Mineralogical Association of Canada [MA]

 CC:701B  Sunday  0800h

Mantle Conditions, Diamond Genesis, and the Kimberlite Sample I

Presiding:  R L Flemming, The University of Western Ontario; T Stachel, University of Alberta; D Schulze, University of Toronto


Diamond Morphology: Link to Metasomatic Events in the Mantle or Record of Evolution of Kimberlitic Fluid?

* Fedortchouk, Y (yana@dal.ca), Department of Earth Sciences, Dalhousie University, Life Sciences Centre Dalhousie University, Halifax, NS B3H 4J1, Canada

Morphology and surface features on diamonds show tremendous variation even within a single kimberlite body reflecting a complex history of growth and dissolution. But does the diamond surface record the conditions in the several mantle sources sampled by the rising kimberlite magma, or evolution of the fluid system in the kimberlite magma itself? To address this question I revised morphological classification of diamonds from several kimberlite pipes from EKATI Mine property, N.W.T., Canada. The novelty of the approach, compared to the existing classifications, is in utilizing a random but large dataset of diamond dissolution experiments accumulated by several researchers including myself. These experiments have shown that similar forms (e.g. trigon etch pits) can be produced in a variety of conditions and environments, whereas their shape and size would depend on the reactant. Similarly, different types of resorption features always form together and can be used for deriving the composition of oxidizing fluid. The proposed classification method is focused on relating various types of diamond surfaces to the composition and conditions of oxidizing media. The study uses parcels of micro-and macro-diamonds (total of 125 carats) from Misery, Grizzly, Leslie and Koala kimberlites, EKATI Mine property, Northwest Territories, Canada. Only octahedron and hexoctahedron diamonds were selected (total ~600 stones). Diamond surfaces were studied using an optical and Field- Emission Scanning Electron Microscope to define resorption elements - simple surface features. These elements were identified for each of the three categories: 1) present on octahedral faces (well-preserved diamonds), 2) present on hexoctahedral faces (rounded resorbed diamonds), and 3) frosting (micro-features). Consistent associations of several elements define Resorption Types of diamonds, which form during a single oxidizing event. We further relate these types to the composition of the C-H-O + chlorides fluid and the temperature for the conditions where experimental data exists. We make an attempt to identify kimberlite vs. mantle resorption events. Initial results of this study show that diamonds carrying only pure features characteristic for H2O-rich oxidizing fluid compose a minor proportion of diamond populations. Some kimberlites, Grizzly and Leslie, have these features modified likely due to the fluid loss. Each diamond population contains several distinct types of diamond surfaces requiring complex history of different resorption events. The large variation of resorption conditions within a single diamond population suggests the importance of mantle resorption.


In-situ Analysis of Diamonds and Their Mineral Inclusions From the Lynx Kimberlite Dyke Complex, Central Quebec

* Van Rythoven, A (adrian.vanrythoven@utoronto.ca), Department of Geology, University of Toronto, Erindale College, 3359 Mississauga rd. N., Mississauga, ON L5L 1C6, Canada
McCandless, T E
EM: , Stornoway Diamond Corporation, 116-980 West 1st st., North Vancouver, BC V7P 3N4, Canada
Schulze, D J (daniel.schulze@utoronto.ca), Department of Geology, University of Toronto, Erindale College, 3359 Mississauga rd. N., Mississauga, ON L5L 1C6, Canada
Bellis, A
EM: , Stornoway Diamond Corporation, 116-980 West 1st st., North Vancouver, BC V7P 3N4, Canada
Taylor, L A
EM: , Planetary Geosciences Institute, Department of Earth and Planetary Sciences, University of Tennessee, 1412 Circle Drive, Knoxville, TN 37996-1410, United States
Liu, Y
EM: , Planetary Geosciences Institute, Department of Earth and Planetary Sciences, University of Tennessee, 1412 Circle Drive, Knoxville, TN 37996-1410, United States

Twenty diamonds from the 522 Ma Lynx kimberlite dyke complex were selected from 442 stones in the 1.47- 3.45mm (+3 to +11 DTC) sieve class on the basis of visible inclusions. The 442 diamonds are part of a larger population of 6598 stones produced from 34 t and 494 t bulk samples taken in 2005 and 2007, respectively. The twenty diamonds all have octahedral primary growth forms. Three macles occur, as does one example of two intergrown octahedra connected along their {111} faces. Two samples are coarse intergrowths of octahedra. Most of the diamonds display a significant degree of resorption and range from octahedra with rounded corners and edges to tetrahexahedroida. Shield and serrate laminae, and hillocks are the most common resorption-related surface features. Nineteen of the samples have light brown to brown colouration. After their external morphology was examined, the diamonds were cut and polished along a single plane to expose included mineral grains for compositional analysis and to image internal structure. Cathodoluminescence imaging reveals deformation lamellae in the majority of the diamonds. A subset of these stones show deformation lamellae truncated by growth/resorption zones and in some cases intersection of planes of different orientation. Oscillatory planar growth patterns are the most common. However, examples of simple homogeneous, complex planar, and complex undulating growth zones occur. Inclusions, particularly olivine, typically occur in core/early growth regions of the diamonds. Of the twenty diamonds, sixteen have primary inclusions. The inclusion suite is largely peridotitic. Seventeen forsteritic olivine inclusions occur in ten diamonds and have molar Mg/(Mg+Fe)= 0.916-0.933. Seven Cr-diopside inclusions occur in one diamond (2.2-2.3 wt. % Cr2O3). Four Cr-pyropes (Cr/(Cr+Al) = 0.28-0.41) occur in three diamonds. Two enstatite inclusions (Mg/(Mg+Fe) = 0.938-0.94) occur in two diamonds. One heterogeneous inclusion of monosulfide solid solution was also found (13.6-20.1 wt. % Ni, 3.9-9.6 wt. % Cu). One stone containing a grain of omphacite (0.01 wt. % K2O, 4.1 wt. % Na2O) is the only eclogitic diamond in the group. The diopside inclusions have nearly identical compositions that indicate equilibration conditions in the range of 58-60 kbar and 1250-1280°C that plot on the 41 mW/m2 geotherm. The least forsteritic olivine inclusion analysed was also found in the diamond with the diopside inclusions, suggesting a fairly fertile lherzolitic region of mantle at 180-190 km depth. The garnet data indicate both strongly harzburgitic (G10, 12.4- 13.7 wt. % Cr2O3, 3.7-4.4 wt. % CaO) and more lherzolitic (G10-G9 boundary, 8.9 wt. % Cr2O3, 5.8 wt. % CaO) parageneses.


New and Unusual Mineral Assemblages Discovered in Diamond from Juina, Brazil, Using FIB/TEM

* Wirth, R (wirth@gfz-potsdam.de), Helmholtz Centre Potsdam GFZ, 3.3 Experimental Geochemistry and Mineral Physics, Potsdam, D 14473, Germany
Kaminsky, F (felixvkaminsky@cs.com), KM Diamond Exploration Ltd., 2446 Shadoldt Lane, West Vancouver, Vancouver, BC V7S3J1, Canada
Matsyuk, S (smatsyuk@hotmail.com), Institute of Geochemistry, Mineralogy and Ore Formation, Palladin Av., Kyiv-142, 03680, Ukraine

Inclusions in diamond carry valuable information about their location in the mantle, the PT-conditions of diamond growth, and the medium the diamond has grown from. Since several years, even nanometre-sized inclusions in diamond are accessible combining site-specific focused ion beam (FIB) sample preparation with TEM investigation. Nanometre-sized inclusions are especially interesting because they might have preserved their original crystal structure thus providing PT estimates. Often, the TEM samples contain complete mineral assemblages, especially when the individual phases are nanometre-sized. Micro- and nanoinclusions in diamonds from Juina, Brazil have been investigated by TEM methods. TEM combines structural (electron diffraction), microstructural and chemical information (EDX and EELS analysis) allowing the identification of even nanometre-sized inclusions in diamond. TEM sample preparation from Laser-cut stones was performed with focused ion beam technique (FIB). The TEM foils 15 x 10 x 0.2 μm in size contain sometimes micrometer-sized inclusions, which are composed of different mineral associations most of them never reported before. The individual mineral phases present in the inclusions have been identified by electron diffraction and chemical composition. It is a common feature of the investigated inclusions that they are composed of a carbonate matrix with different mineral assemblages embedded. Some characteristic mineral assemblages in these inclusions are: 1. Calcite + wollastonite II (high) + cuspidine (fluorine-rich silicate) + monticellite + noncrystalline material composed of carbon + Ca + Si +O, most likely a quench phase. 2. Calcite + wollastonite II (high) + Ca-rich garnet (andradite-kimzeyite-schorlomite group) + nyerereite + apatite enriched in La, Ce and Nd + diamond. Such mineral assemblages have never been reported before as mineral inclusions in diamond. These mineral assemblages might indicate the presence of a carbonatitic melt during diamond formation and growth.


Carbonate, Halide, and Other New Mineral Inclusions in Diamond and Deep-Seated Carbonatitic Magma

* Kaminsky, F (felixvkaminsky@cs.com), KM Diamond Exploration Ltd., 2446 Shadbolt Lane, West Vancouver, BC V7S 3J1, Canada
Wirth, R (wirth@gfz-potsdam.de), Helmholtz Centre Potsdam GFZ German Research Centre for Geosciences, Experimental Geochemistry and Mineral Physics, Telegrafenberg, Potsdam, D14473, Germany
Matsyuk, S (amatsyuk@hotmail.com), Institute of Gechemistry, Mineralogy and Ore Formation, National Academy of Sciences of Ukraine, Palladin Av., Kyiv, 03680, Ukraine

A series of uncommon micro- and nano-inclusions was identified in diamonds from the Juina area: carbonates, halides, and others. Carbonates are represented by calcite (with Sr and Ba), K-rich nyerereite (K2O = 10.0-13.78 wt. %), and nahcolite. Halides are NaCl, KCl, CaCl2 and PbCl2. Minerals of the periclase- wüstite series belong to two separate groups: wüstite and Mg-wüstite with Mg# = 1.9-15.3, and Fe-periclase and periclase with Mg# = 84.9-92.1. Wollastonite-II (high, Ca: Si = 0.992) has a triclinic structure. Ca-rich garnet has a noticeable admixture of Zr; it belongs to the andradite - kimzeyite - schorlomite group. Two types of spinel were distinguished among mineral inclusions in diamond: zoned magnesioferrite (with Mg# varying from 13.5 in a core to 90.8 in a rim) and Fe-spinel (magnetite). Olivine (Mg# = 93.6), intergrown with nyerereite, forms elongated, lath-shaped crystal and, probably, is a retrograde transformation of ringwoodite or wadsleyite. Some apatite grains are enriched in La, Ce and Nd. Among other minerals, there are anhydrite, cuspidine, phlogopite, TiO2 with an α-PbO2 structure, native Fe. All inclusions are polymineralic solid inclusions. These minerals form a carbonatitic-type mineral association in diamond which may have been originated in lower mantle and/or transition zone. Wüstite inclusions with Mg# = 1.9-3.4, according to the experimental data, may have been formed in the lowermost mantle. The source for the observed carbonatitic-type mineral association in diamond is deep-seated carbonatitic, most likely natrocarbonatitic magma.


Crystal-Chemical Correlations in Chromites from Kimberlitic and Non-Kimberlitic Sources.

Freckelton, C N (Candace.Freckelton@Ontario.ca), Ontario Geological Survey, Mininstry of Northern Development and Mines, Sudbury, ON P3E 6B5,
Freckelton, C N (Candace.Freckelton@Ontario.ca), Department of Earth Sciences, University of Western Ontario, London, ON N6A 5B7, Canada
* Flemming, R L (rflemmin@uwo.ca), Department of Earth Sciences, University of Western Ontario, London, ON N6A 5B7, Canada

This study explores the utility of micro X-ray diffraction (μXRD) as a tool for diamond exploration, as a compliment to current industry-standard techniques such as electron probe microanalysis (EPMA). Here we examine chromite. As one of the first phases to crystallize in mantle rocks, it is a useful indicator of upper mantle magmatic conditions in rocks that have been sampled by kimberlites. In addition, chromite does not alter easily from chemical and physical weathering processes. As such, chromite is a useful kimberlite indicator mineral in diamond exploration. We present correlations between crystal structure (unit cell) and chemical composition of chromite, (Fe,Mg)[Cr, Al]2O4, using correlated μXRD and EPMA data for 133 chromites from a three source locations: Two kimberlite sources and one non-kimberlitic source from an Archean granite/greenstone terrain. Quantitative analysis was performed using Electron Probe Microanalysis (EPMA) at Mineral Services, South Africa, prior to the loan of the samples. Randomly-oriented chromite grains, approximately 500 μm in diameter, were analyzed as previously mounted for EPMA. Micro X-ray-diffraction was performed using a Bruker D8-Discover Diffractometer, with θ-θ geometry, with CuKα radiation, operating at 40 kV and 40 mA, with nominal beam diameter of 500 μm. The data were collected in omega scan mode. Two dimensional General Area Detector Diffraction System (GADDS) images were collected for 20 minutes per image, and integrated to produce one-dimensional plots of intensity versus 2θ, for subsequent unit cell refinement using CELREF. Although all samples in this study were considered to be 'chromite', a plot of Cr/(Cr+Al) versus Fe2+/(Fe2++Mg) shows extensive substitution among four dominant members: chromite (FeCr2O4), magnesio-chromite (MgCr2O4), spinel (MgAl2O4), and hercynite (FeAl2O4), where Mg and Fe2+ substitute for one another on the tetrahedral site, and Cr and Al substitute for one another on the octahedral site. Our data are widely variable as compared to the field occupied by chromite inclusions in diamonds (high Cr and Mg (∼60 wt %) and very low Ti (∼0.40 wt %). Plots of the unit cell parameter, ao, versus composition demonstrate a decrease in unit cell size with increasing Al content (and corresponding decrease in Cr content), consistent with a smaller cation radius for Al versus Cr (Al=0.675 Å and Cr=0.905 Å). The trend in unit cell size is unlikely to be effected by Mg-Fe substitution because of the very small difference in their tetrahedral cation radii (Fe2+=0.835 Å and Mg=0.86 Å). Initial plots of composition versus unit cell parameter were clearly able to distinguish a difference between unit cell of kimberlitic chromites and non-kimberlitic chromites. The significantly higher Cr content in kimberlitic chromites (radius=0.905 Å), and correspondingly higher Al content in non-kimberlitic chromites (radius=0.675 Å), results in a striking bimodal distribution in unit cell parameter, ao, where kimberlitic chromites have a larger unit cell (> 8.3 Å) than non-kimberlitic chromites (< 8.3 Å). This preliminary data provides a useful starting point for screening minerals from naturally relevant chromite solid solutions using their corresponding unit cell parameters. Future work will examine which site substitutions (octahedral versus tetrahedral) are affecting the unit cell as well as the effect of cation order-disorder on unit cell parameters.


Crystal-chemical Relationships in Diamond-indicating Peridotitic and Eclogitic Garnets

* Harwood, B P (bpharwoo@uwo.ca), Department of Earth Sciences, University of Western Ontario, London, ON. N6A 5B7, Canada
Flemming, R L (bpharwoo@uwo.ca), Department of Earth Sciences, University of Western Ontario, London, ON. N6A 5B7, Canada

This research examines the relationship between the crystal structure and the chemical composition of a suite of peridotitic and eclogitic garnets. Chemical composition is one of the major controls on crystal structure, and as such, crystal structure may be used as a compliment to chemical composition. The industry-accepted method for classifying kimberlite indicator minerals (KIMs) relies on electron probe micro-analysis (EPMA). Micro X-ray diffraction (μXRD) allows rapid in situ analysis of single crystals and is compatible with a wide variety of sample formats. Approximately 600 garnet samples were analyzed from kimberlites in North America and Southern Africa. The garnets were classified based on their geochemistry, according to the scheme of Grütter et al., (2004). They were largely G9 and G10 garnets, but enough G1, G3 and G4 garnets were analyzed to define the presence or absence of valid trends; a few G5 and G12 garnets were also present. In G9 and G10 garnets a positive correlation was found between the unit cell dimension a0 and Ca substitution for Mg in the X-site, as well as between unit cell dimension a0 and Cr substitution for Al in the Y-site. In both cases the larger cations (Ca and Cr) increased the unit cell dimension. A well-defined trend was also observed between a0 and Ca in G3 and G4 garnets. Garnets from each paragenesis typically followed similar trends, regardless of locality. G9 and G10 garnets plot on a trend parallel to the pyrope (Mg3Al2Si3O12) knorringite (Mg3Cr2Si3O12) join when the unit cell is plotted versus Cr content, indicating substitution of Cr into their crystal structures. When unit cell is plotted versus Ca content G9 garnets trend from pyrope towards a knorringite-uvarovite solid solution ((Mg1.66Ca1.33)Cr2Si3O12), as expected. Surprisingly, G10 garnets trend towards a more calcic composition ((Mg1.0Ca2.0)Cr2Si3O12). This trend is not yet understood. G3 garnets follow a trend from a pyrope-almandine solid solution towards grossular, indicating substitution of Ca for Mg and Fe is the most significant control on the unit cell of G3 garnets. A simple formula was developed that accurately predicts the observed trends for G3, G9 and G10 garnets, when a0 is compared to both Ca and Cr content. This was developed based on the observation that specific cation substitutions result in a consistent change in a0 for the end member garnet pairs involving those substitutions. The difference in a0 between pyrope-knorringite is ∼0.15 Å and the same difference in a0 is found between grossular-uvarovite. Similarly, the difference in a0 between pyrope- grossular and knorringite-uvarovite is ∼0.4 Å. The G3, G9 and G10 trends can be predicted very accurately based on the concentration of both Ca and Cr as they contribute linearly to the unit cell dimension, provided a starting intercept for the unit cell dimension is known. Extensive solid solution in upper mantle garnets produces significant overlap in the unit cell dimensions of garnets from different parageneses, even those with distinct chemical compositions. This overlap limits the utility of the unit cell dimension as a single discriminator of upper mantle garnet type. However, the distinct trends in unit cell versus composition plots seen for G3, G9 and G10 garnets may provide clues to other factors such as pressure-temperature of formation. Furthermore, this technique may have application to garnets used as indicator minerals in skarns, where variation in chemical composition between almandine and uvarovite is used to vector exploration efforts.


Understanding garnet variability: Application of geometallurgy to diamonds and exploration

* Hoal, K O (khoal@mines.edu), Advanced Mineralogy Research Center, Colorado School of Mines, 1310 Maple St #240, Golden, CO 80401, United States
Appleby, S K (sappleby@mines.edu), Advanced Mineralogy Research Center, Colorado School of Mines, 1310 Maple St #240, Golden, CO 80401, United States
Stammer, J G (jstammer@mines.edu), Advanced Mineralogy Research Center, Colorado School of Mines, 1310 Maple St #240, Golden, CO 80401, United States

Peridotitic and eclogitic garnets are a fundamental component in understanding mantle petrology, diamond petrogenesis, and the ascent of mantle materials in kimberlites. They are also critical in exploration programs, as the presence of mantle garnets at the earth's surface provides an indication of dispersion from a deeply derived magmatic carrier. The composition of these garnets further is used as an indicator of diamond prospectivity, on the basis of comparison with garnet compositions known to be in some degree of equilibrium with diamonds. For mantle xenoliths and kimberlites, optical microscopy, electron microprobe analysis (EPMA), and scanning electron microscopy (SEM) are the main tools used for understanding key mineralogical and textural variability relationships. Mineralogy and texture reflect diamond genesis, metasomatic alteration, fluid migration and manifestation, volcanological processes, peridotite disaggregation, and other manifestations of mantle processes that are observable, describable, and applicable in exploration and mining. Mineralogy and texture studies lead to further questions that are better addressed by higher resolution chemical analysis of isotopes and rare earth elements, or luminescence. Understanding mineralogical and textural variability is the primary geological input for geometallurgy (geomet), the field integrating the earth sciences with the extractive industries. The framework for geomet encompasses geology, mineralogy, deposit modeling and extraction methods for the optimum value return of resources, and it relies on the fact that the mineralogy and texture of rocks influence subsequent interpretation and downstream applications. Developments in this area have been made possible by the new generation of high-speed SEM-based quantitative mineralogical instruments, enabling the statistical assessment of thousands of grains or particles, or samples, and their application to models for exploration, ore deposits, or geomet. For diamonds, this means identification and quantification of large mineralogical and textural data sets, and gives the geologist more involvement in model development. In this study, peridotitic and eclogitic garnets were examined in situ and as xenocrysts to gain understanding of the mineralogical and textural variability of the grains using SEM-based quantitative mineralogy. For concentrate garnets, the new technology presented here is the development of mineral definitions that reflect SEM counts and correlate with EPMA data. Internal compositional variability is mapped across individual grains as compared to EPMA spot analysis; designations of G10-G9 compositions, for example, are more complex when viewed in terms of individual internal grain compositional variability. The new mineral lists based on percentages of Ca-Cr count rates are compared to unknown garnets from exploration samples, and digitally categorized into bins reflecting potential diamond prospectivity or secondary alteration, as desired. The high analysis rate (approx. 150 determinations/second) means the SEM-based technique can be faster and produce more statistical information for the geologist who is making the model assessment in the field. Combined with new nontoxic mineral separation methodology in the field and software on the geologist's laptop, a great deal of interpretation can be accommodated in the field, at a reduced cost for shipping large volumes of samples to a central laboratory. Geomet for diamonds provides the mechanism for thinking of the entirety of a project, and using the geological and mineralogical information to predict process implications.


The Diamondiferous Lithospheric Mantle Underlying the Eastern Superior Craton: Evidence From Mantle Xenoliths From the Renard Kimberlites, Quebec

* Hunt, L (lchunt@ualberta.ca), University of Alberta, Department of Earth & Atmospheric Sciences, Edmonton, AB T6G 2E3, Canada
Stachel, T (tstachel@ualberta.ca), University of Alberta, Department of Earth & Atmospheric Sciences, Edmonton, AB T6G 2E3, Canada
Armstrong, J P (JArmstrong@stornowaydiamonds.com), Stornoway Diamond Corporation, Unit 116-980 West 1st Street, North Vancouver, BC V6C 2T6, Canada
Simonetti, A (simonetti.3@nd.edu), University of Notre Dame, Department of Civil Engineering and Geological Sciences, Notre Dame, IN 46556, United States

The Renard kimberlite cluster consists of nine pipes located within a 2km2 area in the northern Otish Mountains of Quebec. The pipes are named Renards 1 to 10, with subsequent investigation revealing Renards 5 and 6 to join at depth (now Renard 65). The pipes are located within the eastern portion of the Superior craton, emplaced into Archean granitic and gneissic host rocks of the Opinica Subprovince (Percival, 2007). Amphibolite grade metamorphism, locally passing into the granulite facies (Percival et al., 1994) occurred in late Archean time (Moorhead et al., 2003). Radiometric dating of the hypabyssal Renard 1 kimberlite indicates Neoproterozoic emplacement, with a 206Pb/238U model age of 631.6±3.5 Ma (2σ) (Birkett et al., 2004). A later study on the main phases in Renard 2 and 3 gave a similar emplacement, with a 206Pb/238U model age of 640.5±2.8Ma (Fitzgerald et al., 2008). This makes this kimberlite district one of the oldest in Canada, similar in eruption age to the Wemindji kimberlites (629±29Ma: Letendre et al., 2003). These events are broadly coeval with the conversion from subduction magmatism to rifting in northern Laurentia (Birkett et al., 2004). The bodies are part of a late Neoproterozoic to Cambrian kimberlite field in eastern Canada (Girard, 2001; Moorhead et al, 2002; Letendre et al., 2003) and fit into the north-east of the Eocambrian/Cambrian Labrador Sea Province of Heaman et al. (2004). To better understand the diamondiferous lithospheric mantle beneath the Renard kimberlites, 116 microxenoliths and xenocrysts were analysed. The samples were dominantly peridotitic, composed primarily of purple garnet, emerald green clinopyroxene and olivine, with a few pink and red garnets. A minor eclogitic component comprises predominantly orange garnets and lesser amounts of clinopyroxene. A detailed study on the major, minor and trace element composition of xenolith minerals is currently underway. All but three of the clinopyroxenes analysed to date plot into the on-craton garnet peridotite field of Ramsay (1992), and follow the garnet peridotite trend of Grütter (2008). Using the single pyroxene geothermobarometer of Nimis and Taylor (2000), the clinopyroxene grains fall along a 38mW/m2 model geotherm. However, the majority fall on the low pressure side of the diamond graphite transition. Initial analysis on the garnet grains show that the majority plots in the on craton lherzolite field (G9A) of Grütter et al. (2006). A smaller eclogite population is also present, along with a minor harzburgitic (G10) population. Using the manganese in garnet thermometer of Creighton (2008) the majority of grains fall in the diamond window (T>950°C). This indicates a currently unexplained disconnect between clinopyroxene and garnet geothermobarometry. The newly developed technique of in situ Pb-Pb dating of clinopyroxene xenocrysts (Schmidberger et al. 2007) was applied to the microxenoliths. Initial results indicate an age of ∼2.7 Ga for the subcratonic lithospheric mantle beneath Renard. This date is significant, coinciding with the beginning of the break up of Vaalbara and a major phase of continental crust generation. Also at 2.7 Ga, Kenorland (including the Superior Province) was formed by accretion of granitoid-greenstone terranes at convergent margins (Barley et al., 2005).