems and Department of Earth and Planetary Sciences,
Macquarie University, North Ryde 2109, NSW, Australia
c
Earth System Science Research Centre, Johannes Gutenberg University, Mainz, Germany
d
Canadian Centre for Isotopic Microanalysis, Department of Earth and Atmospheric Sciences, University of Alberta, Edmonton, Canada
Received 8 November 2014; accepted in revised form 17 June 2015; Available online 24 June 2015
Abstract
Major and trace elements as well as the first oxygen isotopes are reported on eclogite xenoliths from the Mesoproterozoic
KL2 and P3 kimberlite pipes of the Wajrakarur kimberlite field, Eastern Dharwar craton, Southern India. Garnets in
kyanite-bearing samples are rich in grossular, whereas they are predominantly pyrope-almandines in the bimineralic
(kyanite-free) samples. The kyanite eclogite from the P3 pipe is more Mg-rich than those from KL2 pipe. Equilibration tem-
peratures indicate derivation from 4.5 to 5.3 GPa and 1060 to 1220 °C for the KL2 samples and 3.6 GPa, 918 °C for the P3
sample.
Garnet rare earth element patterns show two characteristic types, one with relatively low and flat heavy rare earth element
patterns: Wajrakarur Group 1 and a second with lower light to heavy rare earth element ratios: Wajrakarur Group 2. Most
samples in Wajrakarur Group 1 show pronounced positive Eu anomalies in garnet and positive Eu and Sr anomalies in the
reconstructed whole rock trace element patterns; these are among the strongest anomalies in eclogite xenoliths worldwide. In
contrast, Wajrakarur Group 2 samples show only subtle positive Eu anomalies.
Oxygen isotopic ratios of garnets range between +5.3& and +7.8& d18O. This range extends significantly beyond the
range for unchanged mantle. Similar to many other eclogite suites worldwide, the Wajrakarur Group 1 and Group 2 eclogite
suites shows evidence for an origin as crustal gabbroic material, likely once part of the oceanic crust, which was subducted and
imbricated under the Eastern Dharwar craton. Their surface origin therefore lends support to geodynamic models that favor
amalgamation of the Dharwar craton by subduction.
Ó 2015 Elsevier Ltd. All rights reserved.
⇑ Corresponding author at: Australian Research Council Centre

1. INTRODUCTION
of Excellence for Core to Crust Fluid Systems and Department of
Earth and Planetary Sciences, Macquarie University, North Ryde
Eclogite xenoliths erupted with kimberlites provide
2109, NSW, Australia. Tel.: +61 298508428; fax: +61 298506904.
E-mail address: [email protected] (D.E. Jacob). information on the composition and nature of deeper parts
1
Present address: Department of Geology, University of of the lithospheric mantle. It is therefore no surprise that
Johannesburg, PO Box 524, Auckland Park 2006, South Africa. eclogites have long been the subject of special study since
http://dx.doi.org/10.1016/j.gca.2015.06.023
0016-7037/Ó 2015 Elsevier Ltd. All rights reserved.
166 A.N. Dongre et al. / Geochimica et Cosmochimica Acta 166 (2015) 165–188
their discovery by Hauy (1822). Eclogite xenoliths are usu- The craton is divided into two blocks, the Eastern
ally interpreted either as fragments of subducted oceanic Dharwar craton (EDC) and Western Dharwar craton
crust representing gabbroic or basaltic protoliths or oceanic (WDC), which have apparently distinct histories of crustal
crustal cumulates emplaced tectonically into the litho- evolution (Swaminath et al., 1976; Chadwick et al., 2000;
spheric mantle (Shervais et al., 1988; Taylor and Neal, Ramakrishnan and Vaidyanadhan, 2010). Based on age
1989; Neal et al., 1990; Jerde et al., 1993; Jacob et al., data compilations, a recent study divided the craton into
1994; Viljoen et al., 1996; Snyder et al., 1997; Jacob and three crustal provinces by subdividing the Eastern
Foley, 1999; Barth et al., 2001; Shirey et al., 2001; Jacob, Dharwar craton into a central and an eastern province
2004; Usui et al., 2006; Nikitina et al., 2014), or as (Peucat et al., 2013). The prominent shear zone found along
high-pressure, magmatic cumulates (MacGregor and the eastern margin of the Chitradurga greenstone belt, the
Carter, 1970; Hatton, 1978; Smyth et al., 1989; Viljoen Chitradurga Boundary Fault (CBF, Fig. 1) marks the
et al., 1996; Snyder et al., 1997; Barth et al., 2002; Taylor boundary between these blocks (Chadwick et al., 2003;
et al., 2003; Heaman et al., 2006; Appleyard et al., 2009; Jayananda et al., 2006; Chardon and Jayananda, 2008;
Jacob et al., 2009). Some eclogites have been linked to par- Chardon et al., 2011). Most of the juvenile crust in the
tial melting and melt-mantle reaction processes associated WDC formed at 3.4–3.0 Ga, whereas the EDC is domi-
with subduction, representing residues of tonalite melt nated by 2.7–2.5 Ga crust (Chadwick et al., 2000). A major
extraction from basaltic crust (Ireland et al., 1994; part of the exposed WDC is occupied by largely unclassified
Rollinson, 1997; Jacob and Foley, 1999; Barth et al., gneissic granitoids termed the ‘older gneissic complex’
2002; Rapp et al., 2003; Smart et al., 2009; Tappe et al., (Radhakrishna and Vaidyanadhan, 1997), composed
2011; Pernet-Fisher et al., 2014; Smit et al., 2014). mainly of polyphase TTGs that range from 3.4 to 3.0 Ga
In contrast to xenolith populations at most other kim- in age (Ramakrishnan et al., 1984; Taylor et al., 1984;
berlite occurrences, eclogites dominate over peridotite in Jayananda et al., 2000; Bidyananda et al., 2003, 2011).
the mantle xenolith population in the on-craton KL2 kim- The EDC, in contrast, is composed prominently of linear
berlite pipe of the Wajrakarur kimberlite field, Eastern N–S trending belts of Neoarchaean tonalitic to granodi-
Dharwar craton, Southern India (Fig. 1). However, com- oritic gneissic rocks (younger gneissic complexes) inter-
pared to eclogite xenoliths from South African, Siberian leaved with narrow 2.7 Ga greenstone belts (Chadwick
and Canadian cratons, available data and studies on eclog- et al., 2000; Jayananda et al., 2000; Ramakrishnan and
ite xenoliths from India are sparse. A few authors, includ- Vaidyanadhan, 2010; Bidyananda et al., 2011).
ing Patel et al. (2006) and Patel et al. (2009), have studied Geochronological data in the Central Dharwar craton
eclogite and other mafic xenoliths from the Eastern where kimberlites are erupted show younger TTG gneisses
Dharwar craton, focussing on their major element chem- mainly range from 2.7 to 2.6 Ga, with rare remnants of
istry and petrology. No detailed studies of the trace element older 3.37–3.0 Ga migmatitic gneisses (Peucat et al., 1993,
geochemistry and stable isotopes of these eclogites have 2013; Balakrishnan et al., 1999; Jayananda et al., 2000,
been published. 2013a,b; Rogers et al., 2007).
This study reports the first coherent set of major and A considerable part of the Dharwar craton represents
trace element compositions as well as of garnet d18O values juvenile addition to the crust from the mantle (Dey,
obtained by Secondary Ion Mass Spectrometry (SIMS) of 2013). A thicker lithospheric keel (260–280 km) under the
eclogite xenoliths from the KL2 and P3 kimberlite pipes WDC and comparatively thinner lithospheric mantle keel
from the Wajrakarur kimberlite field. These are used to (200 km) beneath the EDC was inferred based on seismic
investigate the concealed evolutionary record of eclogites data (Gupta et al., 2003a). However, later studies using seis-
and to constrain possible protoliths. mic receiver function data indicate a lithospheric thickness
6100 km beneath most of Peninsular India (Kumar et al.,
2. GEOLOGICAL SETTING 2007). Therefore, the thick lithospheric root sampled by
the 1.1 Ga Wajrakarur kimberlites may have been
The Archean Dharwar craton comprises granite–green- destroyed during the breakup of Gondwanaland thus
stone terranes and gneissic basement of tonalite–trondhje allowing rapid northward movement of the Indian conti-
mite–granodiorite (TTG) composition known as the nent (Griffin et al., 2009).
Peninsular Gneisses (Naqvi and Rogers, 1987). The craton Kimberlites present in the Dharwar craton are restricted
is bordered in the east by the Proterozoic Eastern Ghats to the EDC, east of the Chitradurga shear zone and are
Mobile Belt (EGMB), in the northeast by the Archean mainly distributed in three fields; the Wajrakarur kimberlite
Bastar craton, and is covered to the northwest by the field (WKF) in the southern part of EDC, the Narayanpet
Cretaceous-Tertiary lava flows of the Deccan Traps kimberlite field (NKF) in the northern part of EDC and the
(Fig. 1). A prominent feature of the Dharwar craton is Raichur kimberlite field (RKF) located between these
the north–south trending, 400 km long and 20–30 km (Fig. 1). Most of the WKF pipes are diamondiferous, but
wide cluster of plutons known as the Closepet Granite, diamonds have not been found in the NKF and RKF pipes
dated at 2.51 Ga (Friend and Nutman, 1991). (Neelakantam, 2001). The WKF contains more than 30
Sedimentary rocks of Meso- to Neoproterozoic intracra- individual kimberlite pipes spread out over four distinct
tonic sedimentary basins such as the Cuddapah basin clusters, namely Wajrakarur, Chigicherla, Kalyandurg
unconformably overlie the granite–greenstone terrane in and Timmasamudram (Nayak and Kudari, 1999;
the eastern part. Choudary et al., 2007). Radiometric ages determined using
A.N. Dongre et al. / Geochimica et Cosmochimica Acta 166 (2015) 165–188 167
Fig. 1. Geological map of southern India showing location of the KL2 and P3 kimberlite pipes modified after Patel et al. (2009). EDC:
Eastern Dharwar craton, WDC: Western Dharwar craton, BC: Bastar craton, EGMB: Eastern Ghat mobile belt, CG: Closepet granite, GG:
Godavari graben, NKF: Narayanpet kimberlite field, RKF: Raichur kimberlite field, WKF: Wajrakarur kimberlite field. CBF: Chitradurga
boundary fault.
Ar/Ar and U–Pb isotope systems on mineral separates (phl- to white in color. Many samples show prominent fluid
ogopite and perovskite) demonstrate that kimberlite tracks across minerals, associated with bleaching of the
emplacement in all fields in the Eastern Dharwar craton clinopyroxenes (Fig. 2a); none of the samples were found
took place during the Mesoproterozoic, at ca. 1.1 Ga to contain diamonds.
(Gopalan and Kumar, 2008; Osborne et al., 2011; In most kimberlites of the EDC, eclogitic xenoliths are
Chalapathi Rao et al., 2013). Therefore studies of kimber- greatly subordinate in number to peridotite xenoliths
lites and their xenoliths EDC offer excellent opportunities (Ganguly and Bhattacharya, 1987; Nehru and Reddy,
to obtain insight into Proterozoic lithosphere composition 1989). However, the KL2 pipe in the Kalyandurg cluster
bearing on the formation of Gondwana and the breakup of WKF is unusual, as eclogites constitute more than 95%
of the Rodinia supercontinent. of the xenolith population. Such dominance of eclogitic
compositions in a kimberlitic xenolith suite is otherwise
3. MATERIALS AND METHODS only known from the Roberts Victor kimberlite in South
Africa (MacGregor and Carter, 1970) and the
3.1. Samples and Petrography Zagadochnaya pipe in Siberia (Sobolev, 1977). Eclogite
xenoliths of pipe KL2 include both bimineralic and
Twenty-eight eclogite xenoliths from the Wajrakarur kyanite-bearing types (Rao et al., 2001; Anand et al.,
kimberlite field were examined in this study (27 from KL2 2008; Misra et al., 2008). Patel et al. (2006, 2009) observed
and 1 from P-3 pipes). All eclogites are coarse to medium three types of eclogite xenoliths in this pipe: bimineralic
grained, massive and show compositional banding. The eclogites, enstatite-bearing eclogites, and celsian-bearing
xenoliths vary from angular, sub-angular to oval in shape kyanite eclogites. Babu et al. (2008) also reported coesi-
and are from 1.5 to 4 cm. Hand specimens are grayish black te/quartz and coesite/quartz–kyanite-bearing varieties.
The suite studied here only consists of bimineralic and or bleached along cross-cutting fluid tracks (Fig. 2a). Very
kyanite-bearing eclogites with or without accessory rutile. few completely fresh grains are seen, but many grains con-
All samples are affected by modal kimberlitic alteration tain unaltered cores. In kyanite-bearing samples, clinopy-
and it is possible that quartz, which can be present in eclog- roxenes are particularly strongly altered (Fig. 3a) forming
ite xenoliths (Jacob et al., 2003), was not observed in this a whitish mass consisting of a range of sub-micrometer
suite because it was consumed during this late-stage minerals including zeolites (Berg, 1968; Chinner and
carbonate-rich alteration event. Cornell, 1974; Anand et al., 2008).
Garnets are subhedral to anhedral and show orange, Kyanite varies in shape and occurs as angular elongated
brown to honey brown color (Fig. 2a, b). Some subhedral to skeletal grains and as angular inclusions in clinopyroxene
or rounded garnets are set in a matrix of clinopyroxene (Fig. 2c, e). In some samples, however, kyanite shows
(Group I following the textural classification of curved grain boundaries that are indicative of subsolidus
MacGregor and Carter, 1970; Fig. 2a), whereas other gar- replacement by clinopyroxene (Fig. 3a, b, Lappin and
nets have straight outlines and interlocking fabrics Dawson, 1975). Patel et al. (2006, 2009) found that kyanites
(Group II; Fig. 2b). They are isotropic, fractured and in some of their samples displayed evidence of
rimmed by kelyphite (Fig. 2d). Clinopyroxenes are pale intra-crystalline strain such as bent cleavage traces, undu-
green to white in color (Fig. 2b), appear cloudy in lose extinction, and deformation bands; similar to those
plane-polarized light and are altered along grain boundaries found in kyanite-bearing eclogite xenoliths from other
Fig. 2. Eclogite xenoliths in thick and thin sections (a) Petrographic Group I (after MacGregor and Carter, 1970) eclogite xenolith with
rounded garnets set in a clinopyroxene matrix; (b) Petrographic Group II eclogite xenolith showing straight grain boundaries and interlocking
fabric in garnets and clinopyroxenes; (c) eclogite xenoliths with tiny, elongated kyanite crystals (ky); (d) Garnets showing kelyphitic rims (in
PPL); (e) Eclogite xenolith with fresh clinopyroxene crystal (under crossed nicols); (f) Rutile crystal with ilmenite exsolution (in PPL).
localities. Rutile is observed as a discrete phase procedure (Armstrong, 1995). Detection limits were gener-
only in some of the “bimineralic” eclogites and shows ally between 0.01 and 0.07 wt.%. The operating accelerating
exsolution lamellae of ilmenite (Fig. 2f), similar to voltage was 15 kV and 12 nA for all phases, and the beam
previously described samples from this locality (Patel size was 2 lm. The elemental maps were collected at 20 kV
et al., 2009). and 30 nA and a beam diameter of 5 lm (step size 5 lm).
Trace elements in garnet, clinopyroxene, kyanite and rutile
3.2. Analytical techniques were analyzed in situ by Laser Ablation ICP-MS with an
Agilent 7500ce quadrupole ICP-MS coupled with an
Major element concentrations were determined with a esi/NWR193 nm wavelength excimer laser following meth-
JEOL JXA 8900 RL microprobe at the Johannes ods described in Jacob (2006). Ablation was carried out at
Gutenberg University, Mainz, Germany, using wavelength 6.6 J/cm2 and 10 Hz, using He as the carrier gas. All mea-
dispersive analysis and a range of natural and synthetic surements were performed on polished thin sections using
standards. The data were corrected using the CITZAF a spot size diameter of 100 lm. The commercial software
Fig. 3. Electron microprobe element distribution maps of eclogite xenoliths showing resorption of kyanite by clinopyroxene and garnet (a, b).
Note that clinopyroxenes in (a) are strongly altered and fine-grained alteration phases give the original outline of the grain replacing the
kyanite. Panels (c) to (f) are element distribution maps of an enlarged area shown in (b). They illustrate replacement of kyanite (red in c) by
celsian feldspar rich in barium (f). Scale bar in (c–f) is 200 lm. (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.)
Table 1
170
Major (in wt.%) and trace element (in ppm) compositions and d18O (&) values for garnets. b.d.l. = below detection limit, n.m. = not measured. WG 1 = Wajrakarur Group 1, WG 2 = Wajrakarur
Group 2. Paragenesis is indicated by ky = kyanite-, rt = rutile-bearing or bi = bimineralic. Samples without trace element data have no group assignment. Temperatures are calculated using the
geothermometer by Ellis and Green (1979) projected on the Wajrakarur geotherm (Karmalkar et al., 2009) resulting in a pressure estimate (P1). P2 are pressures estimates using the geobarometer by
Beyer et al. (2015). See text for further discussion.
KL2–1 KL2–4 KL2–5 KL2–7 KL2–8 KL2–9 KL2–10 KL2–20 KL2–37 KL2–44 KL2–48 KL2–49 KL2–50 KL2–61 P3–25
Paragenesis ky, rt ky ky ky ky ky ky ky ky ky bi, rt ky bi bi Ky
Classification WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1
SiO2 41.51 40.31 41.04 41.40 40.67 40.44 41.09 41.20 41.54 40.97 40.01 42.08 41.55 40.56 41.81
TiO2 0.16 0.24 0.15 0.12 0.15 0.13 0.23 0.11 0.16 0.25 0.25 0.15 0.16 0.17 0.04
Al2O3 22.48 22.01 22.33 22.37 22.25 21.91 21.76 22.62 22.78 22.07 21.42 22.61 21.85 22.15 22.85
A.N. Dongre et al. / Geochimica et Cosmochimica Acta 166 (2015) 165–188

Cr2O3 0.03 0.07 0.04 0.08 0.05 0.04 0.17 0.06 0.04 0.08 0.17 0.04 0.05 0.06 0.14
FeO 7.61 19.40 11.76 10.43 10.89 11.76 13.78 9.95 7.73 19.60 16.86 7.66 10.91 12.20 8.01
MnO 0.18 0.42 0.30 0.24 0.25 0.30 0.26 0.23 0.19 0.42 0.13 0.19 0.27 0.26 0.18
MgO 10.81 9.93 9.26 11.69 8.33 5.28 7.86 9.17 11.50 10.08 6.16 11.09 9.53 7.90 15.59
CaO 17.00 8.44 15.31 13.08 17.22 18.60 15.07 16.89 16.50 7.76 14.78 16.60 15.20 16.30 10.79
Na2O 0.06 0.12 0.05 0.05 0.06 0.05 0.09 0.03 0.05 0.12 0.07 0.05 0.07 0.06 0.04
K2O b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. 0.01 b.d.l. b.d.l. b.d.l.
Total 99.84 100.93 100.22 99.46 99.87 98.50 100.31 100.27 100.49 101.35 99.85 100.47 99.58 99.66 99.46
T (°C) 1200 1080 1140 1110 1160 1220 1130 1150 1170 1070 1160 1170 1180 1180 918
P1 (GPa) 5.3 4.5 4.8 4.7 5.0 5.4 4.8 4.9 5.0 4.5 5.0 5.0 5.1 5.1 3.6
P2 (GPa) 5.4 4.6 4.9 4.8 5.1 5.5 5.0 5.1 5.2 4.5 5.6 5.4 5.8 5.8 3.6
d18O 5.3 ± 0.2 6.5 ± 0.2 5.4 ± 0.2 n.m. 5.5 ± 0.2 5.6 ± 0.2 5.8 ± 0.2 5.4 ± 0.2 5.4 ± 0.2 6.7 ± 0.3 5.7 ± 0.2 5.3 ± 0.2 5.4 ± 0.2 5.6 ± 0.2 5.5 ± 0.2
Li 0.398 0.241 0.475 0.580 0.402 0.302 0.585 0.279 0.448 0.194 3.61 0.450 0.226 6.27 0.438
Rb <0.04 0.02 <0.02 0.02 <0.02 <0.03 0.01 <0.04 0.03 0.02 <0.5 <0.04 <0.05 <0.04 <0.04
Th 0.01 0.01 0.01 0.03 0.01 0.003 0.01 0.02 0.01 0.01 0.114 0.01 <0.006 0.188 0.01
U 0.02 0.005 0.04 0.04 0.03 0.01 0.01 0.01 0.03 0.01 0.306 0.03 <0.004 0.623 0.02
Nb 0.04 0.02 0.07 0.03 0.11 0.02 0.004 0.02 0.07 0.03 0.07 0.07 0.03 1.41 0.01
Ta 0.003 0.002 0.005 0.01 0.002 0.004 0.004 0.003 0.01 0.002 0.02 0.004 <0.004 0.04 <0.01
La 0.03 0.03 0.05 0.04 0.04 0.02 0.08 0.07 0.05 0.02 0.877 0.04 0.01 0.693 0.02
Ce 0.465 0.238 0.648 0.465 0.577 0.585 0.863 0.612 0.594 0.263 7.55 0.542 0.36 6.52 0.22
Pr 0.19 0.090 0.250 0.189 0.200 0.257 0.306 0.168 0.199 0.094 1.73 0.195 0.17 2.30 0.10
Sr 2.95 1.32 2.52 2.98 5.55 7.66 10.4 6.03 2.82 1.37 18.9 2.88 4.23 22.5 <0.8
Nd 1.72 0.771 2.37 1.65 1.65 2.30 2.99 1.33 1.65 0.791 11.3 1.58 1.30 21.82 1.14
Sm 0.710 0.393 1.04 0.730 0.701 0.862 1.33 0.616 0.772 0.356 2.62 0.769 0.779 8.93 0.774
Zr 5.74 3.26 6.26 4.16 2.64 3.60 6.76 2.62 6.05 2.86 3.05 5.73 4.34 43.5 1.90
Hf 0.065 0.046 0.085 0.090 0.044 0.039 0.104 0.056 0.093 0.043 0.141 0.084 0.091 0.598 0.023
Eu 0.481 0.249 0.792 0.520 0.700 1.03 1.17 0.691 0.493 0.238 2.76 0.451 0.652 11.2 0.429
Ti 0.094 0.143 0.090 0.072 0.090 0.079 0.135 0.066 0.096 0.150 0.148 0.088 0.060 0.102 0.024
Gd 0.965 0.517 1.35 1.04 0.889 1.04 1.43 0.820 1.05 0.495 2.06 0.969 0.874 11.8 0.834
Tb 0.174 0.097 0.236 0.185 0.175 0.184 0.233 0.139 0.183 0.090 0.260 0.186 0.174 1.93 0.101
Dy 1.21 0.656 1.66 1.32 1.31 1.37 1.63 0.900 1.30 0.622 1.88 1.23 1.16 13.2 0.578
Y 6.54 3.60 9.52 7.40 7.34 7.69 8.86 4.88 6.88 3.31 10.9 6.95 6.91 74.3 3.22
Ho 0.258 0.144 0.370 0.292 0.292 0.304 0.342 0.190 0.277 0.129 0.325 0.249 0.297 2.94 0.119
Er 0.699 0.388 1.10 0.863 0.900 0.894 0.968 0.557 0.789 0.377 0.938 0.794 0.826 8.44 0.378
Tm 0.100 0.054 0.150 0.117 0.136 0.126 0.136 0.078 0.109 0.052 0.098 0.100 0.095 1.23 0.054
Yb 0.700 0.388 1.08 0.857 0.997 0.948 0.878 0.576 0.777 0.353 0.878 0.761 0.819 8.81 0.425
Lu 0.092 0.053 0.166 0.127 0.157 0.126 0.129 0.086 0.111 0.055 0.098 0.106 0.136 1.28 0.074
Sc 37.1 20.4 53.0 53.8 46.1 38.4 38.0 32.0 37.5 17.9 178 40.5 49.7 440 29.4
V 77.3 40.1 65.7 75.6 63.5 60.0 87.7 67.3 76.2 35.2 8078 75.1 98.6 698 35.6
Cr 78.5 60.8 146 355 365 133 902 132 96.1 42.6 11134 79.5 175 2321 797
Mn 1497 784 2237 1833 1966 2181 1873 1769 1456 678 10846 1480 1911 22775 1349
Ni 35.7 18.1 28.0 38.9 22.4 15.3 32.4 21.1 35.1 15.9 656 35.4 30.8 196 69.6
KL2–2 KL2–24 KL2–33 KL2–36 KL2–52 Kl2–54 KL2–56 KL2–57 KL2–60 KL2–17 KL2–29 KL2–42 KL2–43
Paragenesis ky, rt bi bi bi, rt bi bi bi, rt bi, rt bi bi bi bi bi
Classification WG2 WG2 WG2 WG2 WG2 WG2 WG2 WG2 WG2 – – – –

SiO2 41.39 40.75 40.86 40.40 40.48 40.58 39.85 39.93 40.09 40.79 40.99 40.90 39.90
TiO2 0.14 0.21 0.24 0.25 0.22 0.23 0.22 0.24 0.22 0.22 0.13 0.24 0.22
Al2O3 22.72 22.09 21.58 22.30 21.83 21.74 22.00 21.84 21.69 22.38 22.62 21.90 21.96
Cr2O3 0.05 0.08 0.09 0.07 0.08 0.12 0.11 0.14 0.09 0.09 0.04 0.07 0.10
FeO 7.60 19.51 19.23 19.30 19.20 19.19 19.20 19.30 19.47 16.37 13.11 18.80 19.37
MnO 0.19 0.40 0.43 0.44 0.42 0.41 0.43 0.43 0.43 0.45 0.41 0.45 0.43
MgO 11.38 10.07 9.53 9.93 9.82 9.80 9.92 9.70 9.84 13.48 12.77 9.98 10.00
CaO 15.61 7.89 8.30 7.81 8.20 8.19 8.08 8.21 7.93 5.89 9.70 8.01 7.93
Na2O 0.05 0.12 0.12 0.13 0.13 0.13 0.11 0.12 0.13 0.11 0.08 0.13 0.12
K2O b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l.
Total 99.13 101.12 100.39 100.63 100.38 100.39 99.91 99.92 99.89 99.78 99.85 100.48 100.03
T (°C) 1160 1070 1070 1070 1090 1090 1090 1090 1090 1160 1210 1070 1070
P1 (GPa) 5.0 4.5 4.5 4.5 4.6 4.6 4.6 4.6 4.6 5.0 5.3 4.5 4.5
P2 (GPa) 5.1 5.3 4.6 4.6 5.0 5.0 5.1 5.2 5.4 5.3 5.4 5.3 4.3
d18O 6.3 ± 0.2 7.8 ± 0.2 7.4 ± 0.4 7.3 ± 0.3 7.7 ± 0.2 7.6 ± 0.3 7.7 ± 0.2 7.5 ± 0.2 7.6 ± 0.2 n.m. n.m. n.m. n.m.
Li 1.70 0.689 13.1 0.996 1.27 1.00 0.984 0.955 1.07 n.m. n.m. n.m. n.m.
Rb 0.384 <0.05 <0.4 <0.04 0.128 <0.04 <0.05 0.06 0.04 n.m. n.m. n.m. n.m.
Th 0.02 0.01 <0.1 0.03 0.02 <0.006 0.01 0.02 0.003 n.m. n.m. n.m. n.m.
U 0.01 0.00 0.06 0.004 0.01 <0.008 0.005 0.01 <0.006 n.m. n.m. n.m. n.m.
Nb 0.04 0.01 0.05 0.02 0.005 0.01 0.01 0.03 <0.01 n.m. n.m. n.m. n.m.
Ta 0.01 0.004 0.02 0.004 0.002 <0.06 0.01 0.002 0.002 n.m. n.m. n.m. n.m.
La 0.06 0.01 0.14 0.03 0.204 0.01 0.03 0.04 0.01 n.m. n.m. n.m. n.m.
Ce 0.266 0.10 1.30 0.11 0.326 0.09 0.132 0.227 0.09 n.m. n.m. n.m. n.m.
Pr 0.10 0.06 0.52 0.05 0.07 0.05 0.06 0.06 0.04 n.m. n.m. n.m. n.m.
Sr 1.35 0.897 8.12 1.24 1.73 1.43 1.59 3.12 1.32 n.m. n.m. n.m. n.m.
Nd 1.56 0.938 10.28 0.752 0.888 0.775 0.708 0.764 0.621 n.m. n.m. n.m. n.m.
Sm 2.79 1.40 16.14 1.35 1.39 1.47 1.24 1.28 1.17 n.m. n.m. n.m. n.m.
Zr 56.9 26.17 353 26.7 28.6 29.7 27.3 28.1 26.3 n.m. n.m. n.m. n.m.
Hf 0.720 0.345 4.39 0.366 0.360 0.377 0.373 0.368 0.325 n.m. n.m. n.m. n.m.
Eu 2.02 0.934 13.4 0.905 0.939 0.987 0.880 0.985 0.801 n.m. n.m. n.m. n.m.
Ti 0.086 0.128 0.142 0.142 0.132 0.138 0.129 0.140 0.132 n.m. n.m. n.m. n.m.
Gd 9.00 3.50 51.2 3.90 4.09 4.06 3.49 4.13 3.41 n.m. n.m. n.m. n.m.
Tb 1.82 0.813 10.5 0.838 0.917 0.911 0.838 0.908 0.793 n.m. n.m. n.m. n.m.
Dy 13.5 6.01 86.9 6.28 6.66 6.48 6.28 6.53 6.16 n.m. n.m. n.m. n.m.
(continued on next page)
171
GLITTER 4.0 (Macquarie University; Griffin et al., 2008)

KL2–43
was used for data reduction with NIST SRM 612
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
bi
(Jochum and Nehring, 2006) as the external standard and
– 43
Ca as the internal standard using values for CaO from
the electron microprobe analyses for clinopyroxene and
KL2–42
garnet, and 47Ti determined as TiO2 for the rutiles.

n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
Identical element suites were measured in garnet, clinopy-
bi
–
roxene and kyanite (see Table 1), but for kyanites only
those elements that yielded concentrations above the detec-
KL2–29
tion limit are reported in Table 3. USGS glass BCR2-G was

n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
measured as an unknown in every session as quality control
bi
–
and always yielded values within 10% of the published val-

ues (Supplementary Table 1). Care was taken to analyze
KL2–17
only pristine unaltered grain areas, which was an issue par-

n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
n.m.
ticularly with the clinopyroxenes. Prior to each spot analy-
bi
–
sis the target area was checked in reflected and in

transmitted light to ensure clear, crack-free and unaltered
KL2–60
spots. The time-resolved signals for each individual analysis

WG2
0.642
0.708
3239
were screened for inclusions or cracks (identified by spikes

36.5
1.47
4.37
4.58
55.8
39.5
121
482
bi
in the time-resolved signal) and affected areas or complete

analyses were discarded accordingly.
Oxygen isotopic ratios of hand-picked pristine garnet
KL2–57
WG2
grains were analyzed by secondary ion mass spectrometry

0.649
0.705
bi, rt
3259
38.9
1.53
4.59
4.86
56.1
39.7
123
837
(SIMS) using a Cameca IMS 1280 instrument at the

Canadian Centre for Isotopic Microanalysis in
Edmonton following methods given in Ickert and Stern
KL2–56
(2013). Three to five garnet grains per sample were

WG2
0.650
0.701
bi, rt
3274
36.6
1.45
4.38
4.71
58.8
40.6
124
636
mounted in epoxy and placed within 5 mm of the center

of the 25 mm diameter mount to avoid analytical artifacts
as a function of sample location. In-house reference mate-
rials UAG (S0068, pyrope) and UAJM (S0088B, grossu-
Kl2–54
WG2
0.672
0.723
3230
lar) were co-mounted with the unknowns. After

39.3
1.50
4.50
4.79
59.3
40.7
127
669
bi
polishing, the grains were imaged with backscattered elec-

trons using a Zeiss EVO MA15 scanning electron micro-
scope (SEM) to assist in the choice of optimal SIMS spot
KL2–52
localization free of inclusions or cracks. Backscattered

WG2
0.692
0.717
3283
39.1
1.54
4.57
4.76
55.2
41.2
133
385
images of all garnet grains showing that all grains are

bi
internally chemically homogeneous are shown in

Supplementary Fig. 1. A 20 keV, 3–4 nA Cs+ primary
KL2–36
beam of 15 lm diameter was utilized, along with low

WG2
37.08
0.642
0.692
bi, rt
3152
1.40
4.35
4.52
54.4
40.2
131
311
energy electrons for charge compensation. The analytical

session comprised analyses of four unknowns interspersed
with analyses of the two reference materials, each taking
4 min. Secondary ions (18O , 16O ) were measured
KL2–33
37608
WG2
1516
5333
simultaneously in Faraday cups at high mass resolution

18.3
57.5
8.22
55.9
8.88
493
694
478
bi
(m/Dm 2275 for 18O ) to avoid isobaric interferences.

Matrix corrections utilized molar Ca/(Ca + Fe + Mg)
(see Ickert and Stern, 2013) determined on separate frag-
KL2–24
ments by EPMA as mentioned previously, as these were

WG2
35.88
0.618
0.678
3098
1.40
4.20
4.50
54.5
46.9
123
357
uniform in composition within the samples. The quadra-

bi
tic matrix calibration curve parameters of Ickert and

Stern (2013) were adjusted to be consistent with the refer-
KL2–2
ence value for S0088B. The analytical uncertainties in

ky, rt
WG2
5916
d18O include all known sources, including matrix correc-

79.2
3.04
9.13
1.37
9.87
1.53
82.1
116
226
723
Table 1 (continued)
tions, and for 112 spot analyses of the unknowns the

median is ±0.28& (95% confidence). Four to five spots
Classification
on two to four grains per sample were measured and

Paragenesis
average d18OVSMOW of these analyses for each sample

are given in Table 1.
Mn
Tm
Ho
Yb
Lu
Cr
Ni
Er
Sc
Y
V
4. RESULTS
4.1. Mineral chemistry
4.1.1. Major elements

Garnet shows large variations in Ca–Mg–Fe contents
(Table 1). Garnets from samples from pipe KL2 are pre-
dominantly pyrope–almandines (Py21Alm26Grs53–
Py50Alm34Grs16) and overlap in composition with the
eclogite xenolith suite from Roberts Victor, South Africa
(Fig. 4a), which has a typical distribution for on-craton
eclogite xenoliths worldwide (Jacob, 2004). The samples
have low Cr2O3 (0.04–0.17 wt.%) and TiO2 (0.04–
0.25 wt.%). While there are no distinct low-MgO and
high-MgO suites in this set of Indian samples as reported
from some other kimberlite pipes (Koidu, Sierra Leone;
Kaalvallei, South Africa; Kasai Craton, northern Angola;
Victor, Jericho, and Diavik, all in Canada: Viljoen, 1994;
Barth et al., 2001; Aulbach et al., 2007; Smart et al.,
2009; Smit et al., 2014), the kyanite-bearing samples are,
on average, richer in grossular whereas the bimineralic
eclogites have higher almandine. One kyanite eclogite
(KL2–9) shows higher than 50% grossular component,
and is thus a grospydite (Sobolev et al., 1977). Garnets from
kyanite-bearing eclogites fall in Group B and Group C of
Taylor and Neal (1989) while those from bimineralic eclog-
ites fall in the Group B field. The single kyanite-bearing
sample from pipe P3 shows the highest pyrope component
(56%) and falls in the Group A field. Compared to earlier
work on eclogites from Wajrakarur, the suite studied here
overlaps compositionally with the literature data
(Fig. 4a), but does not contain the Mg-rich eclogite compo-
nent described by Patel et al. (2006, 2009). However, most
of the samples described by Patel et al. (2006, 2009) except
for the orthopyroxene-bearing eclogite (open symbol in
Fig. 4a) are from a different kimberlite cluster (cluster Fig. 4. (a) Compositional variation of garnets from kyanite-
bearing samples (black squares) and bimineralic eclogites (black
“P”), which may have sampled a different population of
spots) compared to literature data for Dharwar ecogite xenoliths in
eclogites. gray (Patel et al., 2006, 2009: gray squares = kyanite eclogite, gray
One sample from the KL2 pipe (KL2–29, Table 2) was spots = bimineralic eclogites, white spot = opx-bearing eclogite).
found to contain two chemically distinct clinopyroxene Field for garnets from on-craton Roberts Victor eclogite xenoliths,
populations (Table 2), one having higher Al2O3 (15.8 vs South Africa from Jacob (2004). Group A, Band C subdivisions
10.57 wt.%), higher Na2O (6.13 vs 8.23 wt.%, correspond- from Taylor and Neal (1989), line at 50% Ca marks the grospydite
ing to 59 vs 41 mol% jadeite) and lower CaO (10.5 vs classification (Sobolev, 1977). Dark gray square: kyanite-bearing
13.9 wt.%) content than the other. The Cr2O3 content in P3–25. (b) Na2O vs MgO variation in clinopyroxenes from eclogite
both types is generally less than 0.1 wt.% (Table 2). All xenoliths (symbols as in Fig. 4a). Fields for on-craton Roberts
samples show low Ti content (<0.25 wt.% TiO2), but a Victor kimberlite and kyanite eclogites from off-craton Rietfontein
kimberlite from South Africa are from Appleyard et al. (2009).
number of them were found to contain rutile. The Na2O
Group A, B and C subdivisions are from Taylor and Neal (1989).
content is less than 0.09 wt.% in garnets from
kyanite-bearing eclogites (except in KL2–4 and KL2–44
having 0.12 wt.% Na2O) and is >0.09 in all bimineralic reflected by the jadeite content of the clinopyroxenes of
eclogites. High Na2O contents of garnets in eclogite xeno- Jd30–73, which is typical for eclogite xenolith suites world-
liths from kimberlites in conjunction with high K2O con- wide, however toward the upper end of omphacite compo-
tents in clinopyroxenes usually reflect higher pressures of sitions (Fig. 4b). In contrast, clinopyroxene in the sample
equilibration in eclogites (Harlow, 1997). However, from pipe P3 has the lowest jadeite content and is highest
clinopyroxenes here have K2O concentrations below in MgO, thus plotting in field A (Fig. 4b) of the classifica-
0.08 wt.%, which is the threshold for classification as tion by Taylor and Neal (1989), consistent with the coexist-
high-pressure (and potentially diamondiferous) eclogite ing garnet in this sample.
(McCandless and Gurney, 1989) implying that the high Clinopyroxenes from kyanite-bearing eclogites are rich
Na2O concentrations in garnets in this sample set are a bulk in Al2O3 (10.9–15.8 wt.%) while bimineralic samples
composition effect rather than a pressure effect. This is also have lower Al2O3 (9.89–12.1 wt.%). Even though
kyanite-bearing, P3 pipe eclogite shows a fairly low Al2O3 eclogites are derived from depths mostly just above the dia-
content (7.94 wt.%), all clinopyroxenes have dominantly mond stability field in the Earth’s mantle.
octahedrally coordinated Al with ratios of
Al[6]/Al[4] > 11. This far exceeds the threshold value of 2, 4.1.3. Trace elements
which is the boundary between pyroxenites and eclogites Garnet rare earth element patterns are Light Rare Earth
given by Aoki and Shiba (1973) and is consistent with the Element (LREE) depleted with flat Heavy Rare Earth
high-pressure origin of the rocks. As for garnets, clinopy- Element (HREE) patterns (Fig. 5a, b). Abundances are typ-
roxenes from kyanite bearing eclogites fall in Group B ical for eclogite xenoliths worldwide (Jacob, 2004; Tappe
and C, those from bimineralic eclogites fall in Group B et al., 2011; Pernet-Fisher et al., 2014; Smart et al., 2014;
while P3 pipe eclogite plots in Group A field (Taylor and Smit et al., 2014) except for sample KL2–33, which has
Neal, 1989) (Fig. 4b). anomalously high HREE abundances for an eclogite (ca.
Pervasive late-stage metasomatism affected all samples 350 times chondritic; Fig. 5b). Most garnets have positive
studied here, and is most pronounced in clinopyroxenes europium anomalies, irrespective of the presence or absence
(Fig. 3a, b). Along the pathways of the metasomatic agent, of kyanite; the most pronounced anomalies are displayed
omphacites were transformed into diopside-augite and by a sample in the bimineralic group (KL2–48;
kyanite into celsian feldspar (Fig. 3f). Barium-rich feldspars (Eu/Eu*)N = 3.5). P3 pipe eclogite shows similar abun-
are not uncommon in eclogite xenoliths from kimberlites dances and a pattern like the kyanite-bearing eclogites from
(Smyth et al., 1984; Peltonen et al., 2002; Schmickler pipe KL2.
et al., 2004; Patel et al., 2006) and are usually interpreted REE patterns of garnets are shown grouped into two
as the result of reactions between kimberlite and eclogite. types, termed Wajrakarur Group 1 with prominent positive
The kyanites are very low in Cr2O3 (0.038–0.051 wt.%) Eu-anomalies (Fig. 5a: (Eu/Eu*)N = 1.6–3.5) and
and contain small amounts of FeO (0.12–0.27 wt.%; Wajrakarur Group 2 with small positive Eu-anomalies
Table 3). (Fig. 5b: (Eu/Eu*)N = 1.1–1.3). Both groups contain
kyanite-bearing and kyanite-free samples. Garnets in
4.1.2. Geothermobarometry Wajrakarur Group 1 (strong positive Eu anomalies) have
Suitable geobarometers for bimineralic eclogites are relatively flat HREE patterns with chondrite-normalized
scarce. A suitable possibility for deriving pressures of equi- Lu/Gd ratios mostly close to 1 (0.7–1.4, average 0.95;
libration is the projection of equilibration temperatures Fig. 6) and slightly depleted LREE with
onto a geotherm obtained independently. The garnet– chondrite-normalized Sm/La ratios between 20 and 85
clinopyroxene Mg–Fe2+ exchange geothermometer by (average 34). Wajrakarur Group 2 has mostly subtly posi-
Ellis and Green (1979) is calibrated for the range of tively sloped HREE patterns (average Lu/GdN = 1.3) and
Mg-numbers usually found in eclogite xenoliths from the more strongly depleted LREE patterns (average
mantle, but all iron is calculated as Fe2+ and therefore val- Sm/LaN = 120; Figs. 5b, 6). Group 1 has two outliers with
ues given here are maximum temperatures. Temperature higher Lu/Gd ratios (Fig. 6).
estimates were projected onto the Wajrakarur geotherm Clinopyroxenes also form two groups of REE patterns;
(Karmalkar et al., 2009) and yield 1070–1220 °C, averaging Wajrakarur Group 1 with enriched LREE and steeply
1128 °C. This corresponds to pressures of 4.5–5.4 GPa for negative-sloping HREE patterns (LaN/TbN > 1, Fig. 5c),
the KL2 pipe and 3.6 GPa, 918 °C for the P3 pipe, which are the counterparts of the Group 1 garnets with flat
(Table 1, Supplementary Fig. 1). Kyanite-bearing and HREE patterns (Fig. 5a), and Wajrakarur Group 2 with
kyanite-free samples overlap in pressure and temperature, depleted LREE and enriched MREE (LaN/TbN < 1,
but kyanite-bearing samples plot toward the high tempera- Fig. 5d). Clinopyroxenes from Group 2 are mostly associ-
ture end of this range (1070–1220 °C, average 1147 °C). ated with Group 2 garnet showing stronger LREE deple-
Patel et al. (2006, 2009) and Babu et al. (2008) calculated tion (Fig. 5b). This grouping, just as the garnet grouping,
temperatures of 900–1100 °C and 950–1225 °C at similar is not without exception, as clinopyroxenes from sample
pressures. A single opx-bearing eclogite xenolith described KL2–61 (Fig. 5c, asterisk) with a pattern typical of
by Patel et al. (2006) yielded 3.6 GPa at 870 °C using the Group 2 coexists with garnet (Fig. 5a, asterisk) typical for
Brey and Köhler (1990) calibration, which is within, but Group 1. Clinopyroxenes in Group 1 have prominent pos-
at the lower end of the range obtained for the eclogite suite. itive Eu-anomalies ((Eu/Eu*)N up to 2.6) mirroring those of
Recently Beyer et al. (2015) proposed a new experimentally garnets in Group 1.
calibrated geobarometer for eclogite xenoliths. On applying First row transition elements like Sc, Cr and Ni have
this barometer samples from KL2 pipe show derivation higher concentrations in garnets and clinopyroxenes of
from pressures of 4.2–5.8 GPa and 3.6 GPa for P3 pipe Wajrakarur Group 2 eclogites than in the Wajrakarur
eclogite xenoliths (Table 1). Large uncertainties are due to Group 1 eclogite suite (except KL2–48 which shows
dependence mostly on the precision of silicon determina- anomalously high concentrations; Suppl. Table 2).
tions by electron microprobe (Beyer et al., 2015). Wajrakarur Group 1 eclogite garnets and clinopyroxenes
Nevertheless, most of the pressures determined with this have lower Sc (grt: 17.9–440 ppm, cpx: 6.7–95.9 ppm), Cr
geobarometer agree well with those derived from the pro- (grt: 42.6–11,134 ppm, cpx: 57.3–11,131 ppm) and Ni (grt:
jection on the geotherm or are within 20% of them 15.3–656 ppm, cpx: 163–5651 ppm) concentrations than
(Table 1). Thus, geothermobarometry shows that these the Wajrakarur Group 2 eclogites: Sc (grt: 54.4–694 ppm,
Table 2
Major (in wt.%) and trace element compositions (in ppm) in clinopyroxenes. Abbreviations as in Table 1.
KL2–1 KL2–4 KL2–5 KL2–7 KL2–8 KL2–9 KL2–10 KL2–20 KL2–37 KL2–44 KL2–48 KL2–49 KL2–50 KL2–61 P3–25
Paragenesis Ky, rt ky ky ky ky ky ky ky ky ky bi, rt ky bi bi ky
Classification WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1 WG1
SiO2 54.41 55.26 54.70 55.17 54.80 55.13 56.05 55.03 54.93 56.05 56.26 55.60 55.99 55.16 54.4
TiO2 0.17 0.41 0.19 0.18 0.15 0.11 0.30 0.12 0.18 0.38 0.31 0.18 0.24 0.14 0.15
Al2O3 12.17 10.95 14.03 13.02 14.23 15.80 14.63 13.62 12.17 11.05 16.18 12.06 13.34 14.32 7.94
Cr2O3 0.05 0.10 0.04 0.07 0.06 0.04 0.16 0.05 0.03 0.09 0.18 0.04 0.06 0.05 0.15
FeO 1.57 4.93 2.19 1.86 2.10 2.78 2.60 1.90 1.47 4.98 3.41 1.49 2.22 2.58 1.20
MnO 0.02 0.05 0.03 0.01 0.02 0.05 0.02 0.02 0.01 0.07 0.02 0.01 0.05 0.02 0.01

MgO 9.80 8.54 7.92 8.79 7.88 6.26 6.90 8.61 9.88 8.49 5.40 9.92 8.08 7.45 12.67
CaO 16.54 12.92 13.99 14.5 13.99 12.73 12.37 14.98 16.50 12.22 11.03 16.43 13.96 11.80 19.6
Na2O 4.20 5.93 5.54 5.5 5.58 6.20 6.55 5.15 4.38 5.88 6.86 4.41 5.73 7.00 2.796
K2O 0.03 0.05 0.02 b.d.l. 0.04 0.04 0.03 0.04 0.04 b.d.l. 0.01 0.04 0.05 0.04 0.02
Total 98.95 99.14 98.64 99.10 98.87 99.11 99.63 99.55 99.64 99.21 99.65 100.23 99.71 98.56 98.92
Li 8.03 6.40 10.4 13.9 9.25 12.4 10.4 6.80 8.98 5.92 85.3 8.04 9.84 93.2 8.27
Rb 0.154 0.04 0.03 0.02 <0.03 <0.1 0.04 0.03 0.10 <0.02 <1 <0.02 <0.04 1.20 0.17
Th 0.09 0.02 0.03 0.16 0.11 0.01 0.01 0.04 0.06 0.04 0.242 0.11 <0.01 0.05 0.17
U 0.02 0.01 0.01 0.03 0.04 0.01 <0.003 0.003 0.01 0.01 0.185 0.02 <0.01 0.03 0.04
Nb 0.203 0.10 0.11 0.12 0.04 0.02 0.00 0.03 0.17 0.12 0.200 0.22 0.06 2.74 0.14
Ta 0.03 0.02 0.03 0.02 0.01 0.01 <0.003 0.01 0.04 0.02 0.06 0.03 0.01 0.53 0.02
La 0.992 0.65 0.65 0.73 0.82 0.21 0.62 0.57 1.05 0.76 5.78 1.03 0.27 2.10 2.72
Ce 3.52 2.41 2.55 2.82 2.45 0.94 2.07 1.68 3.69 2.75 16.01 3.41 1.60 10.8 6.92
Pr 0.499 0.34 0.45 0.52 0.27 0.13 0.33 0.20 0.52 0.39 1.71 0.47 0.28 2.36 1.08
Sr 268 168 162 325 332 250 464 405 265 197 797 265 309 627 159
Nd 1.98 1.35 2.50 2.07 0.908 0.538 1.48 0.798 2.04 1.52 4.52 1.98 1.09 13.4 4.78
Sm 0.206 0.148 0.449 0.206 0.110 0.075 0.193 0.093 0.262 0.207 0.805 0.238 0.139 2.84 0.673
Zr 7.63 5.34 8.53 4.19 1.97 2.58 5.46 2.14 7.71 5.49 2.64 9.18 3.61 248 4.21
Hf 0.281 0.209 0.218 0.198 0.117 0.091 0.217 0.112 0.321 0.236 0.175 0.300 0.251 4.13 0.129
Eu 0.083 0.053 0.138 0.107 0.061 0.036 0.119 0.067 0.091 0.072 0.300 0.069 0.070 0.940 0.185
Ti 0.100 0.244 0.111 0.108 0.090 0.067 0.178 0.072 0.108 0.228 0.187 0.110 0.150 0.084 0.088
Gd 0.081 0.064 0.207 0.118 0.065 0.040 0.068 0.067 0.130 0.106 0.447 0.116 0.081 2.17 0.216
Tb 0.008 0.004 0.014 0.009 0.008 0.003 0.009 0.007 0.014 0.009 0.056 0.015 0.009 0.168 0.015
Dy 0.020 0.018 0.038 0.054 0.035 <0.06 0.030 0.024 0.059 0.035 <1 0.072 0.038 0.736 0.039
Y 0.128 0.107 0.153 0.169 0.097 0.058 0.136 0.075 0.169 0.115 0.110 0.125 0.119 2.66 0.112
Ho 0.005 0.003 0.006 0.008 0.004 <0.03 0.006 0.004 0.010 0.005 <0.3 0.008 0.007 0.145 0.006
Er 0.009 0.010 <0.01 0.017 0.012 <0.07 0.014 <0.01 0.018 0.012 <0.5 <0.07 0.020 <0.6 0.017
Tm <0.001 <0.003 <0.01 0.005 <0.003 <0.003 0.004 <0.003 <0.003 <0.003 <0.2 <0.02 0.002 <0.1 <0.02
Yb 0.006 <0.02 <0.02 0.015 0.017 <0.005 0.014 <0.02 0.018 <0.02 <0.8 <0.1 0.016 0.110 <0.02
Lu <0.003 0.0005 <0.01 0.003 0.002 <0.02 0.002 0.005 0.002 <0.004 <0.2 <0.01 0.003 <0.01 0.002
Sc 8.41 6.70 11.6 10.6 11.1 10.4 9.06 8.05 9.13 6.27 58.2 9.31 10.0 95.9 11.5
V 158 109 161 206 129 99.9 200 142 159 118 18440 158 239 1394 139
Cr 67.0 66.2 146 372 240 106 877 118 79.5 57.3 11131 67.1 168 1995 865
Mn 133 91.3 188 168 169 178 164 152 138 95.5 871 129 165 2140 125
Ni 305 197 256 305 197 163 256 173 289 207 5651 297 243 1695 861
175
176
KL2–2 KL2–24 KL2–33 KL2–36 KL2–52 Kl2–54 KL2–56 Kl2–57 KL2–60 KL2–17 Kl2–29a KL2–29b KL2–42 KL2–43
Paragenesis Ky, rt bi bi bi, rt bi bi bi, rt bi, rt bi bi bi bi bi bi
Classification WG2 WG2 WG2 WG2 WG2 WG2 WG2 WG2 WG2 - - - - -
SiO2 55.02 56.20 56.23 55.80 55.64 55.81 55.98 55.46 55.70 55.53 55.52 55.31 56.00 54.50
TiO2 0.20 0.39 0.39 0.39 0.36 0.39 0.39 0.39 0.39 0.44 0.19 0.19 0.40 0.39
Al2O3 12.10 11.13 10.92 11.30 11.08 11.03 11.09 11.01 10.95 9.89 10.57 15.84 11.20 11.02
Cr2O3 0.04 0.08 0.09 0.09 0.09 0.13 0.12 0.15 0.11 0.10 0.08 0.07 0.08 0.10
FeO 1.47 5.00 4.95 4.97 4.89 4.97 5.02 5.03 5.03 4.57 3.39 2.16 4.70 4.92
MnO 0.02 0.05 0.06 0.06 0.05 0.06 0.05 0.06 0.07 0.07 0.04 0.04 0.07 0.06
MgO 9.89 8.49 8.32 8.47 8.22 8.34 8.52 8.37 8.27 9.59 9.66 6.63 8.29 8.40

CaO 15.05 12.14 12.66 12.3 12.6 12.7 12.98 13.02 12.70 12.76 13.96 10.46 12.7 12.81
Na2O 4.13 5.82 6.29 6.66 6.35 6.39 5.79 5.75 6.15 6.16 6.13 8.23 6.54 6.08
K2O 0.04 b.d.l. 0.05 b.d.l. b.d.l. b.d.l. 0.05 0.05 0.04 0.03 0.02 0.02 b.d.l. 0.05
Total 97.95 99.30 100.00 100.04 99.28 99.82 99.98 99.29 99.41 99.16 99.56 98.94 99.98 98.34
Li 10.7 8.99 114 9.11 9.13 9.13 9.63 9.17 9.30 n.m. n.m. n.m. n.m. n.m.
Rb 0.03 <0.03 <0.3 0.08 <0.1 <0.2 0.04 0.06 <0.08 n.m. n.m. n.m. n.m. n.m.
Th 0.004 0.004 <0.1 0.01 0.003 0.004 0.004 0.01 0.002 n.m. n.m. n.m. n.m. n.m.
U 0.003 0.002 <0.1 0.004 0.003 <0.01 0.003 <0.01 <0.01 n.m. n.m. n.m. n.m. n.m.
Nb 0.02 0.02 0.177 0.03 0.01 0.01 0.03 0.01 0.02 n.m. n.m. n.m. n.m. n.m.
Ta 0.002 0.005 0.08 0.01 <0.01 0.001 0.004 <0.01 0.004 n.m. n.m. n.m. n.m. n.m.
La 0.299 0.218 2.98 0.37 0.23 0.23 0.26 0.24 0.27 n.m. n.m. n.m. n.m. n.m.
Ce 1.36 1.17 14.45 1.29 1.14 1.14 1.20 1.15 1.18 n.m. n.m. n.m. n.m. n.m.
Pr 0.293 0.24 2.83 0.25 0.22 0.24 0.25 0.24 0.25 n.m. n.m. n.m. n.m. n.m.
Sr 149 124 852 122 125 125 129 130 126 n.m. n.m. n.m. n.m. n.m.
Nd 1.92 1.47 18.4 1.60 1.53 1.52 1.60 1.49 1.57 n.m. n.m. n.m. n.m. n.m.
Sm 0.804 0.610 6.413 0.571 0.564 0.615 0.652 0.604 0.618 n.m. n.m. n.m. n.m. n.m.
Zr 51.5 36.8 519 41.6 41.3 41.7 41.8 45.6 44.6 n.m. n.m. n.m. n.m. n.m.
Hf 2.25 1.59 22.3 1.76 1.72 1.71 1.72 1.98 1.88 n.m. n.m. n.m. n.m. n.m.
Eu 0.330 0.277 3.41 0.231 0.235 0.269 0.268 0.256 0.223 n.m. n.m. n.m. n.m. n.m.
Ti 0.118 0.234 0.232 0.234 0.216 0.234 0.236 0.232 0.234 n.m. n.m. n.m. n.m. n.m.
Gd 0.840 0.614 7.76 0.619 0.591 0.661 0.701 0.618 0.690 n.m. n.m. n.m. n.m. n.m.
Tb 0.109 0.082 1.15 0.076 0.078 0.083 0.098 0.074 0.085 n.m. n.m. n.m. n.m. n.m.
Dy 0.532 0.384 4.86 0.391 0.382 0.404 0.438 0.445 0.357 n.m. n.m. n.m. n.m. n.m.
Y 1.88 1.38 19.3 1.55 1.48 1.51 1.53 1.60 1.69 n.m. n.m. n.m. n.m. n.m.
Ho 0.074 0.054 0.689 0.045 0.049 0.059 0.059 0.055 0.051 n.m. n.m. n.m. n.m. n.m.
Er 0.151 0.117 1.24 0.088 0.105 0.121 0.125 0.090 0.118 n.m. n.m. n.m. n.m. n.m.
Tm 0.021 0.014 0.157 0.010 0.007 0.010 0.015 0.009 0.012 n.m. n.m. n.m. n.m. n.m.
Yb 0.098 0.065 0.759 0.046 0.061 0.055 0.075 0.047 0.054 n.m. n.m. n.m. n.m. n.m.
Lu 0.013 0.008 <0.1 <0.01 0.005 0.008 0.009 <0.2 <0.01 n.m. n.m. n.m. n.m. n.m.
Sc 24.3 17.3 229 19.1 19.5 20.0 19.5 22.0 20.2 n.m. n.m. n.m. n.m. n.m.
V 457 362 4738 387 390 371 386 374 387 n.m. n.m. n.m. n.m. n.m.
Cr 509 410 6005 360 458 671 722 934 562 n.m. n.m. n.m. n.m. n.m.
Mn 506 463 5185 454 428 428 453 445 446 n.m. n.m. n.m. n.m. n.m.
Ni 364 296 3917 299 306 312 325 323 318 n.m. n.m. n.m. n.m. n.m.
cpx: 17.3–229 ppm), Cr (grt: 311–5,333 ppm, cpx: 360–

KL2–2 6005 ppm) and Ni (grt: 39.5–478 ppm, cpx: 296–
WG2
37.19
62.17
99.70
b.d.l.
n.m.
n.m.
n.m.
n.m.
n.m.
0.05
0.06
0.14
0.03
0.04
3,917 ppm). Lithium concentrations in clinopyroxenes
range between 5.9 and 93.8 ppm in Group 1 while Group
2 have higher Li abundances of 9.1–114 ppm (Table 2).
Kyanites are typically poor in trace elements and con-
P3–25
WG1
61.90
99.30
1680
37.1
0.04
0.05
0.16
0.01
0.04
0.01
2.51
59.9
3.45
191
tain 111–403 ppm Cr (P3–25 kyanite has 1680 ppm Cr) in
addition to Ti, V, Sc and low concentrations of Ni
(Table 3).
KL2–49
Rutiles are mostly present in the Wajrakarur Group 2

WG1
37.28
61.92
99.49
b.d.l.
b.d.l.
eclogite suite (except for sample KL2–48 in Group 1).
0.04
0.05
0.13
0.05
1.34
64.7
2.98
280
111
Their Zr/Hf ratios are 38 to 39, approximately chondritic
(KL2–48 is subchondritic) and are similar or higher to
rutiles from Koidu eclogites (Barth et al., 2001). Nb/Ta
KL2–44
ratios are strongly superchondritic as in many other eclogite

WG1
37.07
61.87
99.26
0.04
0.05
0.16
0.01
0.04
0.01
1.83
69.9
2.47
302
132
xenolith suites (Rudnick et al., 2000; Jacob et al., 2009;
Aulbach et al., 2011), ranging between 65 and 134 and
are higher than Nb/Ta ratios in Koidu rutiles. The highest
KL2–37
Nb/Ta ratio is found in Group 1 sample KL2–48. Uranium

WG1
36.77
62.09
99.15
b.d.l.
0.05
0.04
0.13
0.05
0.01
1.81
67.7
2.90
concentrations (0.39–3.53 ppm) are similar, and Ni (3.24–

292
148
45 ppm) contents are lower than in Koidu rutiles (Barth

et al., 2001).
KL2–20
WG1
62.03
99.87
4.1.4. Garnet oxygen isotopes

b.d.l.
37.5
0.05
0.06
0.17
0.04
0.01
2.05
52.9
1.98
182
190
Oxygen isotopic compositions (& d18OVSMOW) in pris-

tine garnet grains range between +5.3& and +7.8&
Major and trace element compositions of kyanites in wt.% and ppm, respectively. Abbreviations as in Table 1.
(Table 1). This range extends significantly beyond the value

KL2–10
of unaltered mantle equilibrated at high temperature

WG1
37.07
61.87
99.26
n.m.
n.m.
n.m.
n.m.
n.m.
0.04
0.05
0.16
0.01
0.04
0.01
(5.5 ± 0.4&, Mattey et al., 1994). Wajrakarur Group 1

and Group 2 eclogites are not significantly distinct in
d18O values, but highest values are found in the bimineralic
KL2–9
samples of Wajrakarur Group 2 eclogites (Table 1).

WG1
61.98
99.13
b.d.l.
36.8
0.02
0.04
0.27
0.02
0.01
1.14
35.1
1.44
The range is most similar to eclogites from the Orapa

129
166
kimberlite (Fig. 8), which is a locality with abundant inde-

pendent evidence for recycled crustal material (e.g.
Farquhar et al., 2002). In contrast, it is greater than the
KL2–8
WG1
37.07
61.87
99.25
restricted d18O-range in Kuruman xenoliths, which are

0.04
0.05
0.16
0.01
0.04
0.01
2.21
50.7
1.61
205
403
interpreted to originate as lithospheric pyroxenites that

were later transferred to deeper levels (Schmickler et al.,
2004). The Kuruman xenoliths present a suite that was
KL2–7
WG1
37.07
61.87
99.25
never at the Earth’s surface and has not experienced seawa-

n.m.
n.m.
n.m.
n.m.
n.m.
0.04
0.05
0.16
0.01
0.04
0.01
ter alteration.
The dataset was tested for linear correlations between
d18O-values and other geochemical parameters of garnet
KL2–5
WG1
37.07
61.87
99.25
and reconstructed whole rock compositions, particularly

0.04
0.05
0.16
0.01
0.04
0.01
1.22
55.1
2.26
210
221
calcium concentrations, Mg-number, bulk Sr-anomaly,

Zr/Hf and Nb/Ta ratios for data quality assessment. No
correlations were found. Instead, two groups can be differ-
KL2–4
entiated coinciding with the geochemical classification of

WG1
37.07
61.87
99.25
0.04
0.05
0.16
0.01
0.04
0.01
2.36
71.2
2.62
308
165
the samples: the Wajrakarur Group 1 eclogite samples with

higher positive bulk Eu-anomalies (1.45–3.23) have a lower
range of d18O values (5.3–6.7), whereas Wajrakarur Group
KL2–1
2 eclogites (small Eu-anomalies; 1.11–1.33) have compara-

WG1
36.86
61.06
98.20
0.03
0.04
0.13
0.02
0.05
0.01
1.14
61.4
2.86
267
109
tively high d18O values (6.3–7.8). Garnets with high FeO

contents in this suite tend to have high d18O. This was
observed earlier by Jacob et al. (1994, 1998), who described
a bimodal trend of d18O with FeO contents of garnets and
Classification
clinopyroxenes where samples with highest FeO concentra-

Table 3
tions showed the most extreme d18O-values, both negative

Cr2O3
Al2O3
Total
MnO
MgO
TiO2
CaO
SiO2
FeO
and positive (see Fig. 1 in Jacob et al., 1998).

Cr
Ni
Sc
Ti
V
Fig. 5. Chondrite normalized rare earth element patterns in garnets (a and b) and clinopyroxenes (c and d). Normalizing values are from Sun
and McDonough (1989). Samples are divided in two groups, with Wajrakarur Group 1 samples showing strong positive Eu-anomalies and flat
HREE patterns in garnet (a), while Group 2 has garnets with smaller positive Eu-anomalies, lower LREE/HREE ratios and slightly increasing
HREE patterns (b). Corresponding clinopyroxenes in Group 1 have enriched LREE and steep-sloping HREE patterns (LaN/TbN > 1, c),
while those of Group 2 have low LREE/HREE ratios (d). Cpx in sample KL2–61 (marked with an asterisk in c) is an exception as this cpx has
a pattern typical for the second group, but coexists with garnet from the first group. Samples in the first group (a, b) are KL2–1, KL2–4, KL2–
5, KL2–7, KL2–8, KL2–9, KL2–10, KL2–20, KL2–37, KL2–44, KL2–48, KL2–49, KL2–50, KL2–61 and P3–25. The second group (c, d)
comprises KL2–2, KL2–24, KL2–33, KL2–36, KL2–52, KL2–54, KL2–56, KL2–57, KL2–60. Kyanite-bearing and bimineralic eclogites occur
in both groups. Kl2–17, KL2–29, KL2–42, KL2–43 remain unassigned, because trace elements in these samples were not analyzed.
4.1.5. Reconstructed whole rock geochemistry bulk rock compositions were calculated by recombining
The pristine chemical composition of the bulk eclogite major element analyses and mineral modal amounts
cannot be measured directly, but can be reconstructed using (Suppl. Table 2). It is difficult to accurately determine
pristine mineral compositions and modal estimations to modal abundances in small coarse-grained xenolithic sam-
remove potential contamination and alteration effects of ples, therefore equal amounts of clinopyroxene and garnet
host kimberlite magma on its xenoliths (e.g. Jerde et al., were assumed for the bimineralic samples, and 45:45:10%
1993; Barth et al., 2001; Heaman et al., 2006). Applying this cpx:grt:ky for the kyanite-bearing eclogites in accordance
method to this eclogite suite is critical as they show perva- with petrographic observations. Rutile was not included
sive alteration of the primary phases. Clean approximate in the reconstruction as it only occurred in trace amounts
and it is therefore difficult to determine its mode accurately
enough. This method presents a trade-off between the
imprecision of modal estimates for such small and
coarse-grained samples and the xenolith bulk analysis that
usually includes cryptic and modal metasomatic alteration,
particularly prominent in this suite.
The magnesium numbers obtained in this way are rela-
tively insensitive to modal variations of garnet and clinopy-
roxene, while the bulk Al content changes considerably
with the modal amount of garnet, which introduces a large
uncertainty for Al. Among the trace elements, the incom-
patible elements are usually strongly affected by metasoma-
tism and benefit most from this reconstruction (Barth et al.,
2001), while the largest uncertainties are introduced for the
High Field Strength Elements (HFSE) hosted in accessory
rutile which was not included in the reconstruction. In con-
Fig. 6. Chondrite-normalized Sm/La ratios in garnet plotted vs
trast, the patterns of reconstructed REE abundances are
chondrite-normalized Lu/Gd ratios as a measure for the slope of
relatively insensitive to this reconstruction method (Jerde
the LREE and the HREE patterns. The Wajrakarur Group 1
garnets with strong positive Eu-anomalies (open circles, patterns et al., 1993; Aulbach et al., 2007; Smart et al., 2009).
shown in Fig. 6a) have mainly flat HREE and slightly depleted Both groups of eclogites in this study contain
LREE patterns, while Wajrakarur Group 2 garnets with small kyanite-bearing varieties. Reconstructed bulk Al2O3 con-
positive Eu-anomalies (black spots, patterns in Fig. 6b) have subtly tents predictably set the kyanite-bearing eclogites apart
positive sloping HREE patterns with more strongly depleted LREE from the bimineralic samples. While the former range
patterns. Note two outliers for Group 1.
between 20.04 and 23.16 wt.%, the latter fall at 14.89–

18.80 wt.% Al2O3. Magnesium numbers (43.0–76.6) and
CaO contents (8.99–17.40 wt.%) overlap between these
two types of samples (Suppl. Table 2).
Reconstructed whole rock spidergrams normalized to
primitive mantle values (Sun and McDonough, 1989) are
consistent with the two groups of mineral trace element pat-
terns. Most bulk rock samples with flat and only weakly
depleted LREE patterns and slightly decreasing HREE
abundances are Wajrakarur Group 1 eclogites while signif-
icantly depleted LREE and increasing and high HREE
abundances are typical of Wajrakarur Group 2 eclogites
(Fig. 7). Bulk rock spidergrams show trace element abun-
dances mostly between about 0.1 and 10 times primitive
mantle (Fig. 7). Three bimineralic samples (KL2–33,
KL2–48, KL2–61) are noteworthy because almost all their Fig. 8. Ranges of published d18O data for eclogite xenoliths suites
compatible and incompatible trace element contents are from various kimberlites in Siberia, South Africa, Canada and
higher than in the rest of the sample suite. This is also char- USA compared with d18O data for the modern day Samail
acteristic of both garnet and clinopyroxene and so is not ophiolite. The average oxygen isotopic composition in unaltered
due to an effect of modal mineralogy. It is also noted that peridotite is indicated as 5.5 ± 0.4& (Mattey et al., 1994). Data
a number of bimineralic samples from Group 2 have very sources: Udachnaya: Jacob et al. (1994), Snyder et al. (1997);
Orapa: Viljoen et al. (1996); Roberts Victor: MacGregor and
similar compositions suggesting that these may be frag-
Manton (1986); Sloan: Schulze et al. (2003), Jericho/Muskox:
ments of an originally contiguous larger xenolith. All recon-
Smart et al. (2014); Kuruman: Schmickler et al. (2004);
structed bulk compositions from Wajrakarur Group 1 Komsomolskaya: Pernet-Fisher et al. (2014); Samail Ophiolite:
eclogites have strong positive Sr and Eu anomalies Gregory and Taylor (1981).
(Fig. 7) compared to Wajrakarur Group 2. Compared to
trace element data from eclogite xenoliths worldwide the
Wajrakarur suite shows the consistently strongest positive it is possible that these troughs are introduced by the bulk
Sr and Eu anomalies yet described (Jacob, 2004). rock calculation method that did not take rutile into
Wajrakarur Group 2 eclogites (Fig. 7a) exhibit negative account. However, only one of the five samples containing
anomalies at Ti, Zr and Hf, while Wajrakarur Group 1 rutile displays a trough at Zr and Hf in the reconstructed
eclogites do not (Fig. 7b), consistent with the presence of spidergram (Fig. 7a). Rutile in this particular sample
rutile in former. As rutile is a major host for these elements, (KL2–48) has the lowest Zr and Hf concentrations and
adding even 5% modal rutile (which would be unusually
high for an eclogite xenolith) to the reconstructed bulk
composition does not increase the Zr and Hf bulk concen-
trations to flatten out the troughs. The presence of rutile
cannot be excluded in other samples, but this rough esti-
mate shows that the negative Zr and Hf anomalies are an
intrinsic feature or the bulk rock composition and are not
artefacts introduced by the recalculation method.
Nickel and Cr contents are much lower in Wajrakarur
Group 1 (Ni: 80–945 ppm, Cr: 45–2158 ppm) than in
Group 2 eclogites (Ni: 169–2197 ppm, Cr: 383–5669 ppm).
Higher compatible element concentrations along with more
depleted LREE indicate a more mafic composition of the
Wajrakarur Group 2 eclogite suite originating from a
higher proportion of mafic minerals in the protolith (see
below).
5. DISCUSSION
The Wajrakarur eclogite suite bears close similarities

with other eclogite xenolith suites from on-craton settings
worldwide (e.g. Jacob, 2004; Aulbach et al., 2007).
Fig. 7. Primitive mantle normalized trace element patterns for
reconstituted bulk rock compositions of Wajrakarur eclogite Petrography, equilibration temperatures, d18O values,
xenoliths. Normalizing values from Sun and McDonough (1989). major and trace element compositions are indistinguishable
Panel A shows patterns for samples of Group 1 as defined by their from the range shown by many other eclogite xenolith
REE patterns (see Fig. 6), panel B shows the samples from Group suites. However, the suite is set apart from other eclogite
2. Asterisk marks intermediate sample KL2–61. occurrences in that every single sample shows positive
strontium and europium anomalies (particularly pro- modal banding of the rocks. Petrographic evidence for this
nounced in Wajrakarur Group 1) and these anomalies are reaction is observed in the Wajrakarur samples (Fig. 3).
among the strongest yet reported for eclogite xenoliths Consequently, it could be speculated that some of the
worldwide (Fig. 9). bimineralic samples in this suite, particularly those with
Furthermore, all samples in this study are strongly high garnet modes originally contained kyanite (Lappin
altered by late-stage metasomatism by the kimberlitic melt: and Dawson, 1975).
the reaction of omphacites to form diopside-augites along
fluid tracks is visible even in hand specimen (Fig. 2a) and 5.2. Geochemical evidence for low-pressure protoliths
celsian feldspar is introduced by reaction of kyanite
(Fig. 3). Although this is occasionally observed in other One of the major questions for eclogite xenoliths from
eclogite xenolith suites, the extent seen here is unusual. kimberlites is whether they are metamorphosed
Lastly, although many eclogite xenolith localities contain low-pressure rocks, more specifically subducted oceanic
at least some kyanite-bearing samples, suites with such a crust (e.g. Helmstaedt and Doig, 1975; Jacob et al., 1994)
high abundance of kyanite eclogite are uncommon. or high-pressure garnet pyroxenites from the Earth’s
mantle (O’Hara and Yoder, 1967). Evidence for either is
5.1. Kyanite in eclogite xenoliths from the mantle – a neither always obvious nor conclusive due to multi-stage
metastable phase chemical alteration of the rocks by partial melting and
mantle metasomatism (e.g. Gréau et al., 2011; Huang
The presence of kyanite in eclogite xenoliths is key evi- et al., 2012).
dence for a crustal origin of these rocks (Nixon et al., The Wajrakarur eclogites bear a number of clear geo-
1963). It is a typical crustal mineral that is not in equilib- chemical indicators for an origin as low-pressure basaltic
rium with the peridotitic mantle because of its high Al2O3 or gabbroic rocks in addition to the presence of kyanite.
content, and high-pressure mantle melts cannot precipitate Many of these geochemical features are inherited from pla-
kyanite. Instead, kyanite forms upon metamorphism during gioclase after breakdown and subsequent metamorphic
plagioclase breakdown in gabbroic and basaltic rocks. Once subsolidus element redistribution (e.g. Scarrow et al.,
formed, it is stable to mantle pressures and temperatures 2000). They comprise high Al bulk concentrations, high
and may persist unless a flux medium that can facilitate dif- jadeite contents in the clinopyroxenes, strong positive euro-
fusion and reaction is present (e.g. Wayte et al., 1989). pium and strontium anomalies and flat HREE patterns in
Kyanite should react out at pressures and temperatures of the garnets.
the mantle lithosphere either by reaction with olivine in The high Al2O3 contents of all Group 2 (20–23 wt.%)
peridotitic or pyroxenitic environments or by reaction with and many Group 1 samples (14.9–18.8 wt.%) are above
clinopyroxene and garnet in eclogite (Lappin and Dawson, the upper limits for basaltic rocks with the exception of
1975). The kyanite component can be accommodated by arc-related high-alumina basalts (Panjasawatwong et al.,
clinopyroxene and the garnet solid-solution series, which 1995), implying a plagioclase-rich protolith such as gabbro.
leads to a shift in the mineral mode upon consumption of The more mafic compositions may be derived from accu-
kyanite toward higher garnet contents and can result in mulation of mafic minerals in gabbroic magma chambers.
The presence of positive Eu and Sr anomalies is com-
monly used as evidence for plagioclase in the protolith of
eclogites (e.g. Jacob et al., 2003), or even for eclogites with
gabbroic protoliths in the source of basalts (Sobolev et al.,
2000). This is due to the greater compatibility of Sr and
divalent Eu in the plagioclase structure (Drake and Weill,
1975). Modeling REE patterns using partition coefficients
for basaltic mineral/melt pairs (Paster et al., 1974;
Fujimaki et al., 1984; Beattie, 1993, 1994; Bindeman
et al., 1998) show that strong positive Eu anomalies in
the bulk appear in gabbros when plagioclase crystallizes
with olivine. Instead, co-crystallization of plagioclase with
high amounts of cpx, as in melagabbros and melabasalts
suppresses these due to the higher contents of Eu in cpx
as opposed to olivine. The partition coefficient for Sr in pla-
gioclase is about fifty times higher than for other phases, so
Fig. 9. Strontium anomaly (Sr* = (Pr + Nd)/2) vs europium that a significant positive Sr anomaly remains in the bulk
anomaly (Eu* = (Sm + Gd)/2) variation for eclogite xenoliths,
even in the absence of a noticeable positive Eu anomaly
showing the extreme values for Wajrakarur samples. All values are
at lower modal plagioclase abundances. Comparing the
normalized to primitive mantle values (Sun and McDonough;
1989). Data sources are Koidu: Barth et al. (2001), Kuruman: CIPW norms for reconstructed Wajrakarur eclogite bulk
Schmickler et al. (2004), Udachnaya: Jacob et al. (1994), compositions with the highest (KL2–44, Group 1) and the
Bultfontein (Kimberley): Jacob et al. (2009), Roberts Victor coesite smallest Eu anomalies (KL2–36, Group 2) gives a rough
eclogites: Jacob et al. (2003), Jericho/Muskox: Smart et al. (2014). estimate for the amount of plagioclase in the protoliths
(Jacob et al., 2003). This yields 68 wt.% feldspar for KL2– 5.3. Oxygen isotopic evidence
44 and 56 wt.% for KL2–36 as well as a ratio of feldspar to
mafic constituents (ol + cpx) in KL2–44 that is twice that of Unequivocal experimental evidence established that at
KL2–36. This underpins our interpretation that the pro- temperatures and pressures of the Earth’s mantle fraction-
toliths of Wajrakarur Group 2 eclogites are more mafic ation of oxygen isotopes is minimal and cannot explain
than those of Group 1. These values compare well with the large ranges in d18O observed in eclogite xenolith suites
results derived from the same modeling for unmelted (Clayton et al., 1975). Given their likely derivation from
coesite-bearing Roberts Victor eclogites, which are inter- metamorphosed gabbroic rocks, the best modern analogue
preted to be derived from oceanic gabbro protoliths to the samples is oceanic crust whose oxygen isotopic com-
(Jacob et al., 2003). position was altered by seawater prior to subduction.
Furthermore, the extreme Sr and Eu anomalies in Typically, this alteration style creates an extreme variation
Wajrakarur Group 1 eclogites favor gabbro over basalt as of d18O values that extends to values both below and above
precursor rocks for this group, which may also explain mantle values (Fig. 8; Gregory and Taylor, 1981; Bach
the restriction of d18O-values compared to the Roberts et al., 2001). Especially values lower than the mantle values
Victor eclogite xenolith suite (Fig. 8), because oceanic gab- in eclogite xenoliths are considered indicative for oceanic
bros comprise the lower portions of the oceanic crust and crustal seawater alteration at elevated temperatures and
are therefore typically not exposed to low-temperature sea- high water–rock ratios. d18O values higher than 5.5& as
water alteration (Gregory and Taylor, 1981; Bach et al., the range observed here are typical for zeolite facies sea-
2001). floor alteration and many cratonic eclogite xenoliths from
Wajrakarur Group 2 eclogites with higher d18O values kimberlites share this signature (e.g. Jacob, 2004; Spetsius
show smaller Eu anomalies and generally higher trace ele- et al., 2008; Pernet-Fisher et al., 2014).
ment concentrations indicating a more mafic character of Nevertheless, rocks residing in the deep lithosphere
the protoliths compared to Group 1, with higher clinopy- could also acquire a high d18O overprint by interaction with
roxene to plagioclase ratios, thus mela-gabbroic in charac- metasomatic fluids (Gréau et al., 2011; Huang et al., 2012)
ter. The large variation in first row transition trace elements as long as these metasomatic fluids were themselves derived
(Sc, V, Cr; Tables 1 and 2) within this group implies some from crustal material, otherwise they would not have an
variation in the modal mineralogy of cpx or olivine their oxygen isotopic signature different from the mantle.
protoliths, which are the major hosts for these elements in Evidence for interaction with such a fluid at depth within
gabbros and basalts. This means that the eclogites grouped the cratonic lithosphere can be found preserved as
together in the Wajrakarur Group 2 are not derived from small-scale d18O heterogeneity or zonation in a peridotitic
identical protoliths, but rather from a suite of protoliths breccia xenolith (Zhang et al., 2000) but is generally rare.
that are petrogenetically related. Evidence such as heterogeneous d18O-values or correlations
Higher d18O-values are seen in eclogites (e.g. Jacob and between d18O-values and major or trace elements is yet to
Foley, 1999; Jacob et al., 2003; Pernet-Fisher et al., 2014) be presented for cratonic eclogite xenoliths. Given the lack
whose protoliths were altered at high water/rock ratios by of a fractionation mechanism for oxygen isotopes at litho-
colder water closer to the ocean crust surface under zeolite spheric depths, these fluids must ultimately be derived from
facies conditions (Gregory and Taylor, 1981; Bach et al., recycled material. Conversely, it should also be noted that
2001). These rocks are typically ocean floor basalts, but rocks with an apparently unchanged oxygen isotopic signa-
may sometimes also comprise oceanic gabbros exposed to ture may still have undergone seawater alteration but at a
seawater alteration along fractures or shear zones. Zeolite different (lower) water/rock ratio (e.g. Jacob et al., 1994).
facies alteration can also cause enrichment in Li
(Woodland et al., 2002), which is consistent with higher 5.4. Evidence for metasomatism and melting
Li concentrations in the Group 2 eclogites compared with
Group 1 samples that are derived from less altered Wajrakarur Group 1 eclogites have some of the most
protoliths. extreme positive Eu and Sr anomalies yet described
High-pressure garnet pyroxenites, which are sometimes (Fig. 9), which points to significant modal amounts of pla-
proposed as protoliths of subcratonic eclogite xenoliths, gioclase in the protoliths. This feature can only be pre-
are cumulates from melts of mantle peridotite and have dif- served if post-metamorphic geochemical alteration by
ferent mineralogy to eclogites. This is because olivine is the infiltration and/or melting during subduction was restricted
liquidus phase at the high pressures relevant for yielding (i.e. cryptic metasomatism). Alteration of cpx, grt and
protoliths for subcratonic eclogite in the diamond stability kyanite to diopside, celsian feldspar and other phases
field (O’Hara and Yoder, 1967). Experimental melts from observed in thin section (Figs. 2 and 3) is caused by later
peridotites at pressures of 4–5 GPa are too MgO-rich to stage, post-peak metamorphic processes.
match the bulk compositions of the eclogites studied here Most eclogite xenolith samples have clinopyroxenes with
(Walter, 1998). Furthermore, garnets in equilibrium with LREE enriched patterns (Fig. 5c), which we interpret to be
such melts have characteristic HREE-enriched patterns pre- inherited from the gabbroic protolith rather than due to
dicted by their partition coefficients (Green, 1994), while carbonatitic metasomatism. Carbonatite metasomatism
garnets in the eclogite suite here significantly deviate from has been shown not only to affect the LREE, but also the
those, reflecting subsolidus metamorphic trace element HFSE with a shift toward superchondritic Zr/Hf ratios
redistribution (Gonzaga et al., 2010). (e.g. Rudnick et al., 1993; Moine et al., 2004). Most
Zr/Hf ratios in the reconstructed bulk rocks are between 28

and 49 and correlate neither with slope nor content of the
LREE, discounting a significant role of carbonatite metaso-
matism in the history of the eclogites studied here.
Significant metasomatism by silicate (MARID-style) melts
can also be ruled out, because interaction of such melts with
eclogites forms phlogopite and generates characteristic
Zr/Hf ratios in garnet which are not seen in the
Wajrakarur suite (Jacob et al., 2009).
For crystal chemical reasons rutile typically displays
subchondritic Nb/Ta ratios (Foley et al., 2002), while the
rutiles in this suite of eclogites are strongly superchondritic.
The processes leading to high Nb/Ta ratios in rutiles from
subcratonic eclogite xenoliths are still unknown (Hills and
Haggerty, 1989; Rudnick et al., 2000; Barth et al., 2001).
Aulbach et al. (2008) argued for a metasomatic source of
the high Nb/Ta ratios of mantle rutiles on the basis of Fig. 10. Reconstructed bulk MgO vs SiO2 of the Wajrakarur
strongly superchondritic Nb/Ta ratios being correlated eclogite xenoliths. Fields for experimentally produced eclogite
with unradiogenic Hf isotopic ratios. While no Hf isotopic residues upon TTG melting and metabasalt starting compositions
data on the Wajrakarur rutiles exist, KL2–48 rutile with the are from Rapp and Watson (1995), oceanic cumulates from Pernet-
highest Nb/Ta ratios among the samples studied here also Fisher et al. (2014). Note the general affinity of Wajrakarur Group
has a significantly higher uranium concentration, which 1 eclogites with metabasalts, whereas most Wajrakarur Group 2
suggests addition of a metasomatic component to the orig- eclogites plot to higher MgO but do not show a trend toward
inal rutile composition. experimentally produced eclogite residues.
In the absence of strong mantle metasomatic signatures
in the trace element patterns of the pristine areas of the sil- LREE patterns from an unmelted Wajrakarur Group 1
icates (i.e. those not affected by late stage alteration) it can eclogite source composition. Loss of such a significant melt
be speculated that the high Nb/Ta ratios in the rutiles are fraction is not consistent with the major element composi-
related to metasomatism during the earlier stages of sub- tions (Fig. 10).
duction, perhaps in the amphibolite facies where the stabil- It is instead possible that the LREE systematics are
ity field of rutile is intersected (e.g. Ionov and Hofmann, inherited characteristics of the gabbroic protoliths reflecting
1995). combined effects of variable parental liquid compositions
The reconstructed trace element compositions with and fractional crystallization (e.g. Benoit et al., 1996).
higher total REE concentrations in Wajrakarur Group 2 Deconvoluting the pre-metamorphic history of the
eclogites are consistent with protoliths with higher cpx/pla- Wajrakarur eclogites any further, however, would be
gioclase ratios for this group of samples. Compared to entirely speculative at this time.
Wajrakarur Group 1 samples, most of the Group 2 samples
have steeper positive slopes in their garnet LREE pattern 5.5. Implications for the geodynamic evolution of the
(Fig. 6) and higher HREE concentrations in the garnets Dharwar craton
(Fig. 5b). This may suggest that these samples experienced
a partial melting event that is not seen in the Group 1 sam- The tectonic evolution of the Dharwar craton is a matter
ples (e.g. Barth et al., 2001; Smart et al., 2014). However, of debate. Models put forward range between uniformitar-
comparing the reconstructed bulk major element composi- ian plate tectonic amalgamation by subduction in the
tion of the Wajrakarur eclogites with experimental data Neoarchean (Chadwick et al., 2000, 2007; Moyen et al.,
for TTG generation by eclogite melting (Fig. 10, Rapp 2001; Ram Mohan et al., 2013) and other models support-
and Watson, 1995) discounts any significant partial melting ing formation by mantle plumes (Jayananda et al., 2000),
event for the Wajrakarur Group 2 eclogites. Both groups sagduction (Choukroune et al., 1995; Chardon et al.,
have broadly similar SiO2 contents and form a continuous 1998) or a combination of subduction and plume-related
range in MgO contents with the Group 2 samples plotting processes (Naqvi et al., 2002, 2006; Manikyamba and
toward the MgO-rich end of this range (Fig. 10). In con- Kerrich, 2012). Geochemical evidence for a convergent
trast, eclogites that formed as residues after TTG melting, margin setting for the Dharwar craton is becoming increas-
which are typical melts from eclogites, plot to distinctively ingly favored in the literature: felsic volcanics and grani-
lower SiO2 concentrations than those observed in the toids in the northern EDC greenstone belts have adakitic
Wajrakarur suite (e.g. Rudnick, 1995; Rapp and Watson, characteristics including Sr/Y (>40) and La/Yb (>20) ratios
1995; Barth et al., 2001; Tappe et al., 2011). indicative of slab melting (Naqvi et al., 2006, 2008;
These major element characteristics of the samples rule Prabhakar et al., 2009). Boninites are found in the
out the possibility that the steeper sloped LREE patterns Archean Gadwal greenstone belt of EDC (Manikyamba
of the Wajrakarur Group 2 eclogites are caused by signifi- et al., 2005) and negative HFSE anomalies in the
cant partial melting. Our modeling indicates that between Mesoproterozoic Siddanpalli kimberlites have been inter-
15% and 40% melt loss would be required to generate the preted to imply a subducted component in their mantle
source (Chalapathi Rao et al., 2010). Recently, granitoids Dharwar craton (equivalent to Central Dharwar province)
and volcanic rocks in three greenstone belts in the northern and Western Dharwar Craton (Chadwick et al., 2003;
part of EDC (Ram Mohan et al., 2013) have been shown to Jayananda et al., 2006; Chardon and Jayananda, 2008;
carry arc signatures (negative Nb, Ti anomalies; relative Chardon et al., 2011). This shear zone is commonly inter-
HFSE depletion and LILE enrichment), which point to a preted as the suture of accretion of both cratons along
subduction regime with the involvement of recycled older which an eastward-directed convergence with subduction
crust. and collision, which amalgamated the central and the west-
Chadwick et al. (2000) showed that the calc-alkaline ern parts of the craton (Mishra and Prajapati, 2003). The
Closepet Granite bodies (Fig. 1) are compatible with a sub- Wajrakarur Group 1 eclogite xenoliths thus may represent
duction zone origin and used gradually younging ages the subducted remnants of the oceanic crust between these
along this zone as evidence for oblique sinistral convergence two cratons (Babu et al., 2008).
in the Neoarchean. Furthermore, the Dharwar craton Subduction-related subcratonic eclogites appear in the
shows a systematic younging toward the east (Naqvi, global geodynamic record at around 3 Ga and their appear-
2005), which would be generally expected for an imprint ance is thought to mark the global onset of modern style
of subduction and consecutive accretion. Chardon et al. plate tectonics (Shirey & Richardson, 2011). Eclogite xeno-
(2011) showed that the cohesive textural fabric pattern lith suites from cratons around the world have been inter-
throughout the Dharwar craton is consistent with the preted as vestiges of the stabilization of cratons in the
effects of eastward subduction on the overriding block. Archean (e.g. Jacob et al., 1994; Pearson et al., 1995;
Finally, eastward subduction is suggested through various Tappe et al., 2011) or subsequent amalgamation along their
geophysical observations (Gupta et al., 2003b; Gokarn margins in the Proterozoic (e.g. Smart et al., 2014).
et al., 2004). Dating of the Dharwar eclogite xenoliths has been
Recent studies divided the Dharwar craton into three unsuccessful up to now. Published Re–Os model ages range
crustal provinces (East, Central and West; Dey, 2013; from 1.7 to >4 Ga with low Os (29–100 ppt) and very low
Peucat et al., 2013) based on differing crustal ages. While Re (9.6–92 ppt) contents (Anand et al., 2008). This very
it is not clear if these three provinces represent three indi- large spread in model ages may either reflect tectonother-
vidual amalgamated blocks (Jayananda et al., 2013b), they mal disturbances or Re–Os heterogeneity of the protoliths
share a 2.56–2.50 Ga high-temperature tectonometamor- or effects of the extensive kimberlitic alteration, but does
phic event that consolidated and eventually shaped the not bear on the age of eclogite metamorphism.
Dharwar craton (Peucat et al., 1993; Jayananda et al., In the absence of radiometric ages for the studied sub-
2000; Vasudev et al., 2000; Chadwick et al., 2001, 2007; cratonic eclogite suites their role in the consolidation or a
Rogers et al., 2007). later amalgamation of the Dharwar craton remains enig-
The Wajrakarur kimberlite erupted in the Central matic at this time. Nevertheless, since the Wajrakarur kim-
Dharwar province that stretches from the berlites that transported these xenoliths to the surface are
Kolar-Kadiri-Hungund greenstone belts in the east to the 1.1 Ga in age (Gopalan and Kumar, 2008; Osborne et al.,
Chitradurga boundary shear zone in the west (Peucat 2011; Chalapathi Rao et al., 2013), our findings underpin
et al., 2013; Fig. 1). The Central Dharwar craton (CDC) the role of subduction in the formation of peninsular
shows two episodes of TTG crust formation, namely older India and discount purely plume related models.
TTGs (>3.0 Ga) and younger TTGs (<3 Ga) intruded by
the 2.5 Ga Closepet granite and its equivalents, which coin- 6. CONCLUSIONS
cide with the 2.5 Ga tectonometamorphic event that consol-
idated the craton. Furthermore, the CDC records earlier The Wajrakarur KL2 pipe contains a suite of eclogite
thermal events at ca. 2.62 and 3.20 Ga (Jayananda et al., xenoliths dominated by kyanite-bearing eclogites. P3 pipe
2013a,b) which ultimately coincide with the peaks of two eclogite shows geochemical similarities to kyanite-bearing
distinct TTG crust formation events. eclogites from pipe KL2, but is richer in Mg.
Eclogite xenolith suites from a number of cratons world- Some of the eclogites show extreme positive strontium
wide are often considered as residues of TTG melt forma- and europium anomalies, among the highest yet reported
tion incorporated into the lithosphere along convergent for eclogite xenoliths worldwide. Based on these geochemi-
margins (Ireland et al., 1994; Rollinson, 1997; Jacob and cal characteristics the xenoliths are grouped into
Foley, 1999; Smit et al., 2014; Pernet-Fisher et al., 2014; Wajrakarur Group 1 having pronounced Eu anomalies
Tappe et al., 2011). For example, eclogite xenoliths brought and Wajrakarur Group 2 with minor Eu anomalies. The
up from the subcratonic lithosphere by kimberlites in the geochemical signatures in Wajrakarur Group 1 samples
Kaapvaal craton are interpreted to be remnants of the ocea- are characteristic for high modal amounts of plagioclase
nic basin between the Witwatersrand and the Kimberley in the magmatic precursor rock and are preserved because
blocks, which collided at 2.93–2.88 Ga along the post-metamorphic geochemical alteration and/or partial
Colesberg magnetic lineament (Moser et al., 2001; melting during subduction was restricted. They clearly
Schmitz et al., 2004). argue for an origin of these eclogites as low-pressure gab-
The eclogite xenoliths studied here are erupted in kim- broic rocks. Wajrakarur Group 2 eclogites show geochem-
berlites situated in the Central Dharwar province (Eastern ical features consistent with higher clinopyroxene to
Dharwar craton, Fig. 1) to the east of the Chitradurga plagioclase ratios in the protoliths. The range of high
shear zone that marks the contact between Eastern d18O values is distinct from unchanged peridotitic mantle
material. This is a characteristic feature they share with Aulbach S., Pearson N. J., O’Reilly S. Y. and Doyle B. J. (2007)
many other cratonic eclogite xenolith populations, which Origins of xenolithic eclogites and pyroxenites from the Central
exhibit d18O values significantly outside the mantle range Slave Craton, Canada. J. Petrol. 48, 1843–1873.
(e.g. Eiler, 2001; Jacob, 2004; Spetsius et al., 2008; Aulbach S., O’Reilly S. Y., Griffin W. F. and Pearson N. J. (2008)
Subcontinental lithospheric mantle origin of high niobium/tan-
Pernet-Fisher et al., 2014). This is generally taken to indi-
talum ratios in eclogites. Nat. Geosci. 1, 468–472.
cate seawater alteration of their oceanic crustal protoliths Aulbach S., O’Reilly S. Y. and Pearson N. (2011) Constraints from
prior to subduction and is based on a broad analogy with eclogite and MARID xenoliths on origins of mantle Zr/Hf, Nb/
ophiolites (Gregory and Taylor, 1981) as well as drilled pro- Ta variablility. Contrib. Miner. Petrol. 162, 1047–1062.
files through modern oceanic crust (Bach et al., 2001). Babu E. V. S. S. K., Griffin W. L., Panda S. and O’Reilly S. Y.
Therefore, like many other cratonic eclogite xenolith suites (2008) Eclogite xenoliths from the kimberlites of the Eastern
worldwide, the Wajrakarur suite shows clear evidence for Dharwar Craton, south India: material representing ancient
an origin as crustal gabbroic material, while an origin as crust of the Western Dharwar Craton? In 9th International
high-pressure cumulates or solidified mafic melts can be Kimberlite Conference, Frankfurt. 9IKC-A-00161 (Ext. abstr.).
ruled out. Bach W., Alt J. C., Niu Y., Humphris S. E., Erzinger J. and Dick
H. J. B. (2001) The geochemical consequences of late-stage low-
Eclogite bodies are widespread in the subcratonic litho-
grade alteration of lower ocean crust at the SW Indian Ridge:
sphere, into which they were incorporated upon conver- results from ODP Hole 735B (Leg 176). Geochim. Cosmochim.
gence and collision (e.g. Pearson et al., 1995; Jacob, 2004 Acta 65, 3267–3287.
and references therein). They thus represent the vestiges Balakrishnan S., Rajamani V. and Hansen G. N. (1999) U–Pb ages
of craton amalgamation and stabilization and are thought of zircon and titanite from the Ramagiri area, southern India:
to be among the earliest witnesses of modern plate tectonic evidence for accretionary origin of the eastern Dharwar Craton
processes (Shirey & Richardson, 2011; Jacob and Foley, during the late Archean. J. Geol. 107, 69–86.
1999). Our findings that the Dharwar eclogite xenoliths rep- Barth M. G., Rudnick R. L., Horn I., McDonough W. F., Spicuzza
resent metamorphic gabbros and most likely subducted sea- M., Valley J. R. and Haggerty S. E. (2001) Geochemistry of
water altered oceanic rocks thus supports plate tectonic xenolithic eclogites from West Africa: Part I. A link between
low MgO eclogites and Archean crust formation. Geochim.
models for the amalgamation and growth of the Dharwar
Cosmochim. Acta 65, 1499–1527.
craton by subduction and argue against purely Barth M. G., Rudnick R. L., Horn I., McDonough W. F., Spicuzza
plume-related models. M., Valley J. R. and Haggerty S. E. (2002) Geochemistry of
xenolithic eclogites from West Africa: Part II. Origins of the
ACKNOWLEDGMENTS high MgO eclogites. Geochim. Cosmochim. Acta 66, 4325–4345.
Beattie P. (1993) Olivine-melt and orthopyroxene-melt equilibria.
N. Groschopf is thanked for expert help with the electron Contrib. Miner. Petrol. 115, 103–111.
microprobe work. Earth System Research Centre, Johannes Beattie P. (1994) Systematics and energetics of trace element
Gutenberg-University, Mainz is thanked for financial support to partitioning between olivine and silicate melt: implications for
A.D. The authors thank S. Aulbach, S. Tappe, G. Howarth, two the nature of mineral/melt partitioning. Chem. Geol. 117, 57–
anonymous referees and the editor D. Ionov for their constructive 71.
reviews. This is contribution 459 from the ARC Centre of Benoit M., Polve M. and Ceuleneer G. (1996) Trace element and
Excellence for Core to Crust Fluid Systems (http://www.ccfs.mq. isotopic characterization of mafic cumulate in a fossil mantle
edu.au). diapir (Oman ophiolite). Chem. Geol. 134, 199–214.
Berg G. W. (1968) Secondary alteration in eclogites from kimber-
lite pipes. Am. Mineral. 53, 1336–1346.
APPENDIX A. SUPPLEMENTARY DATA Beyer C., Frost D. J. and Miyajima N. (2015) Experimental
calibration of a garnet-clinopyroxene geobarometer for mantle
Supplementary data associated with this article can be eclogites. Contrib. Miner. Petrol. 169, 1–21.
Bidyananda M., Deomurari M. P. and Goswami J. M. (2003)
found, in the online version, at http://dx.doi.org/10.1016/ 207
Pb–208Pb ages of zircons from the Nuggehalli schist belt,
j.gca.2015.06.023.
Dharwar Craton, South India. Curr. Sci. 85, 1482–1485.
Bidyananda M., Goswami J. N. and Srinivasan R. (2011) Pb–Pb
REFERENCES zircon ages of Archaean metasediments and gneisses from the
Dharwar craton, southern India: implications for the antiquity
Anand M., Mishra K. C., Paul D. K., Ishikawa A. and Pearson D. of the eastern Dharwar craton. J. Earth Syst. Sci. 120, 643–661.
G. (2008) Trace element signatures of kyanite eclogites from a Bindeman I. N., Davis A. M. and Drake M. J. (1998) Ion
Southern Indian kimberlite. In 9th International Kimberlite microprobe study of plagioclase-basalt partition experiments at
Conference, Frankfurt. 9IKC-A-00322 (Ext. abstr.). natural concentration levels of trace elements. Geochim.
Aoki K. and Shiba I. (1973) Pyroxenes from lherzolite inclusions of Cosmochim. Acta 62, 1175–1193.
Itinomegata, Japan. Lithos 6, 41–51. Brey G. P. and Köhler T. (1990) Geothermobarometry in four-
Appleyard C. M., Bell D. R. and Le Roex A. P. (2009) Petrology phase lherzolites: II. New thermobarometers, and practical
and geochemistry of eclogite xenoliths from the Rietfontein assessment of existing thermobarometers. J. Petrol. 31, 1353–
kimberlite, Northern Cape, South Africa. Contrib. Miner. 1378.
Petrol. 154, 309–333. Chadwick B., Vasudev V. N. and Hegde G. V. (2000) The Dharwar
Armstrong A. (1995) CITZAF- A package of correction programs craton, southern India, interpreted as the result of Late
for the quantitative electron microbeam X-ray analysis of thick Archaean oblique convergence. Precambr. Res. 99, 91–111.
polished materials, thin films, and particles. Microbeam Anal. 4, Chadwick B., Hedge G. V., Nutman A. P. and Vasudev V. N.
177–200. (2001) Syenite emplacement during accretion of the Late
Archaean Dharwar batholith, south India: SHRIMP U/Pb age Friend C. R. L. and Nutman A. P. (1991) SHRIMP U–Pb
and structure of the Koppal pluton, Karnataka. J. Geol. Soc. geochronology of the Closepet granite and peninsular gneisses,
India 58, 381–390. Karnataka, South of India. J. Geol. Soc. India 38, 357–368.
Chadwick B., Vasudev V. N. and Hegde G. V. (2003) The Fujimaki H., Tatsumoto M. and Aoki K. (1984) Partition
Chitradurga schist belt and its adjacent plutonic rocks, north- coefficients of Hf, Zr and REE between phenocrysts and
west of Tungabhadra, Karnataka: a duplex in the late Archaean groundmasses Proc. 14th Lunar Planet. Sci. Conf.. J. Geophys.
convergent setting of Dharwar craton. J. Geol. Soc. India 69, 5– Res. 89, B662–B672 (supplement).
24. Ganguly J. and Bhattacharya P. K. (1987) Xenoliths in Proterozoic
Chadwick B., Vasudev V. N., Hegde G. V. and Nutman A. (2007) kimberlites from southern India: petrology and geophysical
Structure and SHRIMP U/Pb ages of granites adjacent to the implications. In Mantle Xenoliths (ed. P. H. Nixon). Wiley, New
Chitradurga schist belt: implications for Neoarchaean conver- York, pp. 249–266.
gence in the Dharwar craton, southern India. J. Geol. Soc. India Gokarn S. G., Gupta G. and Rao C. K. (2004) Geoelectric
69, 5–24. structure of the Dharwar Craton from magnetotelluric studies:
Chalapathi Rao N. V., Dongre A. N., Kamde G., Srivastava R. K., archean suture identified along the Chitradurga–Gadag schist
Sridhar M. and Kaminsky F. E. (2010) Petrology, geochemistry belt. Geophys. J. Int. 158, 712–758.
and genesis of newly discovered Mesoproterozoic highly Gonzaga R. G., Lowry D., Jacob D. E., LeRoex A., Schulze D.
magnesian, calcite-rich kimberlites from Siddanpalli, Eastern and Menzies M. A. (2010) Eclogites and garnet pyroxenites:
Dharwar craton, Southern India: products of subduction- similarities and differences. J. Volcanol. Geoth. Res. 190, 235–
related magmatic sources? Mineral. Petrol. 98, 313–328. 247.
Chalapathi Rao N. V., Creaser R. A., Lehmann B. and Panwar B. Gopalan K. and Kumar Ani (2008) Phlogopite K–Ca dating of
K. (2013) Re–Os isotopic study of Indian kimberlites and Narayanpet kimberlites, South India: implications to the
lamproites: implications for mantle source regions and cratonic discordance between their Rb–Sr, Ar/Ar ages. Precambr. Res.
evolution. Chem. Geol. 353, 36–47. 167, 377–382.
Chardon D. and Jayananda M. (2008) A 3D field perspective on Gréau Y., Huang J. X., Griffin W. L., Renac C., Alard O. and
deformation, flow and growth of the lower continental crust. O’Reilly S. Y. (2011) Type I eclogites from Roberts Victor
Tectonics 27, TC1014. http://dx.doi.org/10.1029/ kimberlites: products of extensive mantle metasomatism.
2007TC002120. Geochim. Cosmochim. Acta 75, 6927–6954.
Chardon D., Choukroune P. and Jayananda M. (1998) Sinking of Green T. (1994) Experimental studies of trace-element partitioning
the Dharwar basin (South India): implications for Archaean applicable to igneous petrogenesis—Sedona 16 years later.
tectonics. Precambr. Res. 9, 15–39. Chem. Geol. 117, 1–36.
Chardon D., Jayananda M. and Peucat J.-J. (2011) Lateral Gregory R. T. and Taylor H. P. (1981) An oxygen isotope profile in
constrictional flow of hot orogenic crust: insights from the a section of cretaceous oceanic crust, Samail ophiolite, Oman:
Neoarchean of south India, geological and geophysical impli- evidence for d18O buffering of the oceans by deep (>5 km)
cations for orogenic plateaux. Geochem. Geophys. Geosyst. 12, seawater-hydrothermal circulation at mid-ocean ridges. J.
Q02005. http://dx.doi.org/10.1029/2010GC003398. Geophys. Res. 86, 2737–2755.
Chinner G. A. and Cornell D. H. (1974) Evidence of kimberlite– Griffin W. L., Powell W. J., Pearson N. J. and O’Reilly S. Y. (2008)
grospydite reaction. Contrib. Miner. Petrol. 45, 153–160. GLITTER: data reduction software for laser ablation ICP-MS
Choudary V. S., Rau T. K., Bhaskara Rao K. S., Sridhar M. and (appendix). In Laser Ablation-ICP-MS in the Earth Sciences
Sinha K. K. (2007) Timmasamudram kimberlite cluster, (ed. P. Sylvester), pp. 308–311. Short Course Series 40.
Wajrakarur kimberlite field, Anantapur district, Andhra Mineralogical Association of Canada (MAC).
Pradesh. J. Geol. Soc. India 69, 597–610. Griffin W. L., Kobussen A. F., Babu E. V. S. S. K., O’Reilly S. Y.,
Choukroune P., Bouhallier H. and Arndt N. T. (1995) Soft Norris R. and Sengupta P. (2009) A translithospheric suture in
lithosphere during periods of Archaean crustal growth or the vanished 1 Ga lithospheric root of south India: evidence
crustal reworking. In Early Precambrian Processes (eds. M. P. from contrasting lithospheric sections in the Dharwar craton.
Coward and A. C. Ries). Geological Society Special Publication Lithos 112S, 1109–1119.
95, pp. 67–85. Gupta S., Rai S. S., Prakasam K. S., Srinagesh D., Bansal B. K.,
Clayton R. N., Goldsmith J. R., Karel V. J., Mayeda T. K. and Chadha R. K., Priestley K. and Gaur V. K. (2003a) The nature
Newton R. C. (1975) Limits on the effect of pressure on isotopic of the crust in southern India: implications for Precambrian
fractionation. Geochim. Cosmochim. Acta 39, 1197–1201. crustal evolution. Geophys. Res. Lett. 30, 1419. http://
Dey S. (2013) Evolution of Archean crust in Dharwar craton: the dx.doi.org/10.1029/2002GL016770.
Nd isotope record. Precambr. Res. 227, 227–246. Gupta S., Rai S. S., Prakasam K. S., Srinagesh D., Chadha R. K.,
Drake M. and Weill D. (1975) Partition of Sr, Ba, Ca, Y, Eu2+, Priestley K. and Gaur V. K. (2003b) First evidence for
Eu3+ and other REE between plagioclase feldspar and mag- anomalous thick crust beneath mid-Archean western Dharwar
matic liquid: an experimental study. Geochim. Cosmochim. Acta Craton. Curr. Sci. 84, 1219–1226.
39, 689–712. Hatton C. J. (1978) The geochemistry and origin of xenoliths from
Eiler J. M. (2001) Oxygen isotope variations of basaltic lavas and the Roberts Victor Mine. Ph. D. thesis, University of Cape
upper mantle rocks. Rev. Mineral. Geochem. 43, 319–364. Town, South Africa.
Ellis D. J. and Green D. H. (1979) An experimental study of the Harlow G. F. (1997) K in clinopyroxene at high pressure
effect of Ca upon garnet-clinopyroxene Fe–Mg exchange and temperature an experimental study. Am. Mineral. 82,
equilibria. Contrib. Miner. Petrol. 71, 13–22. 259–269.
Farquhar J., Wing B. and McKeegan K. D. (2002) Insight into Hauy R. J. (1822). Traité de minéralogie. Seconde édition, revue,
crust-mantle coupling from anomalous Delta S-33 of sulfide corrigée et considérablement augmentée par l’auteur. Bachelier
inclusions in diamonds. Geochim. Cosmochim. Acta 66, A225. et Huzard, Paris, 4 vol. in-8 + atlas, 548.
Foley S., Tiepolo M. and Vannucci R. (2002) Growth of early Heaman L. M., Creaser R. A., Cookenboo H. O. and Chacko T.
continental crust by melting of amphibolite in subduction (2006) Multi stage modification of the Northern Slave mantle
zones. Nature 417, 837–840. lithosphere: evidence from zircon and diamond bearing eclogite
xenoliths entrained in Jericho kimberlite, Canada. J. Petrol. 47, regional metamorphism and magmatism in the Dharwar
821–858. craton, southern India. J. Asian Earth Sci. 78, 18–38.
Helmstaedt H. and Doig R. (1975) Eclogite nodules from kimber- Jerde E. A., Taylor L. A., Crozaz G., Sobolev N. V. and Sobolev
lite pipes of the Colorado Plateau-samples of subducted V. N. (1993) Diamondiferous eclogites from Yakutia, Siberia:
Franciscan-type oceanic lithosphere. In Physics Chemistry of evidence for a diversity of protoliths. Contrib. Miner. Petrol.
the Earth, vol. 9 (eds. L. H. Ahrens, J. B. Dawson, A. R. 114, 189–202.
Duncan and A. J. Erlank). pp. 95–111. Jochum K. P. and Nehring F. (2006). GeoReM preferred values.
Hills D. V. and Haggerty S. E. (1989) Petrochemistry of eclogites <http://georem.mpch-mainz.gwdg.de/>.
from Koidu kimberlite complex, Sierra Leone. Contrib. Miner. Karmalkar N. R., Duraiswami R. A., Chalapathi Rao N. V. and
Petrol. 103, 397–422. Paul D. K. (2009) Mantle derived mafic ultramafic xenoliths
Huang J. X., Gréau Y., Griffin W. L., O’Reilly S. Y. and Pearson and the nature of Indian subcontinental lithosphere. J. Geol.
N. J. (2012) Multi-stage origin of Roberts Victor eclogites: Soc. India 73, 657–679.
progressive metasomatism and its isotopic effects. Lithos 142, Kumar P., Yuan S., Ravi Kumar M., Kind R., Li X. and Chadha
161–181. K. (2007) The rapid drift of the Indian tectonic plate. Nature
Ickert R. B. and Stern R. A. (2013) Matrix corrections and error 449, 894–897.
analysis in high precision SIMS 18O/16O measurements of Ca– Lappin M. A. and Dawson J. B. (1975) Two Roberts Victor
Mg–Fe garnet. Geostand. Geoanal. Res. 37, 429–448. cumulate eclogites and their re-equilibration. Phys. Chem.
Ionov D. A. and Hofmann A. W. (1995) Nb–Ta-rich mantle Earth 9, 351–365.
amphiboles and micas: implications for subduction-related McCandless T. E. and Gurney J. J. (1989) Sodium in garnet and
metasomatic trace element fractionations. Earth Planet. Sci. potassium in clinopyroxene; criteria for classifying mantle
Lett. 131, 341–356. eclogites. In Kimberlites and Related Rocks (eds. J. Ross, A.
Ireland T. R., Rudnick R. L. and Spetsius Z. (1994) Trace elements L. Jaques, J. Ferguson, D. H. Green, S. Y. O’Reilly, R. V.
in diamond inclusions from eclogites reveal link to Archean Danchin and A. J. A. Janse). Special Publication-Geological
granites. Earth Planet. Sci. Lett. 128, 199–213. Society of Australia 14, pp. 827–832.
Jacob D. E. (2004) Nature and origin of eclogite xenoliths in MacGregor I. D. and Carter J. L. (1970) The chemistry of
kimberlites. Lithos 77, 295–316. clinopyroxenes and garnets of eclogite and peridotite xenoliths
Jacob D. E. (2006) High sensitivity analysis of trace element poor from the Roberts Victor mine, South Africa. Phys. Earth
geological reference glasses by Laser Ablation Inductively Planet. Inter. 3, 391–397.
Coupled Plasma Mass Spectrometry. Geostand. Geoanal. Res. MacGregor I. D. and Manton W. I. (1986) Roberts Victor
30, 221–235. eclogites: ancient oceanic crust. J. Geophys. Res. 91, 14063–
Jacob D. E. and Foley S. F. (1999) Evidence for Archean ocean 14079.
crust with low high field strength element signature from Manikyamba C. and Kerrich R. (2012) Eastern Dharwar Craton,
diamondiferous eclogite xenolith. Lithos 48, 317–336. India: continental lithosphere growth by accretion of diverse
Jacob D. E., Jagoutz E., Lowry D., Mattey D. and Kudrjavtseva plume and arc terranes. Geosci. Front. 3, 225–240.
G. (1994) Diamondiferrous eclogites from Siberia: remnants of Manikyamba C., Naqvi S. M., Subba Rao D. V., Ram Mohan M.,
Archean oceanic crust. Geochim. Cosmochim. Acta 58, 5191– Khanna T. C., Rao T. G. and Reddy G. L. N. (2005) Boninites
5207. from the Neoarchaean Gadwal greenstone belt, Eastern
Jacob D. E., Jagoutz E., Lowry D. and Zinngrebe E. (1998) Dharwar Craton, India: implications for Archaean subduction
Comment on “The Origins of Yakutian Eclogite Xenoliths” by processes. Earth Planet. Sci. Lett. 230, 65–83.
G.A. Snyder, L.A. Taylor, G. Crozaz, A.N. Halliday, B.L. Mattey D., Lowry D. and MacPherson C. (1994) Oxygen isotope
Beard, V.N. Sobolev and N.V. Sobolev. J. Petrol. 39, 1527– composition of mantle peridotite. Earth Planet. Sci. Lett. 128,
1533. 231–241.
Jacob D. E., Schmickler B. and Schulze D. J. (2003) Trace element Mishra D. C. and Prajapati S. K. (2003) A plausible model for
geochemistry of coesite-bearing eclogites from the Roberts evolution of schist belts and granite plutons of Dharwar
Victor kimberlite, Kaapvaal craton. Lithos 71, 337–351. Craton, India and Madagascar during 3.0–2.5 Ga: insight from
Jacob D. E., Viljoen K. S. and Grassineau N. V. (2009) Eclogite gravity modeling constrained in part from seismic studies.
xenoliths from Kimberley, South Africa – a case study of Gondwana Res. 6, 501–511.
mantle metasomatism in eclogites. Lithos 112, 1002–1013. Misra K. C., Anand M., and Paul D. K. (2008) Metasomatized
Jayananda M., Moyen J. F., Martin H., Peucat J.-J., Auvray B. kyanite-eclogites xenoliths from a southern Indian kimberlite.
and Mahabaleshwar B. (2000) Late Archean (2550–2520 Ma) In 9th International Kimberlite Conference, Frankfurt. 9IKC-A-
juvenile magmatism in the Eastern Dharwar Craton, Southern 00264 (Ext. Abstr.).
India: constrains from geochronology, Nd–Sr isotopes and Moine B. N., Gregoire M., O’Reilly S. Y., Delpech G., Sheppard S.
whole rock geochemistry. Precambr. Res. 99, 225–254. M., Lorand J. P., Renac C., Giret A. and Cottin J. Y. (2004)
Jayananda M., Chardon D., Peucat J.-J. and Capdevila R. (2006) Carbonatite melt in oceanic upper mantle beneath the
2.61 Ga potassic granites and crustal reworking in the western Kerguelen Archipelago. Lithos 75, 239–252.
Dharwar craton, southern India: tectonic, geochronologic and Moser D. E., Flowers R. M. and Hart R. J. (2001) Birth of the
tectonic constraints. Precambr. Res. 150, 1–26. Kaapvaal tectosphere 3.08 billion years ago. Science 291, 465–
Jayananda M., Peucat J.-J., Chardon D., Krishna Rao B., Fanning 468.
C. M. and Corfu F. (2013a) Neoarchean greenstone volcanism Moyen J. F., Martin H. and Jayananda M. (2001) Multi-element
and continental growth, Dharwar Craton, southern India: geochemical modeling of crust-mantle interactions during late-
constraints from SIMS U–Pb zircon geochronology and Nd Archaean crustal growth: the Closepet granite (South India).
isotopes. Precambr. Res. 227, 55–76. Precambr. Res. 112, 87–105.
Jayananda M., Tsutsumi Y., Miyazaki T., Gireesh R. V., Kapfo Naqvi S. M. (2005) Geology and Evolution of the Indian Plate
Kowe-, Tushipokla, Hidaka H. and Kano T. (2013b) (From Hadean to Holocene – 4 Ga to 4 Ka). Capital Publishing
Geochronological constraints on Meso- and Neoarchean Company, New Delhi.
Naqvi S. M. and Rogers J. J. W. (1987) Precambrian Geology of Pernet-Fisher J. F., Howarth G. H., Liu Y., Barry P. H., Carmody
India. Oxford University Press, New York, 223. L., Valley J. W., Bodnar R. J., Spetsius Z. V. and Taylor L. A.
Naqvi S. M., Manikyamba C., Ganeshwara Rao T., Subba Rao D. (2014) Komsomolskaya diamondiferous eclogites: evidence for
V., Ram Mohan M. and Srinivasa Rama D. (2002) oceanic crustal protoliths. Contrib. Miner. Petrol. 167, 1–17.
Geochemical and isotopic constraints of Neoarchaean fossil Peucat J. J., Mahabaleswar B. and Jayananda M. (1993) Age of
plume for evolution of volcanic rocks of Sandur greenstone younger tonalitic mag-matism and granulitic metamorphism in
belt, India. J. Geol. Soc. India 60, 27–56. the South Indian transition zone (Krishnagiri area): compar-
Naqvi S. M., Khan R. M. K., Manikyamba C., Ram Mohan M. ison with older Peninsular Gneisses from the Gorur-Hassan
and Khanna T. C. (2006) Geochemistry of the Neoarchaean area. J. Metamorph. Geol. 11, 879–888.
high-Mg basalts, boninites and adakites from the Kushtagi- Peucat J. J., Jayananda M., Chardon D., Capdevila R., Fanning C.
Hungund greenstone belt of the Eastern Dharwar Craton M. and Paquette J. L. (2013) The lower crust of Dharwar
(EDC): implications for the tectonic setting. J. Asian Earth Sci. craton, southern India: patchwork of Archean granulitic
27, 25–44. domains. Precambr. Res. 227, 4–28.
Naqvi S. M., Sarma D. S., Sawkar R. H., Mohan M. R. and Prabhakar B. C., Jayananda M., Mohammad Shereef N. and Kano
Prathap J. G. R. (2008) Role of adakitic magmatism and T. (2009) Petrology and geochemistry of late Archaean gran-
subduction in gold endowment of Dharwar Neoarchaean itoids in the northern part of eastern Dharwar craton, south
greenstone belts, India. J. Geol. Soc. India 71, 875–888. India: implications for transitional geodynamic settings. J.
Nayak S. S. and Kudari S. A. D. (1999) Discovery of diamond- Geol. Soc. India 74, 299–317.
bearing kimberlite in Kalyandurg area, Anantapur district, Radhakrishna B. P. and Vaidyanadhan R. (1997) Geology of
Andhra Pradesh. Curr. Sci. 76, 1077–1079. Karnataka. Geological Society of India, Bangalore, 345.
Neal C. R., Taylor L. A., Davidson J. P., Holden P., Halliday A. Ramakrishnan M. and Vaidyanadhan R. (2010) Geology of India,
N., Nixon P. H., Paces J. B., Clayton R. N. and Mayeda T. K. vol. 1, second ed. Geological Society of India, Bangalore.
(1990) Eclogites with oceanic crustal and mantle signature Ramakrishnan M., Moorbath S., Taylor P. N., Anantha Iyer G. V.
from the Bellsbank kimberlite, South Africa, part 2: Sr, and Viswanatha M. N. (1984) Rb–Sr and Pb–Pb whole-rock
Nd and O isotope geochemistry. Earth Planet. Sci. Lett. 99, isochron ages of basement gneisses in Karnataka craton. J.
362–379. Geol. Soc. India 25, 20–34.
Neelakantam S. (2001) Exploration for diamonds in southern Ram Mohan M., Piercey S. J., Kamber B. S. and Sarma D. S.
India. Geol. Surv. India Spec. Publ. 58, 521–555. (2013) Subduction related tectonic evolution of the Neoarchean
Nehru C. E. and Reddy A. K. (1989) Ultramafic xenoliths from eastern Dharwar craton, southern India: new geochemical and
Vajrakarur kimberlites, India. Geol. Soc. Aust. Spec. Publ. 14, isotopic constraints. Precambr. Res. 227, 204–226.
745–758. Rao K. R. P., Rao K. N., Dhakate M. V. and Nayak S. S. (2001)
Nikitina L. P., Korolev N. M., Zinchenko V. N. and Tunga Felix J. Petrology and mineralogy of mantle xenoliths of Wajrakarur
(2014) Eclogites from the upper mantle beneath the Kasai and Narayanpet kimberlite fields. Andhra Pradesh, India. Geol.
Craton (Western Africa): petrography, whole-rock geochem- Surv. India Spec. Publ. 58, 577–591.
istry and U–Pb zircon age. Precambr. Res. 249, 13–32. Rapp R. P. and Watson E. B. (1995) Dehydration melting of
Nixon P. H., Knorring O. V. and Rooke J. M. (1963) Kimberlites metabasalt at 8–32 kbar: implications for continental growth
and associated inclusions of Basutoland: A mineralogical and and crust-mantle recycling. J. Petrol. 36, 891–931.
geochemical study. Am. Mineral. 48, 1090–1132. Rapp R. P., Shimizu N. and Norman M. D. (2003) Growth of early
O’Hara M. J. and Yoder H. S. (1967) Formation and fractionation continental crust by partial melting of eclogite. Nature 425,
of basic magmas at high pressure. Scott. J. Geol. 3, 67–117. 605–609.
Osborne I., Sherlock S., Anand M. and Argles T. (2011) New Ar– Rogers A. J., Kolb J. F., Meyer M. and Armstrong R. A. (2007)
Ar ages of southern Indian kimberlites and a lamproite and Tectono-magmatic evolution of the Hutti-Maski Greenstone
their geochemical evolution. Precambr. Res. 189, 91–103. Belt, India: constrained using geochemical and geochronolog-
Panjasawatwong Y., Danyushevsky L. V., Crawford A. J. and ical data. J. Asian Earth Sci. 31, 55–70.
Harris K. L. (1995) An experimental study of the effects of melt Rollinson H. (1997) Eclogite xenoliths in West African kimberlites
composition no plagioclase/melt equilibria at 5 and 10 kb: as residues from Archean granitoid crust formation. Nature
implications for the origin of magmatic high-An plagioclase. 389, 173–176.
Contrib. Miner. Petrol. 118, 420–432. Rudnick R. L. (1995) Eclogite xenoliths: samples of Archean ocean
Paster H., Schauwecker D. S. and Haskin L. A. (1974) The floor. In 6th Inter. Kimberlite (Ext. Abstr.), 473–475.
behavior of some trace elements during solidification of the Rudnick R. I., McDonough W. F. and Chappell B. (1993)
Skaergaard layered series. Geochim. Cosmochim. Acta 38, 1549– Carbonatite metasomatism in the northern Tanzanian mantle:
1577. petrographic and geochemical characteristics. Earth Planet. Sci.
Patel S. C., Ravi S., Thakur S. S., Rao T. K. and Subbarao K. V. Lett. 128, 259–270.
(2006) Eclogite xenoliths from Wajrakarur kimberlites, south- Rudnick R. L., Barth M., Horn I. and McDonough W. F. (2000)
ern India. Mineral. Petrol. 88, 363–380. Rutile bearing refractory eclogites: missing link between con-
Patel S. C., Ravi S., Anilkumar Y., Naik A., Thakur S. S., Pati J. tinents and depleted mantle. Science 287, 278–281.
K. and Nayak S. S. (2009) Mafic xenoliths in Proterozoic Scarrow J. H., Spadea P., Cortesogno L., Savelieva G. N. and
kimberlites from Eastern Dharwar Craton, India: minerology Gaggero L. (2000) Geochemistry of garnet metagabbros from
and P–T regime. J. Asian Earth Sci. 34, 336–346. the Mindyak ophiolite massif, Southern Urals. Ofiolit i25, 103–
Pearson D. G., Snyder G. A., Shirey S. B., Taylor L. A., Carlson R. 115.
W. and Sobolev N. V. (1995) Archaean Re–Os age for Siberian Schmickler B., Jacob D. E. and Foley S. F. (2004) Eclogite
eclogites and constraints on Archaean tectonics. Nature 374, xenoliths from the Kuruman kimberlites, South Africa: geo-
711–713. chemical fingerprinting of deep subduction and cumulate
Peltonen P., Kinnunen K. A. and Huhma H. (2002) Petrology of processes. Lithos 75, 173–207.
two diamondiferous eclogite xenoliths from the Lahtojoki Schmitz M. D., Bowring S. A., de Wit M. J. and Gartz V. (2004)
kimberlite pipe, eastern Finland. Lithos 63, 151–164. Subduction and terrane collision stabilize the western Kaapvaal
craton tectosphere 2.9 billion years ago. Earth Planet. Sci. Lett. oxygen isotopes from the Nyurbinskaya pipe, Yakutia. Eur.
222, 363–376. J. Mineral. 20, 375–385.
Schulze D. J., Valley J. W., Viljoen K. S. and Spicuzza M. J. (2003) Sun S. S. and McDonough W. F. (1989) Chemical and isotopic
Oxygen isotope composition of mantle eclogites. In 8th systematics of oceanic basalts: implications for mantle compo-
International Kimberlite Conference. (Ext. abstr.). sition and processes. Geol. Soc. Lond., Spec. Publ. 42, 313–345.
Shervais J. W., Taylor L. A., Lugmair G. W., Clayton R. N., Swaminath J., Ramakrishnan M. and Vishwanatha M. N. (1976)
Mayeda T. K. and Korotev R. L. (1988) Early Proterozoic Dharwar stratigraphic model and Karnataka Craton evolution.
oceanic crust and the evolution of subcontinental mantle: Rec. Geol. Surv. India 107, 149–175.
eclogites and related rocks from southern Africa. Geol. Soc. Tappe S., Smart K. A., Pearson D. G., Steenfelt A. and Simonetti
Am. Bull. 100, 411–423. A. (2011) Craton formation in Late Archean subduction zones
Shirey S. B. and Richardson S. H. (2011) Start of the Wilson Cycle revealed by first Greenland eclogites. Geology 39, 1103–1106.
at 3 Ga shown by diamonds from subcontinental mantle. Taylor L. A. and Neal C. R. (1989) Eclogites with oceanic crustal
Science 333, 434–436. and mantle signatures from the Bellsbank kimberlite, South
Shirey S. B., Carlson R. W., Richardson S. H., Menzies A. H., Africa, Part 1: mineralogy, petrography, and whole-rock
Gurney J. J., Pearson D. G., Harris J. W. and Wiechert U. chemistry. J. Geol. 97, 551–567.
(2001) Archean emplacement of eclogitic components into the Taylor P. N., Chadwick B., Moorbath S., Ramakrishnan M. and
lithospheric mantle during formation of the Kaapvaal Craton. Vishwanatha M. N. (1984) Petrography, chemistry and isotopic
Geophys. Res. Lett. 28, 2509–2512. ages of Peninsular Gneiss, Dharwar acid volcanic rocks and the
Smart K. A., Heaman L. M., Chacko T., Simonetti A., Kopylova Chitradurga granite with special reference the late Archaean
M., Mah D. and Daniels D. (2009) The origin of high-MgO evolution of the Karnataka craton. Precambr. Res. 23, 349–375.
diamond eclogites from the Jericho Kimberlite, Canada. Earth Taylor L. A., Snyder G. A., Keller R., Remley D. A., Anand M.,
Planet. Sci. Lett. 284, 527–537. Wiesli R., Valley J. W. and Sobolev N. V. (2003) Petrogenesis
Smart K. A., Chacko T., Simonetti A., Sharp Z. D. and Heaman L. of group A eclogites and websterites: evidence from the
M. (2014) A record of Paleoproterozoic subduction preserved Obnazhennaya kimberlite, Yakutia. Contrib. Miner. Petrol.
in the Northern Slave cratonic mantle: Sr–Pb–O isotope and 145, 424–443.
trace-element investigations of eclogite xenoliths of the Jericho Usui T., Nakamura E. and Helmstaedt H. (2006) Petrology and
and Muskox kimberlites. J. Petrol. 55, 549–583. geochemistry of eclogite xenoliths from Colorado plateau:
Smit K. V., Stachel T., Creaser R. A., Ickert R. B., Dufrane S. A., implications for the evolution of subducted oceanic crust. J.
Stern R. A. and Seller M. (2014) Origin of eclogite and Petrol. 47, 929–946.
pyroxenite xenoliths from the Victor kimberlite, Canada, and Vasudev V. N., Chadwick B., Nutman A. P. and Hedge G. V.
implications for Superior craton formation. Geochim. (2000) Rapid development of the Late Archaean Hutti schist-
Cosmochim. Acta 125, 308–337. belt, Northern Karnataka: implications of new field data and
Smyth J. R., McCormick T. C. and Caporuscio F. A. (1984) SHRIMP U–Pb zircon ages. J. Geol. Soc. India 55, 529–540.
Petrology of a suite of eclogite inclusions from the Bobbejaan Viljoen K. S. (1994) The petrology and geochemistry of a suite of
kimberlite: I. Two unique corundum grospydites. In Kimberlites mantle-derived eclogite xenoliths from the Kaalvallei kimber-
II: The Mantle and Crust–Mantle Relationships (ed. J. lite, South Africa. Ph. D. thesis, Witwatersrand, Johannesburg.
Kornprobst). Elsevier, Amsterdam, pp. 109–119. Viljoen K. S., Smith C. B. and Sharp Z. D. (1996) Stable and
Smyth J. R., Caporuscio F. A. and McCormick T. C. (1989) radiogenic isotope study of eclogite xenoliths from the Orapa
Mantle eclogites: evidence of igneous fractionation in the kimberlite, Botswana. Chem. Geol. 131, 235–255.
mantle. Earth Planet. Sci. Lett. 93, 133–141. Walter M. J. (1998) Melting of garnet peridotite and the origin of
Snyder G. A., Taylor L. A., Crozaz G., Halliday A. N., Beard B. komatiite and depleted lithosphere. J. Petrol. 39, 29–60.
L., Sobolev V. N. and Sobolev N. V. (1997) The origins of Wayte R. J., Worden R. H., Rubie D. C. and Droop G. T. R.
Yakutian eclogite xenoliths. J. Petrol. 38, 85–113. (1989) A TEM study of disequilibrium plagioclase breakdown
Sobolev N. V. (1977) Deep-seated inclusions in kimberlites and the at high pressure: the role of infiltrating fluid. Contrib. Miner.
problem of the composition of the upper mantle. American Petrol. 101, 426–437.
Geophysical Union, Washington, DC, 279. Woodland A. B., Seitz H., Marschall R., Olker B. and Ludwig T.
Sobolev A. V., Hofmann A. W. and Nikogosian I. K. (2000) (2002) Li abundances in eclogite minerals: a clue to a crustal or
Recycled oceanic crust observed in “ghost plagioclase” within mantle origin? Contrib. Miner. Petrol. 143, 587–601.
the source of Mauna Loa lavas. Nature 404, 986–990. Zhang H.-H., Mattey D. P., Grassineau B., Lowry D., Brownless
Sobolev N. V., Pokhilenko N. P., Lavrentev Y. G. and Yefimova M., Gurney J. J. and Menzies M. A. (2000) Recent fluid
E. S. (1977) Deep-seated xenoliths, xenocrysts in kimberlites processes in the Kaapvaal Craton, South Africa: coupled
and crystalline inclusions in diamonds from “Udachnaya” oxygen isotope and trace element disequilibrium in polymict
kimberlite pipe, Yakutia. In 2nd Internat. Kimberlite Conf., peridotites. Earth Planet. Sci. Lett. 176, 57–72.
unpaginated. (Ext. abstr.).
Spetsius Z. V., Taylor L. A., Valley J. W., Deangelis M. T.,
Associate editor: Dmitri A. Ionov
Spicuzza M., Ivanov A. D. and Banzerak V. I. (2008)
Diamondiferous xenoliths from crustal subduction: garnet

Dongre 2015

Uploaded by

Copyright:

Available Formats

Dongre 2015

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dongre 2015

Uploaded by

Copyright:

Available Formats

Available online at www.sciencedirect.

Subduction-related origin of eclogite xenoliths from

⇑ Corresponding author at: Australian Research Council Centre

A.N. Dongre et al. / Geochimica et Cosmochimica Acta 166 (2015) 165–188

A.N. Dongre et al. / Geochimica et Cosmochimica Acta 166 (2015) 165–188

GLITTER 4.0 (Macquarie University; Griﬃn et al., 2008)

garnet, and 47Ti determined as TiO2 for the rutiles.

tion limit are reported in Table 3. USGS glass BCR2-G was

and always yielded values within 10% of the published val-

only pristine unaltered grain areas, which was an issue par-

sis the target area was checked in reﬂected and in

spots. The time-resolved signals for each individual analysis

were screened for inclusions or cracks (identiﬁed by spikes

in the time-resolved signal) and aﬀected areas or complete

grains were analyzed by secondary ion mass spectrometry

(SIMS) using a Cameca IMS 1280 instrument at the

(2013). Three to ﬁve garnet grains per sample were

mounted in epoxy and placed within 5 mm of the center

lar) were co-mounted with the unknowns. After

polishing, the grains were imaged with backscattered elec-

localization free of inclusions or cracks. Backscattered

images of all garnet grains showing that all grains are

internally chemically homogeneous are shown in

beam of 15 lm diameter was utilized, along with low

energy electrons for charge compensation. The analytical

simultaneously in Faraday cups at high mass resolution

(m/Dm  2275 for 18O ) to avoid isobaric interferences.

ments by EPMA as mentioned previously, as these were

uniform in composition within the samples. The quadra-

tic matrix calibration curve parameters of Ickert and

ence value for S0088B. The analytical uncertainties in

d18O include all known sources, including matrix correc-

tions, and for 112 spot analyses of the unknowns the

on two to four grains per sample were measured and

average d18OVSMOW of these analyses for each sample

4.1. Mineral chemistry

4.1.1. Major elements

A.N. Dongre et al. / Geochimica et Cosmochimica Acta 166 (2015) 165–188

A.N. Dongre et al. / Geochimica et Cosmochimica Acta 166 (2015) 165–188

cpx: 17.3–229 ppm), Cr (grt: 311–5,333 ppm, cpx: 360–

Rutiles are mostly present in the Wajrakarur Group 2

ratios are strongly superchondritic as in many other eclogite

Nb/Ta ratio is found in Group 1 sample KL2–48. Uranium

concentrations (0.39–3.53 ppm) are similar, and Ni (3.24–

45 ppm) contents are lower than in Koidu rutiles (Barth

4.1.4. Garnet oxygen isotopes

Oxygen isotopic compositions (& d18OVSMOW) in pris-

(Table 1). This range extends signiﬁcantly beyond the value

of unaltered mantle equilibrated at high temperature

(5.5 ± 0.4&, Mattey et al., 1994). Wajrakarur Group 1

samples of Wajrakarur Group 2 eclogites (Table 1).

The range is most similar to eclogites from the Orapa

kimberlite (Fig. 8), which is a locality with abundant inde-

restricted d18O-range in Kuruman xenoliths, which are

interpreted to originate as lithospheric pyroxenites that

never at the Earth’s surface and has not experienced seawa-

and reconstructed whole rock compositions, particularly

calcium concentrations, Mg-number, bulk Sr-anomaly,

entiated coinciding with the geochemical classiﬁcation of

beam of 15 lm diameter was utilized, along with low

(m/Dm 2275 for 18O ) to avoid isobaric interferences.