garnet peridotite xenoliths and xenocrysts from the monk hill kimberlite, south australia: insights...

Garnet Peridotite Xenoliths and Xenocrystsfrom the Monk Hill Kimberlite, South Australia:Insights into the Lithospheric Mantle beneaththe Adelaide Fold Belt

RALF TAPPERT1*, JOHN FODEN1, KARLIS MUEHLENBACHS2 ANDKEVIN WILLS3

1GEOLOGY AND GEOPHYSICS, SCHOOL OF EARTH AND ENVIRONMENTAL SCIENCES, UNIVERSITY OF ADELAIDE,

ADELAIDE, 5005, SA, AUSTRALIA2DEPARTMENT OF EARTH AND ATMOSPHERIC SCIENCES, 1^26 EARTH SCIENCE BUILDING, UNIVERSITY OF ALBERTA,

EDMONTON, ALBERTA, T6G 2E3, CANADA3FLINDERS MINES LTD., NORWOOD, 5000, SA, AUSTRALIA

RECEIVED SEPTEMBER 7, 2010; ACCEPTED JUNE 30, 2011ADVANCE ACCESS PUBLICATION AUGUST 12, 2011

A total of 73 peridotitic mantle xenoliths and a set of garnet and

clinopyroxene xenocrysts from the recently discovered Jurassic Monk

Hill kimberlite (UCO-H77A) in South Australia were used to con-

strain the thermal and compositional structure of the lithospheric

mantle beneath the Adelaide Fold Belt, which is located at the south-

eastern margin of the Australian craton. The xenoliths contain

mostly lherzolitic mineral assemblages (garnetþ cpxþ opx� chro-

mite), but lack preserved olivine as a result of alteration. Pressure

and temperature estimates for the suite of xenoliths (73 samples)

follow an array from �1·2 GPa and 6508C to �5· 0 GPa and

13008C, which reflects the conductive geothermal gradient for this

region at the time of kimberlite emplacement (Jurassic, �189

Ma). Based on the projected intercept of the geotherm with the

mantle adiabat, the maximum depth of the lithospheric mantle be-

neath the Monk Hill kimberlite is estimated to be around 160^

180 km, with the base of the lithosphere lying marginally outside the

diamond stability field. The results challenge previously proposed

paleogeotherms for this region, which are either significantly hotter

or significantly cooler. Sm^Nd isotope data for high-T garnet and

clinopyroxene megacrysts define a robust isochron (189�17 Ma),

which reflects theJurassic emplacement age of the Monk Hill kim-

berlite. This indicates that minerals from deeper parts of the

lithosphere were in isotopic equilibrium and exhumed during the

kimberlite eruption from temperatures above the Sm^Nd closure tem-

perature for garnet and clinopyroxene.Within the suite of peridotite

xenoliths from Monk Hill, abundant low-T (510008C) xenoliths

can be distinguished from a less common high-T (410008C) popula-tion. The high-T xenolith population is characterized by

titanium-enriched compositions, suggesting that the deeper parts of

the lithosphere were affected by pervasive melt metasomatism.This

interpretation is supported by the trace element compositions (rare

earth elements, high field strength elements) of the garnet and clino-

pyroxene xenocrysts.

KEY WORDS: mantle; xenolith; peridotite; lherzolite; kimberlite;

geothermobarometry

I NTRODUCTIONThe Australian continent is composed of two main tectonicunits with very distinct geological histories. The westernpart of Australia encompasses the Australian craton,which consists of Archaean blocks and fold belts that are

*Corresponding author. Present address: Department of Earth andAtmospheric Sciences, University of Alberta, Edmonton, AlbertaT6G2E3, Canada.Telephone:þ1780-492-2827.E-mail: [email protected]

� The Author 2011. Published by Oxford University Press. Allrights reserved. For Permissions, please e-mail: [email protected]

JOURNALOFPETROLOGY VOLUME 52 NUMBER10 PAGES1965^1986 2011 doi:10.1093/petrology/egr036 at U

niversity of Alberta on M

ay 9, 2013http://petrology.oxfordjournals.org/

Dow

nloaded from

http://petrology.oxfordjournals.org/

primarily of Paleoproterozoic age. The eastern part ofAustralia consists of the much younger Tasman Fold Belt(Palaeozoic^Mesozoic), which formed since theDelamerian orogeny (Cambrian^Ordovician) by accre-tion of terranes to the Australian craton (Plumb, 1979;Foden et al., 2006). The contact between the Australiancraton and theTasman Fold Belt is marked by the TasmanLine, which is interpreted as the rifted margin of theProterozoic supercontinent Rodinia (Veevers, 1984; Fig. 1).The compositional and thermal structure of the litho-sphere beneath the Tasman Fold Belt has been extensivelystudied through xenoliths, which were brought to the sur-face by widespread Jurassic to Cenozoic basaltic volcanismin Eastern Australia, including Queensland, New SouthWales, Victoria, and southern South Australia (Fergusonet al., 1977; Sutherland & Hollis, 1982; O’Reilly & Griffin,1985). The composition and the thermal structure of the

lithosphere on the western, cratonic side of the TasmanLine is less well constrained, primarily as a result of therelative scarcity of volcanic rocks carrying suitable mantlexenolith assemblages. Although xenoliths from the litho-sphere of the Australian craton have been recovered fromkimberlites in the Adelaide Fold Belt, most of these xeno-liths are of crustal or uppermost mantle origin (Ferguson& Sheraton, 1979; Pearson & O’Reilly, 1991; Pearson et al.,1991). Thus far, samples from the deeper lithosphericmantle beneath this region have been restricted to isolatedxenocrystic mineral grains, such as garnet, clinopyroxene,ilmenite, chromite, and rare diamond (Scott Smith et al.,1984; Gaul et al., 2003;Tappert et al., 2007, 2009). Initial pet-rological studies of mantle xenocrysts from the AdelaideFold Belt suggested that, during the Jurassic kimberlite vol-canism, the region was underlain by a cool lithosphericmantle with a geothermal gradient similar to Archaean

Fig. 1. Map of south^central Australia showing the location of kimberlites and basaltic rocks carrying mantle xenoliths.

JOURNAL OF PETROLOGY VOLUME 52 NUMBER 10 OCTOBER 2011

1966

at University of A

lberta on May 9, 2013

http://petrology.oxfordjournals.org/D

ownloaded from


cratonic settings (Gaul et al., 2003). These results, however,are in contrast to those based on lower crustal xenoliths(eclogites and mafic granulites) and shallow uppermantle xenoliths, which were in part derived from thesame kimberlites as the xenocrysts, but which indicate amuch higher geothermal gradient (Pearson et al., 1991).The identification of a unique suite of garnet peridotite

xenoliths in the recently discovered Monk Hill kimberlite(UCO-H77A) in South Australia now allows a muchmore robust estimate of the P^Trelations.The petrologicaldata presented in this study from this suite of xenolithshave allowed us to constrain the thermal and composition-al structure of the lithosphere beneath the Australiancraton in this region at the time of kimberlite emplacementin the Jurassic.

SAMPLES AND ANALYTICALTECHNIQUESA total of 73 xenoliths from the Monk Hill kimberlite wereselected for this study. They were recovered from heavymineral concentrates of around 100 tonnes of processedkimberlite. The xenoliths are small (partly determined bythe processing methods), ranging in size from 0·5 to1·5 cm (Fig. 2). The studied samples were selected from sev-eral hundred specimens based on the visual presence of un-altered coexisting garnet (purplish-red), orthopyroxene(yellowish-brown), and clinopyroxene (bright green). Twoxenoliths without visible clinopyroxene were also includedin this study. Xenoliths containing only garnet and clino-pyroxene, which represent the majority of the samples,were not included. The additional presence or absence ofchromite was not used as selection criterion. All xenolithswere mounted in epoxy resin and polished until at leastone crystal of each of the coexisting minerals was exposedto the surface.In addition to the polymineralic xenoliths, a large

number of single mantle xenocrysts were recovered fromthe heavy mineral concentrates from the Monk Hill kim-berlite, including garnet, clinopyroxene, orthopyroxeneand chromite. The majority of the xenocrysts are visuallyand compositionally indistinguishable from their counter-parts in the xenoliths, indicating that they may representfragments of disintegrated xenoliths. Besides thesexenolith-like xenocrysts, which are characterized by rela-tively small sizes (1^5mm), a much coarser (41cm) popu-lation of garnet, clinopyroxene, and orthopyroxenemegacrysts is present in the mineral concentrates. Thesemegacrysts are distinctive not only because of their overalllarger grain sizes. Compared with the bulk of their xeno-lith and smaller xenocryst counterparts, the clinopyroxenemegacrysts are distinctively paler green and the garnetmegacrysts generally less purple in colour. Unlike the

clinopyroxene megacrysts, the garnet megacrysts are usu-ally coated by a thin layer (51mm) of kelyphite.Major and minor element compositions of the mineral

assemblages within the xenoliths and selected xenocrystswere determined using a CAMECA SX-51 electron micro-probe at the University of Adelaide. For each mineralgrain, 2^4 point analyses were averaged. Where possible,multiple grains of each mineral were analysed. The meas-urements were performed using a15 kVacceleration voltageand 20 nA probe current. Depending on the element, peakand background counting times each ranged from 20 to40 s, respectively. The concentration were subsequentlycorrected using the PAP method. The detection limit forthe oxide species is of the order of �200 ppm (exceptNa2O: �400 ppm). Accuracy and precision were tested onsecondary standards (synthetic and natural minerals) andare within �1·0% relative for the major elements.Concentrations of 27 trace and minor elements (Na, Sc,

Ti, Mn, Ni, Ga, Sr,Y, Zr, Nb, Ba, REE, Hf,Ta) were ana-lysed in situ for a limited number of epoxy-mountedgarnet and clinopyroxene xenocrysts by laser ablationinductively coupled plasma mass spectrometry(LA-ICP-MS). The Agilent 7500cs ICP-MS system at theUniversity of Adelaide was coupled to a New Wave Nd^YAG UV-laser (213 nm) system. The counting time foreach sample was typically 100 s. The NIST 612 glass stand-ard, which was analysed after every 5^10 samples, wasused to calibrate the element concentrations of the garnetsand clinopyroxenes. Each analysis was normalized usingCa values determined by electron microprobe.Two garnet and two clinopyroxene megacrysts were

selected for Sm^Nd isotope analysis. The samples wereselected based on the absence of apparent fractures andother signs of severe alteration. The size of the crystalswas between 1·5 and 2 cm. The Sm^Nd isotope dilution

Fig. 2. Examples of garnet-peridotite xenoliths from the Monk Hillkimberlite in hand specimen.

TAPPERT et al. PERIDOTITE XENOLITHS, MONK HILL KIMBERLITE

1967

at University of A



ownloaded from


thermal ionization mass spectrometry (ID-TIMS) ana-lyses were performed on coarsely crushed mineral frag-ments, which were visibly free of inclusions and fractures.Minor coatings on the surface of some of the mineral frag-ments were removed by multiple rounds of acid treatment(boiling HCl) and ultrasonic cleaning. About 100mg ofclinopyroxene and garnet (after leaching and cleaning)were dissolved with a mixture of HF and HNO3 incapped vials overnight. Nd isotopic compositions weremeasured at the University of Adelaide by TIMS on aFinnigan MAT 262 system in multi-dynamic mode (seeElburg et al., 2003). All samples were spiked with a mixed149Sm�150Nd spike. Nd isotopic ratios were monitored bymeasuring J & M specpure Nd2O3, and this yielded a143Nd/144Nd ratio of 0·511569�22 (2s, n¼ 25). The valuefor BCR-1 at the time was 0·512590�28, and LaJolla gave0·511800�22. Nd blanks were less than 500 pg, and Smless than 200 pg. Reproducibility of the 147Sm/144Nd ratiowas better than 0·8%.

RESULTSMineral compositionsBecause of the small size of the xenoliths and the relativelylarge grain sizes of the minerals in the xenoliths, modalmineral abundances vary greatly between samples (Fig. 2).This restricts the possibility of classifying the xenolithspetrographically. Another complication with their classifi-cation is the absence of olivine in all samples. This absencecan be attributed to the alteration that took place withinthe host kimberlite after emplacement. Textural evidencesuggests that olivine was present in the original xenolith as-semblages, but has been completely transformed into ser-pentine and other low-temperature alteration products(e.g. clay minerals). The presence of relict serpentine onthe surface of some of the xenoliths (Fig. 2) supports the as-sumption that olivine was part the original xenolith assem-blages, but the bulk of the serpentine and other alterationminerals was probably removed during the heavy mineralrecovery process. Owing to the petrographic challenges,the classification of the xenoliths is primarily based on thecomposition of the garnets, which in peridotites are sensi-tive indicators of the mineralogical composition of thebulk-rock (Gurney & Switzer, 1973; Sobolev et al., 1973;Dawson & Stephens, 1975; Schulze, 2003; Gru« tter et al.,2004).No significant compositional zoning within any of the

minerals was observed, and, within the analytical errors,multiple grains of the same mineral from each xenolithare compositionally identical. This confirms that the xeno-liths are fully equilibrated samples in equilibrium with theP^T conditions of the Jurassic lithospheric mantle (con-ductive) geotherm.

Garnet

Garnets from the Monk Hill xenolith suite have compos-itions that are typical for peridotitic garnets, withMg-numbers [Mg/(MgþFe)] between 0·78 and 0·84,chromium contents between 1·0 and 7·1wt % Cr2O3, andcalcium contents between 4·6 and 7·0 wt % CaO. Basedon the relationship between their calcium and chromiumcontents, the bulk of the garnets can be classified as lherzo-litic, or G9 garnets, which is consistent with the presenceof clinopyroxene and orthopyroxene in theses samples(Fig. 3a). Based on the classification scheme of Gru« tteret al. (2004), the garnets in one xenolith were classified aswehrlitic, or G12 garnets. Another xenolith contained gar-nets that were classified as ‘G4’ garnets. The G4 class in-corporates a range of garnet types, such as low-Caeclogitic, websteritic, pyroxenitic, and low-Cr peridotiticgarnets, which show considerable compositional overlaps.Because the G4 garnets in this study fall just marginallyoutside the field of lherzolitic garnets (Fig. 3a), and theirMg-number (0·80) overlaps with typical peridotitic gar-nets, we classified these garnets loosely as low-Cr perido-titic garnets. The garnets from two other xenoliths wereclassified as low-Cr megacrystic garnets (G1), based on theGru« tter et al. (2004) classification scheme. These garnetsare characterized by relatively high titanium contents of�0·5wt % TiO2 (Table 1). However, they also have highMg-numbers (0·84), which indicates that they are compos-itionally more closely linked to the lherzolitic garnetsrather than to typical megacrysts. Besides that, they alsolack the large grain size of typical megacrysts. To acknow-ledge their link to the other peridotitic garnets in thisstudy, we refer to them here as high-Ti peridotitic garnets,rather than as low-Cr megacrysts.The major and minor element composition of the iso-

lated garnet xenocrysts (Fig. 3b) was found to be similarto those of the garnets from the xenoliths, with lherzoliticgarnets dominating over low-Cr peridotitic and high-Tiperidotitic garnets. However, no isolated wehrlitic garnetswere present. The chondrite-normalized rare earth elem-ent patterns (REEN) of the various garnet classes werealso found to be similar (Fig. 4,Table 2).The garnets gener-ally have subchondritic to chondritic concentrations of Laand Ce (light REE; LREE). With increasing atomicnumber, the REEN continuously increase towards �10^80times chondritic concentrations of Yb and Lu (heavyREE; HREE). A few garnets were found to have slightlysinusoidal patterns (Fig. 4). The REEN of the Monk Hillxenocryst garnets are overall typical for garnets derivedfrom fertile peridotites (lherzolites) (Gregoire et al., 2003;Viljoen et al., 2004). This interpretation is supported bythe relationship between their Zr andYcontents, which in-dicates that the garnets were mainly derived from fertilelherzolites that had experienced variable degrees of meltmetasomatism (Fig. 5). The most pervasive melt


1968

at University of A



ownloaded from


metasomatism, however, appears to be restricted to thehigh-Ti peridotitic garnet suite, which overall have higherZr and Y contents than the bulk of the lherzolitic andlow-Cr peridotitic garnets (Fig. 5, Table 2).The garnet megacrysts that were selected for Sm^Nd

isotope analysis were compositionally classified as low-Crmegacrysts (G1). Their titanium contents (0·76, 0·88wt %TiO2) are higher than those of any xenolith garnets(Table 5). Compared with the high-Ti peridotitic garnetsfrom the xenoliths, these garnets are also slightly moreiron rich (Mg-numbers 0·79- 0·82) and, therefore, compos-itionally more similar to typical megacrysts in kimberlitesworldwide.

Pyroxenes

Clinopyroxenes from xenoliths have Mg-numbers in therange 0·90^0·95. The clinopyroxene with the lowestMg-number is associated with garnet of low-Cr megacrys-tic (G1) composition. The Ca/(CaþMg) ratio of thisclinopyroxene (0·38) is also lower than that of any otherclinopyroxene (range 0·45^0·49). It has also the highesttitanium content (0·33wt % TiO2). All other clinopyrox-enes are in the range �0· 01 to 0·27wt % TiO2.Chromium and aluminium contents of the clinopyr-oxenes are in the range 0·44^1·47wt % Cr2O3 and1·54^2·82wt % Al2O3, respectively. Sodium contents arebetween 0·47 and 1·87wt % Na2O (Table 1).As in case of the garnets, the major and minor element

compositions of isolated clinopyroxene xenocrysts (n¼ 93)are comparable with the composition of their xenolithcounterparts (Table 3), with Mg-numbers between 0·87and 0·96, Ca/(CaþMg) between 0·35 and 0·52,TiO2 con-tents in the range �0· 01 to 0·45wt %, Al2O3 contents be-tween 0·10 and 2·93wt %, Cr2O3 contents between 0·20and 1·88wt %, and Na2O contents between 0·38 and1·74wt %. Similar to the garnets, the REEN patterns ofthe clinopyroxenes are rather uniform, with high LREEconcentrations (8^80 times chondritic) and a continuousdecrease towards the HREE. It is notable, however, thatclinopyroxenes with low titanium contents (50·1wt %TiO2) have generally higher LREE, whereas clinopyrox-enes with higher titanium (40·1wt % TiO2) tend to havelower concentrations of LREE (Fig. 6).Orthopyroxenes from the Monk Hill xenoliths have

Mg-numbers in the range 0·90^0·92. Chromium is in therange 0·18^0·49wt % Cr2O3, and aluminium in the range1·14^1·74wt % Al2O3.

Chromite

Chromite is present in 21 of the studied xenoliths.Chromite compositions range from 39·2 to 55·2wt %Cr2O3 and 10·6 to 27·9wt % Al2O3. Their Cr-numbers[Cr/(CrþAl)] are between 48·5 and 77·0. Titanium inchromite is generally low (50·72wt % TiO2). Only onesample contained chromite with a slightly higher TiO2

content of 1·34wt % TiO2. The Mg-number of the chro-mites is in the range 0·58^0·67.

GeothermobarometryEquilibration temperatures and pressures for the peridotitexenoliths were determined using the compositions of coex-isting silicate minerals. Experimentally calibrated geother-mobarometers for various mineral exchange reactionswere used to test that all the minerals within a particularxenolith were in equilibrium. The geothermobarometersapplied were restricted to those that are more widelyused and that are calibrated for fertile peridotite compos-itions. The geothermometer calibrations used are from

Fig. 3. Cr2O3 vs CaO (wt %) variation in garnets from peridotitexenoliths (a) and isolated garnet xenocrysts (b) from the Monk Hillkimberlite. Compositional fields are adopted from Gru« tter et al.(2004).


1969

at University of A



ownloaded from


Table 1: Representative compositions (in wt %) of minerals from the Monk Hill garnet peridotites

Sample: C-X1-MXL3 C-X2-MXL1 C-X2-MXL4 C-X4-MXL2

Paragenesis: grt–cpx–opx–chr grt–cpx–opx grt–cpx–opx–chr grt–cpx–opx–sp

Type: lherzolitic high-Ti peridotitic lherzolitic lherzolitic

Mineral: Grt Cpx Opx Chr Grt Cpx Opx Grt Cpx Opx Chr Grt Cpx Opx Chr

P2O5 0·03 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 0·03 �0·02 �0·02 0·03 �0·02 �0·02 �0·02

SiO2 41·1 54·7 57·0 0·08 41·7 53·9 57·0 41·4 54·2 56·5 0·03 41·5 53·8 57·0 0·03

TiO2 0·11 0·14 0·08 1·34 0·46 0·33 0·23 0·03 0·08 0·05 0·36 0·04 0·03 �0·02 0·23

Al2O3 18·3 1·54 1·14 10·6 20·6 2·70 1·59 21·90 2·14 1·57 19·7 21·76 2·40 1·54 19·5

Cr2O3 7·07 1·30 0·49 53·2 3·01 0·87 0·32 2·21 1·05 0·38 44·7 2·42 1·18 0·44 45·3

MgO 19·5 17·8 34·9 12·7 21·4 19·3 33·6 20·3 17·1 35·4 13·2 20·5 16·9 35·4 13·3

CaO 6·99 20·5 0·67 0·00 5·24 16·7 1·29 5·55 22·6 0·41 �0·02 5·55 22·0 0·39 �0·02

MnO 0·35 0·10 0·11 0·12 0·26 0·13 0·13 0·48 0·10 0·12 0·06 0·43 0·07 0·15 0·11

FeO 7·13 2·65 5·20 19·8 7·05 4·06 6·18 7·86 2·15 5·34 19·8 8·03 2·31 5·43 20·1

NiO �0·02 0·07 0·11 0·17 0·04 0·09 0·17 0·03 0·04 0·11 0·18 �0·02 �0·02 0·09 0·17

Na2O �0·04 1·28 0·13 �0·04 �0·04 1·44 0·23 �0·04 1·00 0·06 �0·04 �0·04 1·27 0·05 �0·04

K2O �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02

Total 100·6 100·1 99·9 98·1 99·8 99·5 100·8 99·8 100·6 100·0 98·0 100·3 100·0 100·4 98·6

P 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00

Si 2·97 3·95 3·93 0·01 2·98 3·90 3·92 2·97 3·91 3·89 0·00 2·96 3·90 3·90 0·00

Ti 0·01 0·01 0·00 0·10 0·02 0·02 0·01 0·00 0·00 0·00 0·03 0·00 0·00 0·00 0·02

Al 1·56 0·13 0·09 1·26 1·74 0·23 0·13 1·85 0·18 0·13 2·23 1·83 0·21 0·12 2·19

Cr 0·40 0·07 0·03 4·21 0·17 0·05 0·02 0·13 0·06 0·02 3·40 0·14 0·07 0·02 3·43

Mg 2·10 1·92 3·59 1·89 2·28 2·08 3·44 2·17 1·84 3·63 1·89 2·18 1·83 3·61 1·89

Ca 0·54 1·59 0·05 0·00 0·40 1·29 0·09 0·43 1·75 0·03 0·00 0·42 1·71 0·03 0·00

Mn 0·02 0·01 0·01 0·01 0·02 0·01 0·01 0·03 0·01 0·01 0·01 0·03 0·00 0·01 0·01

Fe2þ 0·43 0·16 0·30 1·66 0·42 0·25 0·36 0·47 0·13 0·31 1·59 0·48 0·14 0·31 1·61

Ni 0·00 0·00 0·01 0·01 0·00 0·01 0·01 0·00 0·00 0·01 0·01 0·00 0·00 0·00 0·01

Na 0·00 0·18 0·02 0·00 0·00 0·20 0·03 0·00 0·14 0·01 0·00 0·00 0·18 0·01 0·00

K 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00

[O] 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12

Mg-no. 0·83 0·92 0·92 0·53 0·84 0·89 0·91 0·82 0·93 0·92 0·54 0·82 0·93 0·92 0·54

Cr-no. 0·21 0·36 0·22 0·77 0·09 0·18 0·12 0·06 0·25 0·14 0·60 0·07 0·25 0·16 0·61

Sample: C-X4-MXL3 C-X5-MXL2 C-X5-MXL3 C-X6-MXL1

Paragenesis: grt–cpx–opx–chr grt–cpx–opx grt–opx grt–cpx–opx–chr

Type: lherzolitic low-Cr peridotitic high-Ti peridotitic wehrlitic

Mineral: Grt Cpx Opx Chr Grt Cpx Opx Grt Opx Grt Cpx Opx Chr

P2O5 0·04 �0·02 �0·02 �0·02 0·06 �0·02 �0·02 0·03 �0·02 �0·02 �0·02 �0·02 �0·02

SiO2 41·3 54·0 56·7 0·04 41·3 54·4 57·5 41·8 56·3 41·5 53·2 56·3 0·04

TiO2 0·04 0·06 0·05 0·23 0·03 0·05 0·04 0·53 0·19 0·05 0·17 0·09 0·56

Al2O3 21·7 2·36 1·56 19·6 22·4 2·16 1·46 20·5 1·64 22·4 1·60 1·66 27·9

Cr2O3 2·52 1·10 0·40 45·1 0·98 0·44 0·20 3·08 0·35 1·58 0·63 0·35 39·2

MgO 20·5 17·1 35·8 13·2 19·4 17·1 34·9 21·3 33·8 19·7 17·6 35·4 14·5

CaO 5·60 22·12 0·39 �0·02 5·35 22·5 0·42 5·26 1·29 5·85 23·9 0·34 �0·02

MnO 0·44 0·05 0·13 0·13 0·49 0·09 0·17 0·28 0·12 0·57 0·06 0·12 0·11

FeO 8·05 2·24 5·41 20·0 9·64 2·81 6·84 7·46 6·33 8·57 1·76 5·90 16·7

NiO �0·02 �0·02 0·10 0·15 �0·02 0·04 0·08 0·04 0·13 �0·02 0·04 0·04 0·16

Na2O �0·04 1·19 �0·04 0·03 �0·04 1·01 0·03 �0·04 0·24 �0·04 0·47 �0·04 �0·04

K2O �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02

Total 100·3 100·2 100·6 98·5 99·7 100·7 101·6 100·4 100·4 100·2 99·5 100·2 99·3

P 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00

Si 2·95 3·91 3·88 0·00 2·97 3·93 3·91 2·98 3·89 2·97 3·89 3·88 0·00

Ti 0·00 0·00 0·00 0·02 0·00 0·00 0·00 0·03 0·01 0·00 0·01 0·00 0·04

Al 1·83 0·20 0·13 2·21 1·90 0·18 0·12 1·72 0·13 1·89 0·14 0·13 2·98

Cr 0·14 0·06 0·02 3·42 0·06 0·02 0·01 0·17 0·02 0·09 0·04 0·02 2·81

Mg 2·19 1·84 3·65 1·88 2·08 1·84 3·54 2·27 3·48 2·10 1·92 3·63 1·95

Ca 0·43 1·72 0·03 0·00 0·41 1·74 0·03 0·40 0·10 0·45 1·87 0·03 0·00

Mn 0·03 0·00 0·01 0·01 0·03 0·01 0·01 0·02 0·01 0·03 0·00 0·01 0·01

Fe2þ 0·48 0·14 0·31 1·61 0·58 0·17 0·39 0·44 0·37 0·51 0·11 0·34 1·26

Ni 0·00 0·00 0·01 0·01 0·00 0·00 0·00 0·00 0·01 0·00 0·00 0·00 0·01

Na 0·00 0·17 0·01 0·01 0·00 0·14 0·00 0·01 0·03 0·00 0·07 0·00 0·00

K 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00

[O] 12 12 12 12 12 12 12 12 12 12 12 12 12

Mg-no. 0·82 0·93 0·92 0·54 0·78 0·92 0·90 0·84 0·90 0·80 0·95 0·91 0·61

Cr-no. 0·07 0·24 0·15 0·61 0·03 0·12 0·08 0·09 0·13 0·05 0·21 0·13 0·49

(continued)


1970

at University of A



ownloaded from


Table 1: Continued

Sample: C-X6-MXL2 C-X7-MXL1 C-X7-MXL5 D-X1-MXL1 D-X3-MXL2

Paragenesis: grt–cpx–opx–chr grt–opx grt–cpx–opx grt–cpx–opx grt–cpx–opx

Type: lherzolitic lherzolitic lherzolitic lherzolitic lherzolitic

Mineral: Grt Cpx Opx Chr Grt Opx Grt Cpx Opx Grt Cpx Opx Grt Cpx Opx

P2O5 0·03 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 0·03 �0·02 0·03 �0·02 �0·02 0·03 �0·02 �0·02

SiO2 41·5 54·1 56·5 0·04 41·7 56·8 42·1 54·2 57·2 41·1 53·4 56·5 41·2 53·0 55·4

TiO2 0·06 �0·02 0·08 0·13 0·26 0·15 �0·02 �0·02 �0·02 0·09 0·19 0·09 0·08 0·20 0·11

Al2O3 20·5 1·82 1·23 14·6 22·19 1·25 21·9 1·90 1·62 21·5 2·50 1·58 21·4 2·46 1·53

Cr2O3 4·36 1·47 0·47 55·2 1·60 0·19 2·08 0·82 0·37 2·37 1·23 0·39 2·47 1·28 0·44

MgO 20·3 17·5 35·5 14·0 21·2 34·5 20·0 17·3 35·3 20·2 16·6 35·1 20·4 16·7 35·2

CaO 6·02 21·8 0·68 �0·02 4·57 0·74 5·63 22·1 0·39 5·49 21·9 0·39 5·48 21·8 0·42

MnO 0·40 0·06 0·14 0·11 0·31 0·07 0·44 0·05 0·16 0·43 0·07 0·12 0·43 0·03 0·09

FeO 7·26 2·00 5·12 14·8 7·98 6·03 7·95 2·09 5·38 8·02 2·24 5·46 8·10 2·37 5·59

NiO 0·03 0·09 0·13 0·08 �0·02 0·14 �0·02 0·06 0·10 �0·02 0·03 0·03 �0·02 0·04 0·06

Na2O �0·04 1·07 0·06 �0·04 0·06 0·14 �0·04 0·77 0·03 �0·04 1·25 �0·04 �0·04 1·27 0·06

K2O �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02

Total 100·4 99·9 99·9 98·9 99·9 100·0 100·2 100·3 100·6 99·3 99·4 99·7 99·6 99·1 98·9

P 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00

Si 2·97 3·92 3·90 0·00 2·97 3·92 3·00 3·92 3·91 2·97 3·90 3·90 2·96 3·88 3·87

Ti 0·00 0·00 0·00 0·01 0·01 0·01 0·00 0·00 0·00 0·00 0·01 0·00 0·00 0·01 0·01

Al 1·73 0·16 0·10 1·65 1·86 0·10 1·84 0·16 0·13 1·83 0·21 0·13 1·82 0·21 0·13

Cr 0·25 0·08 0·03 4·19 0·09 0·01 0·12 0·05 0·02 0·14 0·07 0·02 0·14 0·07 0·02

Mg 2·17 1·89 3·64 2·01 2·26 3·55 2·13 1·87 3·59 2·17 1·81 3·61 2·18 1·82 3·66

Ca 0·46 1·69 0·05 0·00 0·35 0·06 0·43 1·78 0·03 0·42 1·71 0·03 0·42 1·71 0·03

Mn 0·02 0·00 0·01 0·01 0·02 0·00 0·03 0·00 0·01 0·03 0·00 0·01 0·03 0·00 0·01

Fe2þ 0·43 0·12 0·30 1·19 0·48 0·35 0·47 0·13 0·31 0·48 0·14 0·31 0·49 0·15 0·33

Ni 0·00 0·01 0·01 0·01 0·00 0·01 0·00 0·00 0·01 0·00 0·00 0·00 0·00 0·00 0·00

Na 0·00 0·15 0·01 0·00 0·01 0·02 0·00 0·11 0·00 0·01 0·18 0·01 0·00 0·18 0·01

K 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00

[O] 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12

Mg-no. 0·83 0·94 0·93 0·63 0·83 0·91 0·82 0·94 0·92 0·82 0·93 0·92 0·82 0·93 0·92

Cr-no. 0·13 0·35 0·20 0·72 0·05 0·09 0·06 0·22 0·13 0·07 0·25 0·14 0·07 0·26 0·16

Sample: D-X3-MXL3 D-X4-MXL5 D-X5-MXL5 D-X5-MXL6

Paragenesis: grt–cpx–opx grt–cpx–opx grt–cpx–opx–chr grt–cpx–opx–chr

Type: lherzolitic lherzolitic lherzolitic lherzolitic

Mineral: Grt Cpx Opx Grt Cpx Opx Grt Cpx Opx Chr Grt Cpx Opx Chr

P2O5 0·04 �0·02 �0·02 0·04 �0·02 �0·02 0·03 �0·02 �0·02 �0·02 0·04 �0·02 �0·02 �0·02

SiO2 41·46 53·47 56·56 41·92 54·54 56·88 41·97 53·26 57·42 0·03 42·24 53·78 57·23 0·03

TiO2 0·10 0·21 0·08 0·26 0·26 0·11 0·09 0·21 0·08 0·67 0·10 0·22 0·08 0·67

Al2O3 21·43 2·53 1·58 21·37 2·76 1·20 21·69 2·58 1·54 19·57 21·70 2·58 1·61 19·64

Cr2O3 2·52 1·28 0·36 2·67 1·44 0·32 2·35 1·32 0·38 43·50 2·50 1·31 0·43 43·31

MgO 20·38 16·56 35·12 21·27 17·03 34·90 20·33 16·66 35·51 12·97 20·22 16·71 35·23 13·10

CaO 5·50 21·78 0·40 4·92 19·32 0·70 5·57 21·74 0·41 0·03 5·51 21·80 0·40 �0·02

MnO 0·39 0·03 0·12 0·32 0·09 0·10 0·43 0·07 0·14 0·14 0·39 0·05 0·12 0·14

FeO 8·11 2·34 5·59 7·49 2·73 5·76 8·10 2·24 5·65 20·95 8·04 2·24 5·68 21·21

NiO �0·02 0·03 0·13 �0·02 �0·02 0·11 �0·02 0·06 0·09 0·20 �0·02 �0·02 0·10 0·09

Na2O �0·04 1·28 0·05 �0·04 1·87 0·14 �0·04 1·32 0·05 �0·04 �0·04 1·31 0·06 �0·04

K2O �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02

Total 99·9 99·5 100·0 100·3 100·1 100·2 100·6 99·5 101·3 98·1 100·8 100·0 100·9 98·2

P 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00

Si 2·97 3·90 3·90 2·98 3·93 3·92 2·98 3·89 3·90 0·00 3·00 3·90 3·90 0·00

Ti 0·01 0·01 0·00 0·01 0·01 0·01 0·01 0·01 0·00 0·05 0·01 0·01 0·00 0·05

Al 1·81 0·22 0·13 1·79 0·23 0·10 1·82 0·22 0·12 2·22 1·81 0·22 0·13 2·23

Cr 0·14 0·07 0·02 0·15 0·08 0·02 0·13 0·08 0·02 3·32 0·14 0·08 0·02 3·30

Mg 2·18 1·80 3·61 2·25 1·83 3·58 2·16 1·81 3·60 1·86 2·14 1·81 3·58 1·88

Ca 0·42 1·70 0·03 0·37 1·49 0·05 0·42 1·70 0·03 0·00 0·42 1·69 0·03 0·00

Mn 0·02 0·00 0·01 0·02 0·01 0·01 0·03 0·00 0·01 0·01 0·02 0·00 0·01 0·01

Fe2þ 0·49 0·14 0·32 0·45 0·16 0·33 0·48 0·14 0·32 1·69 0·48 0·14 0·32 1·71

Ni 0·00 0·00 0·01 0·00 0·00 0·01 0·00 0·00 0·00 0·02 0·00 0·00 0·01 0·01

Na 0·00 0·18 0·01 0·00 0·26 0·02 0·01 0·19 0·01 0·00 0·00 0·18 0·01 0·00

K 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00

[O] 12 12 12 12 12 12 12 12 12 12 12 12 12 12

Mg-no. 0·82 0·93 0·92 0·84 0·92 0·92 0·82 0·93 0·92 0·52 0·82 0·93 0·92 0·52

Cr-no. 0·07 0·25 0·13 0·08 0·26 0·15 0·07 0·26 0·14 0·60 0·07 0·25 0·15 0·60


1971

at University of A



ownloaded from


Harley (1984) (THar) and Krogh (1988) (TKro). Thesegeothermometers are based on the temperature depend-ence of the iron and magnesium exchange between garnetand orthopyroxene (THar), and between garnet and clino-pyroxene (TKro). Temperatures were also calculated withthe geothermometers of Brey & Ko« hler (1990) (TBKN)andTaylor (1998) (TTay), which utilize the calcium contentof orthopyroxene in equilibrium with clinopyroxene (Breyet al., 1990). Equilibration pressures were determined usingthe geobarometer of Brey & Ko« hler (1990) (PBKN) (forTHar, TKro, and TBKN) and Nickel & Green (1985) (forTTay). These geobarometers are based on the aluminiumcontent of orthopyroxene in equilibrium with garnet(Brey et al., 1990). In addition to using geothermobarom-eters that are based on the equilibrium between differentminerals, P^Testimates were also calculated using the em-pirical single-cpx geothermobarometers of Nimis &Taylor (2000) (TNT, PNT). In all cases, the P^Tconditionsfor each of the mineral assemblages were calculatediteratively.The P^Tresults for the various geothermometer^geoba-

rometer combinations are shown in Fig.7 and are summar-ized inTable 4. The calculated temperature conditions forthe xenoliths range from �650 to414008C, and the pres-sure conditions range from 1·1 to45·0GPa, depending onthe geothermobarometer combination used. Despite somevariations in the P^T conditions for single xenoliths, it isnotable that the results of the geothermobarometry

calculations produce tightly constrained arrays with simi-lar trajectories for each of the selected geothermometer^geobarometer combinations. Particularly theTBKN^PBKN

and TTay^PNG combinations produce consistent results.It is notable, however, that the single-cpx geothermobaro-metry results in overall higher pressure estimates for thexenoliths at the lower end of the P^T array (Fig. 7e), andthe temperatures based onTHar are systematically higherat the low-Tend compared with other geothermometer es-timates (Fig. 7d).Among the xenoliths, a small high-T group (410008C)

can be distinguished from the tightly clustered and morecommon low-T group (510008C). The high-T group alsocontains the high-Ti peridotite xenoliths. Chromite-bearing assemblages are present throughout most of theP^T range. P^Tdata given in the following discussion aregenerally based on the TBKN^PBKN geothermobarometercombination, which is based on an internally consistentset of experimental data, and which has been tested to bereliable over a wide range of P^T conditions (Brey &Ko« hler, 1990; Smith, 1999).

Cr^Al relationships (xenoliths)The coexisting minerals within the xenoliths exhibitstrongly correlated compositional relationships, particu-larly in their Cr-numbers [Cr/(CrþAl)] (Fig. 8). All min-erals show a positive linear correlation in theirCr-numbers, with the exception of the two highest

Fig. 4. Chondrite-normalized rare earth element patterns of garnet xenocrysts from the Monk Hill kimberlite. Chondrite composition fromMcDonough & Sun (1995).


1972

at University of A



ownloaded from


Table 2: Composition of selected garnet xenocrysts from Monk Hill

Sample: 20 29 3 6 10 74 9 22 24 75 26 71

Class: G9 G9 G9 G9 G9 G9 G9 G1 G1 G1 G4 G4

wt %

P2O5 0·03 �0·02 0·05 0·06 �0·02 �0·02 0·03 0·05 0·03 0·05 �0·02 �0·02

SiO2 40·9 41·6 41·9 41·9 41·4 41·6 41·4 41·2 41·5 41·6 41·2 41·1

TiO2 0·17 0·23 0·17 0·03 0·19 0·03 0·18 0·86 0·61 0·89 0·02 0·03

Al2O3 20·8 22·1 19·8 21·7 22·2 22·0 22·2 21·5 20·8 21·1 22·6 23·5

Cr2O3 3·08 1·64 4·17 2·43 1·65 1·83 1·58 0·62 2·03 1·49 0·87 0·40

MgO 20·9 21·3 21·2 20·3 21·6 20·2 21·4 20·0 21·6 20·8 19·3 21·9

CaO 5·40 4·57 5·33 5·75 4·69 5·64 4·66 5·07 5·11 5·16 5·01 4·91

MnO 0·44 0·27 0·28 0·40 0·32 0·46 0·27 0·29 0·30 0·25 0·44 0·34

FeO 7·06 7·72 7·20 7·73 8·00 7·99 7·68 10·19 7·73 8·68 9·65 6·99

Na2O �0·04 �0·04 �0·04 �0·04 0·05 �0·04 �0·04 0·08 0·06 0·07 0·05 �0·04

K2O �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02

Total 98·8 99·5 100·2 100·3 100·0 99·8 99·5 99·9 99·8 100·1 99·2 99·3

P 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00

Si 2·96 2·97 3·00 2·98 2·95 2·98 2·96 2·97 2·97 2·98 2·98 2·93

Ti 0·01 0·01 0·01 0·00 0·01 0·00 0·01 0·05 0·03 0·05 0·00 0·00

Al 1·77 1·86 1·67 1·82 1·86 1·85 1·87 1·82 1·76 1·78 1·92 1·98

Cr 0·18 0·09 0·24 0·14 0·09 0·10 0·09 0·04 0·12 0·08 0·05 0·02

Mg 2·25 2·27 2·26 2·16 2·29 2·16 2·28 2·15 2·30 2·21 2·08 2·33

Ca 0·42 0·35 0·41 0·44 0·36 0·43 0·36 0·39 0·39 0·39 0·39 0·37

Mn 0·03 0·02 0·02 0·02 0·02 0·03 0·02 0·02 0·02 0·02 0·03 0·02

Fe2þ 0·43 0·46 0·43 0·46 0·48 0·48 0·46 0·61 0·46 0·52 0·58 0·42

Na 0·00 0·01 0·01 0·00 0·01 0·00 0·01 0·01 0·01 0·01 0·01 0·00

K 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00

[O] 12 12 12 12 12 12 12 12 12 12 12 12

Mg-no. 84·0 83·1 84·0 82·4 82·8 81·9 83·3 77·8 83·3 81·0 78·1 84·8

ppm

Na 176 252 196 95 249 73 265 487 366 431 128 67

Sc 158 103 107 161 95 187 95 103 107 99 114 186

Ti 686 1583 1379 109 1538 140 1715 5972 4289 6287 561 277

Ni 32 56 67 32 59 25 58 77 122 109 22 28

Ga 3·3 7·5 7·5 2·4 8·2 2·3 7·8 11 9·1 9·4 3·8 2·6

Sr 0·13 0·03 0·40 0·26 0·20 0·09 0·07 0·43 0·54 0·58 0·07 0·08

Y 45 18 13 14 19 32 19 41 32 41 15 36

Zr 142 4·9 80 53 4·7 58 10 92 53 94 51 17

Nb 0·15 0·19 0·43 0·22 0·15 0·16 0·18 0·10 0·15 0·19 0·08 0·11

La 0·01 0·01 0·04 0·03 0·01 0·01 0·01 0·04 0·04 0·05 b.d. 0·01

Ce 0·12 0·04 0·48 0·24 0·03 0·15 0·06 0·33 0·36 0·44 0·09 0·12

Pr 0·07 b.d. 0·19 0·11 0·01 0·08 0·02 0·14 0·13 0·17 0·04 0·07

Nd 0·87 0·13 1·95 1·37 0·12 1·18 0·27 1·53 1·44 1·94 0·55 0·87

Sm 1·24 0·28 1·54 1·23 0·23 1·46 0·43 1·52 1·21 1·76 0·78 0·62

Eu 0·72 0·20 0·67 0·58 0·20 0·74 0·23 0·76 0·57 0·85 0·46 0·28

Gd 3·74 1·15 2·43 2·38 1·11 2·98 1·25 3·77 2·86 3·77 2·12 1·46

Tb 0·88 0·30 0·38 0·44 0·29 0·63 0·30 0·83 0·59 0·82 0·41 0·37

Dy 7·64 2·81 2·50 2·86 2·79 5·27 2·92 7·07 5·07 6·86 3·07 4·08

Ho 1·75 0·68 0·51 0·48 0·69 1·25 0·71 1·60 1·21 1·66 0·60 1·36

Er 5·41 2·27 1·49 1·26 2·31 4·07 2·29 5·12 3·80 4·86 1·51 5·56

Tm 0·83 0·34 0·25 0·19 0·37 0·67 0·34 0·76 0·57 0·74 0·23 1·04

Yb 5·88 2·42 1·78 1·60 2·73 5·38 2·71 5·08 4·14 5·34 1·50 8·86

Lu 0·92 0·40 0·28 0·28 0·45 0·93 0·41 0·77 0·64 0·78 0·26 1·54

Hf 2·02 0·19 1·52 0·61 0·18 0·44 0·24 2·32 1·38 2·37 0·58 0·16

Major and minor element concentrations are from EPMA; trace element data are from LA-ICP-MS. B.d., below detectionlimit.


1973

at University of A



ownloaded from


chromium garnets, which plot to much higher Cr-numbers(Fig. 8b^d). A minor offset from the Cr-number trend wasalso found for the high-titanium peridotitic garnets(Fig. 8c and d).It is notable that xenoliths containing minerals with low

Cr-number do not contain chromite. The presence ofchromite in the studied sample set is restricted to assem-blages with a minimum Cr-number of �0·19 for cpx,�0·11 for opx, and �0· 05 for garnet. At higherCr-numbers, chromite-bearing and chromite-free xeno-liths seem to coexist. However, it is important to keep inmind that chromite may have been part of the sourcelithologies of the seemingly chromite-free xenoliths, but itis not represented because of the relatively small size ofthe xenoliths.The correlation in Cr-number between minerals in the

Monk Hill xenoliths, and the lack of compositionalzoning, suggests that the xenoliths are well equilibrated,and that the Cr-number of each mineral is constrained bythe bulk composition of its protolith and by the P^Tcondi-tions during equilibration. The lack of chromite in xeno-liths that contain minerals with low Cr-numbers indicatesthat chromium in these xenoliths is incorporated (i.e.

buffered) primarily by the silicates garnet, clinopyroxene,and orthopyroxene. Consequently, chromite is presentonly in xenoliths with overall high-Cr-number assem-blages. Even though olivine is not present in the MonkHill xenoliths suite, it is unlikely to have a significant influ-ence on the geochemical behaviour of chromium and alu-minium in the peridotites.Based on experimental evidence, the Cr-number of coex-

isting garnet and spinel (chromite) in peridotites dependsstrongly on the equilibration pressure, and in the case ofgarnet, to some extent on the temperature (Doroshevet al., 1997; Girnis & Brey, 1999). This means that theCr-numbers for single minerals within a xenolith are ex-pected to increase with increasing pressure (i.e. depth oforigin). This increase, however, can continue only as longas chromite is present in the assemblage. Once chromite isexhausted, the Cr-number of the remaining silicate min-erals is unlikely to change significantly with increasingpressure. This means that the Cr-number of minerals inchromite-bearing assemblages should reflect the equilib-rium P^T conditions, whereas the Cr-number of mineralsin chromite-free assemblages is constrained only by theCr-number of the source rock (protolith).

Fig. 5. Zr vsY for garnet xenocrysts from the Monk Hill kimberlite. Compositional fields from Griffin et al. (1999).


1974

at University of A



ownloaded from


Table 3: Composition of selected clinopyroxene xenocrysts

Sample: 24 17 16 98 17 22 23 91 21 92 13 16

wt %

P2O5 �0·02 �0·02 �0·02 �0·02 0·03 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02

SiO2 53·6 53·4 53·7 53·8 54·4 52·9 53·5 53·9 53·5 54·1 54·2 53·8

TiO2 �0·02 0·04 0·06 0·07 0·07 0·12 0·13 0·15 0·20 0·27 0·37 0·41

Al2O3 1·85 2·93 2·17 2·00 1·85 2·33 2·30 2·01 2·46 0·59 2·79 1·68

Cr2O3 0·86 0·87 1·01 0·83 0·95 1·16 1·18 0·88 1·17 1·25 0·64 1·52

MgO 17·3 16·6 17·1 17·2 16·4 16·7 16·9 16·5 16·3 17·5 19·7 16·3

CaO 23·4 22·9 22·6 22·3 22·6 22·2 22·4 21·4 21·3 21·8 15·7 19·3

MnO 0·05 0·09 0·03 0·10 0·10 0·10 0·09 0·09 0·09 0·13 0·12 0·08

FeO 2·02 1·81 2·14 2·11 2·13 2·11 2·21 2·87 2·11 2·44 4·50 4·01

Na2O 0·83 1·11 1·01 0·97 0·85 1·25 1·15 1·25 1·36 0·99 1·56 1·68

K2O �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 �0·02 0·03 �0·02

Total 99·8 99·8 99·8 99·4 99·4 98·8 99·9 99·2 98·6 99·1 99·7 98·8

P 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00

Si 3·90 3·88 3·90 3·92 3·96 3·89 3·89 3·94 3·93 3·97 3·91 3·96

Ti 0·00 0·00 0·00 0·00 0·00 0·01 0·01 0·01 0·01 0·02 0·02 0·02

Al 0·16 0·25 0·19 0·17 0·16 0·20 0·20 0·17 0·21 0·05 0·24 0·15

Cr 0·05 0·05 0·06 0·05 0·05 0·07 0·07 0·05 0·07 0·07 0·04 0·09

Mg 1·87 1·80 1·85 1·87 1·78 1·83 1·83 1·80 1·79 1·91 2·12 1·78

Ca 1·83 1·78 1·76 1·74 1·76 1·75 1·75 1·68 1·68 1·71 1·22 1·52

Mn 0·00 0·01 0·00 0·01 0·01 0·01 0·01 0·01 0·01 0·01 0·01 0·01

Fe2þ 0·12 0·11 0·13 0·13 0·13 0·13 0·13 0·18 0·13 0·15 0·27 0·25

Na 0·12 0·16 0·14 0·14 0·12 0·18 0·16 0·18 0·19 0·14 0·22 0·24

K 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00 0·00

[O] 12 12 12 12 12 12 12 12 12 12 12 12

Mg 93·9 94·2 93·4 93·6 93·2 93·4 93·1 91·1 93·2 92·7 88·7 87·9

ppm

Sc 56 108 46 82 63 49 31 57 121 71 18 36

Ti 111 252 472 337 347 632 527 1187 1066 1492 2323 2343

Mn 568 525 507 572 642 531 490 688 618 596 857 939

Ni 294 313 296 294 311 294 306 286 314 256 540 338

Ga 1·29 1·72 1·93 1·63 1·88 2·21 2·14 3·89 1·69 1·87 4·59 7·08

Sr 125 121 157 144 310 137 102 219 140 140 103 116

Y 4·01 4·59 2·87 2·53 2·47 1·86 1·00 3·36 2·85 2·17 4·33 4·06

Zr 28 17 49 35 25 55 29 47 61 30 9·3 34

Nb 0·54 1·50 1·63 0·80 0·44 0·50 0·48 0·46 0·45 0·12 0·16 0·20

Ba 0·0 0·2 9·0 b.d. 0·0 b.d. 0·0 0·1 0·0 0·1 0·3 0·2

La 7·76 15·24 3·45 12·1 7·93 3·13 5·28 4·05 2·35 2·11 1·94 2·00

Ce 21·8 29·3 11·1 32·7 24·9 11·3 13·6 13·4 10·2 7·06 6·11 7·91

Pr 3·49 3·51 2·10 4·36 3·61 2·05 2·24 2·36 1·91 1·27 1·11 1·39

Nd 16·1 13·6 11·1 17·2 14·9 10·2 10·6 11·7 9·62 6·48 5·99 7·37

Sm 3·17 2·15 2·55 2·66 2·64 2·13 1·82 2·74 2·33 1·55 1·61 1·92

Eu 0·84 0·48 0·74 0·70 0·72 0·59 0·44 0·80 0·66 0·46 0·49 0·63

Gd 2·09 1·58 1·94 1·73 1·66 1·52 1·03 2·15 1·73 1·21 1·53 1·84

Tb 0·25 0·18 0·22 0·19 0·18 0·15 0·09 0·23 0·20 0·14 0·20 0·23

Dy 1·17 0·98 0·96 0·79 0·82 0·63 0·30 1·03 0·91 0·66 1·11 1·22

Ho 0·17 0·19 0·12 0·11 0·11 0·08 0·04 0·14 0·14 0·11 0·18 0·18

Er 0·38 0·46 0·25 0·25 0·24 0·16 0·09 0·29 0·25 0·19 0·44 0·35

Tm 0·05 0·07 0·03 0·02 0·02 0·02 0·01 0·03 0·03 0·02 0·05 0·04

Yb 0·28 0·52 0·14 0·14 0·12 0·07 0·04 0·17 0·17 0·12 0·24 0·19

Lu 0·05 0·09 0·01 0·02 0·02 0·01 b.d. 0·02 0·02 0·02 0·03 0·02

Hf 0·53 0·26 1·85 1·44 0·47 2·48 0·61 1·39 2·59 1·81 0·50 1·75

Ta 0·11 0·24 0·17 0·11 0·10 0·12 0·10 0·10 0·11 b.d. 0·01 0·02

Major and minor element concentrations are from EPMA; trace element data are from LA-ICP-MS.


1975

at University of A



ownloaded from


The Cr-number of the coexisting minerals in the MonkHill xenoliths as a function of pressure is illustrated inFig. 8. Despite some scatter, it is notable that theCr-numbers in orthopyroxene and clinopyroxene inchromite-bearing xenoliths show a linear correlation withpressure (Fig. 9a and b). Chromites, as expected, alsoshow a strong correlation between Cr-number and pres-sure (Fig. 9c). This correlation supports the notion ofGirnis & Brey (1999) that the Cr-number of chromites canbe used as an independent geobarometer (e.g. for isolatedchromite grains in heavy mineral concentrates from kim-berlites). The Cr-numbers of garnets in chromite-bearingxenoliths show a moderate pressure dependence at pres-sures53GPa. At pressures43GPa, the Cr-number of gar-nets increases dramatically compared with the Cr-numberof the other minerals in the assemblage (Fig. 9). This in-crease is related to the higher solubility of the knorringite(Mg3Cr2Si4O12) component in the garnet structure athigh pressures (Irifune et al., 1982). If chromite is not pre-sent in the assemblage, garnet retains a much lowerCr-number, which plots onto the equilibration lines out-lined in Fig. 8b^d, even at very high pressures.

Sm^Nd isotopesThe garnet and clinopyroxene megacrysts selected for iso-topic analysis have very different 147Sm/144Nd ratios(garnet �0·5; cpx �0·16) and produce a statisticallyrobust isochron with an age of 189�17 Ma (Fig. 10), an ini-tial 143Nd/144Nd ratio of 0·512581 (�0· 000041), and an

eNd value of 3·6 (Table 3). This value is significantlylower than the expected value for the depleted mantle inthe Jurassic, which would be around eNd¼ 8·7 (Goldsteinet al., 1984), and indicates that the lithospheric mantle hasexperienced geochemical enrichment. Because thebulk-rock compositions for the peridotite xenoliths andmegacrysts are not known, it is not possible to accuratelycalculate a depleted mantle model age. However, assumingthe 147Sm/144Nd ratio of the enriched mantle lies betweenthat of the clinopyroxenes (0·164) and that of typical kim-berlites in this region (�0· 085; Foden et al., 2002), the en-riched lithospheric mantle has a depleted mantle modelage in the range 0·51^1·11 Ga. This time interval includesthe period of the break-up of Rodinia and the rifted pas-sive margin history of this part of Gondwana (Jenkins,1990; Foden et al., 2002).In addition to the Sm^Nd isotope analyses, equilibration

temperatures for the clinopyroxene megacrysts were calcu-lated using the single-cpx geothermobarometer of Nimis& Taylor (2000) (Table 2). The high temperatures(412008C) and pressures (44·2GPa) obtained suggest thatthe megacrysts were derived from the deepest parts of thelithospheric mantle.

DISCUSS IONWell-preserved peridotite samples from the lithosphericmantle beneath the Australian craton are extremely rare.For this reason the garnet peridotite xenoliths and

Fig. 6. Chondrite-normalized rare earth element patterns of clinopyroxene xenocrysts from the Monk Hill kimberlite.


1976

at University of A



ownloaded from


Fig. 7. P^Testimates for the Monk Hill garnet peridotite xenoliths using various geothermometer and geobarometer combinations. (SeeTable 2for abbreviations.) The diagonal lines are for comparison purposes only.


1977

at University of A



ownloaded from


xenocrysts from the Monk Hill kimberlite provide valuableand unique insights into the thermal structure and thecomposition of the lithospheric mantle beneath theAdelaide Fold Belt and the margin of the Australiancraton. The sample set covers a significant depth rangefrom the beginning of the garnet stability field in

peridotites, at depths of �50 km, to the base of the litho-sphere, which almost reaches the diamond stability fieldat depths of �160^180 km (Fig. 11). The high abundance oflherzolitic xenoliths at Monk Hill is consistent with previ-ous observations on isolated xenocryst minerals, whichwere recovered from heavy mineral concentrates of South

Table 4: Calculated temperatures (8C) and pressures (Gpa) for the Monk Hill garnet peridotites, using various geotherm-

ometer^geobarometer combinations

Xenolith ID Paragenesis Type TBKN PBKN TTay PNG TKro PBKN THar PBKN TNT PNT

@ PBKN @ TBKN @ PNG @ TTay @ PBKN @ TKro @ PBKN @ THar @ PNT @ TNT

B-X1-MXL1 grt–cpx–opx lherz. 835 2·25 832 2·45 861 2·38 895 2·55 828 2·91

C-X1-MXL1 grt–cpx–opx lherz. 856 2·38 839 2·46 958 2·93 926 2·75 834 2·66

C-X1-MXL2 grt–cpx–opx–chr lherz. 724 1·64 723 1·81 829 2·14 894 2·45 721 2·43





C-X2-MXL1 grt–cpx–opx high-Ti 1327 4·84 1270 4·50 1470 5·70 1346 4·96 1269 4·48






















C-X5-MXL2 grt–cpx–opx low-Cr 842 2·38 829 2·45 901 2·70 930 2·85 823 2·67

C-X5-MXL3 grt–opx high-Ti n.a. n.a. n.a. n.a. n.a. n.a. 1261 4·30 n.a. n.a.


C-X6-MXL1 grt–cpx–opx–chr wehrl. 637 1·12 645 1·54 735 1·67 901 2·61 641 2·15


C-X7-MXL1 grt–opx lherz. n.a. n.a. n.a. n.a. n.a. n.a. 1115 4·08 n.a. n.a.

(continued)


1978

at University of A



ownloaded from


Australian kimberlites (Gaul et al., 2003). Xenoliths andxenocrysts indicate that the lithospheric mantle beneaththe Adelaide Fold Belt is predominantly composed of fertile(i.e. clinopyroxene-saturated) peridotite. The separation

of the Monk Hill xenolith suite into low-T (510008C) andhigh-T (410008C) groups is consistent with observationson peridotite xenoliths from other cratons, where the low-T group is commonly represented by coarse-grained

Table 4: Continued

Xenolith ID Paragenesis Type TBKN PBKN TTay PNG TKro PBKN THar PBKN TNT PNT

@ PBKN @ TBKN @ PNG @ TTay @ PBKN @ TKro @ PBKN @ THar @ PNT @ TNT





D-X1-MXL1 grt–cpx–opx lherz. 826 2·26 794 2·34 923 2·76 918 2·73 786 2·70


D-X1-MXL11 grt–cpx–opx–chr lherz. 868 2·52 827 2·55 981 3·12 933 2·86 819 2·74
































BKN, Brey et al. (1990); Tay, Taylor (1998); NG, Nickel & Green (1985); Har, Harley (1984); Kro, Krogh (1988); NT, Nimis& Taylor (2000). n.a., not applicable.


1979

at University of A



ownloaded from


peridotite xenoliths and the high-T group is linked torecrystallized, finer grained, sheared peridotites (Carswell& Gibb, 1987; Boyd et al., 1997, 2004). Because of the lackof olivine and the small size of the xenoliths, such texturalinferences cannot be made for the Monk Hill sample suite.

The deepest part of the lithospheric mantle beneath theMonk Hill kimberlite is represented by titanium-enrichedsamples, which include a small group of high-titaniumperidotitic xenoliths and the megacrysts (Fig. 11). The en-riched nature of these samples indicates that the base ofthe lithosphere has been affected by metasomatism. Thehigh-temperature origin and the compositional character-istics of the megacrysts from Monk Hill are consistentwith observations on megacryst suites from other cratonsand support the notion that megacrysts generally formclose to the base of the lithosphere (Schulze, 1987; Bell &Moore, 2004).The geothermal gradient derived from the Monk Hill

xenoliths is significantly lower than the typical geothermalgradient calculated for the lithospheric mantle beneathsoutheastern Australia (i.e. theTasman Fold Belt; O’Reilly& Griffin, 1985; Fig. 11). This means that the geotherm re-flects the cratonic or marginally cratonic setting of theMonk Hill kimberlite. The Monk Hill geotherm, however,is still considerably higher than geotherms beneath typicalArchean cratonic nuclei, which generally follow geother-mal gradients with projected surface heat flow values inthe range 38^42 mW m�2 (Pollack & Chapman, 1977;Fig. 11). The P^T data from the garnet-peridotites indi-cate that the base of the lithosphere lies just marginallyoutside the diamond stability field (Fig. 11), which is con-sistent with the absence of diamonds in the Monk Hillkimberlites. This observation, however, also suggests thata slight increase in lithosphere thickness or a slightlylower geothermal gradient can account for the presence oflithospheric diamonds in the Eurelia kimberlites, whichare located only �100 km NWof Monk Hill.Previously proposed geotherms for the lithosphere be-

neath the Adelaide Fold Belt differ greatly from thegeotherm presented here. A much higher geothermal gra-dient has been determined for the eastern margin of theAustralian craton by Pearson et al. (1991) based on P^Testi-mates from lower crustal xenoliths. These xenoliths includesamples from the Calcutteroo and Pine Creek kimberlites,which are located in close proximity (�10 km) to theMonk Hill kimberlite, and, therefore, should reflect thesame thermal conditions as those presented here (Fig. 1).Temperatures along this proposed geotherm, however, are�2508C higher compared with the Monk Hill geotherm(Fig. 11). A reason for this severe discrepancy may lie inthe fact that Pearson et al. (1991) combined P^T data forlower crustal xenolith samples from kimberlites in SouthAustralia with data for samples from Kayrunnera in NewSouth Wales, which is located �400 km NE of the SouthAustralian kimberlite field (Fig. 1). More important thanthe mere distance between the sites is the fact that the vol-canic rocks at Kayrunnera are not located on theAustralian craton, but rather they are located on or justeast of the Tasman Line (O’Reilly & Griffin, 1985; Direen

Table 5: Composition and pressure estimates for the garnet

and clinopyroxene megacrysts from Monk Hill

Sample: Sm–Nd isotope samples

GRT-1 GRT-2 CPX-3 CPX-4

P2O5 0·03 0·03 �0·02 �0·02

SiO2 41·5 41·0 53·5 53·9

TiO2 0·76 0·88 0·43 0·34

Al2O3 21·3 21·3 2·82 2·79

Cr2O3 2·25 1·09 0·37 0·71

MgO 21·0 20·4 19·0 20·0

CaO 5·33 5·11 16·9 15·9

MnO 0·22 0·31 0·15 0·09

FeO 8·13 9·86 5·11 4·64

NiO 0·04 �0·02 0·08 0·07

Na2O 0·09 0·07 1·45 1·50

K2O �0·02 �0·02 0·03 0·04

Total 100·5 100·2 99·9 100·0

P 0·00 0·00 0·00 0·00

Si 2·95 2·95 3·88 3·88

Ti 0·04 0·05 0·02 0·02

Al 1·79 1·81 0·24 0·24

Cr 0·13 0·06 0·02 0·04

Mg 2·23 2·19 2·06 2·15

Ca 0·41 0·39 1·31 1·23

Mn 0·01 0·02 0·01 0·01

Fe2þ 0·48 0·59 0·31 0·28

Ni 0·00 0·00 0·00 0·00

Na 0·01 0·01 0·20 0·21

K 0·00 0·00 0·00 0·00

Total 8·05 8·07 8·07 8·07

[O] 12 12 12 12

Mg 0·82 0·79 0·87 0·88

Nd (ppm) 1·82 1·52 6·98 5·63

Sm (ppm) 1·50 1·32 1·89 1·47

Sm147/Nd144 0·4990 0·5246 0·1639 0·1576

Nd143/Nd144 0·513190 0·513239 0·512781 0·512780

eNd 10·8 11·73 2·79 2·77

Nd143/Nd144i 0·512573 0·512590 0·512578 0·512585

eNdi 3·48 3·81 3·58 3·71

tDM (Ma) 22 44 1111 990

TNT (8C) – – 1234 1291

PNT (GPa) – – 4·27 4·61


1980

at University of A



ownloaded from


& Crawford, 2003; Fig. 1). The Kayrunnera volcanic rocksalso have older, Permian, emplacement ages comparedwith the Jurassic kimberlites in South Australia (Gleadow& Edwards,1978; Stracke et al.,1979). In addition, their dis-tinct chemistry indicates that they are alkaline basaltsrather than kimberlites (Ferguson & Sheraton, 1979). Thelower crustal xenoliths from Kayrunnera, therefore, mayreflect a significantly different, and probably a muchhotter geothermal gradient compared with the samplesfrom the South Australian kimberlite field. Given the

lower pressure and hence shallower origin of the xenolithsanalysed by Pearson et al. (1991) it is also possible thatthese were not equilibrated with the conductive geothermat the time of their eruption.Additional geothermal gradients for the lithospheric

mantle beneath the Adelaide Fold Belt have been deter-mined by Gaul et al. (2003), using kimberlite-derived peri-dotitic garnet xenocrysts. The P^T data, in this case, arebased on the semi-empirical nickel in garnet (TNi)geothermometer (Griffin et al., 1989; Ryan et al., 1996) in

Fig. 8. Correlation in Cr# of the mineral components in the Monk Hill garnet peridotite xenoliths. Grey boxes mark fields of chromite-freeassemblages.


1981

at University of A



ownloaded from


combination with the PCr geothermometer of Ryan et al.(1996). Although the garnet xenocrysts for that study werealso recovered from kimberlites in close proximity toMonk Hill, the results indicate a significantly coolergeotherm compared with the Monk Hill data and contrasteven more with the crustal xenolith-based geotherm ofPearson et al. (1991) (Fig. 11). A geothermal gradient of�42 mW m�2 surface heat flow was determined for sam-ples from various kimberlites, which were combined underthe location name Burra, and include the aforementionedCalcutteroo and Pine Creek kimberlites, as well as

kimberlites at Mittopita, Terowie, Wanna Gorge, andMungibbie. An even lower geothermal gradient of 39 mWm�2 surface heat flow was proposed for the lithosphericmantle sampled by the diamondiferous Eurelia kimberlites(Orroroo; Gaul et al., 2003; Fig. 1). This deviation from ourresults may be due to the fact that only a few data pointswere used to define the geothermal gradients, resulting ina high level of uncertainty. It has also been noted else-where that the approach of usingTNi in combination withPCr on garnet xenocrysts underestimates the geothermalgradient (Gru« tter et al., 2006).

Fig. 9. Cr# vs equilibration pressure (PBKN at TBKN) for orthopyroxene, clinopyroxene, chromite and garnet from the Monk Hill garnetperidotite xenoliths. Dotted lines mark the approximate limits of chromite saturation. Filled symbols are for spinel-bearing xenoliths.


1982

at University of A



ownloaded from


Fig. 10. Sm^Nd isochron for garnet and clinopyroxene megacrysts from the Monk Hill kimberlite (seeTable 3). Data-point error ellipses are 2s.

Fig. 11. P^Tarray for the garnet peridotite xenoliths from Monk Hill kimberlite, based on PBKN^TBKN (see Fig. 8). The dashed line throughthe data marks the palaeo-geothermal gradient for the lithospheric mantle beneath Monk Hill. The geothermal gradient for southeasternAustralia and previously proposed geothermal gradients for the eastern margin of the Australian craton and Burra are shown for comparison.The garnet^spinel transition for lherzolite compositions is taken from O’Neill (1981). Fine dotted lines are model conductive geotherms for vari-ous surface heat flow values, based on Pollack & Chapman (1977). The diamond^graphite transition is adopted from Kennedy & Kennedy(1976). A mantle adiabat is shown for a surface (potetial) temperature of 13508C.


1983

at University of A



ownloaded from


Although the xenoliths from Monk Hill represent alarge part of the lithospheric mantle, they do not provideinformation about the shallower (550 km) parts of thelithosphere. The results from previous studies on xenolithsin South Australian kimberlites, however, indicate thatlower crustal rocks, which are dominated by eclogites andmafic granulites, may extend to depths of more than50 km (Ferguson et al., 1979; Pearson et al., 1991). Becausegarnet-bearing peridotites extend to relatively shallowdepths, it is unlikely that garnet-free, spinel-bearing peri-dotites are a widespread constituent of the lithosphericmantle in this region.

Age of the geothermçkimberliteemplacementThe emplacement age of the Monk Hill kimberlite hasbeen determined as 189·5�2·4 Ma using U^Pb datingtechniques on groundmass perovskite (Tappert, R., et al.unpublished data). Similar Jurassic ages have beenreported for some of the other kimberlites in SouthAustralia (Stracke et al., 1979; Scott Smith et al., 1984;Wyatt et al., 1994). The fact that the Sm^Nd isotope com-positions of the garnet and clinopyroxene megacrysts pro-duce an isochron with an age (189�17 Ma) identical tothe age of the host kimberlite suggests that the megacrystswere in isotopic equilibrium at the time of the kimberliteemplacement, and were above the closure temperature forSm^Nd in garnet and clinopyroxene. Sm^Nd mineral iso-chrons that reflect the eruption age of the host kimberliteare not uncommon, and have previously been observedfor mineral separates from peridotite xenoliths from othercratonic regions (Richardson et al., 1985; Pearson et al.,1995; Pearson, 1999). Considering that the garnet andclinopyroxene megacrysts were sampled as random singlecrystals (i.e. without any paragenetic constraints), thissuggests that large parts of the deeper lithospheric mantlebeneath Monk Hill were in isotopic equilibrium at thetime of kimberlite emplacement.

CONCLUSIONSGarnet peridotite xenoliths from the Monk Hill kimberlitereflect the composition and the paleo-geothermal gradientof the lithospheric mantle beneath the Adelaide FoldBelt at the eastern margin of the Australian craton.The geothermal gradient derived from multiple mineralequilibria is well constrained and extends from �1·2GPaand 6508C almost to the graphite^diamond transition at�5·0GPa and 13008C, indicating a lithosphere thicknessof 160^180 km. The resulting geotherm is consistent withthe transitional setting of the Adelaide Fold Belt betweenthe Archean Gawler Block to the west and the non-cratonicTasman Fold Belt to the east. The presence and absence ofchromite in the mineral assemblages accounts for much ofthe compositional variability (particularly in chromium)

within the predominantly lherzolitic xenolith suite.Additional enrichment, mainly in titanium and iron, is re-stricted to the high-temperature (410008C) group of xeno-liths and to the macrocrysts (i.e. to samples from the baseof the lithosphere). Sm^Nd data for clinopyroxene andgarnet megacrysts define an isochron consistent with theJurassic (189 Ma) emplacement age of the Monk Hillkimberlite.

ACKNOWLEDGEMENTSThe assistance of David Bruce with the Sm^Nd isotopeanalyses is gratefully acknowledged. Reviews from PaoloNimis and two anonymous reviewers helped to improvean earlier version of this paper.

FUNDINGThis work was supported through a linkage grant(LP0667689) by the Australian Research Council (ARC),the Department of Primary Industries and ResourcesSouth Australia (PIRSA), and Flinders Mines Ltd.

REFERENCESBell, D. R. & Moore, R. O. (2004). Deep chemical structure of the

southern African mantle from kimberlite megacrysts. South AfricanJournal of Geology 107, 59^80.

Boyd, F. R., Pokhilenko, N. P., Pearson, D. G., Mertzman, S. A.,Sobolev, N.V. & Finger, L.W. (1997). Composition of the Siberiancratonic mantle: evidence from Udachnaya peridotite xenoliths.Contributions to Mineralogy and Petrology 128, 228^246.

Boyd, F. R., Pearson, D. G., Hoal, K. O., Hoal, B. G., Nixon, P. H.,Kingston, M. J. & Mertzman, S. A. (2004). Garnet lherzolitesfrom Louwrensia, Namibia: bulk composition and P/T relations.Lithos 77, 573^592.

Brey, G. P. & Ko« hler, T. (1990). Geothermobarometry in four-phaselherzolites II. New thermobarometers, and practical assessment ofexisting thermobarometers. Journal of Petrology 31, 1353^1378.

Brey, G. P., Ko« hler,T. & Nickel, K. G. (1990). Geothermobarometry infour-phase lherzolites I. Experimental results from 10 to 60 kb.Journal of Petrology 31, 1313^1352.

Carswell, D. A. & Gibb, F. G. F. (1987). Garnet lherzolite xenoliths inthe kimberlites of northern Lesothoçrevised P^T equilibrationconditions and upper mantle paleogeotherm. Contributions to

Mineralogy and Petrology 97, 473^487.Dawson, J. B. & Stephens,W. E. (1975). Statistical classification of gar-

nets from kimberlite and associated xenoliths. Journal of Geology 83,589^607.

Direen, N. G. & Crawford, A. J. (2003). TheTasman Line: where is it,what is it, and is it Australia’s Rodinian breakup boundary?AustralianJournal of Earth Sciences 50, 491^502.

Doroshev, A. M., Brey, G. P., Girnis, A. V., Turkin, A. I. &Kogarko, L. N. (1997). Pyrope^knorringite garnets in the Earth’smantle: experiments in the MgO^Al2O3^SiO2^Cr2O3 system.Geologiya i Geofizika 38, 523^545.

Elburg, M. A., Bons, P. D., Foden, J. & Brugger, J. (2003). A newlydefined Late Ordovician magmatic^thermal event in the Mt.Painter Province, northern Flinders Ranges, South Australia.AustralianJournal of Earth Sciences 50, 611^631.


1984

at University of A



ownloaded from


Ferguson, J. & Sheraton, J.W. (1979). Petrogenesis of kimberlitic rocksand associated xenoliths of southeastern Australia. In: Boyd, F. R.& Meyer, H. O. A. (eds) Kimberlites, Diatremes, and Diamonds: their

Geology, Petrology and Geochemistry. Proceedings of the 2nd International

Kimberlite Conference. Washington, DC: American GeophysicalUnion, pp. 140^160.

Ferguson, J., Ellis, D. J. & England, R. N. (1977). Unique spinel^garnet lherzolite inclusion in kimberlite from Australia. Geology 5,278^280.

Ferguson, J., Arculus, R. J. & Joyce, J. (1979). Kimberlite and kimber-litic intrusives of southeastern Australia: a review. BMR Journal of

Australian Geology and Geophysics 4, 227^241.Foden, J., Song, S. H., Turner, S., Elburg, M., Smith, P. B., Van der

Steldt, B. & Van Penglis, D. (2002). Geochemical evolution of litho-spheric mantle beneath S.E. South Australia. Chemical Geology 182,663^695.

Foden, J., Elburg, M., Dougherty-Page, J. & Burtt, A. (2006). Thetiming and duration of the Delamerian orogeny: correlation withthe Ross Orogen and implications for Gondwana assembly.Journal of Geology 114, 189^210.

Gaul, O. F., O’Reilly, S.Y. & Griffin,W. L. (2003). Lithosphere struc-ture and evolution in southeastern Australia. In: Hillis, R. R. &Mu« ller, D. (eds) Evolution and Dynamics of the Australian Plate.

Geological Society of Australia, Special Publication 22, and Geological

Society of America, Special Papers 372, 185^202.Girnis, A.V. & Brey, G. P. (1999). Garnet^spinelôlivineôrthopyrox-

ene equilibria in the FeO^MgOÂl2O3^SiO2^Cr2O3 system: II.Thermodynamic analysis. EuropeanJournal of Mineralogy 11, 619^636.

Gleadow, A. J.W. & Edwards, A. C. (1978). Mineralogical note: fissiontrack age of a basic inclusion from the Kayrunnera kimberliticbreccia pipe. Journal of the Geological Society of Australia 25, 359.

Goldstein, S. L., O’Nions, R. K. & Hamilton, P. J. (1984). A Sm^Ndisotopic study of atmospheric dusts and particulates from majorriver systems. Earth and Planetary Science Letters 70, 221^236.

Gregoire, M., Bell, D. R. & Le Roex, A. (2003). Garnet lherzolitesfrom the Kaapvaal craton (South Africa): trace element evidencefor a metasomatic history. Journal of Petrology 44, 629^657.

Griffin,W. L., Cousens, D. R., Ryan, C. G., Sie, S. H. & Suter, G. F.(1989). Ni in chrome pyrope garnets: a new geothermometer.Contributions to Mineralogy and Petrology 103, 199^202.

Griffin, W. L., Shee, S. R., Ryan, C. G., Win, T. T. & Wyatt, B. A.(1999). Harzburgite to lherzolite and back again: metasomatic pro-cesses in ultramafic xenoliths from the Wesselton kimberlite,Kimberley, South Africa. Contributions to Mineralogy and Petrology

134, 232^250.Gru« tter, H. S., Gurney, J. J., Menzies, A. H. & Winter, F. (2004). An

updated classification scheme for mantle-derived garnet, for useby diamond explorers. Lithos 77, 841^857.

Gru« tter, H., Latti, D. & Menzies, A. (2006). Cr-saturation arrays inconcentrate garnet compositions from kimberlite and their use inmantle barometry. Journal of Petrology 47, 801^820.

Gurney, J. J. & Switzer, G. S. (1973). The discovery of garnets closelyrelated to diamonds in the Finsch pipe, South Africa. Contributionsto Mineralogy and Petrology 39, 103^116.

Harley, S. L. (1984). An experimental study of the partitioning of ironand magnesium between garnet and orthopyroxene. Contributionsto Mineralogy and Petrology 86, 359^373.

Irifune,T., Ohtani, E. & Kumazawa, M. (1982). Stability field of knor-ringite Mg3Cr2Si3O12 at high-pressure and its implication to theoccurrence of Cr-rich pyrope in the upper mantle. Physics of theEarth and Planetary Interiors 27, 263^272.

Jenkins, R. J. F. (1990). The Adelaide Fold Belt: tectonic reappraisal.In: Jago, J. B. & Moore, P. S. (eds) The Evolution of a Late

PrecambrianÊarly Palaeozoic Rift Complex: the Adelaide Geosyncline.

Geological Society of Australia, Special Publication 16, 395^420.Kennedy, C. S. & Kennedy, G. C. (1976). The equilibrium boundary

between graphite and diamond. Journal of Geophysical Research 81,2467^2470.

Krogh, E. (1988). The garnet^clinopyroxene iron^magnesiumgeothermometerça reinterpretation of existing experimentaldata. Contributions to Mineralogy and Petrology 99, 44^48.

McDonough,W. F. & Sun, S. S. (1995). The composition of the Earth.Chemical Geology 120, 223^253.

Nickel, K. G. & Green, D. H. (1985). Empirical geothermobarometryfor garnet peridotites and implications for the nature of the litho-sphere, kimberlites and diamonds. Earth and Planetary Science Letters73, 158^170.

Nimis, P. & Taylor, W. R. (2000). Single clinopyroxene geothermo-barometry for peridotites. Part I. Calibration and testing of a Crin cpx barometer and enstatite in cpx thermometer. Contributions toMineralogy and Petrology 139, 541^554.

O’Neill, H. S. C. (1981). The transition between spinel lherzolite andgarnet lherzolite, and its use as a geobarometer. Contributions to

Mineralogy and Petrology 77, 185^195.O’Reilly, S.Y. & Griffin,W. L. (1985). A xenolith-derived geotherm for

southeastern Australia and its geophysical implications.Tectonophysics 111, 41^63.

Pearson, D. G. (1999). The age of continental roots. Lithos 48, 171^194.Pearson, N. J. & O’Reilly, S. Y. (1991). Thermobarometry and P^T^t

pathsçthe granulite to eclogite transition in lower crustal xeno-liths from eastern Australia. Journal of Metamorphic Geology 9,349^359.

Pearson, N. J., O’Reilly, S.Y. & Griffin,W. L. (1991). The granulite toeclogite transition beneath the eastern margin of the Australiancraton. EuropeanJournal of Mineralogy 3, 293^322.

Pearson, D. G., Carlson, R.W., Shirey, S. B., Boyd, F. R. & Nixon, P.H. (1995). The stabilization of Archean lithospheric mantleçaReÔs isotope study of peridotite xenoliths from the Kaapvaalcraton. Earth and Planetary Science Letters 134, 341^357.

Plumb, A. (1979). The tectonic evolution of Australia. Earth-Science

Reviews 14, 205^249.Pollack, H. N. & Chapman, D. S. (1977). Mantle heat flow. Earth and

Planetary Science Letters 34, 174^184.Richardson, S. H., Erlank, A. J. & Hart, S. R. (1985).

Kimberlite-borne garnet peridotite xenoliths from old enrichedsubcontinental lithosphere. Earth and Planetary Science Letters 75,116^128.

Ryan, C. G., Griffin, W. L. & Pearson, N. J. (1996). Garnetgeothermsçpressure^temperature data from Cr-pyrope garnetxenocrysts in volcanic rocks. Journal of Geophysical Research 101,5611^5625.

Schulze, D. J. (1987). Megacrysts from alkaline volcanic rocks. In:Nixon, P. H. (ed.) Mantle Xenoliths. Chichester: John Wiley,pp. 433^451.

Schulze, D. J. (2003). A classification scheme for mantle-derivedgarnet in kimberlite: a tool for investigating the mantle and explor-ing for diamonds. Lithos 71, 195^213.

Scott Smith, B. H., Danchin, R. V., Harris, J. W. & Stracke, K. J.(1984). Kimberlites near Orroroo, South Australia. In:Kornprobst, J. (ed.) Kimberlites I: Kimberlites and Related Rocks.Amsterdam: Elsevier, pp. 121^142.

Smith, D. (1999). Temperatures and pressures of mineral equilibrationin peridotite xenoliths: review, discussion, and implications. In:Fei, Y., Bertka, C. M. & Mysen, B. O. (eds) Mantle Petrology: Field

Observations and High Pressure Experimentation: a Tribute to Francis R.

(Joe) Boyd. Geochemical Society, Special Publication 6, 171^188.


1985

at University of A



ownloaded from


Sobolev, N. V., Lavrentev, Y. G., Pokhilenko, N. P. & Usova, L. V.(1973). Chrome-rich garnets from the kimberlites of Yakutia andtheir paragenesis. Contributions to Mineralogy and Petrology 40, 39^52.

Stracke, K. J., Ferguson, J. & Black, L. P. (1979). Structural setting ofkimberlites in southeastern Australia. In: Boyd, F. R. & Meyer, H.O. A. (eds) Kimberlites, Diatremes, and Diamonds. Their Geology,

Petrology and Geochemistry. Proceedings of the 2nd International Kimberlite

Conference. Washington, DC: American Geophysical Union,pp. 71^91.

Sutherland, F. L. & Hollis, J. D. (1982). Mantle^lower crust petrologyfrom inclusions in basaltic rocks in eastern Australiaçan outline.Journal of Volcanology and Geothermal Research 14, 1^29.

Tappert, R., Foden, J., Wills, K. & Goryniuk, M. C. (2007).Geochemical variability within the lithospheric mantle beneaththe Adelaide Fold Belt, South Australia. Geochimica et Cosmochimica

Acta 71, A1004.Tappert, R., Foden, J., Stachel, T., Muehlenbachs, K., Tappert, M. &

Wills, K. (2009). The diamonds of South Australia. Lithos

112(Supplement 1), 806^821.

Taylor, W. R. (1998). An experimental test of some geothermometerand geobarometer formulations for upper mantle peridotites withapplication to the thermobarometry of fertile lherzolite and garnetwebsterite. NeuesJahrbuch fu« r Mineralogie, Abhandlungen 172, 381^408.

Veevers, J. J. (1984). Phanerozoic Earth History of Australia. Oxford:Clarendon Press.

Viljoen, K. S., Dobbe, R., Smit, B., Thomassot, E. & Cartigny, P.(2004). Petrology and geochemistry of a diamondiferous lherzolitefrom the Premier diamond mine, South Africa. Lithos 77, 539^552.

Wyatt, B. A., Shee, S. R., Griffin,W. L., Zweistra, P. & Robison, H. R.(1994). The petrology of the Cleve kimberlite, Eyre peninsula,South Australia. In: Meyer, H. O. A. & Leonardos, O. H. (eds)Kimberlites, Related Rocks and Mantle Xenoliths. Proceedings of the 5th

International Kimberlite Conference, Volume 1. Companhia de Pesquisa de

Recursos Minerais, Special Publication 1/A, 62^79.


1986

at University of A



ownloaded from


garnet peridotite xenoliths and xenocrysts from the monk hill kimberlite, south australia: insights...

Documents