Journal of Petrology

This Article Abstract FREE Full Text (PDF) Supplementary data Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (20) Request Permissions Google Scholar Articles by KOPYLOVA, M. G. Articles by CARO, G. Search for Related Content GeoRef GeoRef Citation Social Bookmarking          What's this?

Journal of Petrology | Volume 45 | Number 5 | Pages 1045-1067 | 2004
Journal of Petrology 45(5) © Oxford University Press 2004; all rights reserved.

Mantle Xenoliths from the Southeastern Slave Craton: Evidence for Chemical Zonation in a Thick, Cold Lithosphere

M. G. KOPYLOVA^* and G. CARO

EARTH AND OCEAN SCIENCES DEPARTMENT, THE UNIVERSITY OF BRITISH COLUMBIA, 6339 STORES ROAD, VANCOUVER, B.C., CANADA V6T 1Z4

RECEIVED JANUARY 28, 2003; ACCEPTED OCTOBER 28, 2003

	ABSTRACT

TOP ABSTRACT INTRODUCTION HOST KIMBERLITE SAMPLES PETROGRAPHY ANALYTICAL METHODS MINERAL COMPOSITIONS GEOTHERMOBAROMETRY DISCUSSION CONCLUSIONS SUPPLEMENTARY DATA REFERENCES

 
We present the first data on the petrology of the mantle lithosphere of the Southeastern (SE) Slave craton, Canada. These are based on petrographic, mineralogical and geochemical studies of mantle xenoliths in Pipe 5034 of the Cambrian Gahcho Kué kimberlite cluster. Major types of mantle xenoliths include altered eclogite, coarse garnet or spinel peridotite, and deformed garnet peridotite. The peridotites belong to the low-temperature suite and formed at T=600–1300°C and P= 25–80 kbar in a thick (at least 220–250 km), cool lithosphere. The SE Slave mantle is cooler than the mantle of other Archaean cratons and that below other terranes of the Slave craton. The thick lithosphere and the relatively cool thermal regime provide favourable conditions for formation and preservation of diamonds beneath the SE Slave terrane. Similar to average Archaean mantle worldwide, the SE Slave peridotite is depleted in magmaphile major elements and contains olivine with forsterite content of 91–93·5. With respect to olivine composition and mode, all terranes of the Slave mantle show broadly similar compositions and are relatively orthopyroxene-poor compared with those of the Kaapvaal and Siberian cratons. The SE Slave spinel peridotite is poorer in Al, Ca and Fe, and richer in Mg than deeper garnet peridotite. The greater chemical depletion of the shallow upper mantle is typical of all terranes of the Slave craton and may be common for the subcontinental lithospheric peridotitic mantle in general. Peridotitic xenoliths of the SE Slave craton were impregnated by kimberlitic fluids that caused late-stage recrystallization of primary clinopyroxene, spinel, olivine and spinel-facies orthopyroxene, and formation of interstitial clinopyroxene. This kimberlite-related recrystallization depleted primary pyroxenes and spinel in Al. The kimberlitic fluid was oxidizing, Ti-, Fe- and K-rich, and Na-poor, and introduced serpentine, chlorite, phlogopite and spinel into peridotites at P < 35 kbar.

KEY WORDS: kimberlite xenolith; lithosphere; mantle terrane; chemical zoning; thermobarometry; Slave craton

	INTRODUCTION

TOP ABSTRACT INTRODUCTION HOST KIMBERLITE SAMPLES PETROGRAPHY ANALYTICAL METHODS MINERAL COMPOSITIONS GEOTHERMOBAROMETRY DISCUSSION CONCLUSIONS SUPPLEMENTARY DATA REFERENCES

 
Recent studies of the Slave craton (Northwest Territories, Canada) show the underlying lithosphere to be heterogeneous, and to consist of distinct petrological domains (Griffin et al., 1999b; Grütter et al., 1999; Carbno & Canil, 2002; Davis et al., 2003). Three compositionally distinct, NE-trending lithospheric domains have been defined (Grütter et al., 1999); the southern terrane is divided further into western and eastern parts with contrasting compositions (Carbno & Canil, 2002). This paper presents data on the composition and thermal regime of the Southeastern (SE) Slave lithosphere based on a study of mantle xenoliths in the Gahcho Kué kimberlite cluster (Fig. 1).

View larger version (63K): [in this window] [in a new window]   Fig. 1. Schematic map of the Slave craton (Northwest Territories, Canada; see inset), showing the location of kimberlite pipes (•). Double lines designate the boundaries between northern, central and southern lithospheric domains as distinguished by distinct compositions of garnet in kimberlite concentrates (Grütter et al., 1999). The SW and SE Slave terranes may be separated by the Pb isotopic boundary of Thorpe et al. (1992) (thin continuous line). The darker area is the postulated surface and subsurface extent of the Central Slave Basement Complex (protocraton of Ketchum & Bleeker, 2001). The thick continuous curves show the orientation of pre-2·63 Ga D1 fold structures (Davis et al., 2003); double dashed line traces the northern boundary of the 2·63–2·62 Ga diorite with the granodiorite plutons of the Defeat Suite (Davis et al., 2003). Orientation of the S-wave polarization (bars with arrows) indicates the directions of anisotropy of the mantle (Bank et al., 2000). The spatial location of the conductive mantle anomaly (Jones et al., 2001) is contoured by a thin dashed line. Also shown are the minimum extents of a shallow ultra-depleted layer (horizontal lined pattern) and of the deeper Archaean lherzolitic layer (Griffin et al., 1999b) of the Central Slave (dashed outlines). The thick continuous straight line indicates the cross-section of Fig. 13.

 
The Slave craton has now been studied with sufficient thoroughness to recognize the blocky, quilt-like structure of its underlying mantle. The mantle domains differ from one another in their architecture, bulk composition, conductivity and seismic anisotropy. Within the Slave craton, the northern domain has the most undepleted, lherzolitic composition of the garnet-bearing mantle; harzburgitic (G10) garnet is extremely rare as a xenocrystic component in the Jericho kimberlite and in the Coronation district diamondiferous kimberlite in the far northwestern Slave craton (Grütter et al., 1999; Armstrong, 2002; Davis et al., 2003). Other characteristics common for the northern domain mantle are a high proportion of mantle eclogite (Grütter et al., 1999), and the WSW–ENE directions of mantle anisotropy (Fig. 1) recorded by S-wave polarization (Bank et al., 2000). The central domain stands out because it contains a layer of anomalously conductive mantle at 80–100 km (Jones et al., 2001), and an ultra-depleted layer of harzburgitic composition down to 100–150 km depths (Griffin et al., 1999b). Furthermore, the Central Slave has a significantly lower proportion of eclogitic garnet xenocrysts (Grütter et al., 1999) and a distinct direction of mantle anisotropy (Fig. 1). The SE Slave mantle is dominantly lherzolitic with subordinate eclogite; harzburgitic garnet xenocrysts are not ultra-depleted and show moderate to high Cr₂O₃ contents (Grütter et al., 1999). The southwestern domain is defined on the basis of the geochemistry of garnet xenocrysts in the Dry Bones kimberlite. The domain is ultra-depleted harzburgitic at shallow depths and is underlain by moderately depleted mantle that lacks high-Cr harzburgitic garnet (Carbno & Canil, 2002). Both southern domains are characterized by the WSW–ENE directions of mantle anisotropy (Fig. 1), and relatively low conductivity, which is typical of the lithospheric mantle (Jones et al., 2001).

The boundaries between the mantle domains are poorly constrained and should be revised as more data become available. Some data points used for the present constraints may be anomalous. An example of one such anomalous point may be the Jericho pipe, which is characterized by a thinned crust (35 km vs 37–39 km in other Northern Slave localities; Bank et al., 2000) and its proximity to the Proterozoic Kilihigok basin. The tentative NW–SE divisions between mantle domains match well with the orientation of some other mantle and crustal structures of the craton (Fig. 1). The domain boundaries are roughly parallel to the direction of mantle anisotropy (Bank et al., 2000), to the orientation of an early (pre-2·63 Ga) fold belt in the central and southern Slave and to the northern boundary of the 2·62–2·63 Ga diorites– granodiorite intrusives (Davis et al., 2003). These boundaries, however, cut through the mainly north– south older crustal divisions of the craton (Fig. 1). Major crustal boundaries delineate a 4·03–2·85 Ga composite crustal block of the Slave protocraton (Ketchum & Bleeker, 2001), which is further subdivided along a north– south boundary defined by bedrock geology and Pb isotopes into an eastern younger part and a western older part. It is a deep continuation of this ‘Pb line’ that may separate the SW and SE Slave terranes (Carbno & Canil, 2002).

Our study focuses on the peridotitic upper mantle of the SE Slave and reports its rock types and bulk composition. The data presented in this paper leave no doubt as to the complex character of the SE Slave upper mantle as reflected in its chemical zoning. Our data also reveal that the SE Slave mantle has a relatively cool, thick lithosphere that is favourable for the preservation of diamonds.

	HOST KIMBERLITE

TOP ABSTRACT INTRODUCTION HOST KIMBERLITE SAMPLES PETROGRAPHY ANALYTICAL METHODS MINERAL COMPOSITIONS GEOTHERMOBAROMETRY DISCUSSION CONCLUSIONS SUPPLEMENTARY DATA REFERENCES

 
The Gahcho Kué kimberlite cluster (63°30'N, 109°30'W) is located 75 km south of Aylmer Lake and 275 km ENE of Yellowknife in the southeastern corner of the Slave Craton (Fig. 1). The cluster consists of six diamondiferous kimberlite bodies: the 5034-Kennady, Tesla, Tuzo, Hearn, Wallace and 5034-South. The kimberlites were emplaced into voluminous Archaean granitic rocks that intrude older metavolcanic and metasedimentary rocks of the Yellowknife supergroup. The 5034-Kennady pipe has an irregular shape with surface dimensions of about 120 m by 180 m, and a 35 m wide dyke-like body extends from the pipe some 300 m. The overall surface area of the intrusion is 2·1 ha, and the majority of it is overlain by the shallow water of Kennady Lake. The bulk of this multiphase kimberlite body is hypabyssal macrocrystal, and it contains abundant inclusions of the surrounding granitic rock (Rikhotso et al., 2003). The hypabyssal rock comprises a macrocrystal monticellite-bearing phlogopite serpentine kimberlite and an aphanitic monticellite serpentine kimberlite, both classified as Type IA. The presence of abundant phlogopite, diopside and pectolite in the kimberlite is related to its contamination by country-rock granitoids (Kopylova et al., 2001; Caro et al., 2004).

The 5034-Kennady kimberlite has been dated radiometrically by the Rb–Sr method on phlogopite as Middle Cambrian, at 539 Ma (Heaman et al., 1997). A Cambrian age of kimberlite emplacement is also reported for the Snap Lake kimberlite (522–535 Ma, Agashev et al., 2001), located 100 km WNW of the Gahcho Kué, and thus may be common to all kimberlites of the SE Slave craton. In contrast, in the SW part of the Slave craton, kimberlites were emplaced later, in the Late Ordovician, well before the Late Cretaceous–Eocene epochs of kimberlite formation of the Central Slave craton (Heaman et al., 1997).

	SAMPLES

TOP ABSTRACT INTRODUCTION HOST KIMBERLITE SAMPLES PETROGRAPHY ANALYTICAL METHODS MINERAL COMPOSITIONS GEOTHERMOBAROMETRY DISCUSSION CONCLUSIONS SUPPLEMENTARY DATA REFERENCES

 
Mantle xenoliths in the Gahcho Kué cluster are common only in the hypabyssal macrocrystal kimberlite of the 5034 pipe. They are absent from other magmatic phases of the 5034 kimberlite and are altered in other kimberlite intrusions of the Gahcho Kué cluster. This study is based mainly on xenolith samples recovered from the 1995–1997 large-diameter cores drilled by Canamera Ltd (an operator for Mountain Province Ltd) for bulk sampling in the diamond exploration programme. Three peridotite xenolith samples from the small-diameter drill core of the Hearn pipe were also contributed by De Beers. Selective sampling of mantle and lower-crustal xenoliths during logging provided comprehensive collection of 85 xenoliths. Among them, 52 larger (4–10 cm in diameter) and fresher samples were studied in thin section. These samples, apart from the predominant peridotites, include lower-crustal granulites and gneisses, amphibolites (six thin sections) and altered eclogites (eight thin sections). In the eclogites, all primary minerals, with the exception of garnet (20–30%) and rutile (2%), are completely altered to serpentine, chlorite and phlogopite (Fig. 2d). Secondary mineral grain boundaries indicate that the size of the original clinopyroxene was 0·4–0·6 mm. Garnet grains are large and round, and contain inclusions of needle-like rutile and round chlorite pseudomorphs. The mineralogical and textural characteristics of these rocks are typical of eclogite xenoliths found in the Northern Slave kimberlites (Kopylova et al., 1999b). This fact, together with the absence of lower-crustal eclogites in the Slave craton, suggests a mantle origin for the eclogite from the 5034 pipe. Overall proportions of mantle xenoliths in our collection are 61% coarse peridotite, 18% eclogite, 17% deformed peridotite and 4% orthopyroxenite. A total of 38 peridotite xenoliths were selected for the detailed study.

View larger version (152K): [in this window] [in a new window]   Fig. 2. Microphotographs showing textures of the SE Slave xenoliths. Mineral abbreviations here and below are from Kretz (1983). The scale bar represents 1 mm unless specified otherwise. (a) Coarse–equant texture of a spinel harzburgite xenolith. (b) Mosaic-porphyroclastic texture of a garnet peridotite xenolith. (Note bands of pyroxene neoblasts that define texture as laminated.) Garnets can be intact, non-disrupted, or disrupted and completely kelyphitized. (c) Coarse–equant texture of a garnet harzburgite xenolith. (d) Altered eclogite. Fresh grains of garnet and rutile are set in finer-grained serpentine and chlorite with laths of euhedral phlogopite. (e) Recrystallization of spinel into skeletal microlites is observed only on the side adjacent to a secondary interstitial clinopyroxene. (f) Complete recrystallization of spinel into smaller skeletal microlites inside a phlogopite-rich area of peridotite.

 

	PETROGRAPHY

TOP ABSTRACT INTRODUCTION HOST KIMBERLITE SAMPLES PETROGRAPHY ANALYTICAL METHODS MINERAL COMPOSITIONS GEOTHERMOBAROMETRY DISCUSSION CONCLUSIONS SUPPLEMENTARY DATA REFERENCES

 
Xenoliths of coarse peridotite contain either garnet (69%) or spinel (31%). The absence of spinel–garnet peridotite indicates a gap in sampling of certain mantle depths by the kimberlite host.

Coarse-equant peridotites are typically composed of olivine, orthopyroxene and lesser amounts of clinopyroxene, garnet or spinel (Fig. 2a and c). Most (90%) of the coarse peridotite is harzburgite, containing less than 5 vol. % clinopyroxene; lherzolite, wehrlite and orthopyroxenite are also present. Olivine forms large subhedral grains that may poikilitically enclose subhedral garnet or orthopyroxene. Orthopyroxene is usually less altered than olivine and commonly occurs in olivine-free clusters of anhedral and euhedral grains, which occasionally show thin exsolution lamellae of clinopyroxene and kink banding. Clinopyroxene typically forms small anhedral crystals in interstices between olivine and orthopyroxene. Garnet is present in large round grains with thick kelyphitic rims that may replace 10–100% of the grains. Spinel occurs as small, interstitial, anhedral vermicular dark brown grains (Fig. 2a). Occasionally spinel forms symplectites with orthopyroxene or olivine or euhedral triangular grains. Spinel is also common as smaller, rhombic crystals within or around kelyphitic rims.

Deformed peridotites have porphyroclastic or mosaic-porphyroclastic textures (Harte, 1977). Porphyroclastic rocks are mostly garnet-bearing peridotites containing 10–80% neoblasts of olivine and orthopyroxene. In mosaic-porphyroclastic peridotite (Fig. 2b), olivine and orthopyroxene neoblasts typically make up more than 95 vol. % of the rock, whereas garnet, olivine and pyroxene porphyroclasts are scarce. Neoblasts in mosaic peridotites are markedly smaller than those in porphyroclastic rocks. Mosaic peridotite commonly exhibits a laminated texture, as neoblasts form thin lenses or bands defined by modal variations in olivine and pyroxenes. Bands formed predominantly of orthopyroxene are finer-grained than those in which olivine predominates. Garnet can be either intact and round, as in the non-disrupted texture of Harte (1977), or completely kelyphitized and disaggregated, as in disrupted peridotite.

Xenoliths carry evidence for alteration caused by percolation of kimberlitic fluids and melts. Many samples show veinlets 0·5–0·2 mm thick that extend from the kimberlite host, and are composed of serpentine, chlorite, phlogopite and fine euhedral spinel. These minerals look similar to the corresponding phases in the kimberlite and stand out in peridotitic rocks because of yellow phlogopite and green chlorite. In the most affected samples these kimberlite-related minerals make up no more than 0·1% of the total volume. Clinopyroxene, spinel and, to a lesser degree, orthopyroxene, in direct contact with or in the vicinity of the veinlets, are strongly altered or completely resorbed and replaced by the phlogopite–chlorite– spinel–serpentine aggregate (Fig. 3e). In severely altered pyroxene grains we observe complete recrystallization into aggregates of skeletal microlites with common optical orientations (Fig. 3a) or into new, larger clinopyroxene grains with numerous inclusions of euhedral spinel (Fig. 3f). Less altered samples have spongy rims of uneven thickness on clinopyroxene (Fig. 3b, d and e) that look similar to structures of partial melting and recrystallization (e.g. Dawson, 2002). Other clinopyroxene grains show no apparent signs of these processes but are surrounded and cut by phlogopite bands. Spinel also shows recrystallization into skeletal aggregates (Fig. 2e and f) that are often restricted to one side of a primary grain that is in contact with secondary interstitial minerals (Fig. 2e). This recrystallized spinel has a markedly darker black colour, compared with brown primary spinel, and can occur as larger triangular hollow crystals. Peridotites affected by visible penetration of kimberlite-related material often show interstitial films of secondary microlitic clinopyroxene along grain boundaries of primary minerals (Fig. 3g). These clinopyroxene rims are also observed in xenoliths with no apparent kimberlite veining and in completely serpentinized samples (Fig. 3h). The latter fact attests to a late origin of the rims, which would otherwise be destroyed by retrograde reactions.

View larger version (130K): [in this window] [in a new window]   Fig. 3. Microphotographs showing alteration and recrystallization of clinopyroxene in the SE Slave peridotite. The scale bar represents 1 mm unless specified otherwise. (a) A serpentine–chlorite vein with minor phlogopite and spinel reaches inside a xenolith from the kimberlite. Grains of orthopyroxene and clinopyroxene are recrystallized into fine-grained skeletal aggregates. (b) A concentric texture of crystallization of serpentine, chlorite and spinel in a pocket of kimberlite-related material in a xenolith. A vein of chlorite extends from the pocket to partially melted clinopyroxene. (Note areas of unmelted clinopyroxene on the extreme right and the uneven development of a zone of clinopyroxene melting and recrystallization.) (c) A darker vein of chloritization extends inside a xenolith from the kimberlite to a completely recrystallized clinopyroxene. (d) A large grain of clinopyroxene with an unevenly developed rim owing to secondary melting and recrystallization. (e) A large grain of clinopyroxene with an unevenly developed rim owing to secondary melting and recrystallization. The central part of the grain is resorbed and pseudomorphed by fine-grained phlogopite and spinel; a small island of clinopyroxene on the right has not been affected by the kimberlite-related melting. (f) Recrystallized clinopyroxene contains numerous inclusions of euhedral spinels and is rimmed by kimberlitic phlogopite and spinel. (g) A rim of microlitic secondary clinopyroxene between serpentinized primary olivine and orthopyroxene. (h) A mantle of secondary clinopyroxene marks former grain boundaries in a completely serpentinized coarse peridotite.

 

	ANALYTICAL METHODS

TOP ABSTRACT INTRODUCTION HOST KIMBERLITE SAMPLES PETROGRAPHY ANALYTICAL METHODS MINERAL COMPOSITIONS GEOTHERMOBAROMETRY DISCUSSION CONCLUSIONS SUPPLEMENTARY DATA REFERENCES

 
Minerals in the garnet peridotites were analysed using an automated CAMECA SX-50 microprobe (Department of Earth and Ocean Sciences, University of British Columbia, Canada) and were treated with the ‘PAP’ (Z) on-line correction program. Instrument calibration was performed on a set of electron microprobe (EMP) standards used for analyses of the Northern Slave xenoliths (Kopylova et al., 1999a) to ensure consistency with previous analytical and thermobarometric results. Silicates were analysed at an accelerating voltage of 15 mV and a 20 mA beam current. On-peak counting times were 10 s for major and 20 s for minor elements. Because some thermobarometers used in this study are heavily dependent on concentrations of Al and Ca in orthopyroxene, and the SE Slave orthopyroxenes are extremely depleted in these elements, we determined precision and minimum detection limits for Al and Ca from counting statistics and replicate analyses. Aluminium can be analysed at levels higher than 0·08 wt % with a precision (2) of ±0·06 wt %, and CaO can be analysed at levels higher than 0·02 wt % with a precision of ±0·07 wt %. Precision and minimum detection levels for other elements at these analytical conditions have been given by Kopylova et al. (1999a).

Individual phases in a sample were analysed as 8–15 points in cores and rims of 4–5 grains; phases used for thermobarometry were analysed at points of their mutual contact. Analyses with poor stoichiometry and totals were excluded and mineral compositions were averaged over 3–12 analyses for homogeneous phases or presented as individual analyses for inhomogeneous minerals (Electronic Appendix 1 at http://www.petrology.oupjournals.org.). Where possible, recrystallized pyroxenes and spinel were avoided.

For bulk chemical analysis of xenoliths, only larger (80–120 g) and fresher samples were selected. The samples were ground in a jaw crusher and then in a tungsten-carbide ring mill with minimum grinding times to preserve the original redox state of iron. Bulk compositions of the xenoliths were measured at McGill University Geochemical Laboratories (Montreal, Canada); all elements except ferrous iron were determined by X-ray fluorescence (XRF) spectrometry using a Philips PW2400 spectrometer on fused pellets. Ferrous iron was analysed volumetrically. Two of the larger (>200 g) peridotite xenoliths that belonged to different rock types were split into two 100 g aliquots that were crushed and then analysed separately to investigate sample heterogeneity. The resulting chemical variance was estimated to be ±0·3 wt % for SiO₂ and MgO, and insignificant for all other elements.

Modal mineral abundances were estimated by two methods. In the first method, point-counting was used for mineral modes in 34 thin sections (6–8 cm² in area) of the least altered xenoliths. This size of thin section approaches the minimum area of 7–8 cm² recommended for modal estimations in coarse peridotite to eliminate the overestimations of clustering garnet and pyroxene (Cox et al., 1987). In finer-grained deformed peridotite these areas are undoubtedly representative of the modal mineral abundances. Precision of the point-counting was assessed through replicate analyses of thin sections at 6% for olivine, garnet and spinel, and 3–5% for pyroxenes. The overall variance of mineral modes is also dependent on the intrinsic sample heterogeneity, which was estimated for the Northern Slave peridotites at 3–7 vol. % (Kopylova & Russell, 2000).

Mineral modes were also computed by converting the bulk composition of peridotite samples to an equivalent modal mineralogy using mass balance relationships (Kopylova & Russell, 2000). This conversion has the advantage of quantitative assessment of misfit through the root mean square (RMS) of chemical equations involved (Table 1). Seven samples permitted a comparison between calculated and observed modal mineralogy. Six of these showed an excellent fit, i.e. within 2–3 vol. % of mineral modes. Samples showing larger differences between calculated and observed modes are those having a high degree of misfit (RMS >0·4). For these samples, point-counting of the mode is more reliable. However, for specimens with good fit between the modes we consider the calculated modes to be more accurate as they average a larger volume (30–125 cm³) of sampled material. Our best estimates of mineral abundances, therefore, derive from either calculation or measurement, and are reported in Table 1. We are confident that the modes are representative of the bulk composition of peridotitic mantle below the Gahcho Kué kimberlite, as they are determined on large and fresh xenoliths by two independent methods that show good agreement.

View this table: [in this window] [in a new window]   Table 1: Modal mineral abundances of the SE Slave peridotites

 

	MINERAL COMPOSITIONS

TOP ABSTRACT INTRODUCTION HOST KIMBERLITE SAMPLES PETROGRAPHY ANALYTICAL METHODS MINERAL COMPOSITIONS GEOTHERMOBAROMETRY DISCUSSION CONCLUSIONS SUPPLEMENTARY DATA REFERENCES

 
Olivine
Olivine compositions range in mg-number [molar Mg/(Mg + Fe)] from 0·87 to 0·93. Olivine in spinel peridotite is more Mg-rich (mg-number = 0·925–0·935) on average than olivine in garnet peridotite, but there is no compositional difference between olivines from coarse and deformed garnet peridotites, with both having Fo_91–93 and 0·38 wt % NiO (Fig. 4a). Typically, olivine is homogeneous and porphyroclasts have compositions similar to neoblasts. In some samples (11-2, 2-7, 15-2, 3-5; Electronic Appendix 1) relatively Fe-rich olivine (mg-number = 0·87–0·90) is present as separate grains and in rims of magnesian olivine. In sample 3-5, Fe-rich olivine occurs in recrystallized and partially melted areas of primary Mg-rich olivine grains.

View larger version (21K): [in this window] [in a new window]   Fig. 4. Range of olivine (a) and orthopyroxene (b) compositions in peridotitic xenoliths of the SE Slave Craton. The range of olivine compositions for Archaean low-temperature peridotite is based on median analyses for low-temperature peridotite xenoliths entrained by volcanic rocks on Archaean cratons (Griffin et al., 1999c); a similar high-temperature range is based on average analyses and standard deviations for high-temperature xenoliths of the Kaapvaal, Udachnaya and Northern Slave cratons (Kopylova & Russell, 2000).

 
Orthopyroxene
In common with olivine, orthopyroxene in spinel peridotite is more Mg-rich on average than in garnet peridotite (Fig. 4b), and there is no distinguishable compositional variation between orthopyroxenes from coarse and deformed peridotites. Orthopyroxene in spinel peridotite, on average, has a markedly higher Al₂O₃ content (0·6–2·5 wt %) than orthopyroxene in garnet peridotite (0·25–0·45 wt %). Orthopyroxene grains appear unzoned. Two samples of spinel peridotite that are affected most by the kimberlite-related recrystallization (61-1 and 11-2) contain a homogeneous, relatively low-Al (0·6–1·0 wt % Al₂O₃) orthopyroxene. These samples show very intense recrystallization of pyroxenes into microlitic aggregates, interstitial secondary clinopyroxene, and skeletal, compositionally distinct, secondary spinel and zoned primary spinel. Complete recrystallization of orthopyroxene into secondary, unzoned, low-Al grains is observed only in spinel peridotite.

Clinopyroxene
Clinopyroxene is Cr-diopside, with Cr content ranging from 1 to 2·24 wt % Cr₂O₃ and mg-number = 0·91–0·95. The appearance, composition and patterns of heterogeneity of clinopyroxene correlate with petrographical evidence for kimberlite-related recrystallization. Samples with phlogopite–chlorite veining and interstitial films of clinopyroxene often show zoned clinopyroxene and intra-granular compositional variation (Table 2). All samples with heterogeneous clinopyroxene invariably display partially melted and recrystallized grains. These samples altered by kimberlite fluids contain two generations of clinopyroxene: primary and secondary. Secondary clinopyroxene is found (1) in recrystallized areas of primary grains, (2) as coarse grains intergrown with kimberlitic euhedral spinel or phlogopite, and (3) as microlitic rims along grain boundaries of primary minerals. Secondary clinopyroxene displays markedly distinct and variable compositions (Table 2 and Fig. 5) and is significantly lower in Al₂O₃ and higher in TiO₂ and CaO than primary clinopyroxene (Table 2). Complex grains produced by partial recrystallization of primary clinopyroxene are zoned from high-Al to low-Al secondary clinopyroxene compositions. In many cases this zoning is associated with a concomitant enrichment in CaO and MgO and depletion in Na₂O (Fig. 5). Parallel increases in TiO₂ may or may not be recorded. Chromium variations are uncorrelated with Al₂O₃.

View this table: [in this window] [in a new window]   Table 2: Petrography and compositional variance of pyroxenes and spinel in the SE Slave peridotite xenoliths

 

View larger version (16K): [in this window] [in a new window]   Fig. 5. Clinopyroxene compositions (Na2O vs Al2O3) in peridotitic xenoliths of the SE Slave craton. Open symbols denote compositions of primary clinopyroxene; filled symbols denote recrystallized clinopyroxene in partially melted coarse primary grains and in fine interstitial and fully recrystallized grains. The arrow illustrates the most common chemical evolution that accompanies recrystallization of clinopyroxene in individual samples.

 
Samples visibly altered by kimberlite fluids also contain unzoned clinopyroxene, including both primary high-Al and secondary low-Al clinopyroxene. The former was not apparently affected by highly localized kimberlite-induced recrystallization, whereas the latter was completely regrown (samples 37-1, AK61-1, 36-1, 38-2 in Table 2).

Garnet
Garnets are homogeneous in composition and can be classified as Cr-pyrope. On a CaO–Cr₂O₃ diagram (Fig. 6), garnet plots along the ‘lherzolitic’ trend, reflecting the presence of clinopyroxene in these harzburgites, although in low abundance. There is a strong positive correlation (K = 0·942) between TiO₂ and Na₂O in garnet. TiO₂ does not correlate with any other element in garnet, but correlates with TiO₂ and Na₂O in coexisting pyroxenes. Minor core-to-rim zoning in Cr₂O₃ (<0·3–0·5 wt % Cr₂O₃), Al₂O₃ and CaO is recorded in two samples, 59-1 and 3-2 (Electronic Appendix 1).

View larger version (16K): [in this window] [in a new window]   Fig. 6. Garnet compositions (Cr2O3 vs CaO) in peridotitic xenoliths of the SE Slave Craton. Fields for harzburgitic and lherzolitic garnet are from Gurney & Zweistra (1995).

 
Spinel
Three textural varieties of spinel occur in the peridotite: (1) coarse, primary, anhedral spinel; (2) finer skeletal spinel associated with kimberlite-related alteration; (3) fine-grained euhedral spinel in kelyphitic rims on garnet. All three types of spinel are compositionally distinct. Secondary spinels found in kelyphitic rims and in veins altered by kimberlitic fluid have markedly higher TiO₂ and Fe³⁺ contents. These spinels contain 0·2–2·2 wt % TiO₂ and 0·03–0·085 mol % Fe³⁺ as opposed to 0–0·05 wt % TiO₂ and 0–0·03 mol % Fe³⁺ in the primary spinel. The secondary spinels also stand apart on the Cr₂O₃–MgO plot, as kelyphitic spinel compositions extend to lower Cr₂O₃ and higher MgO contents (Fig. 7). Most primary spinel grains are lower in Cr₂O₃ than spinel equilibrated in the diamond stability field (Fig. 7). The three textural varieties of spinel differ also with respect to compositional zoning and heterogeneity. Intra-grain heterogeneity (for up to 10 wt % Cr₂O₃ and Al₂O₃) is common for kelyphitic spinels. Primary spinel, in contrast, is fairly homogeneous, and the only detectable zoning relates to a coupled Fe–Mg substitution (up to 1 wt % FeO and 2 wt % MgO). Recrystallization from primary to skeletal secondary spinel involves increases in MgO, TiO₂ and Fe³⁺ at the expense of Al₂O₃ and at constant Cr₂O₃ (Fig. 7).

View larger version (15K): [in this window] [in a new window]   Fig. 7. Spinel compositions (Cr2O3 vs MgO) in peridotitic xenoliths of the SE Slave Craton. The field for spinel found as diamond inclusions is from Gurney & Zweistra (1995). An arrow connects primary and skeletal spinel in sample AK61-1 and illustrates a chemical evolution as a result of the kimberlite-related recrystallization.

 

	GEOTHERMOBAROMETRY

TOP ABSTRACT INTRODUCTION HOST KIMBERLITE SAMPLES PETROGRAPHY ANALYTICAL METHODS MINERAL COMPOSITIONS GEOTHERMOBAROMETRY DISCUSSION CONCLUSIONS SUPPLEMENTARY DATA REFERENCES

 
Applying thermobarometry to variably recrystallized peridotite affected by interaction with the host kimberlite presents a major challenge. Analyses of recrystallized and secondary grains of pyroxene, olivine and spinel were not used for constraints on the steady-state geotherm but provide us with ‘temperatures and pressures of disequilibrium’ (Table 2 and Electronic Appendix 1 at http://www.petrology.oupjournals.org). Equilibrium pressure–temperature conditions (Table 3) were computed for representative primary mineral compositions, averaged over 3–6 analyses typically measured in mineral cores (Electronic Appendix 1). Because areas of recrystallization can be found in cores of some clinopyroxene grains and may not be easily discernible, we have investigated the impact of using recrystallized low-Al pyroxenes for the equilibrium P–T array. Temperatures calculated from these pyroxenes show slightly larger scattering in the P–T field than the inferred equilibrium P–T array. Minimum pressures and temperatures are displaced to lower values by 60–100°C and 8 kbar. Maximum pressure and temperature calculated using the Finnerty–Boyd solution are higher by 12 kbar and 100°C, but recrystallized and primary clinopyroxenes yield similar results using the Brey–Köhler thermobarometry. The position of the arrays for primary and recrystallized clinopyroxenes is similar.

View this table: [in this window] [in a new window]   Table 3: Equilibrium pressure and temperature estimates for the SE Slave peridotite xenoliths

 
Equilibrium P–T estimates for analysed garnet peridotites were calculated with the two-pyroxene and olivine– garnet (Ol–Gar) geothermometers recommended for cratonic peridotite (Finnerty & Boyd, 1987; Pearson et al., 1994). Table 3 lists pairs of temperature and pressure estimates made using: (1) the two-pyroxene geothermometer of Brey & Köhler (1990) (BK) and the Al-in-Opx geobarometer of Brey & Köhler (1990) (AlBK); (2) the geothermometer of Finnerty & Boyd (1987) (FB) and the geobarometer of MacGregor (1974) (MG); (3) the geothermometer of O'Neill & Wood (1979) (OW) and the Al-in-Opx geobarometer of Brey & Köhler (1990) (AlBK).

Our analysis shows that the OW Ol–Gar temperature estimates for clinopyroxene-free rocks cannot be used in a single P–T array with two-pyroxene temperature estimates. The OW temperature estimates at depth do not correlate with the FB geothermometer, and are up to 200°C lower than the BK temperatures (Table 3). These results duplicate the findings of Smith (1999), who reported that the OW temperatures match the BK temperatures to ±50° in only 17 out of 55 garnet peridotite xenoliths from the test suite (Smith, 1999, fig. 2B). This can be partly explained by known deficiencies in the BK formulation; at T >1100°C it yields values at least 50–100°C hotter than all widely used geothermometers (Smith, 1999). The OW formulation is also fraught with uncertainty associated with Fe³⁺ in garnet and may close at a different temperature than a two-pyroxene thermometer (Canil & O'Neill, 1996).

Figure 8 compares two resulting P–T arrays for the SE Slave garnet peridotite. The BK solution (Fig. 8b) is shifted towards higher P and T and is characterized by a larger scatter of temperature estimates at any given pressure.

View larger version (33K): [in this window] [in a new window]   Fig. 8. Equilibrium pressure and temperature estimates for the SE Slave mantle xenoliths in comparison with those from other Slave kimberlites. (a) is based on the thermometry of Finnerty & Boyd (1987) and the barometry of MacGregor (1974). (b) uses the formulation of Brey & Köhler (1990). Curves labelled 32, 39 and 48 indicate the spinel–garnet transition for peridotite containing olivine Fo92 and a characteristic SE Slave spinel (Cr2O3 = 32–48 wt %, XFe3+= 0·01) calculated according to O'Neill (1981). Striped P–T fields indicate equilibrium pressures and temperatures for spinel peridotite based on two-pyroxene temperatures for pressures from 20 kbar to a maximum pressure permitted for garnet-free spinel-bearing peridotite with the content of Cr2O3 found in spinel of samples 18-2, 38-4 and 6-4. P–T fields for the Central Slave are based on xenoliths in kimberlites DO18, DO-27 and A154S of Lac de Gras (line of longer dashes; Pearson et al., 1999) and on xenoliths in kimberlites Panda, Leslie, Grizzly, Sable and Pigeon of Lac de Gras (line of shorter dashes; Menzies et al., 2003). A P–T field for the Northern Slave (continuous outline) is based on the Jericho data (Kopylova et al., 1999a). Graphite–diamond equilibrium (indicated by G/D) is from Kennedy & Kennedy (1976).

 
Temperature estimates for spinel peridotite were calculated using the olivine–spinel thermometer (O'Neill & Wall, 1987) and by the BK and FB two-pyroxene thermometers (Tables 2 and 3). We used two-pyroxene temperatures for plotting spinel peridotite to maintain consistency with the P–T arrays for garnet peridotites. Temperatures were calculated for several pressure estimates (Table 3) between 20 and 30 kbar. The maximum pressure estimates are constrained by the absence of garnet in the rocks. Spinel with Cr and Fe³⁺ contents typical of the SE Slave mantle and equilibrated with Fo₉₃ would transform into garnet at P = 28–29 kbar. The minimum pressure estimates are constrained by the intersection of the average cratonic geotherm and Slave geotherms of Fig. 8 with the P–T univariant lines for the spinel peridotites. The striped P–T fields of Fig. 8 indicate our best estimates of equilibrium temperatures and pressures for spinel peridotite that are consistent with the cratonic geothermal regime and with the Cr₂O₃ content of the spinel.

Among spinel peridotites, only three out of seven samples contain high-Al pyroxenes untouched by fluid-assisted recrystallization associated with the kimberlite host; these three form the basis for P–T estimates (Table 3) used for the equilibrium P–T array. The remaining four spinel-bearing specimens contain either completely recrystallized homogeneous orthopyroxene, or clinopyroxene, or both. ‘Disequilibrium’ two-pyroxene temperatures calculated for these specimens extend to unrealistically high values of 1200°C, which are not recorded by the spinel–olivine thermometer (Table 2), and are clearly related to pyroxene compositions alone. The inferred petrogenesis of these rocks will be further discussed below.

Spinel peridotite was formed at 570–680°C and 20–29 kbar, according to the FB thermometer, and at 730–840°C and 24–28 kbar, according to the BK thermometer. A single porphyroclastic spinel peridotite equilibrated at a higher temperature than non-deformed spinel peridotite (Fig. 8). All garnet peridotite, coarse and sheared, formed at 950–1250°C and 55–72 kbar in a cold cratonic mantle deep within the diamond stability field. Mantle samples from intermediate depths of 100–170 km have not been identified in the Gahcho Kué kimberlites.

Chemical and modal composition of the SE Slave mantle in comparison with cratons worldwide
Petrographic evidence for kimberlite veining in the xenoliths warrants caution when interpreting bulk chemical compositions and modes as representative of the peridotitic mantle. To detect possible addition of kimberlitic material to xenoliths, we compared the analysed bulk chemical compositions with the compositions predicted by the modal mineral assemblage. This was done for a subset of samples with independent measurements of bulk chemical composition, mineral composition, and visually estimated mineral modes. Based on this subset of samples, MgO, FeO total, CaO, Al₂O₃, Cr₂O₃ and SiO₂ showed good agreement between calculated and analysed values. Measured contents of TiO₂ and K₂O are always higher (up to 0·13 wt % TiO₂ and 0·4 wt % K₂O) than the values predicted from the model modes and mineral compositions. Na₂O contents are consistently lower, by up to 0·13 wt %, than the Na₂O content forecast by primary mineralogy. We conclude that introduction of kimberlitic material was minor, and chemical and modal data on the rocks are representative of the mantle below the SE Slave. Table 4 lists bulk compositions of peridotite xenoliths by type, together with their average compositions and standard deviations. The bulk compositions of coarse and deformed garnet peridotite are statistically similar and, therefore, were averaged as ‘garnet peridotite’. The mean composition of the spinel peridotite differs from that of the deeper garnet peridotite in being poorer in Al₂O₃, CaO and FeO, and richer in MgO.

View this table: [in this window] [in a new window]   Table 4: Bulk chemical compositions (wt %), average modal compositions and standard deviations of the SE Slave peridotites

 
Modal mineral abundances in most groups of samples (Table 1) are characterized by a broad unimodal distribution. For these samples the averaged modes by rock type are given in Table 4. Several xenoliths (18-2, 1-3 and 41-1) stand out because of their low olivine content. They have high abundances of garnet (15–21 vol. %) and clinopyroxene (7–9 vol. %), and lower modes of olivine (65–68 vol. %) and orthopyroxene (6–9 vol. %).

The SE Slave peridotite compositions are fairly depleted and plot along the trend produced by basalt extraction at low pressure (Fig. 9). Spinel peridotites are markedly poorer in Ca and Al, typical of very depleted spinel peridotites found on other Archaean cratons, and plot close to extremely depleted alpine-type harzburgite. In contrast, the SE Slave garnet-bearing xenoliths contain more Ca and Al, and cluster around average abyssal oceanic peridotite, as do most other garnet peridotites worldwide (Fig. 9). The data points on this plot would move to the right of the oceanic array with an increase in modal clinopyroxene and the Ca/Al ratio. The clinopyroxene mode of 3 vol. % is typical of low-temperature garnet peridotites beneath most cratons, including the SE Slave. All of these cratons plot on the oceanic array. Low-temperature xenoliths from the Kaapvaal and Central Slave craton contain lower amounts of clinopyroxene (0·8–1·8 vol. %), have a lower Ca/Al value of 0·5–7 (Boyd, 1989; Griffin et al., 1999c) and are shifted to the right of the array (Fig. 9b).

View larger version (27K): [in this window] [in a new window]   Fig. 9. Bulk composition of peridotite xenoliths from the SE Slave (a) plotted as wt % Ca vs wt % Al in comparison with the available mean bulk compositions of xenoliths from other Archaean cratons (b). The thick dashed line with an arrow delineates the overall trend expected for the depletion of oceanic peridotite (Boyd, 1989). In (b) a field marked ‘K’ encompasses some statistical parameters of the Kaapvaal low-T garnet peridotite sample, i.e. average and median low-T garnet lherzolite and harzburgite (Boyd, 1989; Griffin et al., 1999c). A field marked ‘CS’ contains the average (Griffin et al., 1999a) and median (Griffin et al., 1999c) values of Central Slave low-T garnet peridotites. Other abbreviations: K, the Kaapvaal mean high-T peridotite and median spinel peridotite (Griffin et al., 1999c); F, the Fennoscandian averages for low-T spinel–garnet and garnet peridotite calculated from data of Peltonen et al. (1999); NS, the Northern Slave average (Kopylova & Russell, 2000) for low-T spinel- and garnet-bearing peridotite; SS, the SE Slave average for coarse spinel and garnet peridotite (Table 4, this work). Bulk compositions of Siberian peridotites are not plotted as they are modified by the late introduction of Ca (Boyd et al., 1997). The average compositions of abyssal peridotite and alpine-type harzburgite are from Boyd (1989).

 
Another useful elemental ratio for worldwide comparison is Mg/Si, which reflects pyroxene/olivine ratio. The Mg/Si value in the SE Slave peridotites (1·26–1·34) is comparable with these ratios in most Archaean peridotites worldwide, i.e. 1·27–1·32 for the Siberian, Fennoscandian and Northern Slave cratons. Extremes in the worldwide datasets are represented by the low-T garnet peridotites from the Kaapvaal (Mg/Si = 1·22; Boyd, 1989) and the Central Slave (Mg/Si = 1·37; Griffin et al., 1999a), which have exceptionally low and high olivine modes, respectively.

The SE Slave peridotites contain 60–88 vol. % olivine with mg-number = 0·89–0·93. Spinel peridotite has more magnesian olivine on average and is conspicuously more depleted than garnet peridotite (Fig. 10). The SE Slave peridotites match well the general range of these parameters for peridotites below Archaean cratons; the SE Slave spinel peridotites plot with Archaean harzburgites, and the SE Slave garnet peridotites resemble Archaean lherzolites (Fig. 10a).

View larger version (35K): [in this window] [in a new window]   Fig. 10. mg-number vs modal olivine (vol. %) for peridotites worldwide and peridotite xenoliths of the Slave craton. mg-number is for olivine where mineral chemical compositions are available or else represents values of bulk whole-rock composition. The latter are calculated assuming all iron as FeO and are generally equivalent to the mg-number of olivine in the same rock. A dashed arrow designates the ‘oceanic trend’ of Boyd (1989). Fields in (a) are for harzburgitic and lherzolitic xenoliths brought up by volcanic rocks on Archaean cratons, for lherzolitic xenoliths brought up by volcanic rocks on Proterozoic cratons, and for lherzolites in areas that experienced Phanerozoic tectonothermal events (Griffin et al., 1999c). Fields in (b) are for Northern Slave peridotites (continuous lines, Kopylova & Russell, 2000) and for the Kaapvaal and Siberian cratons (dashed lines, Boyd, 1989; Boyd et al., 1997, 1999). Open circles with crosses indicate median peridotite compositions of the upper (mg-number=92·8) and lower (mg-number=91·5) layers beneath the Central Slave (Griffin et al., 1999a).

 
Figure 10b contrasts compositions of SE Slave peridotite with corresponding available data for other Slave terranes, as well as the Kaapvaal and Siberian cratons. Data for the Central Slave are the most controversial, as they are derived from point-counts on small or highly altered samples (Griffin et al., 1999a; MacKenzie & Canil, 1999). However, if we accept averaged estimates for the Central Slave (Griffin et al., 1999c) as real, the three terranes of the Slave mantle show broadly similar compositions. Spinel peridotites of the SE Slave plot within the corresponding field for the Northern Slave. Coarse, low-temperature garnet peridotites of the northern, central and SE Slave are also similar. Deformed garnet peridotites (except one sample) of the SE Slave fit within the field for Slave low-temperature garnet peridotites. High-temperature garnet peridotite is present only in the northern terrane. Most peridotites from the Slave are markedly distinct from low-T peridotites from the Kaapvaal and Siberian cratons. Olivine-poor, orthopyroxene-rich, low-T peridotite is uncommon on the Slave craton, and compositions of its olivine extend to less magnesian values. These characteristics are in fact closer in composition to high-T peridotites from the Kaapvaal and Siberian cratons, which straddle Boyd's oceanic array (Fig. 10).

Our data show that modal olivine, garnet and clinopyroxene below the SE Slave are relatively constant with depth. Average orthopyroxene modes decrease slightly from 19 vol. % in spinel peridotite to 15 vol. % in garnet peridotite. This is commonly observed in the mantle and is explained by consumption of orthopyroxene to generate garnet.

	DISCUSSION

TOP ABSTRACT INTRODUCTION HOST KIMBERLITE SAMPLES PETROGRAPHY ANALYTICAL METHODS MINERAL COMPOSITIONS GEOTHERMOBAROMETRY DISCUSSION CONCLUSIONS SUPPLEMENTARY DATA REFERENCES

 
Petrology of the SE Slave peridotitic mantle
Two textural varieties of SE Slave garnet peridotite, coarse and deformed, are identical in mineral and bulk chemical compositions, and formed at similar equilibrium P–T conditions. The highly depleted mineral compositions of all SE Slave peridotites and their Mg-rich olivines and orthopyroxenes are typical of low-T suites of Archaean cratonic peridotite (Fig. 4a), which are interpreted as samples of lithosphere (Boyd, 1987, 1999; Harte & Hawkesworth, 1989); high-T suites of peridotite enriched in asthenospheric components (Fe, Na, Ti) are absent from the SE Slave. The presence of deformed lithospheric peridotite is relatively rare on cratons and is ascribed to deformation in shear zones within the lithosphere (Xu et al., 1993). The lithospheric peridotite of the SE Slave equilibrated at T = 950–1300°C, which is relatively high for a suite of low-T cratonic peridotites. However, similar equilibrium temperatures of low-T suites are also noted for the Central Slave (up to 1350°C; Menzies et al., 2003).

The harzburgites of the SE Slave contain garnet that plots in the lherzolitic field (Fig. 6), which is common in cratonic mantle (e.g. Griffin et al., 1999d; Peltonen et al., 1999). Equilibration with clinopyroxene is also typical for harzburgitic garnet in other parts of the Slave craton, as shown for the Northern (Kopylova et al., 1999a) and Central Slave mantle (MacKenzie & Canil, 1999; Pearson et al., 1999). The absence of high- and medium-Cr harzburgitic garnets from the mantle xenoliths of the Gahcho Kué pipe contrasts with their presence as xenocrysts in the SE Slave kimberlites, including Gahcho Kué (Grütter et al., 1999; Davis et al., 2003). We ascribe this to less representative sampling of the mantle by the xenoliths.

Figures 8 and 11 compare the thermal regime of the SE Slave mantle with other segments of cratonic mantle. Regardless of the geothermobarometric method used, SE Slave garnet peridotites plot at a slightly higher pressure and lower temperature than xenoliths from Northern and Central Slave kimberlites (Fig. 8). In accordance with this, orthopyroxene in garnet peridotites of the SE Slave is poorer in Al₂O₃ than orthopyroxene in similar mantle rocks below other parts of the Slave craton (Fig. 12). The SE Slave is also cooler than the mantle beneath other cratons, e.g. of Kaapvaal, Siberia, Fennoscandia, Tanzania, and Superior and other blocks of the North American craton (Fig. 11), and may be the coldest recognized mantle terrane at depths greater than 180 km.

View larger version (28K): [in this window] [in a new window]   Fig. 11. Pressure [orthopyroxene–garnet barometer of Brey & Köhler (1990)] and temperature [two-pyroxene thermometer of Brey & Köhler (1990)] estimates for the cratonic mantle. P–T fields for the Kaapvaal craton are based on thermobarometry of the Kimberley and Lesotho xenoliths (F. R. Boyd, unpublished data, 1990); for the Siberian craton on the Udachnaya (Boyd et al., 1997) and Mir xenoliths (Roden et al., 1999); for the Fennoscandian craton on the Eastern Finland kimberlites (Kukkonen & Peltonen, 1999; Lehtonen et al., 2003); for the Tanzanian craton on the Labait xenoliths (Lee & Rudnick, 1999); for the Superior craton on the Kirkland Lake xenoliths (Meyer et al., 1994); for the northern Canadian Shield on the Somerset Island kimberlites (Kjarsgaard & Petersen, 1992; Schmidberger & Francis, 1999). The Gahcho Kué xenoliths are shown as squares and diamonds (see Fig. 8 for symbols).

 

View larger version (19K): [in this window] [in a new window]   Fig. 12. A histogram of Al2O3 contents in orthopyroxenes of the Slave garnet peridotites. Data on the SE Slave are from this work; data for Central Slave are from the Lac de Gras pipes (Pearson et al., 1999); data for Northern Slave are from the Jericho pipe (Kopylova et al., 1999a). Analyses of low-Al2O3 recrystallized orthopyroxenes in the SE Slave peridotites were not used in the plot.

 
Our thermobarometric estimates together with data on peridotitic mineral chemistry indicate that the lithosphere of the SE Slave craton sampled by the Gahcho Kué kimberlites is at least 220–250 km deep. This minimum lithospheric thickness is greater than that of the Northern Slave (160–190 km, Kopylova et al., 1999b), the Central Slave (200–210 km, Pearson et al., 1999; Menzies et al., 2003) and the SW Slave (160–190 km, Carbno & Canil, 2002). A thicker, 225 km lithosphere was also reported for the SE Slave based on compositions of diamond inclusions in the Snap Lake kimberlite (Promprated et al., 2003), located 100 km WNW of the Gahcho Kué cluster. The thicker lithosphere and a cooler thermal regime control the larger depth interval where diamond is stable in the SE Slave mantle, i.e. define a larger ‘diamond window’ (Griffin & Ryan, 1995) for this terrane relative to other terranes of the Slave craton.

The apparent differences between the Slave mantle terranes may reflect the lateral heterogeneity of the craton or its temporal evolution. The combined cross-section of Fig. 13 compares a Jurassic time-slice of the Northern Slave with a Tertiary time-slice of the Central Slave, an Ordovician time-slice of the SW Slave and a Cambrian time-slice of the SE Slave. If the contrasting terranes reflect the craton's lateral heterogeneity, this may be rooted in their distinct mantle history. The age of the division between the SE and SW blocks of the craton may be Neoarchaean, as at that time the 2·69 Ga eastern block accreted to the older, Mesoarchaean, western part of the craton (Davis et al., 2003). The SE–NW divisions between other domains of the Slave mantle may have been imposed later, at 2·61–2·64 Ga, as a result of SE-vergent tectonism. The mantle lithosphere may have developed by tectonic imbrication of one or more slabs subducted beneath the craton at this time (Davis et al., 2003). However, evidence for the generation of subcontinental lithospheric mantle by subduction is controversial; it agrees with tectonic (Hoffman, 1990) and geophysical data (reviewed by Snyder, 2002), but is not supported by geochemical analysis (Griffin et al., 1999c). The oldest origin of the Central Slave mantle terrane has been advocated by Aulbach et al. (2001), as it may have resulted from plume subcretion at 3·3 Ga.

View larger version (36K): [in this window] [in a new window]   Fig. 13. The deep structure of the Slave mantle in a N–SE–SW cross-section along the transect line of Fig. 1. Lithology of the northern (Kopylova & Russell, 2000) and southeastern domains is based on the Brey–Köhler thermobarometry, and lithology and compositions of the central and southwestern domain are inferred from Griffin et al. (1999a, 1999b) and Carbno & Canil (2002). mg-numbers are calculated from whole-rock peridotite compositions of the Jericho and Gahcho Kué peridotites from corresponding depths. Question marks indicate depths not sampled by xenoliths or concentrate garnets. Position of the seismic discontinuity X is from Bostock (1998). (See text for description of the mantle blocks.)

 
A possible temporal evolution of the whole Slave mantle also cannot be excluded. The Gahcho Kué pipe gives us the earliest look into Slave cratonic mantle. Perhaps the lithosphere below the entire craton was colder and thicker in the Cambrian, and later evolution led to lithospheric thinning and to hotter geothermal regimes in Ordovician–Jurassic–Tertiary time. In this case, the Slave craton would be much like other cratonic areas with documented thermal erosion of the lithosphere, e.g. the Wyoming province, the Colorado Plateau, the East Sino-Korean Craton and the Kaapvaal craton (Griffin et al., 1998).

Origin of chemical zoning on cratons
A pronounced chemical zoning characterizes all Slave mantle terranes, as summarized in Fig. 13. In all known examples the shallow mantle shows greater chemical depletion than its deeper portion.

Below the SE Slave, spinel peridotite is chemically distinct from the underlying spinel-free mantle, as recorded by the mineral and bulk chemistry of the Gahcho Kué-derived xenoliths. Here the chemical contrast is manifested in lower Fe content and in significant depletion in Al and Ca in the whole-rock compositions (Figs 9 and 10) and in higher mg-numbers in olivine and orthopyroxene (Fig. 4) in the shallow mantle. The gradient of this chemical change is unknown, as there is an absence of samples from intermediate depths.

Below the Northern Slave, one of the major chemical boundaries occurs at a depth of 80–100 km; it separates shallow, more Fe-depleted, garnet-free mantle from deeper, less depleted, garnet-bearing mantle (Kopylova & Russell, 2000). Analysis of whole-rock Re–Os isotopic compositions and chondrite-normalized platinum group element (PGE) patterns shows that the Northern Slave mantle is also stratified with respect to age. The spinel peridotite generally yields older Archaean–Proterozoic T_RD and T_MA model ages. Age of rhenium depletion (T_RD) for Northern Slave spinel peridotite is 1·2–3·2 Ga and 1–3 Ga for garnet peridotite (Irvine, 2002).

An interpretation of whole-rock Re–Os ages has been fundamentally challenged by the recognition that these ages may be underestimated (Alard et al., 2002). However, thorough analysis (Pearson et al., 2004) re-validates the data. It emphasizes the following points. (1) The new technique of dating melt-depletion using individual sulphide inclusions (N. J. Pearson et al., 2002) gives Re–Os model ages that are as old as the oldest Re depletion ages for whole rocks for the Kaapvaal and Siberian xenoliths. (2) A very good fit exists between the ages of crust and mantle in Southern Africa as inferred from whole-rock Re–Os isotope analyses. The latter yield Archaean ages on the craton and Proterozoic ages off-craton (D. G. Pearson et al., 2002). (3) In samples with highly unradiogenic Os isotope compositions and the lowest Pd/Ir_N, Os isotope systematics should be dominated by primary sulphides and hence their bulk-rock Re depletion ages should approximate the melting age of the rock. Based on this, the oldest T_RD ages date the stabilization of the Slave lithosphere and the Northern Slave spinel periodtite is older than the Northern Slave garnet peridotite.

This age pattern probably reflects refertilization of the deeper mantle of incremental downward growth of the lithospheric mantle (Irvine et al., 1999; Irvine, 2002). Thus, chemical layering of the Northern Slave is linked to distinct periods in the stabilization of the craton. Overprinted on this old structure of the Northern Slave mantle are a thin layer of Phanerozoic fertile peridotite enriched in clinopyroxene and garnet and an underlying layer of magmatic pyroxenites (Kopylova & Russell, 2000).

A two-layer lithospheric structure was also reported in the Central Slave on the basis of garnet concentrate data (Griffin et al., 1999a, 1999b) and xenolith data (Pearson et al., 1999; Menzies et al., 2003). The upper, ultra-depleted layer occurs down to depths of 150 km and thins towards the north, south and SW (Griffin et al., 1999a, 1999b). Below Ekati pipes (Lac de Gras area), the ultra-depleted layer is found down to 140 km (Menzies et al., 2003). The shallow layer may be the strongly depleted lithosphere formed at an active convergent margin, whereas the deeper layer represents material that rose diapirically from the lower mantle, as suggested by a high proportion of lower-mantle inclusions in its diamonds (Griffin et al., 1999a). Formation of the ultra-depleted layer may precede or postdate plume subcretion at 3·3 Ga, which has been dated by Re–Os analysis of individual sulphide inclusions in olivine (Aulbach et al., 2000; Pearson et al., 2002).

Evidence for possible chemical layering of the SW Slave terrane (Carbno & Canil, 2002) is most controversial, as it is not based on independent estimates of pressure. A shallow, ultra-depleted layer of mantle lithosphere beneath the SW margin of the craton may be traced by garnet compositions from the Dry Bones kimberlite. Carbno & Canil (2002) found that the onset of order-of-magnitude variations in garnet chemistry occurs at 900°C (estimated from Ni content). If these Ni-in-garnet temperatures reflect a steady-state geotherm, then the geochemical changes occur at 110 km, and all low-temperature garnets came from a shallow ultra-depleted layer. However, the Ni-in-garnet temperatures ‘probably represent heating of the SW Slave mantle before entrainment in their host Dry Bones kimberlite’ (Carbno & Canil, 2002). If so, the correlation of geochemical parameters with temperatures estimated from Ni reflects different degrees of reworking by a hot metasomatizing front rather than a true chemical zoning.

At present we do not know the geometry of mantle domains with distinct bulk compositions and whether the chemically distinct layers continue across all the terranes with variations in depth and thickness. Comparison of olivine and whole-rock mg-numbers across mantle terranes suggests that the shallow depleted layer of the Northern Slave may be the thinned out ultra-depleted layer of the Central Slave. It is also possible that the depleted shallow mantle may stretch beneath the entire southern part of the Slave that encompasses the SW and the SE terranes. The position of the chemical boundary between 110 and 160 km beneath the SE Slave matches well with the hypothetical 110 km deep contact between an ultra-depleted shallow mantle and the deeper mantle mapped by garnet chemistry in the SW Slave. It also fits the location of the X discontinuity recorded in the SW Slave by teleseismic data at 110 km (Bostock, 1998). The deeper layer of the SE Slave characterized by the presence of high-Cr harzburgitic concentrate garnets (Grütter et al., 1999) may continue into the deep lithosphere of the Central Slave where such garnet also occurs (Davis et al., 2003). The layer is, however, absent below the SW Slave.

Petrological evidence amassed today suggests that progressively less depletion of the subcontinental lithospheric peridotitic mantle with depth may be the rule rather than the exception. It was recorded in systematic chemical differences between spinel and garnet peridotites on the Fennoscandian (Peltonen et al., 1999) and Wyoming cratons (Eggler et al., 1988), and in the off-cratonic mantle below the Andes (Kempton et al., 1999). The consistent decrease in degree of depletion with depth inferred from analysis of garnet concentrate data (Gaul et al., 2000; Griffin et al., 2003) was demonstrated for many Archaean, Proterozoic and some Phanerozoic subcontinental mantle columns. This analysis, however, was not based on independent depth estimates but rather on an assumption that the Ni-in-garnet temperatures reflect steady-state geotherms. A contrasting example of compositional changes in mantle xenoliths that is known to exist in the southern part of the North American craton (Dunn et al., 2003) is not gravitationally stable (Jordan, 1978; Poudjom Djomani et al., 2001), and therefore must be transient. Its origin is ascribed to tectonic stacking and the presence of allochthonous mantle terranes (Dunn et al., 2003).

Many possible processes could account for the observed chemical zoning. It could be produced by a single episode of polybaric melting associated with adiabatic decompression of upwelling mantle. Such melting would lead to a progressively higher degree of partial melt extraction at shallower depths (Hirschmann et al., 1999, and references therein). The trends in bulk composition of the peridotitic cratonic mantle could also be explained by an increasing degree of melt-related metasomatism with depth; this interpretation is preferred because of the negative correlation between olivine forsterite content and the abundances of metasomatically introduced Ti, Y, Zr and Ga in garnet (Gaul et al., 2000). Finally, a general pattern of decreasing depletion with depth could be a result of incremental growth of the lithospheric mantle coupled with a secular evolution in its composition (Griffin et al., 1999c). This growth could take place through several episodes of subduction that build up the cratonic roots downward and outward from the craton nucleus (e.g. Snyder, 2002). The younger subducted slabs that are less depleted, and therefore denser, would become stacked below older and more depleted blocks of the mantle. Another possible mechanism of incremental growth is the subcretion of plumes consisting of moderately depleted peridotite and eclogitic material (Gaul et al., 2000). The lack of data on the ages of cratonic peridotites and on the geometry of mantle domains of the Slave craton prevents any definite conclusions on the origin of mantle chemical zoning of the Slave mantle.

Kimberlite-induced recrystallization and formation of low-Al minerals
The latest processes recorded by the SE Slave mantle xenoliths relate to their ascent in the host kimberlitic magma. Percolation of kimberlite melt and/or volatiles exsolved from the host melt (addressed collectively as kimberlite fluid) caused partial melting and recrystallization of primary peridotitic minerals. This recrystallization affected mostly clinopyroxene, to a lesser degree spinel and olivine, and touched only slightly on orthopyroxene in spinel peridotite xenoliths. This recrystallization depleted all Al-bearing primary minerals of Al.

Secondary clinopyroxene, which replaces primary clinopyroxene in melted zones and crystallizes along grain boundaries, is invariably low in Al; most often it is also Na- and Cr-poor and occasionally Ca- and Ti-rich. It resembles low-Al (0·1–1·6 wt % Al₂O₃) phenocrystic clinopyroxene in kimberlites in general (Mitchell, 1986) and in the host 5034-Kennady hypabyssal kimberlite (Caro et al., 2004). Fine-grained interstitial, secondary clinopyroxene, similarly depleted in Na, Al and Cr, was reported in xenoliths from Jericho (Northern Slave), Grizzly (Central Slave; Boyd & Canil, 1997) and Udachnaya (Central Siberia; Boyd et al., 1997). In the Northern Slave, similar to the SE Slave, the development of secondary clinopyroxene is accompanied by a pronounced compositional heterogeneity in primary clinopyroxene (Kopylova et al., 1999a). Boyd et al. (1997) ascribed the occurrence of late-stage clinopyroxene, which was found together with monticellite and serpentine in Siberian peridotite, to low-pressure crystallization during transport and eruption of the kimberlite.

Kimberlite-related recrystallization of primary spinel into aggregates of skeletal microlites is associated with a drop in Al content (up to 3 wt % Al₂O₃) and an addition of Fe³⁺, Mg and Ti. Recrystallization of orthopyroxene into microlitic aggregates or into larger grains also depletes secondary orthopyroxene of Al, with changes in Al₂O₃ up to 2 wt %. This Al depletion is observed only in spinel peridotite, as orthopyroxene in the garnet-facies peridotite is originally very poor in Al₂O₃ (0·7–2 wt % Al₂O₃; Smith, 1999) and may already be equilibrated with kimberlitic fluid with respect to Al. In contrast, the Al₂O₃ content of orthopyroxene in the spinel-facies peridotites typically varies between 2 and 5 wt % (Smith, 1999), so a depletion in Al is easily noticed.

The key to the low-Al nature of secondary minerals recrystallized by the kimberlitic fluid lies in their equilibrium with chlorite, phlogopite and spinel, which develop in veinlets and in patches within the xenoliths. Aluminium preferentially resides in spinel and, under water-saturated conditions, partitions into hydrous minerals such as phlogopite and chlorite. It is well documented that orthopyroxene in equilibrium with hydrous minerals is Al-poor. Such orthopyroxene coexisting with chlorite was found in metamorphosed ultramafic rocks (Evans & Trommsdorff, 1974; Pfeifer, 1987) and in chlorite peridotite xenoliths of the Navajo and Grand Canyon fields (Smith & Riter, 1997). Boyd et al. (1999) examined xenoliths from Southern Africa and noted that crystallization of equilibrated mica or amphibole may reduce the Al₂O₃ content of orthopyroxene by 1 wt %. The low Na content of secondary clinopyroxene supports participation of hydrous fluids in its formation. The partition coefficient for Na between clinopyroxene and aqueous fluids at mantle pressures and temperatures is found to be very low (Ryabchikov et al., 1982).

Pressures and temperatures attending formation of late low-Al phases coexisting with kimberlitic phlogopite, chlorite and spinel are difficult to constrain. In two samples (11-2 and 61-1), secondary low-Al pyroxenes coexist and allow for calculation of two-pyroxene temperatures. However, in sample 11-2, clinopyroxene is extremely Ca-rich and lies outside the compositional range for which the thermometers are calibrated. Sample 61-1 gives T = 1186–1195°C at 20 kbar. These temperatures are much higher than the chlorite stability field at any pressure and therefore signal their incomplete equilibration. At the excess of forsterite and water in the CMASH system, chlorite dehydrates at 880–900°C and 20 kbar (Smith & Riter, 1997) but is stable at P < 35 kbar along the steady-state SE Slave geotherm (Goto & Tatsumi, 1990). This pressure estimate would be lower for possible transient high temperatures associated with kimberlite magmatism. The presence of phlogopite cannot constrain temperatures and pressures of the kimberlite–xenolith interaction, as it is stable to 60 kbar. We conclude that temperatures of interaction between kimberlite host and included xenoliths cannot be unambiguously defined, but pressures were probably lower than 35 kbar.

The addition of 0·1 vol. % of kimberlitic material to xenoliths had an insignificant effect on their bulk composition. The fluid added little Ti and K and subtly withdrew Na, as evidenced by a surplus of Ti and K and a deficit of Na relative to the values predicted from the modes and mineral compositions of primary minerals. Introduced Ti resides in secondary phlogopite, clinopyroxene, spinel and garnet, and added K in phlogopite. This interaction with kimberlitic fluid also leached Na out of peridotite, leaving Na-depleted clinopyroxenes. The fluid must also have been oxidizing and Fe-rich, as inferred by an increased Fe³⁺ in skeletal spinel and high Fe in olivine rims recrystallized by the fluid.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION HOST KIMBERLITE SAMPLES PETROGRAPHY ANALYTICAL METHODS MINERAL COMPOSITIONS GEOTHERMOBAROMETRY DISCUSSION CONCLUSIONS SUPPLEMENTARY DATA REFERENCES

 
(1) A suite of mantle-derived xenoliths of the Gahcho Kué kimberlite cluster of the SE Slave craton includes garnet and spinel coarse peridotite (61%), altered eclogite (18%), deformed garnet peridotite (17%) and orthopyroxenite (4%). Peridotites formed at T = 600–1300°C and P = 25–80 kbar in a thick (at least 220–250 km), relatively cool lithosphere. The SE Slave mantle is cooler than the mantle of other Archaean cratons and that below other terranes of the Slave craton.

(2) The SE Slave spinel peridotite is poorer in Al, Ca and Fe, and richer in Mg, than deeper garnet peridotite. All terranes of the Slave craton share the following common traits: (a) a greater chemical depletion of the spinel peridotites; (b) the presence of relatively orthopyroxene-poor, undepleted, low-T peridotite.

(3) Peridotitic xenoliths carry evidence for alteration caused by percolation of kimberlitic fluids at P < 35 kbar. Recrystallization of clinopyroxene, spinel, olivine and spinel-facies orthopyroxene occurs in close proximity to veinlets of serpentine, chlorite, phlogopite and spinel. The kimberlite-induced recrystallization depleted primary pyroxenes and spinel of Al, which escaped to form secondary hydrous minerals. The hydrous kimberlitic fluid was oxidizing, Ti-, Fe- and K-rich, and Na-poor.

	SUPPLEMENTARY DATA

TOP ABSTRACT INTRODUCTION HOST KIMBERLITE SAMPLES PETROGRAPHY ANALYTICAL METHODS MINERAL COMPOSITIONS GEOTHERMOBAROMETRY DISCUSSION CONCLUSIONS SUPPLEMENTARY DATA REFERENCES

 
Supplementary data for this paper are available on Journal of Petrology online.

	ACKNOWLEDGEMENTS

 
We thank J. Boyd, G. Pearson and B. Davis for sharing unpublished data and for discussions on various aspects of the study. Reviews by D. Canil, B. Griffin, M. Roden, C.-T. Lee and anonymous reviewers made this paper better. We are grateful to P. Kempton for editorial comments and corrections. We are indebted to De Beers Canada and Mountain Province Ltd for access to xenoliths from Tesla and Hearne pipes, and for permission to publish the results. Funding for this research derived from LITHOPROBE Grant (2000–2001) and from NSERC research grant to M.G.K. (2000–current). This paper is LITHOPROBE Contribution 1297.

	FOOTNOTES

 

^* Corresponding author. Telephone: (604) 822-0865. Fax: (604) 822-6088. E-mail: mkopylov{at}eos.ubc.ca

Present address: Laboratoire Géochimie-Cosmochimie (UMR 7579 CNRS), Insitut de Physique du Globe de Paris, Université Denis Diderot, 4 Place Jussieu, 75252 Paris Cedex 05, France.

	REFERENCES

TOP ABSTRACT INTRODUCTION HOST KIMBERLITE SAMPLES PETROGRAPHY ANALYTICAL METHODS MINERAL COMPOSITIONS GEOTHERMOBAROMETRY DISCUSSION CONCLUSIONS SUPPLEMENTARY DATA REFERENCES

 
Agashev, A. M., Pokhilenko, N. P., McDonald, J. A., Takazawa, E., Vavilov, M. A., Sobolev, N. V. & Watanabe, T. (2001). A unique kimberlite–carbonatite primary association on the Snap Lake dyke system, Slave craton: evidence from geochemical and isotopic studies. Abstracts, the Slave–Kaapvaal Workshop, September 2001, Merrickville.

Alard, O., Griffin, W. L., Pearson, N. J., Lorand, J.-P. & O'Reilly, S. Y. (2002). New insights into the Re–Os systematics of sub-continental lithospheric mantle from in situ analysis of sulfides. Earth and Planetary Science Letters 203, 651–663.[CrossRef][Web of Science]

Armstrong, J. P. (2002). Diamond exploration data in the North Slave craton, Nunavut. DIAND NU Open Report 2002-01 (CD-ROM).

Aulbach, S., Griffin, W. L., Pearson, N. J., O'Reilly, S. Y., Doyle, B. J. & Kivi, K. (2001). Re–Os isotope evidence for Meso-Archean mantle beneath 2·7 Ga Contwoyto terrane, Slave craton, Canada: implications for the tectonic history of the Slave craton. Abstracts, the Slave–Kaapvaal Workshop, September 2001, Merrickville.

Bank, C.-G., Bostock, M. G., Ellis, R. M. & Cassidy, J. F. (2000). A reconnaissance teleseismic study of the upper mantle and transition zone beneath the Archean Slave craton in NW Canada. Tectonophysics 319, 151–166.[CrossRef][Web of Science]

Bostock, M. G. (1998). Mantle stratigraphy and evolution of the Slave province. Journal of Geophysical Research 103, 21183–21200.[CrossRef][Web of Science]

Boyd, F. R. (1987). High- and low-temperature garnet peridotite xenoliths and their possible relation to the lithosphere– asthenosphere boundary beneath southern Africa. In: Nixon, P. H. (ed.) Mantle Xenoliths. New York: John Wiley, pp. 403–412.

Boyd, F. R. (1989). Compositional distinction between oceanic and cratonic lithosphere. Earth and Planetary Science Letters 96, 15–26.[CrossRef][Web of Science]

Boyd, F. R. (1999). The origin of cratonic peridotites: a major-element approach. In: Snyder, G. A., Neal, C. R. & Ernst, W. G. (eds) Planetary Petrology and Geochemistry; the Lawrence A. Taylor 60th Birthday Volume. Columbia: Bellwether, pp. 5–14.

Boyd, F. R. & Canil, D. (1997). Peridotite xenoliths from the Slave craton, NWT. Lunar and Planetary Institute Contribution 921, 34.

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

Boyd, F. R., Pearson, D. G. & Mertzman, S. A. (1999). Spinel-facies peridotites from the Kaapvaal root. In: Gurney, J. J. & Richardson, S. R. (eds) Proceedings of the 7th International Kimberlite Conference. Cape Town: Red Roof Design, pp. 40–47.

Brey, G. P. & Köhler, T. (1990). Geothermobarometry in four-phase lherzolites. II. New thermobarometers, and practical assessment of existing thermobarometers. Journal of Petrology 31, 1353–1378.[Abstract/Free Full Text]

Canil, D. & O'Neill, H. S. C. (1996). Distribution of ferric iron in some upper mantle assemblages. Journal of Petrology 37, 609–635.[Abstract/Free Full Text]

Carbno, G. B. & Canil, D. (2002). Mantle structure beneath the SW Slave craton, Canada: constraints from garnet geochemistry in the Drybones Bay kimberlite. Journal of Petrology 43, 129–142.[Abstract/Free Full Text]

Caro, G., Kopylova, M. & Creaser, R. (2004). The hypabyssal 5034 kimberlite of the Gahcho Kue cluster, Southeastern Slave craton, Northwest Territories, Canada: a granite-contaminated Group-I kimberlite. Canadian Mineralogist (in press).

Cox, K. G., Smith, M. R. & Beswetherick, S. (1987). Textural studies of garnet lherzolites: evidence of exsolution origin from high-temperature harzburgites. In: Nixon, P. H. (ed.) Mantle Xenoliths. New York: John Wiley, pp. 537–550.

Davis, W. J., Jones, A. G., Bleeker, W. & Grütter, H. (2003). Lithosphere development in the Slave Craton: a linked crustal and mantle perspective. Lithos 71(2–4), 575–589.[CrossRef][Web of Science]

Dawson, J. B. (2002). Metasomatism and partial melting in upper-mantle peridotite xenoliths from the Lashaine Volcano, Northern Tanzania. Journal of Petrology 43, 1749–1777.[Abstract/Free Full Text]

Dunn, D., Smith, D. & Bergman, S. C. (2003). Mantle xenoliths from the Prairie Creek lamproite province, Arkansas, USA. Extended Abstracts, 8th International Kimberlite Conference, June 2003, Victoria, B.C. (CD).

Eggler, D. H., Meen, J. K., Welt, F., Dudas, F. O., Furlong, K. P., McCallum, M. E. & Carlson, R. W. (1988). Tectonomagmatism of the Wyoming province. Colorado School of Mines Quarterly 83(2), 25–40.[Web of Science]

Evans, B. W. & Trommsdorff, V. (1974). Stability of enstatite + talc and CO₂-metasomatism of metaperidotite, Val d'Erta, Lepontine Alps. American Journal of Science 274, 274–296.[Abstract]

Finnerty, A. A. & Boyd, J. J. (1987). Thermobarometry for garnet peridotites: basis for the determination of thermal and compositional structure of the upper mantle. In: Nixon, P. H. (ed.) Mantle Xenoliths. New York: John Wiley, pp. 381–402.

Gaul, O. F., Griffin, W. L., O'Reilly, S. Y. & Pearson, N. J. (2000). Mapping olivine composition in the lithospheric mantle. Earth and Planetary Science Letters 182, 223–235.[CrossRef][Web of Science]

Goto, A. & Tatsumi, Y. (1990). Stability of chlorite in the upper mantle. American Mineralogist 75, 105–108.[Abstract]

Griffin, W. L. & Ryan, C. C. (1995). Trace elements in indicator minerals: area selection and target evaluation in diamond exploration. Journal of Geochemical Exploration 53, 311–337.[CrossRef][Web of Science]

Griffin, W. L., O'Reilly, S. Y., Ryan, C. G., Gaul, O. & Ionov, D. A. (1998). Secular variations in the composition of subcontinental lithospheric mantle: geophysical and geodynamic implications. In: Braun, J., Dooley, J., Goleby, B., van der Hilst, R. & Klootwijk, C. (eds) Structure and Evolution of the Australian Continent. American Geophysical Union Monograph, Geodynamics, 26, 1–26.

Griffin, W. L., Doyle, B. J., Ryan, C. G., Pearson, N. J., O'Reilly, S. Y., Davies, R., Kivi, K., van Achterbergh, E. & Natapov, L. M. (1999a). Layered mantle lithosphere in the Lac de Gras Area, Slave Craton: composition, structure and origin. Journal of Petrology 40, 705–727.[CrossRef][Web of Science]

Griffin, W. L., Doyle, B. J., Ryan, C. G., Pearson, N. J., O'Reilly, S. Y., Natapov, L., Kivi, K., Kretschmar, U. & Ward, J. (1999b). Lithosphere structure and mantle terranes: Slave Craton, Canada. In: Gurney, J. J. & Richardson, S. R. (eds) Proceedings of the 7th International Kimberlite Conference. Cape Town: Red Roof Design, pp. 299–306.

Griffin, W. L., O'Reilly, S. Y. & Ryan, C. G. (1999c). The composition and origin of subcontinental lithospheric mantle. In: Fei, Y., Bertka, C. & Mysen, B. (eds) Mantle Petrology: Field Observations and High-pressure Experimentation: a Tribute to Francis (Joe) Boyd. Geochemical Society, Special Publication 6, 13–45.

Griffin, W. L., Fisher, N. I., Friedman, J., Ryan, C. G. & O'Reilly, S. Y. (1999d). Cr-pyrope garnets in the lithospheric mantle; I, Composition systematics and relations to tectonic setting. Journal of Petrology 40, 679–704.[CrossRef][Web of Science]

Griffin, W. L., O'Reilly, S. Y., Doyle, B. J., Kivi, K. & Coopersmith, H. G. (2003). Lithospheric mapping beneath the North American plate. Extended Abstracts, 8th International Kimberlite Conference, June 2003, Victoria, B.C. (CD).

Grütter, H. S., Apter, D. B. & Kong, J. (1999). Crust–mantle coupling: evidence from mantle-derived xenocrystic garnets. In: Gurney, J. J. & Richardson, S. R. (eds) Proceedings of the 7th International Kimberlite Conference. Cape Town: Red Roof Design, pp. 307–312.

Gurney, J. J. & Zweistra, P. (1995). The interpretation of the major element compositions of mantle minerals in diamond exploration. Journal of Geochemical Exploration 53, 293–309.[CrossRef][Web of Science]

Harte, B. (1977). Rock nomenclature with particular relation to deformation and recrystallization textures in olivine-bearing xenoliths. Journal of Geology 85, 279–288.[Web of Science]

Harte, B. & Hawkesworth, C. J. (1989). Mantle domains and mantle xenoliths. In: Ross, J., Jaques, A. L., Ferguson, J., Green, D. H., O'Reilly, S. Y., Danchin, R. V. & Janse, A. J. (eds) Kimberlites and Related Rocks. Special Publication of the Geological Society of Australia 14, 649–686.

Heaman, L. M., Kjarsgaard, B., Creaser, R. A., Cookenboo, H. O. & Kretschmar, U. (1997). Multiple episodes of kimberlite magmatism in the Slave Province, North America. In: Cook, F. & Erdmer, (compilers) Slave–Northern Cordillera Lithospheric Evolution (SNORCLE) Transect and Cordilleran Tectonics Workshop Meeting (March 1997), University of Calgary. LITHOPROBE Report 56, 14–17.

Hirschmann, M. M., Asimow, P. D., Ghiorso, M. S. & Stolper, E. M. (1999). Calculation of peridotite partial melting from thermodynamic models of minerals and melts III. Controls on isobaric melt production and the effect of water on melt production. Journal of Petrology 40, 831–851.[CrossRef][Web of Science]

Hoffman, P. F. (1990). Geological constraints in the origin of the mantle root beneath the Canadian shield. Philosophical Transactions of the Royal Society of London, Series A 331, 523–532.[CrossRef]

Irvine, G. J. (2002). Time constraints on the formation of lithospheric mantle beneath cratons: a Re–Os isotope and platinum group element study of peridotite xenoliths from Northern Canada and Lesotho. Ph.D. thesis, University of Durham.

Irvine, G., Kopylova, M. G., Carlson, R. W., Pearson, D. G., Shirey, S. B. & Kjarsgaard, B. A. (1999). Age of the lithospheric mantle beneath and around the Slave craton: a Re–Os isotope study of peridotite xenoliths from the Jericho and Somerset Island kimberlites. Abstracts, 9th V. M. Goldschmidt Conference. Houston, TX: Lunar and Planetary Institute, pp. 134–135.

Jones, A. G., Ferguson, I. J., Chave, A. D., Evans, R. L. & McNeice, G. W. (2001). Electric lithosphere of the Slave craton. Geology 29(5), 423–426.[Abstract/Free Full Text]

Jordan, T. H. (1978). Composition and development of the continental tectosphere. Nature 234, 544–548.

Kempton, P. D., Hawkesworth, C. J., Lopez-Escobar, L. & Ware, A. J. (1999). Spinel ± garnet lherzolite xenoliths from Pali Aike, Part 2: Trace element and isotopic evidence bearing on the evolution of lithospheric mantle beneath southern Patagonia. In: Gurney, J. J., Gurney, J. L., Pascoe, M. D. & Richardson, S. H. (eds) Proceedings of the 7th International Kimberlite Conference. Cape Town: Red Roof Design, pp. 415–428.

Kennedy, C. S. & Kennedy, G. C. (1976). The equilibrium boundary between graphite and diamond. Journal of Geophysical Research 81, 2467–2470.

Ketchum, J. W. F. & Bleeker, W. (2001). 4·03–2·85 Ga growth and modification of the Slave protocraton, NW Canada. Abstracts, the Slave–Kaapvaal Workshop, September 2001, Merrickville.

Kjarsgaard, B. A. & Petersen, T. D. (1992). Kimberlite-derived ultramafic xenoliths from the diamond stability field: a new Cretaceous geotherm for Somerset Island, Northwest Territories. Current Research, Part B; Geological Survey of Canada, Paper 92-1B, 1–6.

Kopylova, M. G. & Russell, J. K. (2000). Chemical stratification of cratonic lithosphere: constraints from the Northern Slave craton, Canada. Earth and Planetary Science Letters 181, 71–87.[CrossRef][Web of Science]

Kopylova, M. G., Russell, J. K. & Cookenboo, H. (1999a). Petrology of peridotite and pyroxenite xenoliths from the Jericho kimberlite: implications for the thermal state of the mantle beneath the Slave craton, Northern Canada. Journal of Petrology 40, 79–104.[CrossRef][Web of Science]

Kopylova, M. G., Russell, J. K. & Cookenboo, H. (1999b). Mapping the lithosphere beneath the North Central Slave Craton. In: Gurney, J. J. & Richardson, S. R. (eds) Proceedings of the 7th International Kimberlite Conference. Cape Town: Red Roof Design, pp. 468–479.

Kopylova, M. G., Caro, G. & Russell, J. K. (2001). Kimberlite and xenoliths from the southern Slave—a terrane with the highest diamond potential in the Slave craton. Extended Abstracts, Slave– Northern Cordillera Lithospheric Evolution (SNORCLE) Transect and Cordilleran Tectonics Workshop Meeting (February 22–25), Pacific Geoscience Centre. LITHOPROBE Report 79, 18–24.

Kretz, R. (1983). Symbols for rock-forming minerals, American Mineralogist 68, 277–279.[Abstract]

Kukkonen, I. T. & Peltonen, P. (1999). Xenolith-controlled geotherm for the central Fennoscandian Shield: implications for lithosphere– asthenosphere relations. Tectonophysics 304, 301–315.[CrossRef][Web of Science]

Lee, C.-T. & Rudnick, R. L. (1999). Compositionally stratified cratonic lithosphere: petrology and geochemistry of peridotite xenoliths from the Labait tuff cone, Tanzania. In: Gurney, J. J. & Richardson, S. R. (eds) Proceedings of the 7th International Kimberlite Conference. Cape Town: Red Roof Design, pp. 503–521.

Lehtonen, M., O'Brien, H., Peltonen, P., Johanson, B. & Pakkanen, L. (2003). Layered mantle at the edge of the Karelian craton: P–T of mantle xenocrystals and xenoliths from Eastern Finland kimberlites. Extended Abstracts, 8th International Kimberlite Conference, June 2003, Victoria, B.C. (CD).

MacGregor, I. D. (1974). The system MgO–Al₂O₃–SiO₂: solubility of Al₂O₃ in enstatite for spinel and garnet peridotite compositions. American Mineralogist 59, 110–119.[Web of Science]

MacKenzie, J. M. & Canil, D. (1999). Composition and thermal evolution of cratonic mantle beneath the central Archean Slave Province, NWT, Canada. Contributions to Mineralogy and Petrology 134, 313–324.[CrossRef][Web of Science]

Menzies, A., Westerlund, K., Gurney, J., Carlson, J., Fung, A. & Nowicki, T. (2003). Peridotitic mantle xenoliths from kimberlites on the Ekati Diamond Mine Property, NWT, Canada. Extended Abstracts, 8th International Kimberlite Conference, June 2003, Victoria, B.C. (CD).

Meyer, H. O. A., Waldman, M. A. & Garwood, B. L. (1994). Mantle xenoliths from kimberlite near Kirkland Lake, Ontario. Canadian Mineralogist 32, 295–306.

Mitchell, R. H. (1986). Kimberlites: Mineralogy, Geochemistry, and Petrology. New York: Plenum.

O'Neill, H. St. C. (1981). The transition between spinel lherzolite and garnet lherzolite, and its use as a geobarometer. Contributions to Mineralogy and Petrology 77, 185–194.[CrossRef][Web of Science]

O'Neill, H. St. C. & Wall, V. J. (1987). The olivine– orthopyroxene–spinel oxygen geobarometer, the nickel precipitation curve, and the oxygen fugacity of the Earth's upper mantle. Journal of Petrology 28, 1169–1191.[Abstract/Free Full Text]

O'Neill, H. St. C. & Wood, B. J. (1979). An experimental study of Fe–Mg partitioning between garnet and olivine and its calibration as a geothermometer. Contributions to Mineralogy and Petrology 70, 59–70; corrected (1980) in Contributions to Mineralogy and Petrology 72, 337.[CrossRef][Web of Science]

Pearson, D. G., Boyd, F. R., Haggerty, S. E., Pasteris, J. D., Field, S. W., Nixon, P. H. & Pokhilenko, N. P. (1994). The characterization and origin of graphite in cratonic lithospheric mantle: a petrological carbon isotope and Raman spectroscopic study. Contributions to Mineralogy and Petrology 115, 449–466.[CrossRef][Web of Science]

Pearson, D. G., Irvine, G. J., Carlson, R. W., Kopylova, M. G. & Ionov, D. A. (2002). The development of lithospheric mantle keels beneath the earliest continents: time constraints using PGE and Re–Os isotope systematics. In: Fowler, C. M. R., Ebinger, C. J. & Hawkesworth, C. J. (eds) The Early Earth: Physical, Chemical and Biological Development. Geological Society, London, Special Publications 199, 65–90.

Pearson, D. G., Irvine, G. J., Ionov, D. A., Boyd, F. R. & Dreibus, G. E. (2004). Re–Os isotope systematics and Platinum Group Element fractionation during mantle melt extraction: a study of massif and xenolith peridotite suites. In: Lorand, J.-P., Reisberg, L. & Alard, O. (eds) Chemical Geology, Highly Siderophile Element Volume. Amsterdam: Elsevier (in press).

Pearson, N. J., Griffin, W. L., Doyle, B. J., O'Reilly, S. Y., van Achtenbergh, E. & Kivi, K. (1999). Xenoliths from kimberlite pipes of the Lac de Gras area, Slave Craton, Canada. In: Gurney, J. J. & Richardson, S. R. (eds) Proceedings of the 7th International Kimberlite Conference. Cape Town: Red Roof Design, pp. 644–658.

Pearson, N. J., Alard, O., Griffin, W. L., Jackson, S. E. & O'Reilly, S. Y. (2002). In situ measurement of Re–Os isotopes in mantle sulfides by laser ablation multicollector–inductively coupled plasma mass spectrometry: analytical methods and preliminary results. Geochimica et Cosmochimica Acta 66, 1037–1050.[CrossRef][Web of Science]

Peltonen, P., Huhma, H., Tyni, M. & Shimizu, N. (1999). Garnet peridotite xenoliths from kimberlites of Finland: nature of the continental mantle at an Archean craton–Proterozoic mobile belt transition. In: Gurney, J. J. & Richardson, S. R. (eds) Proceedings of the 7th International Kimberlite Conference. Cape Town: Red Roof Design, pp. 664–676.

Pfeifer, H.-R. (1987). A model for fluids in metamorphosed ultramafic rocks: IV Metasomatic veins in metaharzburgites of Cima di Gagnone, Valle Versasca, Switzerland. In: Heldgeson, H. V. (ed.) Chemical Transport in Metasomatic Processes. Dordrecht: Reidel, pp. 591–632.

Poudjom Djomani, Y. H., O'Reilly, S. Y., Griffin, W. L. & Morgan, P. (2001). The density structure of subcontinental lithosphere through time. Earth and Planetary Science Letters 184, 605–621.[CrossRef][Web of Science]

Promprated, P., Taylor, L., Floss, C., Malkovets, V., Anand, M., Griffin, W., Pokhilenko, N. & Sobolev, N. (2003). Diamond inclusions from Snap Lake, NWT, Canada. Extended Abstracts, 8th International Kimberlite Conference, June 2003, Victoria, B.C. (CD).

Rikhotso, C. T., Poniatowski, B. T. & Hetman, C. M. (2003). Overview of the exploration, evaluation, and geology of the Gahcho Kue kimberlites, Northwest Territories. In: Kjarsgaard, B. B. (ed.) VIIIth International Kimberlite Conference, Slave Province and Northern Alberta Field Trip Guidebook. Ottawa, Ont.: Geological Survey of Canada, pp. 79–86.

Roden, M. F., Laz'ko, E. E. & Jagoutz, E. (1999). The role of garnet pyroxenites in the Siberian lithosphere: evidence from the Mir Kimberlite. In: Gurney, J. J. & Richardson, S. R. (eds) Proceedings of the 7th International Kimberlite Conference. Cape Town: Red Roof Design, pp. 664–676.

Ryabchikov, I. D., Schreyer, W. & Abraham, K. (1982). Compositions of aqueous fluids in equilibrium with pyroxenes and olivines at mantle pressures and temperatures. Contributions to Mineralogy and Petrology 79, 80–94.[CrossRef][Web of Science]

Schmidberger, S. S. & Francis, D. (1999). Nature of the mantle root beneath the North American Craton: mantle xenolith evidence from Somerset Island kimberlites. Lithos 48, 195–216.[CrossRef][Web of Science]

Smith, D. (1999). Temperatures and pressures of mineral equilibration in peridotite xenoliths: review, discussion, and implications. In: Fei, Y., Bertka, C. & Mysen, B. (eds) Mantle Petrology: Field Observations and High-pressure Experimentation: a Tribute to Francis (Joe) Boyd. Geochemical Society, Special Publication 6, 171–188.

Smith, D. & Riter, J. C. A. (1997). Genesis and evolution of low-Al orthopyroxene in spinel peridotite xenoliths, Grand Canyon field, Arizona, USA. Contributions to Mineralogy and Petrology 127, 391–404.[CrossRef]

Snyder, D. (2002). Lithosphere growth at margins of cratons. Tectonophysics 355, 7–22.[CrossRef][Web of Science]

Thorpe, R. I., Cumming, G. L. & Mortensen, J. K. (1992). A major Pb-isotope boundary in the Slave Province and its probable relation to ancient basement of the western Slave. Canada–NWT MDA Summary Volume. Geological Survey of Canada, Open File 2484, 179–184.

Xu, Y-G., Ross, J. V. & Mercier, J.-C. (1993). The upper mantle beneath the continental rift of Tanlu, Eastern China: evidence for the intra-lithospheric shear zones. Tectonophysics 225, 337–360.[CrossRef][Web of Science]

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

This article has been cited by other articles:

				S. A. Gibson, J. Malarkey, and J. A. Day Melt Depletion and Enrichment beneath the Western Kaapvaal Craton: Evidence from Finsch Peridotite Xenoliths J. Petrology, October 22, 2008; (2008) egn048v1. [Abstract] [Full Text] [PDF]

				C. Bonadiman, M. Coltorti, S. Duggen, L. Paludetti, F. Siena, M. F. Thirlwall, and B. G. J. Upton Palaeozoic subduction-related and kimberlite or carbonatite metasomatism in the Scottish lithospheric mantle Geological Society, London, Special Publications, January 1, 2008; 293(1): 303 - 333. [Abstract] [Full Text] [PDF]

				A. Ulianov, O. Muntener, P. Ulmer, and T. Pettke Entrained Macrocryst Minerals as a Key to the Source Region of Olivine Nephelinites: Humberg, Kaiserstuhl, Germany J. Petrology, June 1, 2007; 48(6): 1079 - 1118. [Abstract] [Full Text] [PDF]

				H. GRUTTER, D. LATTI, and A. MENZIES Cr-Saturation Arrays in Concentrate Garnet Compositions from Kimberlite and their Use in Mantle Barometry J. Petrology, April 1, 2006; 47(4): 801 - 820. [Abstract] [Full Text] [PDF]

				L. M. HEAMAN, R. A. CREASER, H. O. COOKENBOO, and T. CHACKO Multi-Stage Modification of the Northern Slave Mantle Lithosphere: Evidence from Zircon- and Diamond-Bearing Eclogite Xenoliths Entrained in Jericho Kimberlite, Canada J. Petrology, April 1, 2006; 47(4): 821 - 858. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) Supplementary data Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (20) Request Permissions Google Scholar Articles by KOPYLOVA, M. G. Articles by CARO, G. Search for Related Content GeoRef GeoRef Citation Social Bookmarking          What's this?

Online ISSN 1460-2415 - Print ISSN 0022-3530

Copyright © 2009 Oxford University Press

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

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