UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The rare element Little Nahanni Pegmatite Group, NWT : studies of emplacement, and magmatic evolution.. 2010

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
ubc_2010_fall_barnes_elspeth.pdf
ubc_2010_fall_barnes_elspeth.pdf
ubc_2010_fall_barnes_elspeth.pdf [ 126.05MB ]
ubc_2010_fall_barnes_elspeth.pdf
Metadata
JSON: 1.0052534.json
JSON-LD: 1.0052534+ld.json
RDF/XML (Pretty): 1.0052534.xml
RDF/JSON: 1.0052534+rdf.json
Turtle: 1.0052534+rdf-turtle.txt
N-Triples: 1.0052534+rdf-ntriples.txt
Citation
1.0052534.ris

Full Text

   THE RARE ELEMENT LITTLE NAHANNI PEGMATITE GROUP, NWT: STUDIES OF EMPLACEMENT, AND MAGMATIC EVOLUTION FROM GEOCHEMICAL AND LI ISOTOPIC EVIDENCE      by   Elspeth M. Barnes       A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY   in   The Faculty of Graduate Studies  (Geological Sciences)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)    July 2010   © Elspeth M. Barnes, 2010    ii ABSTRACT Rare element pegmatites represent some of the last stages of igneous differentiation and are influential in element redistribution in the upper crust, leading to significant enrichment/depletion of various trace elements. Research into the processes that form these intrusions increases our understanding of the geochemical evolution of silicate earth and improves the potential for successful pegmatite exploration. This study focussed on the dikes comprising the rare element Little Nahanni Pegmatite Group (LNPG), Mackenzie Mountains, northern Canadian Cordillera. These peraluminous dikes have high concentrations of several rare elements, e.g., Li (up to 14,000 ppm), Cs (up to 500 ppm), Ta (up to 190 ppm), and Rb (up to 7,500 ppm). Orientation of the dikes was influenced during emplacement (2-3 kbar, ~400-500 °C) at ~90 Ma (apatite, U-Pb) by pre-existing foliation in the strongly deformed, stratified host rock of the Fork anticlinorium (axial planar cleavage and bedding). Differences in 40Ar/39Ar dates on pegmatite minerals (muscovite 77.1±3.6 and ~80 Ma and lepidolite 65.8±0.8 Ma) indicate the presence of an elevated paleogeothermal gradient (~60°C/km). Structural and contact metamorphic evidence identify a local heat source within the anticlinorium that may have been the source chamber for the dikes. Whole rock trace element concentrations and ratios, mineralogical and textural variations, and fractionation of Li isotopic ratios (δ7Li  = -0.94‰ to +11.36‰) record a range of magmatic fractionation. Approximately 85% of the dikes are spodumene-rich, with discontinuous REEN patterns and low degrees of Li isotope fractionation, the remaining ~15% show greater magmatic fractionation, with little spodumene, and have flat or listric REEN patterns and strongly fractionated Li isotopic ratios. The replication of the REEN patterns by P and F saturation (mineral precipitation and fluid separation), illustrates the influence of flux components on the composition of late stage melts. The Li isotope composition of rapidly crystallized, co-precipitated mineral assemblages appears to show the retention of a kinetic isotopic fractionation signature; providing a potential method to assess the chemical equilibrium of the system. This integrated study advances our understanding of rare element pegmatite formation in several aspects, in particular the role of fluxes in their geochemical evolution.  iii TABLE OF CONTENTS ABSTRACT ......................................................................................................................... ii TABLE OF CONTENTS ........................................................................................................ iii LIST OF TABLES ................................................................................................................ vi LIST OF FIGURES............................................................................................................. viii ACKNOWLEDGEMENTS .................................................................................................... xii CO-AUTHORSHIP STATEMENT ......................................................................................... xiv CHAPTER 1    INTRODUCTION.......................................................................1 1.1 OBJECTIVES ............................................................................................................1 1.2 GEOLOGICAL SETTING ............................................................................................2 1.2.1     Geology ...............................................................................2 1.2.2     Previous work ......................................................................4 1.3 GRANITIC AND RARE ELEMENT PEGMATITES: FORMATION AND USES ...................5 1.4 LITHIUM - ELEMENT AND STABLE ISOTOPES............................................................9 1.4.1 Introduction ........................................................................9 1.4.2 Chemistry..........................................................................11 1.4.3 Analysis ............................................................................12 1.5 OVERVIEW OF THE DISSERTATION.........................................................................13 1.6 CONTRIBUTIONS TO THE PROJECT .........................................................................14 1.7 REFERENCES.........................................................................................................16 CHAPTER 2   THE LITTLE NAHANNI PEGMATITE GROUP, NWT: UPPER CRUSTAL DIKE EMPLACEMENT IN AN OROGENIC REGION..........................23 2.1 INTRODUCTION .....................................................................................................23 2.2 REGIONAL GEOLOGICAL SETTING ........................................................................24 2.3 PREVIOUS WORK ..................................................................................................29 2.4 GEOLOGY .............................................................................................................30 2.4.1     Stratigraphic units..............................................................30 2.4.2     Bedrock structure...............................................................30 2.4.3     Relationship between structure and pegmatites.................37 2.4.4     Metamorphism...................................................................39 2.4.5     Pegmatite mineralogy and primary textures ......................41 2.5 ANALYSIS AND INTERPRETATION .........................................................................42 2.5.1     Analytical methods ............................................................42 2.5.2     Fluid inclusion microthermometry ....................................42 2.5.3     Pressure-temperature constraints .......................................45 2.5.4     U-Pb geochronology of the dikes ......................................45 2.5.5     40Ar/39Ar geochronology of the dikes................................48 2.5.6     Rb/Sr geochronology and initial Sr ratio of the dikes .......53 2.6 DISCUSSION ..........................................................................................................53 2.6.1     Origins of the magma ........................................................56 2.6.2     Geological setting and cooling history ..............................57  iv 2.6.3     Rb/Sr results ......................................................................62 2.6.4     Dike formation...................................................................63 2.6.5     Primary textures reflecting conditions of crystallization...64 2.6.6     Tectonic implications.........................................................66 2.7 CONCLUSIONS.......................................................................................................67 2.8 ACKNOWLEDGEMENTS .........................................................................................68 2.9 REFERENCES.........................................................................................................70 CHAPTER 3    GEOCHEMICAL EVIDENCE OF LATE STAGE MAGMATIC FRACTIONATION IN THE LITTLE NAHANNI PEGMATITE GROUP, NWT....77 3.1 INTRODUCTION .....................................................................................................77 3.2 GEOLOGICAL SETTING..........................................................................................79 3.2.1     Location .............................................................................79 3.2.2 Mineralogy and textures ....................................................82 3.2.2.1 Petrography .......................................................83 3.3 SAMPLING AND ANALYTICAL METHODS ..............................................................95 3.3.1     Sampling............................................................................95 3.3.2     Sample preparation for ICP-MS analysis ........................107 3.3.3     ICP-MS analyses .............................................................109 3.3.4     Rietveld analyses .............................................................109 3.4 RESULTS .............................................................................................................110 3.4.1     Whole rock major elements .............................................110 3.4.2 Whole rock minor and trace elements .............................110 3.4.2.1 Rare earth elements .........................................116 3.4.2.2 Element ratios..................................................121 3.4.3     Mineral separate minor and trace elements .....................123 3.4.4 Rietveld method results ...................................................125 3.5 DISCUSSION ........................................................................................................130 3.5.1     Comparisons of the extent of magmatic fractionation.....130 3.5.2 Rare earth element fractionation......................................133 3.5.3 ‘Immobile’ element pair fractionation.............................136 3.5.4 Crystal/melt and fluid/melt fractionation ........................140 3.6 CONCLUSIONS.....................................................................................................148 3.7 ACKNOWLEDGEMENTS .......................................................................................149 3.8 REFERENCES.......................................................................................................150 CHAPTER 4   STRONG LI ISOTOPE FRACTIONATION IN THE HIGHLY EVOLVED LITTLE NAHANNI PEGMATITE GROUP, NWT....................159 4.1 INTRODUCTION ...................................................................................................159 4.2 GEOLOGY ...........................................................................................................161 4.3 SAMPLES AND ANALYTICAL METHODS ..............................................................164 4.3.1     Sample preparation ..........................................................168 4.3.2     Column chemistry............................................................170 4.3.3     Instrumental set-up ..........................................................172 4.3.4 Analyses...........................................................................172 4.3.4.1 Notation...........................................................172 4.3.4.2 Method protocol ..............................................175  v 4.4 RESULTS .............................................................................................................177                            4.4.1 δ7Li results from whole rock pegmatite, mineral separates and                            regional granitic intrusions ...........................................................177                            4.4.2 Accuracy and precision........................................................183 4.4.2.1 UBC analyses ...................................................183 4.4.2.2 UMD analyses..................................................184 4.5 DISCUSSION ........................................................................................................184 4.5.1 Whole rock analyses: evidence of magmatic and Li isotopic fractionation..................................................................................184  4.5.1.1 UMD analyses.................................................188 4.5.2 Comparison of LNPG δ7Li values with literature ...............189                                  4.5.3 Li isotope fractionation in minerals.....................................189 4.5.3.1 Calculation of whole rock δ7Li signatures assuming equilibrium conditions ...................189 4.5.3.2 Effect of non-equilibrium conditions on mineral δ7Li values......................................................196                            4.5.4 Timescale of pegmatite consolidation and Li diffusion ......199 4.6 CONCLUSIONS.....................................................................................................200 4.7 ACKNOWLEDGEMENTS .......................................................................................201 4.8 REFERENCES.......................................................................................................203 CHAPTER 5    CONCLUSIONS......................................................................210 5.1 EXPLORATION TOOLS..........................................................................................216 5.2 NOTABLE PETROGENETIC ASPECTS OF THE LNPG ..............................................220 5.3 IMPLICATIONS FOR LI STUDIES............................................................................221        5.4 FUTURE RESEARCH..............................................................................................222 5.5 REFERENCES.......................................................................................................225   APPENDICES...............................................................................................228 Appendix 1 Fluid inclusion microthermometry ........................................................229 Appendix 2  U-Pb analysis .........................................................................................230 Appendix 3 Rb/Sr analysis ........................................................................................232 Appendix 4     List of whole rock samples of pegmatitic material from the LNPG collected by Wengzynowski (2002) ......................................................233 Appendix 5   Compilation of δ7Li values from this study and from similar rock types, minerals and standards reported in the literature...................................239 Appendix 6 Composite annotated image of thin sections (cross-polarised) cut across the width of a ~80 cm width dike (using cross-polarized light), sample P389765 .................................................................................................240 Appendix 7 Annotated images of thin sections (photographed using cross-polarized light) from selected LNPG samples displaying various textural and mineralogical details..............................................................................241       vi LIST OF TABLES  CHAPTER 2 Table 2.1 Variation in the lithologies of the Hyland Group crosscut by the LNPG pegmatites ........................................................................31  Table 2.2 Lines of investigation, analytical methods, samples and minerals used and summary of the results obtained...................................44  Table 2.3 Microthermometry summary for fluid inclusions hosted in pegmatitic quartz crystals from sample EB02.............................46  Table 2.4 Table of analytical data, weight, U and Pb concentrations and isotopic ratio results obtained ......................................................49  Table 2.5 Sample information and results from 40Ar/39Ar analyses of minerals and whole rock samples ................................................52  Table 2.6 Sample information and results from Rb/Sr analyses of pegmatitic micas ............................................................................................54  Table 2.7 Mineral sizes, isotopic systems, and relevant closure temperatures of geochronological analyses undertaken for this study..............58  CHAPTER 3 Table 3.1 General petrographic observations from the LNPG samples ......85  Table 3.2 Major element compositions of whole rock LNPG samples .......98  Table 3.3 Trace element compositions of the whole rock LNPG samples, the Macusani glass samples and the LNPG mineral separates........101  Table 3.4 Results from Rietveld method and x-ray powder diffraction data analysis of selected whole rock LNPG samples ........................126  Table 3.5 Measured LNPG concentrations, published REE partition coefficients and the calculated REE concentration after fractionation of the phases .........................................................143  CHAPTER 4 Table 4.1 Table of modal mineral abundance as determined from several LNPG whole rock samples correlated with their δ7Li values and predominant mineralogy............................................................165  Table 4.2 List of eluent compositions for cationic exchange columns 1 and 2 for the separation of Li, and per sample volumes......................171  Table 4.3 Compositions of eluents and their constituents measured by HR- ICP-MS......................................................................................173  vii  Table 4.4 Typical settings achieved after tuning the Nu 021 Plasma ICP-MS in preparation for Li isotope analysis ........................................174  Table 4.5 Results from matrix effect tests, determined by doping L-SVEC and Puratronic® with varying amounts of Al............................178  Table 4.6 Results of δ7Li analysis from PCIGR (UBC) and the University of Maryland....................................................................................179  Table 4.7 Whole rock δ7Li values calculated from the modal mineralogy of the LNPG whole rock samples determined by the Rietveld method and the δ7Li values of rock-forming minerals ...........................191  CHAPTER 5 Table 5.1 Illustrations of the rare element pegmatite model from the LNPG ...................................................................................................217                                viii LIST OF FIGURES  CHAPTER 1 Figure 1.1 Simplified map of the Selwyn Basin and Tintina Gold Belt in Yukon and Northwest Territories and regional geological features of the study area.............................................................................3  Figure 1.2 Schema of generalized compositional and textural evolution of Li- rich, rare-element pegmatites.........................................................7  Figure 1.3 P-T fields of host rocks for rare element pegmatites within the upper crust. ....................................................................................8   CHAPTER 2 Figure 2.1 Geologic map of the region encompassing the study area ..........25  Figure 2.2 Geologic map of the study area showing areas with the highest density of pegmatites, bedding and pegmatite orientations and the location of the cross sections measured for the study..................26  Figure 2.3 Cartoon cross section of the boundary between the Selwyn Basin strata (medium grey) and the crystalline basement of ancestral North America .............................................................................27  Figure 2.4 Illustrations of macro (outcrop) scale folds and the relationship with the pegmatitic dikes .............................................................32  Figure 2.5 Annotated photographs of the varied response to folding in the common rock types of the study area ..........................................33  Figure 2.6 Photograph of the south side of cirque 3 measured on traverse D- D’ .................................................................................................35  Figure 2.7 Equal-area projections displaying the structural data collected throughout the study area.............................................................36  Figure 2.8  Photographs illustrating features of the pegmatitic dikes………38  Figure 2.9 Examples of typical contact metamorphic textures and fabrics in the study area ...............................................................................40 .........................................................................................................  Figure 2.10 Photographs showing some of the common magmatic textures occurring in the dikes of the Little Nahanni Pegmatite Group....43  Figure 2.11 Pressure and temperature conditions of dike crystallisation obtained from quartz by fluid inclusion microthermometry........47   ix Figure 2.12 Geochronological results from U-Pb and 40Ar/39Ar data.............50  Figure 2.13 Geochronological results from Rb/Sr data ..................................55  Figure 2.14 Geochronological results from the LNPG dikes plotted against temperature and depth constraints ...............................................61   CHAPTER 3 Figure 3.1 Geologic map of the study area showing areas with the highest density of pegmatites, and numbered locations for the whole rock samples ........................................................................................80  Figure 3.2 Photographs of the study area (view to northeast) and examples of pegmatites and crystallisation textures ........................................81  Figure 3.3 Examples of hand samples from the LNPG pegmatites ..............84  Figure 3.4 Petrographic illustration of quartz, feldspar and mica.................86  Figure 3.5 Petrographic illustrations of spodumene .....................................88  Figure 3.6 Petrographic illustrations of mica (I)...........................................90  Figure 3.7 Petrographic illustrations of mica (II)..........................................92  Figure 3.8 Petrographic illustration of disequilibrium textures ....................94  Figure 3.9 Cartoon interpretation of the crystallisation sequence of the main rock-forming minerals of the LNPG pegmatites .........................96  Figure 3.10 Photograph of the Macusani glass cobble obtained for this study from the Smithsonian Institution ...............................................108  Figure 3.11 Total Alkali-Silica diagram (TAS) classifying the LNPG whole rock samples, illustrating their typically granitic chemical composition ...............................................................................111  Figure 3.12 Shand’s index diagram (Al2O3/CaO + Na2O + K2O)(molar) displaying the peraluminous composition of the whole rock LNPG samples. .....................................................................................112  Figure 3.13 Diagram illustrating the decreasing trends of A. CaO and MnO. B. Al2O, Na2O and K2O and C. P2O5, F and H2O+ with SiO2 content of the LNPG whole rock samples..............................................113  Figure 3.14 Upper continental crust-normalized (Taylor and McLennan 1985) trace element patterns of the LNPG whole rock samples..........114   x Figure 3.15 Plot of average whole rock LNPG trace element compositions normalized to upper continental crust values (Rudnick and Gao 2003) ..........................................................................................115  Figure 3.16 Rare earth element patterns of the whole rock LNPG samples, normalized to C1 chondrite (Sun and McDonough 1989) separated into 4 groups ..............................................................................117  Figure 3.17 Spatial distribution of the various LNPG whole rock REEN patterns across the study area ....................................................119  Figure 3.18 Two geochemical methods used to differentiate the three groups of distinctive REE patterns ............................................................120  Figure 3.19 Assessment of the degree of magmatic fractionation in the whole rock LNPG samples...................................................................122  Figure 3.20 Rare earth element patterns of LNPG mineral separates, normalized to C1 chondrite (Sun and McDonough 1989).........124  Figure 3.21 Pie chart representations of the whole-rock mineralogy of several LNPG samples determined by the Rietveld method..................128  Figure 3.22 Comparison of the degree of magmatic fractionation observed in the LNPG with examples of Cretaceous magmatism................131  Figure 3.23 LNPG trace element and REE data compared with samples of regional granite and felsite from the O’Grady Batholith...........132  Figure 3.24 Compilation of K/Rb against Cs ppm and K/Rb against Nb/Ta (used as proxies for magmatic differentiation) for LNPG whole rock samples, bulk crust compositions and various evolved granites and pegmatites..............................................................134  Figure 3.25 Compilation of Y/Ho against Zr/Hf and Hf ppm for LNPG whole rock samples, bulk crust compositions and various evolved granites and pegmatites..............................................................138  Figure 3.26 Diagram showing the variation in CaO wt. % against P2O5 wt. % between the three LNPG whole rock REEN groups. .................141  Figure 3.27 Results of modeling crystal fractionation and fluid/melt fractionation of LNPG REEN abundances .................................145  Figure 3.28 Results of modeling LNPG REEN abundances with crystal fractionation of fluoride and monazite ......................................147  CHAPTER 4 Figure 4.1 Geologic map of the Little Nahanni Pegmatite Group area. .....162  xi  Figure 4.2 Chondrite normalized REE patterns of the LNPG whole rock samples illustrating three groups of distinctive patterns............167  Figure 4.3 Whole rock δ7Li values of the LNPG samples plotted against selected trace element concentrations and ratios .......................181  Figure 4.4 Map of the Little Nahanni Pegmatite Group area showing the distribution of the whole rock samples and their δ 7Li values. ..182  Figure 4.5 δ 7Li histograms comparing results from this study with literature values .........................................................................................190  Figure 4.6 Pie chart representations of the whole-rock mineralogy of several LNPG samples determined by the Rietveld method…………..194  Figure 4.7 Cartoon of the different distributions of 6Li and 7Li during mineral growth from a silicate melt ........................................................198  CHAPTER 5 Figure 5.1 Cartoon of the geological setting of the Little Nahanni Pegmatite Group .........................................................................................212  Figure 5.2 Representation of the relationship between the three distinct REEN whole rock patterns, and their approximate δ7Li values……….215                          xii ACKNOWLEDGEMENTS  The completion of this thesis marks the conclusion of one of the most academically and personally challenging periods of my life. But definitely the toughest part was this paragraph, until I found someone had already said, very eloquently, what I wanted to say… ‘May your trails be crooked, winding, lonesome, dangerous, leading to the most amazing view. May your mountains rise into and above the clouds.’. Edward Abbey (1927-1989)  This was not a solitary effort and I would like to acknowledge the role of others in helping me over the last few years. Foremost, my thanks go to my academic supervisory committee: Lori Kennedy, Jim Mortensen and especially Dominique Weis, who guided me through the process. They advised, pushed, challenged, encouraged, cajoled and supported me (financially and academically), and I could not have accomplished this without them. In the same breath I want to thank a one-man personal advisory committee, Philippe Tortell, whose wise words kept me flowing around obstacles and heading in the right direction. Acknowledgement goes to Lee Groat for financial support and access to the LNPG.  Any errors or misconceptions in my dissertation are my own responsibility; any really good stuff was probably the result of discussion or collaboration with others. I was fortunate to have been surrounded by generous, inquisitive and accomplished academicians, researchers and technicians during my time here, many I hope, have ended up as friends. Past and present colleagues from the PCIGR, UBC, are warmly thanked for their professional training, help and advice throughout the research for this dissertation. Especially Bruno Kieffer who undertook all aspects of Rb-Sr geochronological analysis and provided clean-lab technical assistance and training, Jane Barling for her advice and all aspects of MC-ICP-MS training, Bert Mueller for help with the HR-ICPMS and GFAAS analysis and Vivian Lai (assistance with chemistry procedures), Rich Friedman (U-Pb analysis), Janet Gabites (data reduction), Hai Lin (mineral separation), Wilma Pretorius (trace element analysis and data reduction) and Tom Ullrich (40Ar-39Ar analysis). From the EOS X-Ray Diffraction Facility, Jenny Lai and Elisabetta Pani are thanked for preparing and analysing samples for Rietveld analysis, undertaking the data reduction and helping me with data interpretation. Significant advice and comment was provided by Greg Dipple, Ken Hickey, Mati Raudsepp, Kelly Russell, Dick Tosdal from UBC. Many thanks go to Dan Marshall, from Simon Fraser University for training in fluid inclusion analysis. And I would especially like to thank James Scoates, for always answering my questions, for always asking how the battle was going and for letting me camp out in his lab while I finished up.  Considerate and able help in the field was provided by: Tashia ‘Pipes’ Dzikowski, Allison Brand, Peter Weir, Marie-Eve Caron and Anita Lam and they are all sincerely thanked for their efforts. Great Slave Helicopters are acknowledged for their friendly and professional field support, and Inconnu Lodge, YT, is warmly thanked for the enjoyable stopovers, additional helicopter field support and the incredible opportunity to fly through the Cirque of the (unforgettable) Unclimbables.   xiii I am very lucky to have had a strong safety net as I wobbled my way across the Ph.D. high-wire and I want to acknowledge the interest, support and good humour of my siblings (Sue, Alison and Hugh), my close friends at UBC (especially Caroline- Emmanuelle Morisset, Nick and Kylie Williams, Gill and Andy Dean, Alyssa Shiel, Inês Nobre Silva, the inspiring R-E Farrell, Brian Hunt, Andrea Cade, Claire Chamberlain, Katrin Breitsprecher, Amber Henry, Kirsten Rasmussen, Luke Beranek, Stefan Wallier and Desiree Tommasi) and friends from the world outside (Claudia Kinmonth, Sarah Gleeson, Charlotte Smith and Hendrik Falck). And many thanks go to colleagues from the disbanded EOS East Lepisma saccharina appreciation society, Gareth Chalmers, Gary Clarke, Roger Pieters, Larysa Pakhomova and Chris Payne for keeping the daily routine enjoyable, often 7 days a week.  I would finally like to mention my friends Ron and Pat Cavell. Pat encouraged and empathized with me throughout my thesis and very sadly she passed away just before we could celebrate its completion. Here’s to you, Pat.                                  xiv CO-AUTHORSHIP STATEMENT  Chapters 2, 3 and 4 are all co-authored by members of the supervisory committee who provided technical editorial input from the developmental stage onwards. Additional technical input was provided by members of UBC staff and researchers from other institutions as described.  Chapter 2   The Little Nahanni Pegmatite Group, NWT: Upper crustal dike emplacement in an orogenic region Authors: Elspeth M. Barnes, Lori A. Kennedy, James Mortensen, Dan Marshall and Lee A. Groat  The lead author: • designed and planned the structural study with LAK • spent three field seasons mapping, taking notes, and collecting rock samples and structural data (logistical assistance provided by LAG). • undertook literature research • interpreted structural data (with advice from LAK) • measured all fluid inclusions (interpretation of the measurements was undertaken by DM) • prepared rock samples for geochronological analysis and thin section preparation • formulated the interpretation of the regional geothermal history from the geochronological results (with advice from JM) • made all petrographic observations (LAK provided significant assistance with the interpretation) • was responsible for all figures  Field assistance provided by: Tashia Dzikowski, Allison Brand, Peter Weir, Marie-Eve Caron.   PACIFIC CENTRE FOR ISOTOPIC AND GEOCHEMICAL RESEARCH Bruno Kieffer undertook all aspects of Rb-Sr geochronological analysis. Rich Friedman undertook all aspects of U-Pb geochronological analysis. Tom Ullrich undertook all aspects of 40Ar-39Ar geochronological analysis with some assistance in data reduction and interpretation from Janet Gabites.  Chapter 3    Geochemical evidence of late stage magmatic fractionation in the Little Nahanni Pegmatite Group, NWT Authors: Elspeth M. Barnes, Dominique Weis and Lee A. Groat  The lead author: • prepared many samples for chemical analysis (sample digestion and dilution), LAG provided access to most samples • undertook about half HR-ICP-MS data collection and data reduction • took all GFAAS data collection and reduction • undertook literature research • interpreted all data  xv • was responsible for all figures • Dominique Weis contributed to many aspects of the study and manuscript  PACIFIC CENTRE FOR ISOTOPIC AND GEOCHEMICAL RESEARCH Wilma Pretorius provided initial geochemical data including chemistry, analysis and data reduction.  EOS X-RAY DIFFRACTION FACILITY Jenny Lai undertook sample preparation and Rietveld analysis and Elisabetta Pani provided data reduction.  LABORATOIRE DE GEODYNAMIQUE DES CHAINES ALPINES (LGCA), Université Joseph Fourier, Grenoble, France Catherine Chauvel and her team; Christèle Poggi and Sarah Bureau, provided additional geochemical data including chemistry, analysis and data reduction.   Chapter 4  Strong Li isotope fractionation in the highly evolved Little Nahanni Pegmatite Group, NWT Authors: Elspeth M. Barnes, Dominique Weis and Lee A. Groat  The lead author: • formulated the experiment • undertook literature research • developed the method for Li isotopic analysis, based on a published format (Jeffcoate et al. 2004) • set up chemical procedures and coordinated acquisition of labware • undertook all aspects of sample preparation and chemistry, most samples provided by LAG • undertook all data collection and data reduction • interpreted results • was responsible for all figures • Dominique Weis contributed to many aspects of the study and manuscript  PACIFIC CENTRE FOR ISOTOPIC AND GEOCHEMICAL RESEARCH Bruno Kieffer advised on technical aspects of geochemical procedure Jane Barling provided significant technical expertise, designed the analytical protocol and training on the MC-ICPMS and advised on the analytical interpretation.  UNIVERSITY OF MARYLAND Roberta Rudnick provided additional Li isotopic analysis and data reduction.      CHAPTER 1: INTRODUCTION  1.1 OBJECTIVES This dissertation investigates the processes involved in the formation of rare element pegmatites. These rocks are examples of extreme magmatic fractionation and are an intrinsic part of the ongoing process of planetary differentiation as they re-distribute some of the more incompatible elements (e.g., Sn, Li, Cs, Ta, Nb) to the upper levels of the crust. On a human scale this re-distribution makes an abundance of otherwise rare elements (many of which are vital to high technology industries) accessible to exploration and potential economic development. The subject of this study was the rare element Little Nahanni Pegmatite Group (LNPG) located in the Mackenzie Mountains, Northwest Territories, Canada. These dikes are highly fractionated granitic pegmatites and have been categorized by Černý and Ercit (2005) as members of the albite-spodumene subclass of Li-enriched rare element pegmatites (REL-Li; Rare-ELement-Li). Encompassed within an area of ~2.5 km ×  12 km, more than 200 sub-vertical pegmatitic dikes of the LNPG are exposed throughout an elevation of ~ 400 m. This dissertation focuses on: • the influence of geological structures and setting to the location and emplacement of the LNPG dikes, • the whole rock geochemistry of the dikes, • and the variation in their Li isotopic signatures. The location of the LNPG in the rugged terrain of the Mackenzie Mountains benefited the first two aspects of the study. The good exposure of the pegmatite dikes that cross-cut a series of valleys was a significant factor in the accomplishment of the structural study. It enabled detailed mapping of the pegmatites and their relationship to the host rock structures. The exposure also assisted the collection of large bulk samples (Wengzynowski 2002) facilitating the accurate geochemical analysis of the pegmatites despite their coarse crystallinity. To date, most studies of granitic pegmatites have focused on minerals whereas whole rock pegmatite analysis is much less common. Strong fractionation of Li isotopes between solid and fluid phases makes the application of this isotopic system appropriate to the study of processes occurring in the 1 crystallisation of hydrous magma. With recent improvements to instrumentation, Li isotopes have begun to emerge as a tool for assessing the degree of magmatic fractionation and equilibrium in geochemically evolved silicate melts. A significant part of this project required setting up the procedures for chemical preparation and analysis of Li isotopes in order to measure the Li isotopic ratios of whole rock pegmatites, mineral separates and local granitic intrusions. This method development allowed us to examine Li isotopic variations associated with the geochemistry of the rocks and minerals.  1.2 GEOLOGICAL SETTING 1.2.1 Geology The LNPG lies in the Selwyn Basin area of the Northern Cordillera (Fig 1.1). The host rocks for the pegmatitic dikes are the Precambrian to Lower Cambrian Yusezyu and Narchilla formations of the Hyland Group, siliciclastic and carbonate passive margin facies of the Selwyn Basin; a deep-water sedimentary basin initially formed by the rifting of the Rodinia supercontinent, that persisted from late Precambrian to Middle Devonian time. The eastern margin of the basin migrated further eastwards with time resulting in deeper water offshelf strata commonly overlying shallow water shelf rocks in the study area (Gordey and Anderson 1993, Cook et al. 2004). The Yusezyu Formation, consisting of several km’s depth of relatively shallow water sedimentary rocks, is interpreted to represent submarine fan clastic deposits (Gordey and Anderson 1993). The conformable, overlying Narchilla Formation is composed of deeper water, more carbonate-rich shales. Deep-water deposition continued through the Cambrian and most of the Devonian period before sedimentation flow directions shifted in the Late Devonian period. The March Fault is a significant, long-lived fault east of the LNPG that separates the Precambrian to Lower Cambrian Narchilla and Gull Lake Formations from the Lower Cambrian Vampire Formation to the northeast (Gordey and Anderson 1993). Mountain building replaced sediment accumulation in the Selwyn Basin area in the Late Permian-Early Triassic period (ca. 250 Ma; Beranek and Mortensen 2007). The accretion of the Yukon-Tanana terrane to the west initiated an extended compressional period forming, in the study area, the tightly folded, ridge-forming Fork anticlinorium (~10 km southwest of the March Fault).  After 100 Ma an extensional, dextral 2 British Columbia Al as ka Yukon Northwest Territories Selwyn Basin Denali Fault Tintina Fault 200 km 140° 62° 61° 60° 134° 132°136° 130° 128°138° Tintina Gold Belt O’Grady Batholith ? Fork anticlinorium March Fault 50 km Yukon Northwest Territories Selwyn Basin 130° 128° 62° 63° Extent of the Tintina Gold Belt Area covered by the Selwyn Basin Mid-Cretaceous Tungsten Plutonic Suite (TPS) Mid-Cretaceous intrusions (not TPS) B Study area Figure 1.1 Belt (striped) across eastern Alaska and Yukon to the western part of the Northwest Territories. Black lines are faults, thrust faults have teeth on up thrust side. Grey line is an anticlinal axis. B. (inset of A) shows the distribution of mid Cretaceous intrusions, including members of the Tungsten Plutonic Suite (shown in darker grey) and the O’Grady Batholith. The March Fault and the Fork anticlinorium are major structural elements that occur in the study area (shown as a white box) and are mentioned in the text. Redrawn after Hart 2004 with addition data from Gordey and Anderson 1993 and Mortensen 2000. Simplified map of the Selwyn Basin and Tintina Gold Belt in Yukon and Northwest Territories and regional geological features of the study area. A. Extent of the Selwyn Basin (shaded) and the Tintina Gold et al. et al. A Canada 500 km 3 transcurrent setting prevailed in the western part of the Selwyn Basin (~ 85 Ma; Mair et al. 2006). Reactivation of the March Fault occurred around this time, although the sense of movement is undetermined (Hart and Lewis 2006). The study area lies within the bounds of the Tintina Gold Belt (Fig. 1.1), a discontinuous belt of mineralized Cretaceous intrusions that runs several hundred kilometres across the Yukon Territory and into contiguous Alaska (Mortensen et al. 2000, Hart et al. 2004, Rasmussen et al. 2006). Within a 50 km radius of the LNPG typically small intrusions (<5.4 km2) commonly associated with aplitic or pegmatitic dikes have been previously described as the Selwyn Plutonic Suite, (101 to 93 Ma; Gordey and Anderson 1993) and more recently categorized as members of the Tungsten Suite (94-99 Ma; Rasmussen et al. 2006), highly evolved, areally small intrusions which overlie larger and deeper Tay River -Tungsten and Tay River plutons (Rasmussen et al. 2006, with categories modified from Mortensen et al. 2000 and Hart et al. 2004). The more extensive hornblende-bearing O’Grady Batholith (~ 270 km2; U-Pb, titanite; 92.5 +/- 1.8 Ma; Mortensen, unpublished data) lies ~90 km north of the LNPG. This intrusion is a member of the Tombstone Suite (90-94 Ma; Rasmussen et al. 2006), which tends to be more mafic than the Tungsten Suite with a mantle component. The Nd and Sr isotopic signatures of the Tombstone Suite indicate high levels of crustal contamination, although geochemical variations, east to west, suggest a change in the composition of the source material (Lang 2000). The O’Grady batholith incorporates a 1 km long area of Li mineralization within a zone of pegmatitic granite (Ercit et al. 2003) and is the only other example of high Li abundance in the area apart from the LNPG.  1.2.2 Previous Work The LNPG dikes were discovered in 1961 by D.C. Rotherham of the Canada Tungsten Mining Corporation Ltd. (Rotherham 1962). The recognition of the unusual mineralogy and abundance of strategic elements led to several mineral claims being staked. These lapsed when the exploration focus of the area shifted to the discovery of the Cantung tungsten mine at around the same time. Elevated Li prices in the late 1970s encouraged exploration and staking by Canadian Superior Exploration Ltd (Beavon 1977; Ahlborn 1977), Cominco Limited 4 (Ahlborn 1979, 1980) and Canamera in 1994. These claims were allowed to lapse, and by 2005 the LNPG area was covered by eight adjoining claims. One claim (Cali1) is leased to V. H. Ahlborn, while War Eagle Mining Company Inc. holds the other seven claims (Wengzynowski 2002; news release Feb. 17, 2005). In August 2009, War Eagle entered into an agreement with VM Exploration, who currently have an option to earn up to 80% interest in the property, in consideration of VM Exploration spending $2M before Fall 2012 (management’s discussion and analysis, prepared as of Feb. 26, 2010). From 1977 to 1982, the Geological Survey of Canada undertook a regional geology study (Gordey and Anderson 1993). Groat et al. (1994) outlined the state of pegmatite research in the Northern Cordillera and described the mineralogy and the textural zonation across individual dikes. Mauthner et al. (1994) and Mauthner (1996a and b) undertook mineralogical, geochemical, and geochronological studies of the dikes with additional mineralogical studies undertaken on cassiterite and micas by Rollo (1999) and Pemberton (2002). The most recent publications on the LNPG are a comprehensive survey of the geology and mineralogy by Groat et al. (2003) and an overview of the LNPG and its economic potential written as part of a Mineral and Energy Resource Assessment (MERA) study (Barnes et al. 2007).  1.3 GRANITIC AND RARE ELEMENT PEGMATITES: FORMATION AND USES Pegmatites are the product of magmatic differentiation and most have an overall granitic composition with a basic mineralogy of quartz, feldspar and mica. Granitic pegmatites are generally accepted to be differentiates of magma originally derived from small amounts of partial melting of an undepleted crustal source  (e.g., Černý 1991a), although anatexis without fractionation has also been proposed (Stewart 1978). Several components that can influence pegmatite formation, such as Cl, F, B, Li, P and H2O are only available in abundance in previously unmelted rocks as they are among the first to be released into the melt. Together with other incompatible elements such as Cs and Na they alter the structure of the silicate melt restricting polymerization and crystal nucleation, decreasing viscosity and the liquidus and solidus temperatures, promoting significant undercooling and enabling the melt to remain fluid to < 400° C (e.g., Swanson 1977, Manning 1981, London 1997, Swanson and Fenn 1992, Sowerby and Keppler 5 2002, Sirbescu and Nabelek 2003).  Given the right circumstances, this low viscosity magma can separate and travel rapidly from the plutonic source (Baker 1996) and be driven by its own buoyancy into the country rock. The evolved magma that forms pegmatitic dikes can occur within a magma chamber (as observed in the O’Grady Batholith (Ercit et al. 2003) and several of the smaller regional intrusions close to the LNPG (Gordey and Anderson 1993)) or can escape the main magma body (e.g., the LNPG) and propagate for up to several kilometres into the country rock (Černý 1991a). And during propagation magmatic fractionation can continue resulting in regional compositional zonation (Fig. 1.2; Černý 1991a). The LNPG dikes are members of the REL-Li class and the albite spodumene subclass of pegmatite (Černý and Ercit 2005). These are highly fractionated granitic pegmatites that are typically enriched in Li, Rb, Cs, Be, Ga, Mn, and Y, with Ta > Nb, strongly depleted in rare earth elements (REE) and emplaced into low to moderate pressure (i.e., ≤ 2-4 kbar), upper greenschist facies environments (Fig. 1.3; Černý, 1991a and b). Granitic pegmatites have been called ‘the building blocks for our society’ (London 2005), as they can contain several component minerals and rare elements essential in household products and modern technology. Under suitable conditions magmatic fractionation can concentrate some rare elements in granitic pegmatites up to many hundred times the average crustal abundance. For example, the average crustal concentration of Li is ~16 ppm (Rudnick and Gao 2003), at LNPG Li can exceed 14,000 ppm. Caesium is also highly concentrated in the LNPG (2 ppm average crustal concentration, Rudnick and Gao 2003; up to 500 ppm in the LNPG) and like Li Cs has diverse industrial applications. Lithium is currently used as an alloy in metal production, a flux facilitating lower temperature melting in high-strength ceramics and glass, and a constituent of high-temperature lubricants and pharmaceuticals. The most recent interest in Li is its potential in rechargeable energy storage for cellular phones, digital cameras and computers, and as Li-ion batteries in the increasingly significant field of electric and hybrid vehicles. Applications for Cs include: atomic clocks, drilling fluid, scintillation counters, gamma and X-ray detectors, as a ‘getter’ in vacuum tubes, fibre optics, night vision equipment, photoelectric cells and smoke detectors. Oxide minerals also occurring in granitic pegmatites can contain high concentrations of other ‘rare elements’ (e.g., Ta, 6 Decreasing P and T Barren Be Be, Nb, Ta Li, Be, Ta, Nb Li, Cs, Be, Ta, Nb Increasing fractionation Increasing volatile and incompatible element content Complexity of zonation Extent of replacements Granite source Pe gm at ite pr op a ga tio n u p to se ve ra lk m ’s Pe gm at ite co m po sit io na le vo lu tio n o cc u rr in g du rin g pr op a ga tio n Figure 1.2 Schema of generalized compositional and textural evolution of Li-rich, rare element pegmatites. Redrawn and revised from Trueman and erný (1982).È 7 02 4 6 8 400200 600 800 10 25 °C/km 60 °C/km andalusite kyanite greenschist zeolite 0 T °C km 1 2 3 kbar 0 12 14 4 5 lawsonite-albite-chlorite prehnite- pum pellyite Rare element pegmatite field petalite spodumene amphibolite s il li m a n it e w e t g ra ni te so lid us Figure 1.3 Typical P-T field of the host rocks for rare element pegmatites within the upper crust. Al SiO triple point from Pattison (1992), spodumene-petalite reaction boundary and rare-element pegmatite field redrawn from London (2008) after London (1984) and erný (1991a) respectively; metamorphic facies from Pattison and Tracy (1991); wet granite solidus from Brown (2002). 2 5 È 8 Nb and Sn) that are used in capacitors, alloys and electronic devices, including televisions, computers and toys. However, commercially viable examples of rare element granitic pegmatites tend to have more extensive monomineralic zones than the LNPG (e.g., Tanco mine, Bernic Lake, Manitoba, Canada). And although Li (and other elements) can be highly concentrated in e.g., silicate and phosphate minerals, rare element granitic pegmatites are a minor commercial source of Li, due to the complicated processing of silicate material and the economic viability of other sources (brines, clays and evaporite sediments).  1.4 LITHIUM: ELEMENT AND STABLE ISOTOPES 1.4.1 Introduction Lithium is the third lightest element, with an atomic weight of 6.941 (Coplen et al. 2002). It is moderately incompatible in igneous fractionation processes and has an abundance of 1.6 ppm within bulk silicate Earth (BSE, McDonough 2003), <50 ppm within the upper crust (24 ppm, Rudnick and Gao 2003; 35 ± 11ppm Teng et al. 2004), rising to >10,000 ppm in the rare element pegmatites of this study. Lithium is an alkali metal with an electron configuration of 1s2 2s1 that occurs only in a 1+ ionic state as it has a high second ionization energy (7298.1 kJ·mol−1). It has an ionic radius of 6.8 ×  10-2 nm in tetrahedral coordination and 8.2 ×  10-2 nm in octahedral coordination (Whitaker and Muntus 1970). There are seven short-lived Li radioisotopes (with <1 second half-lives) and two stable isotopes, 6Li and 7Li, which constitute ~7.5% and ~92.5%, respectively, of the natural terrestrial Li abundance. Lithium isotopes can fractionate strongly in the highly evolved magma that forms rare element pegmatites (Teng et al. 2006) but the processes involved are still unclear. Equilibrium and kinetic processes are both significant pathways for Li isotope fractionation in many natural geological environments, e.g., fusion, dissolution, crystallisation and weathering.  Equilibrium fractionation is related to the influence of the atomic weight of an isotope on the vibrational energy of a molecule at rest, and is therefore more significant at lower temperatures (Urey 1947; Chacko et al. 2001). Within a substance in this rest state there is a difference between the vibrational energies (‘zero point energy’, ZPE) of lighter and heavier isotopes, denoted as ΔZPE. Due to the inverse 9 relationship between vibrational frequency and atomic mass, heavier isotopes have lower frequencies. Different substances have different ‘zero point energies’ between isotopes, therefore different ΔZPE. The greater the ΔZPE, the greater the stabilizing effect of the incorporation of heavy isotopes. The entropy of the system is increased as the heavier isotope partitions into the substance with the greater ΔZPE. In atomic structures with low coordination numbers (c.n.), elements tend to be strongly bound and have greater bond energy. This leads to a larger ΔZPE than those with high coordination numbers that are bound less strongly. In the case of Li, 6Li and 7Li will have a greater ΔZPE in tetrahedral configurations (c.n. 4) than in octahedral sites (c.n. 6) and therefore 7Li will preferentially occur in the tetrahedral site. Within Li-rich rare element pegmatites much of the Li is located in octahedral sites in the crystal structures of e.g., clinopyroxene (spodumene) and mica, which will therefore tend to have ‘lighter’ isotopic signatures (enriched in 6Li). Whereas, in doped silicate glass, which can be used as a proxy for natural silicate melt, Henderson (2005) reports that Li occurs in 2 or 4-fold coordination with Si and O. Therefore, 7Li is expected to occur preferentially in the silicate melt. Teng et al. (2006) proposed that in a fractionating magmatic system, with ongoing mineral precipitation preferentially incorporating 6Li, the residual melt would become increasingly enriched in 7Li. At higher temperatures, the influence of ΔZPE in fractionation is reduced (Chacko et al. 2001), as the overall energy budget of the system increases; this has been observed in basalt (Tomascak et al.1999b). Lithium isotope fractionation is strong in highly evolved pegmatites that can cool to ~400 °C prior to consolidation (Sirbescu and Nabelek 2003; Teng et al. 2006). In addition to equilibrium isotope effects, the large mass difference between 6Li and 7Li produces significant kinetic fractionation. This is a unidirectional process that occurs prior to equilibrium being reached between isotopes of different diffusivities, with diffusivity described by the temperature of the system, the medium through which the isotopes are diffusing and the mass of the particles involved. 10 Using the two Li isotopes (7Li and 6Li) to demonstrate the relationship between isotope diffusivity and mass in a phase:  D6/D7=(m7Li/m6Li) β      (1)  where the ratio of diffusivity (D) between 6Li and 7Li in a material is inversely related to the ratio of their masses (m; m7Li/m6Li = 7.016003/6.015121) by the experimentally derived exponent β. Richter et al. (2003) determined that Li has a rate of diffusion orders of magnitude greater than many other elements and defined β = 0.215 for Li isotopes in molten silicates. Measurements on clinopyroxene (Parkinson et al. 2007) suggest similar values for β in silicate minerals producing a ~3% faster diffusion rate for 6Li over 7Li in this mineral. Variations in element concentration equalize by chemical diffusion through and between solids, liquids and gases and this process promotes isotopic fractionation if the isotopes have different diffusivities. This unidirectional process is being increasingly used in geological ‘geospeedometry’ studies to investigate the rates of magmatic and hydrothermal processes (e.g., Richter et al. 1999 and 2003, Coogan et al. 2005, Parkinson et al. 2007, Gallagher and Elliott 2009). Maloney et al. (2008) studied the Li isotopic signature of tourmaline from a Li-rich pegmatite, and suggested that fractionation would occur between a boundary layer encompassing the precipitating crystal (tourmaline) and the rest of the melt.  1.4.2 Chemistry One of the current limitations to Li isotope analysis in Li-rich pegmatites is the need to purify Li prior to isotopic analysis to remove matrix elements that may interfere with the measurement of Li.  While ion exchange chromatography is a commonly used method for this approach, the ease of fractionation between the Li isotopes creates challenges during sample preparation as the resin in the chromatographic column preferentially retains 6Li (Taylor and Urey 1938). This results in 7Li being eluted more readily than 6Li and makes it necessary to collect ~100% of the Li from the columns. If the initial Li portion (higher in 7Li) is not collected, the sample is depleted in 7Li, and the measured δ7Li will be too low. Alternatively, if the last of the Li to come off the column 11 (higher in 6Li) is not collected, the sample will be depleted in 6Li, and the measured δ7Li will be erroneously high. Several ion chromatographic methods have been developed to separate Li from other elements using one or more cationic exchange columns depending on the sample requirements or instrumentation used (e.g., Moriguti and Nakamura 1998, Tomascak et al. 1999a, Nishio and Nakai 2002, Rudnick et al. 2004, Lynton et al. 2005). This study utilized a method developed by Jeffcoate et al. (2004) with the first column removing most major cations, e.g., Fe, Mg etc., followed by a second column to remove Na eluted with Li from the first column. This chromatographic procedure uses organic solvents and aqueous inorganic acids as eluents.  1.4.3 Analysis Lithium isotope ratios have been measured using various methods. Several studies have been undertaken using thermal ionization mass spectrometry (TIMS) (e.g., Chan 1987, You and Chan 1996, Moriguti and Nakamura 1998), quadrupole ICP-MS (e.g., Grégoire et al. 1996, Kŏsler et al. 2001),secondary ion-mass spectrometry (SIMS; e.g., Kasemann et al. 2005, Coogan et al. 2005, Bell et al. 2009), atomic emission spectrometry (AES; Rehman et al. 2009) and neutron activation (NA; e.g., Wölfle et al. 1977, Yamada et al. 1995). Recent technological advances in measurement precision, and instrument availability, have made magnetic sector multi-collector inductively coupled plasma mass spectrometry (MC ICP-MS) increasingly popular (e.g., Nishio and Nakai 2002, Jeffcoate et al. 2004, Rudnick et al. 2004). The separation of Li by ion exchange columns prior to analysis, significantly reduces the potential for mass dependant effects (spectral matrix effects) due to the presence of components (other than Li) in the measured analyte solution. The lightness of Li reduces the number of potential ions or molecular isobars that could produce non- spectral effects during analysis. The only isobaric interferences that may influence the signals of 6Li+ and 7Li+, are 12C++ and 14N++, respectively (Tomascak 2004). As with the chemical separation, Li isotope measurement is complicated by the instrumental ‘mass dependant’ fractionation of 6Li and 7Li. A rapid analytical protocol of blank-standard-sample-standard-blank measurements was followed in this study 12 (following Jeffcoate et al. 2004). Sample measurements were bracketed by the blank measurements taken at the start and end of each protocol to allow for ‘blank correction’, and by measurements of a known standard, typically the international accepted standard L-SVEC. Sample measurements were normalized against the standard to correct for instrumental ‘mass dependant’ fractionation. Instrument stability and sample reproducibility are critical, external precision from MC ICP-MS analyses range from 0.3 to 1.1‰ (2s; Tomascak et al. 1999a, and Jeffcoate et al. 2004). Measurements in this study indicated an external precision of 0.88‰ 2SD over a 3-month period.  1.5 OVERVIEW OF THE DISSERTATION This dissertation was written as a cohesive study of various aspects of rare element pegmatite formation occurring in the dikes of the Little Nahanni Pegmatite Group, Northwest Territories. It has been written in manuscript format with three complementary research chapters (Chapters 2, 3 and 4) that will be published as individual journal papers. As a result there is some repetition between the chapters, although reference to the relevant aspects of other chapters is made. Chapter 2 views the emplacement of the dikes in a structural and regional geological setting. It provides detailed structural data on the relationship between the orientation of the dikes and the predominant country rock foliations. The structural data are interpreted in the context of a petrographic study of the country rock and microthermometric analyses of fluid inclusions. Results from several geochronological systems are used to determine the age of cooling and emplacement of the dikes and to interpret the geothermal history of the area. Chapter 3 investigates the late stage magmatic processes in the highly evolved silicate melt. It reports the major, minor and trace element geochemistry of whole rock and mineral separate samples from the LNPG, plus minor and trace element data of samples of Macusani glass (a natural peraluminous obsidian; Pichavant et al. 1988a and b). In addition, it includes a petrographic study of the dikes and Rietveld diffraction analyses of the whole rock LNPG samples. An interpretation of the magmatic evolution 13 is drawn from the trace element results and is supplemented by geochemical modeling of the whole rock, chondrite-normalised REE patterns. Chapter 4 reports the Li isotopic signature of the whole rock and mineral samples from the LNPG and data from nearby contemporaneous granitic intrusions and samples of the Macusani glass. Isotopic results from the LNPG are viewed in the light of the geochemical data (presented in Chapter 3), to assess the cause of Li isotopic fractionation. A common cause for element fractionation determined in Chapter 3 and Li isotope fractionation is suggested. Chapter 5 provides the conclusion to the dissertation and highlights the significant findings. It looks at the common features of the LNPG dikes within an exploration context and provides additional advice for rare element pegmatite exploration. This is juxtaposed with an outline of the exceptional features of the dikes. The implications of the study on our understanding of elemental Li and its isotopic systematics are discussed. Several suggestions for future research in this field conclude the dissertation.  1.6 CONTRIBUTIONS TO THE PROJECT This dissertation would not have been possible with out the help, advice and cooperation of others. This section outlines the individual contributions made by the lead and co-authors to this study as they pertain to each chapter, followed by the contributions of others to the research in general. The co-authors provided editorial comment throughout. Chapter 2 was co-authored by Prof. Lori Kennedy (UBC), Prof. Jim Mortensen (UBC), Prof. Dan Marshall (Simon Fraser University) and Prof. Lee Groat (UBC). Prof. Kennedy provided assistance in the field, giving advice on the methodology of the structural study, on-campus advising on the metamorphic petrography and providing extensive editorial supervision. Prof. James Mortensen suggested significant field locations, provided comment and oversaw the interpretation of the geochronological results and the regional geology. Prof. Marshall provided the equipment and training for the fluid inclusion studies, calculated the results and significantly helped with their interpretation. Prof. Groat provided financial and logistical support. 14 I undertook fieldwork (with assistance) over several seasons collecting samples, mapping, making observations and taking structural measurements. I took all fluid inclusion microthermometric measurements, made petrographic observations and prepared samples for analysis. From the data, I interpreted the influence of the pre- existing structures in the host rock on the orientation of the dikes and reassessed the regional geothermal history in the light of the geochronological results. The co-authors for Chapters 3 and 4 were Prof. Dominique Weis (UBC), and Prof. Lee Groat.  Prof Weis provided access and technical support to the PCIGR laboratory (UBC), where the majority of the trace element and Li isotope analysis was undertaken, and advice on the interpretation of the results.  Dr Groat provided access to the pre-ground whole rock samples and the geochemical analyses undertaken on them prior to the start of this study. For Chapter 3, I prepared and analysed about half of the samples (the rest were analysed by others at the PCIGR or at the Université Joseph Fourier, Grenoble, France).  I was the sole worker on the Li isotope study of Chapter 4, and undertook method research and development, sample preparation and measurement of the majority of the samples (except for sample analysis undertaken at the University of Maryland, as noted in the text). I also undertook all aspects of the geochemical modeling of the samples and was responsible for drafting all the figures.                    15 1.7 REFERENCES AHLBORN, V.H. (1977): Assessment report on prospecting permit 105 - I-2, by D.C. Rotherham, Feb. 1962.  AHLBORN, V.H. (1979): The LICA pegmatites, Nahanni Spodumene Belt, 1978 field investigations, Cominco Rare Metals Exploration Report.  AHLBORN, V.H. (1980): Assessment report - prospecting and geochemical survey on Lica claims, Selwyn Mountains, NWT, Nahanni Mining Division (105I/12, 62º12'N, 128º50'W), Cominco Limited.  BAKER, D.R. (1996): Granitic melt viscosities: Empirical and configurational entropy models for their calculation. American Mineralogist 81, 126-134.  BARNES, E.M., GROAT, L.A. & FALCK, H. (2007): A review of the late Cretaceous Little Nahanni Pegmatite Group and associated rare-element mineralization in the Selwyn Basin area, Northwest Territories. In Mineral and Energy Resource Assessment of the Greater Nahanni Ecosystem Under Consideration for the Expansion of the Nahanni National Park Reserve, Northwest Territories (D.F. Wright, D. Lemkow and J.R. Harris, eds.), Geological Survey of Canada Open File 5344, 191-202.  BEAVON, R.V. (1977): Preliminary geological survey of the Cali 1-9, 11 mineral claims, Cantung area, District of Nahanni, NWT. Canadian Superior Exploration Ltd.  BELL, D.R., HERVIG, R.L., BUSECK, P.R. & AULBACH, S. (2009): Lithium isotope analysis of olivine by SIMS: Calibration of a matrix effect and application to magmatic phenocrysts. Chemical Geology 258, 5–16.  BERANEK, L.P. & MORTENSEN, J.K. (2007): Investigating a Triassic overlap assemblage in Yukon: On-going field studies and preliminary detrital-zircon age data. In Yukon Exploration and Geology 2006 (D. S. Emond, L. L. Lewis and L. H. Weston, eds.). Yukon Geological Survey, 83-92.  BROWN, M. (2002): Retrograde processes in migmatites and granulites revisited. Journal of Metamorphic Geology 20, 25-40.  ČERNÝ, P. (1991a): Rare-element granitic pegmatites, Part I: Anatomy and internal evolution. Geoscience Canada 18, 49-67.  ČERNÝ, P. (1991b): Rare-element granitic pegmatites, Part II: Regional to global environments and petrogenesis. Geoscience Canada 18, 68-80.  ČERNÝ, P. & ERCIT, T.S. (2005): The classification of granitic pegmatites revisited. The Canadian Mineralogist 43, 2005-2026.  16 CHACKO, T., COLE, D.R. & HORITA, J. (2001): Equilibrium oxygen, hydrogen and carbon isotope fractionation factors applicable to geologic systems. In Stable Isotope Geochemistry (J.W. Valley and D.R. Cole, eds.), Reviews in Mineralogy and Geochemistry 43, 1-62.  CHAN, L.-H. (1987): Lithium isotope analysis by thermal ionization mass spectrometry of lithium tetraborate. Analytical Chemistry 59, 2662-2665. COOGAN, L.A., KASEMANN, S.A. & CHAKRABORTY, S. (2005): Rates of hydrothermal cooling of new oceanic upper crust derived from lithium-geospeedometry. Earth and Planetary Science Letters 241, 415-424. COOK, F.A., CLOWES, R.M., SNYDER, D.B., VAN DER VELDEN, A.J., HALL, K.W., ERDMER, P. & EVENCHICK, C.A. (2004): Precambrian crust beneath the Mesozoic northern Canadian Cordillera discovered by Lithoprobe seismic reflection profiling. Tectonics 23, TC2010, doi:10.1029/2002TC001412. COPLEN, T.B., BÖHLKE, J.K., DE BIÈVRE, P., DING, T., HOLDEN, N.E., HOPPLE, J.A., KROUSE, H.R., LAMBERTY, A., PEISER, H.S., RÉVÉSZ, K., RIEDER, S.E., ROSMAN, K.J.R., ROTH, E., TAYLOR, P.D.P., VOCKE, JR., R.D. & XIAO, Y.K. (2002): Isotope-abundance variations of selected elements (IUPAC Technical Report). Pure Applied Chemistry 74, 1987-2017.  ERCIT, T.S., GROAT, L.A. & GAULT, R.A. (2003): Granitic pegmatites of the O'Grady Batholith, N.W.T., Canada: a case study of the evolution of the elbaite subtype of rare- element granitic pegmatite. The Canadian Mineralogist 41, 117-138.  GALLAGHER, K. & ELLIOTT, T. (2009): Fractionation of lithium isotopes in magmatic systems as a natural consequence of cooling. Earth and Planetary Science Letters 278, 286-296.  GORDEY, S.P. & ANDERSON, R.G. (1993): Evolution of the northern Cordilleran miogeocline, Nahanni map area [105I], Yukon and Northwest Territories. Geological Survey of Canada Memoir 428.  GRÉGOIRE, D.C., ACHESON, B.M. & TAYLOR, R.P. (1996): Measurement of lithium isotope ratios by inductively coupled plasma mass spectrometry: application to geological materials. Journal of Analytical Atomic Spectrometry 11, 765–772.  GROAT, L.A., ERCIT, T.S., RAUDSEPP, M. & MAUTHNER, M.H.F. (1994): Geology and mineralogy of the Little Nahanni Pegmatite Group, part of NTS area 105 I/02. Economic Geology Survey Open File 1994-14. DIAND, NWT.  GROAT, L.A., MULJA, T., MAUTHNER, M.H.F., ERCIT, T.S., RAUDSEPP, M., GAULT, R.A. & ROLLO, H.A. (2003): Geology and mineralogy of the Little Nahanni rare-element granitic pegmatites, Northwest Territories. The Canadian Mineralogist 41, 139-160. 17  HART, C.J.R. & LEWIS, L.L. (2006): Gold mineralization in the upper Hyland River area: a non-magmatic origin. In Yukon Exploration and Geology 2005 (D.S. Emond, G.D. Bradshaw, L.L. Lewis and L.H. Weston, eds.), Yukon Geological Survey, 109-125.  HART, C.J.R., MAIR, J.L., GOLDFARB, R.J. & GROVES, D.I. (2004): Source and redox controls on metallogenic variations in intrusion-related ore systems, Tombstone-Tungsten Belt, Yukon Territory, Canada. Transactions of the Royal Society of Edinburgh: Earth Sciences 95, 339-356.  HENDERSON, G.S. (2005): The structure of silicate melts: a glass perspective. The Canadian Mineralogist 43, 1921-1958.  JEFFCOATE, A.B., ELLIOTT, T., THOMAS, A. & BOUMAN, C. (2004): Precise, small sample size determinations of lithium isotopic compositions of geological reference materials and modern seawater by MC-ICP-MS. Geostandards and Geoanalytical Research 28, 161-172.  KASEMANN, S.A., JEFFCOATE, A.B. & ELLIOTT , T. (2005): Lithium isotope composition of basalt class reference material. Analytical Chemistry 77, 5251-5257.  KŎSLER, J., KUCĚRA, M. & SYLVESTER, P. (2001): Precise measurement of Li isotopes in planktonic foraminiferal tests by quadrupole ICP-MS. Chemical Geology 181, 169–179.  LANG, J.R. (ed.), (2000): Regional and system-scale controls on the formation of copper and/or gold magmatic-hydrothermal mineralization. Final Report, Mineral Deposit Research Unit Special Publication 2.  LONDON, D. (1984): Experimental phase equilibria in the system LiAlSiO4 -SiO2 -H2O; a petrogenetic grid for lithium-rich pegmatites. American Mineralogist 69, 995-1004.  LONDON, D. (1997): Estimating abundances of volatile and other mobile components in evolved silicic melts through mineral–melt equilibria. Journal of Petrology 38, 1691– 1706.  LONDON, D. (2005): Granitic pegmatites: an assessment of current concepts and directions for the future. Lithos 80, 281-303.  LONDON, D. (2008): Pegmatites. Mineralogical Association of Canada, Québec, Canada.  LYNTON, S.J., WALKER, R.J. & CANDELA, P.A. (2005): Lithium isotopes in the system Qz-Ms-fluid: An experimental study. Geochimica et Cosmochimica Acta 69, 3337-3347.  MAIR, J.L., HART, C.J.R. & STEPHENS, J.R. (2006): Deformation history of the northwestern Selwyn Basin, Yukon, Canada; implications for orogen evolution and Mid- Cretaceous magmatism. Geological Society of America Bulletin 118, 304-323. 18  MALONEY J.S., NABELEK, P.I., SIRBESCU, M.-L.C. & HALAMA, R. (2008): Lithium and its isotopes in tourmaline as indicators of the crystallization process in the San Diego County pegmatites, California, USA. European Journal of Mineralogy 20, 905-916.  MANNING, D.A.C. (1981): The effect of fluorine on liquidus phase relationships in the system Qz–Ab–Or with excess water at 1 kb. Contributions to Mineralogy and Petrology 76, 206– 215.  MAUTHNER, M.H.F. (1996a): Mineralogy, geochemistry and geochronology of the Little Nahanni Pegmatite Group, Logan Mountains, southwestern Northwest Territories. M.Sc. thesis, University of British Columbia, Vancouver, British Columbia.  MAUTHNER, M.H.F. (1996b): Geological and geotechnical report on the Little Nahanni Pegmatite Group, Logan Mountains, NWT, Claims Tun 1 to 7, NTS 105I/1 (62º12'N, 128º50'W), September 4, 1995 – September 3, 1996.  MAUTHNER, M.H.F., MORTENSEN, J.K., GROAT, L.A. & ERCIT, T.S. (1994): Geochronology of the Little Nahanni Pegmatite Group, Selwyn Mountains, southwestern Northwest Territories. Canadian Journal of Earth Sciences 32, 2090-2097.  MCDONOUGH, W.F. (2003): Compositional model for the earth’s core. In Treatise on Geochemistry, 2, (K.K. Turekian and H.D. Holland, eds.), Elsevier–Pergamon, Oxford, UK.  MORIGUTI, T. & NAKAMURA, E. (1998): High-yield lithium separation and the precise isotopic analysis for natural rock and aqueous samples. Chemical Geology 145, 91-104.  MORTENSEN, J.K., HART, C.J.R., MURPHY, D.C. & HEFFERNAN, S. (2000): Temporal evolution of Early and mid-Cretaceous magmatism in the Tintina Gold Belt. In The Tintina Gold Belt: Concepts, Exploration and Discoveries (J. Jambor ed.), British Columbia and Yukon Chamber of Mines, 2, 49-57.  NISHIO, Y. & NAKAI, S. (2002): Accurate and precise lithium isotopic determinations of igneous rock samples using multi-collector inductively coupled plasma mass spectrometry. Analytica Chimica Acta 456, 271-281.  PARKINSON, I.J., HAMMOND, S., JAMES, R.H. & ROGERS, N.W. (2007): High-temperature lithium isotope fractionation: insights from lithium isotope diffusion in magmatic systems. Earth and Planetary Science Letters 257, 609-621.  PATTISON, D.M. (1992): Stability of andalusite and sillimanite and the Al2SiO5 triple point: constraints from the Ballachulish Aureole, Scotland. Journal of Geology, 100, 423- 446.  19 PATTISON, D.M. & TRACY, R.J. (1991): Phase equilibria and thermobarometry of metapelites. Reviews in Mineralogy and Geochemistry 26, 105-206.  PEMBERTON, B.C. (2002): The origin of cassiterite at the Little Nahanni Pegmatite Group, Northwest Territories. B.Sc. thesis, University of British Columbia, Vancouver, British Columbia.  PICHAVANT, M., KONTAK, D.J., HERRERA, J.V. & CLARK, A.H. (1988a): The Miocene- Pliocene Macusani Volcanics, SE Peru I. Mineralogy and magmatic evolution of a two- mica aluminosilicate-bearing ignimbrite suite. Contributions to Mineralogy and Petrology 100, 300-324.  PICHAVANT, M., KONTAK, D.J., HERRERA, J.V. & CLARK, A.H. (1988b): The Miocene- Pliocene Macusani Volcanics, SE Peru II. Geochemistry and origin of a felsic peraluminous magma. Contributions to Mineralogy and Petrology 100, 325-338.  RASMUSSEN, K.L., MORTENSEN, J.K. & FALCK, H. (2006): Geochronological and lithogeochemical studies of intrusive rocks in the Nahanni region, southwestern Northwest Territories and southeastern Yukon. In Yukon Exploration and Geology 2005 (D.S. Emond, G.D. Bradshaw, L.L. Lewis and L.H. Weston eds.), Yukon Geological Survey, 287-298.  REHMAN, E.U., REHMAN, S.U. & AHMED, S. (2009): 6Li atom percentage determination by atomic absorption–emission spectrometry using a natural lithium hollow cathode lamp. Applied Spectroscopy 63, 971-973.  RICHTER, F.M., LIANG, Y. & DAVIS, A.M. (1999): Isotope fractionation by diffusion in molten oxides. Geochimica et Cosmochimica Acta 63, 2853–2861.  RICHTER F.M., DAVIS, A.M., DEPAOLO, D.J. & WATSON, E.B. (2003): Isotope fractionation by chemical diffusion between molten basalt and rhyolite. Geochimica et Cosmochimica Acta 67, 3905-3923.  ROLLO, H.A. (1999): Mineralogy and geochemistry of mica minerals from the Little Nahanni Pegmatite Group, Northwest Territories. B.Sc. thesis, University of British Columbia, Vancouver, British Columbia.  ROTHERHAM, D.C. (1962): Assessment report 017568- Hyland River Area, Mackenzie - geophysics, geochemistry, regional geology, Canada Tungsten Mining Corporation Ltd.  RUDNICK, R.L. & GAO, S. (2003): Composition of the continental crust. In Treatise on Geochemistry, 3, (K.K. Turekian, and H.D. Holland, eds.), Elsevier–Pergamon, Oxford, UK.  20 RUDNICK, R.L., TOMASCAK, P.B., NJO, H.B. & GARDNER, L.R. (2004): Extreme lithium isotopic fractionation during continental weathering revealed in saprolites from South Carolina. Chemical Geology 212, 45-57.  SIRBESCU, M.-L. & NABELEK, P.I. (2003): Crystallization conditions and evolution of magmatic fluids in the Harney Peak granite and associated pegmatites, Black Hills, South Dakota—evidence from fluid inclusions. Geochimica et Cosmochimica Acta 67, 2443- 2465.  SOWERBY, J.R. & KEPPLER, H. (2002): The effect of fluorine, boron, and excess sodium on the critical curve in the albite–H2O system. Contributions to Mineralogy and Petrology 143, 32– 37.  STEWART, D. B. (1978): Petrogenesis of lithium-rich pegmatites. American Mineralogist 63, 970-980.  SWANSON, S.E. (1977): Relation of nucleation and crystal-growth rate to the development of granitic textures. American Mineralogist 62, 966-978.  SWANSON, S.E. & FENN, P.M. (1992): The effect of F and Cl on albite crystallization: a model for granitic pegmatites? Canadian Mineralogist 30, 549– 559.  TAYLOR, T.I. & UREY, H.C. (1938): Fractionation of the lithium and potassium isotopes by chemical exchange with zeolites. Journal of Chemical Physics 6, 429-438.  TENG, F.-Z., MCDONOUGH, W.F., RUDNICK, R.L., DALPÉ, C., TOMASCAK, P.B., CHAPPELL, B.W. & GAO, S. (2004): Lithium isotopic composition and concentration of the upper continental crust. Geochimica et. Cosmochimica Acta 68, 4167-4178.  TENG, F.-Z., MCDONOUGH, W.F., RUDNICK R.L., WALKER, R.J. & SIRBESCU, M.-L.C. (2006): Lithium isotopic systematics of granites and pegmatites from the Black Hills, South Dakota. American Mineralogist 91, 1488-1498.  TOMASCAK, P.B. (2004): Developments in the understanding and application of lithium isotopes in the earth and planetary sciences. Reviews in Mineralogy & Geochemistry 55, 153-195.  TOMASCAK, P.B.,CARLSON, R.W. & SHIREY, S.B. (1999a): Accurate and precise determination of Li isotopic compositions by multi-collector sector ICP-MS. Chemical Geology 158, 145-154.  TOMASCAK, P.B., TERA, F., HELZ, R.T. & WALKER R.J. (1999b): The absence of lithium isotope fractionation during basalt differentiation: New measurements by multicollector sector ICP-MS. Geochimica et Cosmochimica Acta 63, 907–910.  21 TRUEMAN, D.L. & ČERNÝ, P. (1982): Exploration for rare-element granitic pegmatites. Granitic Pegmatites in Science and Industry, Mineralogical Association of Canada Short Course Handbook 8, 463-493.  UREY, H.C. (1947): The thermodynamic properties of isotopic substances. Journal of the Chemical Society, 562 – 581.  WENGZYNOWSKI, W.A. (2002): Geological report on the Mac property. War Eagle Mining Company Inc.  WHITAKER, E.J.W. & MUNTUS, R. (1970): Ionic radii for use in geochemistry. Geochimica et Cosmochimica Acta 34, 945-956.  WÖLFLE, R. & NEUBERT, E.A. (1977): Determination of isotopic composition of lithium by neutron activation analysis; comparison with mass spectrometry. Journal of Radioanalytical Chemistry 39, 375-384.  YAMADA, Y., YASUIKE, K., ITOH, M., KIRIYAMA, N., KOMURA, K. & UENO, K. (1995): Neutron activation analysis of 6Li isotopic abundance using a low background liquid scintillation counter. Journal of Radioanalytical and Nuclear Chemistry 201, 381-390.  YOU, C.-F. & CHAN, L.-H. (1996): Precise determination of lithium isotopic composition in low concentration natural samples. Geochimica et Cosmochimica Acta 60, 909-915. 22 CHAPTER 2: THE LITTLE NAHANNI PEGMATITE GROUP, NWT: UPPER CRUSTAL DIKE EMPLACEMENT IN AN OROGENIC REGION*  2.1 INTRODUCTION Structural aspects of pegmatite emplacement are seldom studied and yet many pegmatite fields were emplaced into areas with a well-developed structural fabric (Černý 1991). Pegmatitic dikes can be emplaced syndeformation, and subsequently become folded or faulted. Alternatively, pre-existing structural fabrics such as cleavage, faults or folds may strongly influence the orientation and distribution of pegmatite dikes (Brisbin 1986). For example, pegmatitic dikes in the Bernic Lake area, Manitoba, which include the Tanco pegmatite, were emplaced into a large high-strain area. In this region dilatant zones that developed during deformation provided conduits for magma movement and fundamentally influenced the emplacement of mobile late stage granitic magma (Kremer and Lin 2006). We use field observations to evaluate the relationship between the orientation of sub-vertical, flat-walled dikes of the Cretaceous rare element Little Nahanni Pegmatite Group (LNPG) and folds, bedding and cleavage produced during Mesozoic deformation of metasedimentary host rock. Timing of deformation versus dike emplacement is established from crosscutting relationships and geochronological studies. Dike emplacement is put into the context of the magmatic and metamorphic evolution of the area by combining geochronological analyses, including U-Pb on apatite, 40Ar/39Ar on muscovite and lepidolite, and Rb/Sr on whole rock, feldspar, muscovite and apatite, with petrographic study of porphyroblast development and metamorphic textures. In addition, fluid inclusion microthermometry is used to outline the depth of emplacement and the crystallization temperature of the dikes. The highly evolved nature of the magma responsible for these granitic pegmatite dikes is illustrated by the low K/Rb ratios (4-24) and notable concentrations of several  * A version of this chapter will be submitted for publication. Barnes, E.M., Kennedy, L.A., Mortensen, J.K., Marshall, D. & Groat, L.A. (2010): The Little Nahanni Pegmatite Group, NWT: Upper crustal dike emplacement in an orogenic region. 23 rare elements (e.g., Li, Cs and Ta) have average concentrations several orders of magnitude higher than typical crustal values (Rudnick and Gao 2003) at 6264, 203 and 94 ppm respectively (Barnes et al. 2007). Primary pegmatitic textures, including extreme crystal size, elongate crystal growth perpendicular to host-rock contact and composite pegmatite-aplite layering are indicative of crystallization from evolved volatile- and flux- rich silicate melt, and provide further insight on the conditions prevalent during crystallization.  2.2 REGIONAL GEOLOGICAL SETTING The study area is located mainly on the eastern flank of the Fork Anticline (Figs. 2.1 and 2.2; Gordey and Anderson 1993) in the Mackenzie Mountains, Northwest Territories, close to the Yukon border (505000E 6895000N; NTS 105 I/2). This anticlinal structure is formed from numerous smaller anticlines and is herein informally referred to as the ‘Fork anticlinorium’. It lies near the western margin of the unexposed North American craton (Fig. 2.3), and is underlain by the oldest exposed rock units in the area, the Neoproterozoic Yusezu and overlying Narchilla formations that compose the Hyland Group. The pegmatite dikes are generally northwest-southeast striking, and have a high aspect ratio (approximately 1-2 m wide x 100-1000 m along strike).  They are located within an area of approximately 12 x 2.5 km. Late Pleistocene glacial erosion created a series of straight, roughly northeast trending, steep hanging valleys above tree-line (1500 – 2200 m above sea level) providing 60-70% rock exposure. Pegmatitic dikes have not been discovered further east than the March Fault (Fig. 2.2), where Hyland Group strata are thrust over the Lower Cambrian to Cambro-Ordovician Rabbitkettle and Vampire formations on the east side of the anticlinorium. The majority of the dikes are located within a 2 km long zone between 6895000 m N and 6887000 m N. However, several dikes crop out further north, and the pegmatite field may also extend to the south, below the Quaternary sedimentary cover at the southern tip of the ridge. Gordey and Anderson (1993) provide a general framework for the evolution of the northern Cordilleran miogeocline; shallow to increasingly deeper water offshelf deposits of the host Hyland Group are interpreted as probable upper fan facies deposited 24 or  no i F kA nticli ru m 525000E 6 85 0 0N 8 0 3 0 E 48 0 0 Syncline Anticline 6925000N Pr ot er oz oi c al e P oz oi c oz oi M es c Yusezu Formation Narchilla Formation Vampire Formation Gull Lake Formation Haywire Group Road River Group Rabbitkettle Group Earn Group Cretaceous intrusion 10 km Study site F March ault 1 2 3 4 5 6 7 8 1 2 3 4 5 8 8 4 6 7 7 6 6 2 6 4 6 6 5 6 6 3 Lened pluton Cac pluton Fault Selwyn   Basin Alaska Yukon B.C. LNPG Tintina Fault t Lim i o  Cordilleran f d  eform ation N.W.T. O’Grady Batholith (inset) Figure 2.1 Geologic map of the region encompassing the study area. Inset shows the location of the study area within the Selwyn Basin and Western Canada. Adapted from Gordey and Anderson (1993). 25 504 E 506 E 508 E 68 N 95 6 9  N 8 0  i qu e 3 C r  i uC rq e 4 Cirq ue 5 F r  A icl no i o k nt i r um r h Fa M a c ult 0200 2000 1 0 50 Contour interval 500 metres UTM Coordinates x 1000 0 1000 metres 2000 1 2 4 3 Geological boundary:approximate, inferred Axial trace of anticlinorium, defined Syncline, extrapolated beneath overburden Thrust fault, teeth indicate upthrust side Selected bedding orientations : inclined, vertical  35 7843 Selected pegmatite orientations : inclined, vertical Little Nahanni Pegmatite Group - areas of highest density of dikes Cambro-Ordovician Rabbitkettle Fm.4 Lower Cambrian Vampire Formation3 Precambrian to Lower Cambrian Narchilla Formation 2  Precambrian Yusezu Formation 1 35 A B C D E A’ B’ C’ D’ E’ A A’ Cross section i qu e 2 C r{ H y a  ro up l nd G Figure 2.2 Geologic map of the study area showing areas with the highest density of pegmatites, bedding and pegmatite orientations and the location of the cross sections measured for the study. Modified from Groat et al. (2003). 26    Ancestral North America NESW m M es oz oi c co pr es si on LNPG metasedimentary strata crystalline basement not to scale Figure 2.3 Cartoon cross section of the boundary between the Selwyn Basin strata and the crystalline basement of ancestral North America. By mid-Cretaceous time the Selwyn Basin (medium grey) had undergone intense compressional deformation. Modified from Mair et al. (2006). 27 on the eastward migrating margin of the Selwyn Basin. Accretion of the Yukon-Tanana terrane to the southwest during early Mesozoic time (Beranek and Mortensen 2007) resulted in the development of a northwest-southeast trending fabric in the eastern Selwyn Basin (Fig. 2.1), oriented subparallel to the original margin of the basin along the North American craton. Compression from the Late Permian-Early Triassic to post mid- Cretaceous produced intensive thin-skinned tectonic deformation and crustal thickening in the Selwyn Fold Belt (Gordey and Anderson 1993; Hart and Lewis 2006; Beranek and Mortensen 2007) and reactivated the underlying and long-lived March Fault (Neoproterozoic; Hart and Lewis 2006). A period of S- and I-type magmatism occurred in the region during the Mesozoic orogeny (Mortensen et al. 2000; Hart et al. 2004; Mair et al. 2006; Rasmussen et al. 2007). Cretaceous intrusions proximal to the LNPG have been assigned to the Tungsten and Tay River-Tungsten suites (100-95 Ma based on U-Pb zircon and monazite dating; Rasmussen et al. 2007). These S-type intrusions with sedimentary rock and mantle affinities are typically small (< 15 km2 exposure) and associated with W (Cu-Zn-Mo) and skarn intrusion-related mineralization. The Tungsten suite tends to be more geochemically evolved than the Tay River-Tungsten suite and emplaced at shallower depths although both suites are associated with aplitic and pegmatitic dikes. Slightly younger 40Ar/39Ar biotite ages (~89-86 Ma) were reported by Rasmussen et al. (2007) from the Nahanni Range Road pluton ~30 km northwest of the LNPG. These 40Ar/39Ar biotite ages are interpreted to reflect post-intrusion uplift and cooling and, with the exception of dates obtained from the LNPG dikes, are the youngest record of a thermal event in the area. Another example of intrusion related mineralization in the area is found within the extensive O’Grady Batholith (270 km2) that lies approximately 95 km further north of the LNPG and was emplaced at 94 Ma (Mortensen unpublished data). The O’Grady Batholith is part of the mid-Cretaceous Tombstone Suite (Hart et al. 2004), a linear belt of dominantly alkalic, metaluminous intrusions that cuts the Robert Service Thrust Fault northwest of LNPG and extends, intermittently, southeast to the Big Charlie and Macleod plutons south of Coal River, NWT (Heffernan 2004).  This hornblende-bearing, alkali- feldspar-rich composite intrusion includes a rare example of the elbaite subtype of 28 granitic rare element pegmatite (Ercit et al. 2003) and is the only regional example of lithium mineralization other than the LNPG dikes. In the western Selwyn Basin Mair et al. (2006) recognized otherwise poorly documented sinistral movement on brittle faults that postdates this mid-Cretaceous period of magmatism. Mair et al. (2006) suggest that some extensional deformation possibly as a result of orogenic collapse occurred prior to the more pervasive north-south oriented dextral movement induced by northward movement of the Kula Plate (Engebretson et al. 1985).  2.3 PREVIOUS WORK The pegmatitic dikes investigated here were discovered during a regional reconnaissance program by Canada Tungsten Mining Corporation Ltd. in 1961 (Rotherham 1962). Initial studies by Groat et al. (1994) described the dikes as albite- spodumene type and named the swarm the Little Nahanni Pegmatite Group. The dikes have recently been classified as a type of REL-Li  (Rare-ELement-Li) subclass of complex granitic pegmatitic dikes (Černý and Ercit 2005). Mauthner et al. (1995) determined a U-Pb age of ~82 Ma for manganocolumbite from the dikes, that was interpreted as an emplacement age. Potassium-argon dating of micas from the dikes yielded ages of ~65 Ma which were interpreted to be thermally reset ages. Mineralogical studies and textural observations by Mauthner et al. (1995), Mauthner (1996) and Groat et al. (2003) led Groat et al. (2003) to suggest that the dikes originated from more than one melt composition which became increasingly geochemically evolved and water saturated during crystallization. Examples of the evolving geochemistry of the dikes include a weakly defined zonation in the spodumene- (Li-rich clinopyroxene) rich pegmatitic dikes. This zonation occurs at slightly greater depth than surrounding slightly Ta and Sn enriched spodumene-poor dikes that are more prevalent in the north and south of the study area (Wengzynowski 2002). In addition, at a well-exposed location the abundance of quartz in the dikes is observed to increase with elevation, indicating that the composition of the melt evolved during dike propagation. Recently, Barnes et al. (2007) provided an overview of the geology of the LNPG and the economic potential of the 29 dikes as part of a Mineral and Energy Resource Assessment (MERA) study undertaken on the proposed expansion of the Nahanni National Park Reserve. 2.4 GEOLOGY 2.4.1 Stratigraphic units The Yusezu Formation hosts the majority of the pegmatite swarm and was therefore studied in more detail than the overlying Narchilla Formation (Figs. 2.1 and 2.2). The Yusezu Formation comprises coarse- to fine-grained siliciclastic rock units that included variable amounts of carbonate matrix, suggesting variations in the amount of clastic grain input. The rocks are typically pale grey to black, but become pale brown to tan with Fe-oxidation or increased dolomite content. Beds range in thickness from millimetres to several metres, with massive sandstone beds tending to be thicker than laminar or cross-bedded strata. Asymmetric linear ripples form rare but useful way-up indicators. Bed thicknesses are laterally discontinuous over tens of metres commonly making it impossible to trace individual beds between ridges. A summary of the country rock types observed in the field is included in Table 2.1.  2.4.2 Bedrock structure The study area is situated within the low temperature Selwyn Fold Belt that is characterized by parallel buckle folds with moderately well-developed cleavage. The largest folds mapped in detail in the area are 10–100s of metres in amplitude and wavelength (Fig. 2.4). These folds are parasitic on the larger Fork anticlinorium, which is one of several regional scale folds (strike lengths of 30-40 km, amplitudes of 1-2 km and half wavelengths of 5-9 km; Gordey and Anderson 1993). Smaller parasitic folds of metres or less in amplitude and wavelength (Fig. 2.5) are common. Orientation of the axial planar cleavage (S1) varies significantly with lithology and is mainly attributed to the convergent and divergent ‘fanning’ of cleavage as a function of rock type and competence. Fracture cleavage fans around folds in relatively thick (>30 cm) sandstone beds (Fig. 2.5A). In contrast, phyllitic cleavage is more consistent in orientation and is displayed by micaceous partings and pressure solution cleavage in finer-grained pelitic rocks and pressure solution cleavage in carbonate-rich siliciclastic beds (Fig. 2.5B and 30         Table 2.1 Variation in the lithologies of the Hyland Group crosscut by the LNPG pegmatites.   Rock  type Weathered colour Bed thickness (cm) Sedimentary structures Interbedded with: Notes conglomerate pale grey highly variable massive   blue opalescent quartz sandstone grey to dark grey, pale buff, tan 10 - 100  massive, finely laminated, linear ripples, cross bedding shale, slate oxidised reduction spots, variable mica content (secondary?) siltstone light to dark grey, orange weathered; buff, brown mm or 10 - 30  massive, planar bedding limestone, siltstone with <~30% carbonate  sandstone, 10 - ~30% carbonate matrix dark grey, buff up to 100  cross bedding massive sandstone lenses 10-20 cm thick oxidised reduction spots, bleached weathering siltstone, 10 - ~30% carbonate matrix dark grey, black, buff, tan, 1-10 planar bedding, cross bedding, common graded beds dark grey (sharp base) to buff colour over ~ 4 cm   oxidised reduction spots limey sandstone, > ~30% carbonate matrix buff 10 - 200   slate limestone buff 1-10   sandstone (2 cm thick beds)  shale dark grey, black, white weathered 1-20 cross bedding still visible slate grey, dark grey, light green     cross bedded sandstone (20 cm thick beds) oxidised reduction spots 2 x 7 mm phyllite pale to dark grey mica schist grey     sandstone (10 cm thick beds) +/- andalusite, two generations (large, round; small, euhedral); staurolite, biotite, chloritoid, cordierite 31 S0 A ~100 m B S0 ~80 m Figure 2.4 Illustrations of macro (outcrop) scale folds and the relationship with the pegmatitic dikes. A) Looking north at the eastern end of the ridge between cirque 4 and 5 (see Fig. 2.2) showing a northeast verging asymmetric anticline. B) View looking north across the head of cirque 3 in the plane of a fault (see Fig. 2.2) cutting a syncline with pegmatitic dikes (white) crosscutting the bedding (black) in the footwall and following it in the hanging wall. The dikes are not folded, the slight curvature of their trace is due to the slope of the country rock. There has been minor, if any, heave on the fault since emplacement of the pegmatitic dikes. Fault = white on black line; syncline = black on grey line, both projected above ground surface. 32 33 5C). The majority of folds are concentric, with the strata maintaining constant thickness through the fold; however, a recrystallized carbonate unit in the south of the study area displays similar folding (i.e., thinning of limbs and thickening of hinges). This type of folding is interpreted to result from solution transfer processes, rather than a high temperature effect. The spatial relationship between host rock structures and dike orientation was evaluated by mapping targeted areas at a scale of 1: 2000 across the apex of the Fork anticlinorium (traverse A-A’, Fig. 2.2), on the west limb (traverse B-B’, Fig. 2.2) and the ridges between cirques 2, 3, 4 and 5 on the east limb (traverses C-C’, D-D’, E-E’, Fig. 2.2, see also Fig. 2.6). Country rocks record one major, pervasive deformational event (D1) that folded bedding (S0). The folds are asymmetrical, upright, northeast verging (Figs. 2.4, 2.5 and 2.7A) and are metres to 10s of metres in amplitude and wavelength. A pervasive cleavage (S1), generally dips steeply to the southwest and northeast but can vary either as a mechanical response in lithologies of different competence or due to later deformation (Figs. 2.4, 2.5 and 2.7B). F1 fold hinges and L1 bedding-cleavage (S1) intersection lineations plunge shallowly to northwest and southeast (Fig. 2.7C and D). The doubly plunging fold hinges are interpreted to be a result of a later deformation and may be related to non-pervasive crenulation cleavage (S2) with a similar orientation to S1. In contrast to the subtle post-S1 deformation, the dip of S2 in the coarse crystalline carbonate of the southwest area associated with similar folding is locally almost vertical, suggesting that this area has undergone more extensive later deformation. The March Fault (Figs. 2.1 and 2.2) is a significant and long-lived northwest trending reverse fault that separates the Neoproterozoic to Lower Cambrian Narchilla and Gull Lake formations from the Lower Cambrian Vampire Formation to the northeast (Gordey and Anderson 1993). It lies approximately 5 km east of the pegmatite field and was active during the mid-Cretaceous (Hart and Lewis 2006). Less significant faults (10- 100 m in length) with approximately the same orientation as the March fault are also present, and, occur on both sides of the axial plane of the anticlinorium. Pegmatitic dikes can be oriented parallel to the faults and may be slightly offset. Strike-slip faults were not observed in the field area. 34 75 m D’                                                                                              D (East)                                                                                        (West) El ev tio a n   m )  ( 1900 1700 508045E 6893050N Bedding measured/inferred S  cleavages1 Pegmatites Ridge line Measured structures 507189E 6892827N Figure 2.6 Photograph of the south side of cirque 3 as measured on traverse D-D'. Taken looking south and overlain by a schema of representative structures 35 D) Intersection lineations                 n=11 C) Fold hinges n=37 Contour lines at 3,6,12 and 24% A) S0 n=126 Contour lines at 1,2,4,8 and 16% B) S1 n=146 Contour lines at 1,2,4,8 and 16% E) Pegmatites n=275 Contour lines at 1,2,4,8,16 and 32% Figure 2.7 Equal-area projections displaying the structural data collected throughout the study area. A) Poles to bedding planes. B) Poles to S  cleavage. C) Fold hinge lineations. D) Bedding and S  1 1 foliation intersection lineations. E) Poles to S  cleavage. F) Poles to pegmatites. 2 36  2.4.3 Relationship between structure and pegmatites The >200 Little Nahanni Pegmatite Group dikes are typically planar and flat- walled with an average width of ~1 m (up to 10 m wide; Mauthner 1996).  They strike northwest and southeast and tend to dip steeply 70-90° (Fig. 2.7E). The dikes rarely cross-cut each other, and individual dikes can be followed along strike for several hundred metres. Rafts of the country rock (up to metres in length; Wengzynowski 2002) occur locally within the dikes, and the dikes are locally observed to bulge, split and rejoin. Most of the LNPG dikes occur within an area of approximately 30 km2 on the eastern limb of the Fork anticlinorium where bedding planes are steeper in the upright limb of the asymmetric, overturned folds.  In plan view (Fig. 2.2) the overall distribution of the dikes forms an ‘S-shape’ curving from northwest-southeast to a more east-west direction in the middle of the group. This pattern is a result of topography: it represents the intersection of the moderately-steeply dipping planar dikes with the northeasterly- incised valleys of the area field area.  Without this variation in topography the trace of the dikes would be relatively linear, striking approximately northwest southeast. The nature of the dike-wall rock contact varies with host rock lithology.  The dike/ host rock contacts are commonly sharp, although fluid/volatile flow from the dikes has formed some exomorphic aureoles of silver mica or schorl of ≤10 cm (Groat et al. 1994) within the pelitic host rock. If located in the carbonate strata the dikes can be cut by calcite-rich fractures (Mauthner 1996). The general orientation of the dikes is similar to the orientation of bedding and cleavage (Fig. 2.7) and they do not appear to coalesce at depth as suggested by Young (2007). Field observations determined that individual dikes are parallel to steeply dipping bedding and cleavage planes but cut across the less steep planes.  In any given asymmetric, overturned fold, the dikes will be parallel to steeply dipping bedding and cleavage: however, they cross cut the shallowly dipping limb of folds Large dikes (1-2 metres wide) are typically planar and are not obviously folded or deformed, and pegmatitic veins generally oriented parallel to S1 in tightly folded rocks are not folded indicating dike emplacement post D1 fold formation (Fig. 2.8A). However, 37 S0 S1 eg m t t p a i e ptygmatic stringer B A S0 Figure 2.8 Photographs illustrating features of the pegmatitic dikes. A) A linear pegmatite following the orientation of S  cleavage in interbedded shale and sandstone. Rock hammer handle ~30 cm long 1 for scale. B) Thin vein of feldspar and mica with parallel sides, ptygmatically folded within limestone, pencil for scale. 38 localized areas of minor boudinage, and folding (especially of the thinner, less competent dikes) provide some evidence of syn- to post- emplacement deformation, e.g., thin feldspar and mica veinlets associated with the dikes show ptygmatic folding  (<1 cm width; Fig. 2.8B). A few individual, axial surface parallel faults that offset dikes by 1-2 metres in an apparent normal faulting offset (Fig. 2.4B). The cause of this faulting is not known but may be related to post intrusion relaxation of the country rock. In summary, most dikes were emplaced after the D1 folding event. The presence of folded thin pegmatite veins and offset of dikes along minor faults indicates that a post D1 deformation deformed the dikes.  2.4.4 Metamorphism Subgreenschist facies regional metamorphism is recorded in the majority of the sedimentary strata by widespread slatey to phyllitic axial planar cleavage (S1) defined by muscovite and biotite.  At some localities, S1 cleavages folded and developed a crenulation cleavage (S2) defined by aligned biotite. There is evidence of higher-grade metamorphism in the study area which is not spatially associated with the dikes which may constraints on the relative timing of magmatism thought to be associated with their emplacement.  A brief description of the mineralogy and microstructures of the metasedimentary rock units follows. Interbedded Al-rich strata are more sensitive to variations in pressure and temperature and the formation of strata-bound andalusite (1-4 cm in diameter; displaying small anhedral and larger euhedral morphologies; Fig. 2.9A and B), staurolite (<1 cm in diameter; Fig. 2.9C) and cordierite (Fig. 2.9D) record conditions of contact metamorphism. Similarly andalusite+biotite hornfels is observed in km-wide haloes surrounding nearby mid- Cretaceous plutons (Gordey and Anderson 1993). In the LNPG area the S1 cleavage is crosscut by, and does not appear to wrap around these porphyroblasts, indicating that porphyroblast growth occurred after development of S1 (Passchier and Trouw 2006). In areas where the crenulation cleavage (S2) occurs in the country rock it is continuous through the porphyroblasts; therefore peak metamorphic conditions occurred after D2 39 40 began. In contrast to the S1 cleavage, however, deflection of S2 around the porphyroblasts, that is locally observed, may have formed syn or post D2. Contact metamorphism in less Al-rich strata at several outcrops at LNPG is recorded in the mineral assemblage garnet + amphibole + plagioclase + quartz (GAPQ). This assemblage consists of a poorly zoned Mn almandine, tremolitic pargasite, oligoclase and quartz. The minerals are relatively fine-grained ranging up to 1 mm, but average 0.1 mm. One outcrop yielded the GAPQ assemblage in textural equilibrium with mica, Mg-annite. Application of the GAPQ assemblage as a geobarometer used in combination with fluid inclusion results to provide constraints on the pressure- temperature conditions during dike emplacement is discussed later. Peak metamorphic conditions were achieved during contact metamorphism; the presence of S2 fabrics within prograde porphyroblasts indicates that these conditions were initiated late D2 at the earliest. Whether D2 continued after peak conditions subsided is uncertain, as microtextural evidence of rare pressure shadows and wrapped foliation (similar to the S2 assemblage) around porphyroblasts (Fig. 2.9A and B) could have occurred during or after porphyroblast growth. The random orientation of late porphyroblasts such as biotite or chlorite around staurolite (Fig. 2.9C) or in pinite haloes around cordierite  (Fig. 2.9D) indicates growth after the earlier stress regime had abated.  2.4.5 Pegmatite mineralogy and primary textures The mineralogy of the Little Nahanni Pegmatite Group includes quartz, K- feldspar, plagioclase, spodumene, various micas (lepidolite, muscovite and minor cookeite), columbite-group minerals, cassiterite, tourmaline, beryl, lithiophilite and garnet (Mauthner 1996; Groat et al. 2003) and berlinite. The LNPG dikes display distinctive textures that provide crucial information regarding conditions occurring during crystallization including compositional and grain- size variations and rare miarolitic cavities. Textural and mineralogical banding, mm- to cm- wide, also occurs parallel to the pegmatite/wallrock contact boundary; this banding is commonly defined by repeated bands of small(er) spodumene or quartz crystals. Spodumene-free dikes also display banding parallel to the wall rock contact, with bands of equant mm-sized aplite grains (feldspar and quartz) cross-cutting more pegmatitic 41 (coarse-grained) layers of K-feldspar + quartz ± muscovite ± lepidolite (Fig. 2.10A) which have crystallized towards the centre (Mauthner 1996). Elongate minerals (feldspar, spodumene, quartz) typically crystallize perpendicular to the pegmatite/wallrock contact and single spodumene or feldspar crystals are locally observed to span the width of a dike (approximately 80 cm in Fig. 2.10B). The dikes do not have well defined ‘pocket zones’, distinctive cores or well-developed, consistent mineralogical zonation. Rare miarolitic cavities occur in two forms; planar areas of euhedral crystal terminations approximately 15 x 15 x 0.5 cm  (~110 cm3) between pegmatite-aplite bands; and as rounded cavities approximately 5 cm diameter (~65 cm3; Mauthner 1996) located towards the centre of the dikes. 2.5 ANALYSIS AND INTERPRETATION 2.5.1 Analytical methods Minerals chosen for analysis were first examined petrographically to ensure their primary, magmatic nature, lack of alteration and, in the case of the Rb-Sr analysis, co- precipitation. Fluid inclusion (FI) microthermometric measurements were taken from clear quartz crystals within a pegmatite sample (EB02) using a Linkam THMS-G 600 heating and freezing stage to determine the conditions during crystallization. Geochronological analyses (U-Pb, 40Ar/39Ar and Rb/Sr) were undertaken on mineral separates and whole rock samples (where appropriate) from four pegmatite samples to determine the age of crystallization  (Table 2.2): • EB1305; U-Pb (apatite) and Rb/Sr (muscovite, feldspar, apatite and whole rock), • EB02, EB0509 and EB1304; (micas, quartz from EB02 was also used in FI analysis) The initial Sr ratio of sample EB1305 was calculated from the whole rock, apatite, feldspar and muscovite Rb/Sr data.  2.5.2 Fluid inclusion microthermometry Few primary fluid inclusions (FI) were found in the sample suite that was examined in this study, and only one sample (EB02) out of 31 provided useful inclusions. 42 Afe l sp a d r fe l pa ds r sp o e du m n e 1 3 1 2 3 43 1 1 1 o ie ta i n f r n t o o p g a i e / a ro ck  nt ct e m t t w ll co a orie tation o  n f pegmatite/wallrock contact ce tr l z n e n a o  B Figure 2.10 Photographs showing some of the common magmatic textures occurring in the dikes of the Little Nahanni Pegmatite Group. A) Compositional and grain size banding (numbered 1 to 4) in a quartz, feldspar and mica-rich pegmatite. Bands 1 (coarse-grained), 2 (medium-grained) and 3 (fine- grained, aplitic); feldspar, mica and quartz, band 4; fine-grained quartz and mica. B) Elongate feldspar crystals (some individual crystals outlined) with interstitial quartz, mica and spodumene (some crystals highlighted in black). In both examples the long axes of the minerals are orientated perpendicular to the pegmatite/country rock contact indicated by white arrows. Rock hammer handle ~30 cm long. 43     Table 2.2 Lines of investigation, analytical methods, samples and minerals used and summary of results.    Investigation Method, samples and minerals used Summary of results   Fluid inclusions Rb/Sr U/Pb 40Ar/39Ar Depth and T of pegmatite emplacement Sample: EB02 quartz  7-8 km depth 400-500º C Age of pegmatite emplacement    Sample: EB1305 apatite  90.3±1.9 Ma (MSWD = 2.7) Geothermal gradient   Data not used   Samples: EB02, EB0509a, EB1304e mica ~ 60 °C/km from 40Ar/39Ar analyses Rb/Sr analyses unreliable 44 These occurred in clear euhedral quartz crystals (average 1 cm width x 4 cm length) that were intergrown with large books (>1 cm diameter) of transparent muscovite. The measured fluid inclusions had negative crystal shapes and were either relatively equant or highly elongate. Each fluid inclusion was oriented along crystallographic axes with the long axes (where present) occurring in the direction of crystal growth. The population of fluid inclusions occurred as three-dimensional arrays entrapped during crystal growth. The equant fluid inclusions were 10-20 microns in diameter and the elongate inclusions were up to 20 microns in length and 3-5 microns in width. Microthermometric data from these inclusions are relatively uniform and are interpreted to represent one fluid inclusion population, comprising a brine phase, a liquid carbonic and a carbonic vapor phase at 17 ºC (Table 2.3; Appendix 1).  2.5.3 Pressure-temperature constraints Fluid inclusion isochores were calculated using the programs of Bakker (1999) and the data of Diamond (1992) and have been combined with garnet-amphibole-plagioclase- quartz (GAPQ) mineral equilibria (Fig. 2.11; Kohn and Spear 1989; 1990). The results indicate that the dikes intruded at a depth of approximately 7-8 km depth and at temperatures of 400-500 ºC, assuming that the GAPQ results and the intrusion of the dikes are relatively contemporaneous. This is also consistent with garnet-biotite thermometry using the thermodynamic data of Berman (1988) and the program TWQ (Berman 1991). The general lack of aluminosilicates with garnet and other minerals in the LNPG dikes generally used for thermobarometry preclude any pressure-temperature constraints based on reactions involving the aluminosilicates. However, the absence of kyanite and sillimanite confirms that the country rocks experienced pressure below the aluminosilicate triple junction (Holdaway 1971), and <3.2 kb (Gordey and Anderson 1993).  2.5.4 U-Pb geochronology of the dikes U-Pb isotopic compositions were determined for four fractions of clear, dark turquoise, pegmatitic apatite from sample EB1305 (also used in the Rb/Sr analysis), using a modified single collector VG-54R thermal ionization mass spectrometer equipped with 45    Parameter n range (ºC) average (ºC) stdev (ºC) TmCO2 49 -59.7 to -56.0 -57.9 0.9 Te 2 -44.2 to -40.0 -42.8 2.4 TmICE 5 -9.0 to -4.3 -6.5 1.8 TmCLATH 29  7.8 to 10.6 9.3 0.7 ThCO2 39 23.5 to 27.3  25.8 2.0 TnCLATH 10  -28.6 to -33.1 -30.5 2.0 TnICE 5 -50.0 to -44.7 -47.0 2.4 TnS-CO2 36 -100.6 to -94.5 -97.2 1.7 Table 2.3 Microthermometry summary for fluid inclusions hosted in pegmatitic quartz crystals from sample EB02.TmCO2 is the melting temperature of solid CO2, Te is the eutectic temperature of the brine in equilibrium with ice and clathrate. TmICE is the melting temperature of ice, TnICE is the nucleation temperature of ice, TmCLATH is the melting temperature of clathrate in the presence of brine plus carbonic liquid and vapour, ThCO2 is the homogenization temperature of liquid and vapour CO2, TnS-CO2 is the nucleation of solid CO2. 46 02 4 6 8 400200 600 800 10 0 T °C km kyanite 12 14 16 18 20 22 24 1000 sillimanite aluminosilicate stability fields pargasitic and tremolitic GAPQ garnet biotite thermometry fluid inclusion isochores outline of Fig. 2.13A and B g 8  3Par  +  1 aQz + Py +  2Gr A         3b  +  6An + Tr 3 b       3A  + 6An + 3FeT r aQ 3FePa + 18 z +Gr +Alm 3 + n    Ab  6A  +  4Py + 3FeT r Pa Q r A 3 rg + 18 z +  2G  +  5 lm 3 +  Ab  +  4 Al m   6A n + 3T r  Qz y 3F eP a + 18 a  +  5 P  +  2 G r andalusite 1 2 3 4 5 6 7 8 kbar Figure 2.11 Pressure and temperature conditions of dike crystallisation obtained from quartz byfluid inclusions microthermometry. Combined with constraints on the P-T conditions in the host rock from GAPQ and garnet-biotite thermometry indicating a maxima of ~ 2 kbar (~ 6-9 kilometres depth) and ~ 400-500 °C and related to the hostrock metamorphic grades. Fluid inclusion measurements, this study; thermometry previously undertaken by D. Marshall (unpublished data). Aluminosilicate fields taken from Holdaway (1971). Details of pargasitic and tremolitic GAPQ equilibria from Kohn and Spear (1989 and 1990) and garnet-biotite geothermometry from Berman (1988 and 1991). 47 an analogue Daly photomultiplier (Table 2.4; Appendix 2). A 238U/206Pb isochron age of 94±14 Ma (MSWD = 2.4) was calculated for the four apatite analyses, with initial 206Pb/204Pb = 19.6 ± 0.5. However, the regression can be anchored to the Pb isotopic composition of coexisting alkali feldspar grains (Chamberlain and Bowring 2000), and this regression has been anchored using feldspar compositions of nearby samples taken from Mauthner et al. (1995).  Although the Pb isotopic composition of feldspars (mostly fine-grained albite) that coexist with the dated apatite grains in this study was not determined Mauthner et al. (1995) analyzed feldspars from two separate dikes from near the apatite locality toward the northern end of the pegmatite field and obtained nearly identical compositions, suggesting that the Pb isotopic composition of the magma that formed the pegmatites was reasonably uniform, at least within small areas of the field. A third feldspar sample reported by Mauthner et al. (1995) from a locality approximately 3 km to the southeast yielded a slightly more radiogenic composition.  Regressing the four apatite analyses with those of the two feldspars from the northern end of the field yields an isochron age of 90.3 ± 1.9 Ma with an MSWD of 2.7 (Fig. 2.12A).  Including the third feldspar analysis in the regression yields essentially the same isochron age (89.6 ± 2.5 Ma) but with a much larger MSWD (29).  We believe that the age calculated using only the two northern feldspar samples gives the best estimate for the emplacement age of the pegmatites. The U-Pb system in apatite has been determined to have a relatively high closure temperature (~620 °C, Krogstad and Walker 1994; ~450 °C, Chamberlain and Bowring 2000) although it is infrequently used for U-Pb dating because the typically low ratio of radiogenic to common Pb results in high errors in calculated ages.  2.5.5 40Ar/39Ar geochronology of the dikes Step-heating 40Ar/39Ar dating studies of muscovite and lepidolite grains from three pegmatite samples were employed to determine the age of cooling below the relevant mineral closure temperatures (~425° C muscovite, Harrison et al. 2009; ~300° C lepidolite, Smith et al. 2005). Analytical data and results are presented in Table 2.5. Muscovite sample EB02 produced an inverse isochron age of 77.1 ± 3.6 Ma (MSWD = 0.48), with a 40Ar/36Ar of 243 ± 65 (Fig. 2.12B). Three samples from the same large grain of igneous muscovite (~2 cm diameter) from sample EB0509a produced inverse isochron 48 Table 2.4 Table of analytical data, weight, U and Pb concentrations and isotopic ratio results obtained. A1 271 342 1.70 76.5 17.5 319.2 0.13 24.21 0.15 0.582 A2 608 270 1.61 77.1 40.5 244.4 0.11 23.07 0.27 0.491 A3 549 253 1.65 79.9 38.2 219 0.13 22.7 0.17 0.852 A4 510 243 1.75 94.6 42.2 176.6 0.09 22.18 0.15 0.771 Total common Pb (ng) Th/USample EB1305e (fractions) 1 corrected for 1 pg blank U 2 radiogenic and common Pb, corrrected for spike and 5.5 pg blank Pb; 206Pb/204Pb = 17.4, 206Pb/207Pb=15.0 and 208Pb/204Pb= 36.4 3correlation coefficient ±1sd % 206Pb/204Pb ±1sd % rho3Weight (mg) U ppm1 Pb ppm2 238U/204Pb 49 0 20 40 Cumulative Ar % 39 60 80 100 0 20 40 60 80 100 120 A p p a re n t a g e (M a ) EB02 muscovite B 15 steps corresponding to 99.87% of the Ar39 74.4±1.8 Ma MSWD = 0.61 0.0000 0.0004 0.0008 0.0016 0.0020 0.0024 0.03 0.05 0.07 0.09 77.1±3.6 Ma Ar/ Ar=243±6540 36 MSWD=0.48 36 Ar 40 Ar 39 Ar/ 40 Ar 0.0012 EB1305 apatite A 2 0 6 2 0 4 P b / P b 238 204 U/ Pb 18 20 24 26 400100 300200 90.3 1.9 Ma Initial Pb/ Pb = 19.666 0.045 MSWD=2.7 206 204 ± ± 0 22 Figure 2.12 Geochronological results. A) U/Pb isochron obtained from magmatic apatite in sample EB1305. B) Ar/ Ar spectra from the analysis of mica samples in EB02. Steps used to calculate plateau ages for Ar/ Ar step-heating experiments are marked with an arrow. Initial Ar/ Ar values are within error of the accepted atmospheric value for each analysis. Standard deviation (SD) on all analyses is 2s. 40 39 40 39 40 36 50 0 20 40 60 80 100 39Cumulative Ar % 50 75 100 50 75 100 50 75 3912 steps corresponding to 99.2% of the Ar 396 steps corresponding to 96.40% of the Ar 393 steps corresponding to 90.2% of the Ar 100 79.87±0.83 Ma MSWD=0.85 78.58±0.69 Ma MSWD=1.3 83.8±1.3 Ma MSWD=0.65 0.000 0.001 0.003 0.004 0.04 0.06 0.08 39Ar/40Ar 0.00 0.02 1 2 3 1      83.7±1.9 Ma 40 36Ar/ Ar=297±16 MSWD=0.68 2      78.82±0.86 Ma 40 36Ar/ Ar=291±10 MSWD=1.5 3      80.3±2.1 Ma 40 36Ar/ Ar=289±31 MSWD=1.6 A p re nt a e M ) pa  g  ( a EB0509 muscovite 3 samples C 36Ar 40Ar 0.002 0 20 40 60 80 100 0 20 40 60 80 100 120 n g ) A pp ar e t a e (M a 140 65.89±0.37 Ma MSWD = 1.09  39Cumulative Ar % 3910 steps corresponding to 74.8% of the Ar EB1304 lepidolite D 0.00005 0.00015 0.00025 0.00045 0.079 0.083 0.087 0.091 39Ar/40Ar 65.84±0.82 Ma 40 36Ar/ Ar=299±44 MSWD=1.2 36Ar 40Ar 0.00035 Figure 2.12 contd. 40 39Geochronological results. C) and D) Ar/ Ar spectra from the analysis of mica samples in EB0509 and EB1304 respectively. Steps used to calculate plateau 40 39 40 36ages for Ar/ Ar step-heating experiments are marked with an arrow. Initial Ar/ Ar values are within error of the accepted atmospheric value for each analysis. Standard deviation (SD) on all analyses is 2s. 51   Sample name EB02 EB0509 (1) EB0509 (2) EB0509 (3) EB1304 Sample location Cirque 4 (south side) Cirque 2 (south side) Cirque 2 (south side) Cirque 2 (south side) Cirque 4 (south side) Mineral silver muscovite translucent  muscovite translucent  muscovite translucent  muscovite lilac lepidolite Inverse isochron age (Ma) 77.1±3.6 83.7±1.9  78.82±0.86  80.3±2.1  65.84±0.82 Isochron MSWD 0.48 0.68 1.5 1.6 1.2 Initial 40Ar/36Ar 243±65 297±16 291±10 289±31 299±44 Plateau age (Ma) 74.4±1.8 83.8±1.3 78.58±0.69 79.87±0.83 65.89±0.37 Plateau % of 39Ar 99.87 99.2 96.4 90.2 74.80 Plateau MSWD 0.48 0.65 1.3 0.85 1.09 Integrated age (Ma) 74.05±2.15 83.71±1.35 78.35±0.87 78.81±1.09 67.82±0.76 Standard deviation on the analytical error is 2σ.The parameter J used in the process was determined using Fish Canyon Tuff sanidine, with an age of 28.03±0.18 Ma, as a flux monitor (Renne et al. 1994).   Table 2.5 Sample information and results from 40Ar/39Ar analyses of pegmatitic micas. 52 ages of 83.7 ± 1.9 Ma (MSWD = 0.68) 78.8 ± 0.9 Ma (MSWD=1.5) and 80.3 ± 2.1 Ma (MSWD = 1.6), with calculated initial 40Ar/36Ar ratios of 297 ± 16, 291 ±  0 and 289 ± 31, respectively (Fig. 2.12C).  The lack of agreement in the isochron ages from separate portions of the same muscovite grain indicates that variable Ar loss has affected the grain.  We interpret the oldest age obtained (83.7 ± 1.9 Ma) to be a minimum crystallization age of formation of the muscovite.    The third sample, EB1304e, gives an inverse isochron age for lepidolite of 65.8 ± 0.8 Ma (MSWD = 1.2), with an initial 40Ar/36Ar ratio of 299 ± 44 (Fig. 2.12D).  Initial steps from this sample recorded anomalous compositions and are not included in the age calculation.  2.5.6 Rb/Sr geochronology and initial Sr ratio of the dikes The strontium isotopic compositions of muscovite, apatite, albite and whole rock powder from sample EB1305e were measured on a Thermo Finnigan thermal ionization mass spectrometer (TIMS; Table 2.6, Appendix 3). The four samples yield a four-point Rb/Sr isochron (Fig. 2.13A) with a slope corresponding to an age of 79 ± 11 Ma (MSWD = 61), and initial Sr ratio of 0.742 ± 0.045.  This calculated initial ratio is extremely high in comparison to the nearby mid-Cretaceous, crustally derived, granitic Rudi intrusion  (87Sr/86Sr = 0.72048 ± 0.000018; Heffernan 2004). Excluding the muscovite analysis and recalculating the age using only the apatite and feldspar analyses yields a two-point isochron age of 90.2 ± 2.9 Ma (Fig. 2.13B). These results and concerns about the use of Rb/Sr in dating pegmatites are discussed below.  2.6 DISCUSSION The LNPG provides good field examples of the REL-Li class of pegmatite. This class forms part of the L (lithium) –C (caesium) –T (tantalum) -type pegmatite family (enriched in those elements), which is interpreted to be genetically associated with late orogenic, peraluminous granites from undepleted supracrustal and basement gneiss protoliths (Černý and Ercit 2005) Exceptional 3D exposure of the pegmatitic dikes of the LNPG provides a rare opportunity to evaluate the influence of pre-existing structures on the emplacement of a 53 Table 2.6 Sample information and results from Rb/Sr analyses of minerals and whole rock samples. Sample EB1305 Rb ppm Sr ppm 87Rb/86Sr Error     2% 87Sr/86Sr Error (+/-2s) Apatite 454.1642 1642.8191 0.792049292 0.015840986 0.736132 0.000007 Muscovite 4896.824552 25.1396 594.7051208 11.89410 1.408572 0.000010 Albite 3253.5475 57.3105 166.0134392 3.320268783 0.947973 0.000008 Whole rock 1265.7574 43.8125 84.53004348 1.69060087 0.829320 0.000009 Whole rock duplicate 1285.60333 43.9702 83.52474955 1.670494991 0.829320 0.000009 Standards 87Sr/86Sr Error (+/-2s) Cycles 86Sr/88Sr SRM987 600ng 0.710234 0.000007 124 0.1192 SRM987 600ng 0.710231 0.000009 125 0.1193 54 0.6 0.8 1.0 1.2 1.4 1.6 0 200 400 600 800 87 86Rb/ Sr 87 86 Sr / Sr EB1305a where no error bars are shown error is smaller than symbol size muscovite whole rock apatite 79 ± 11 Ma Initial Sr/ Sr = 0.742 87 86 ± 0.045 MSWD=61 albite apatite albite EB1305a 90.2 ± 2.9 Ma Initial Sr/ Sr = 0.735116 87 86 ± 0.000033 errors are smaller than symbol size 1.0 0.80 0.68 0.96 0.92 0.88 0.84 0.76 0.72 0 40 80 120 200160 87 86Rb/ Sr 87 86 Sr / Sr Figure 2.13 Geochronological results from Rb/Sr analysis of whole rock and mineral separates from sample EB1305a. A ) Four-point errorchron (muscovite, albite, whole rock and apatite) indicating an age of 79 ± 11 Ma (MSWD=61). B) Two-point isochron (albite and apatite) indicating an age of 90.2 2.9 Ma. A B 55 dike swarm of this pegmatite class. The progenitor magmatic body for the LNPG pegmatite dikes has not been identified and probably lies undetected at depth. This is not uncommon and has been postulated for other pegmatite fields (e.g., Cap de Creus; Martin and Vito 2005 and references therein). The granitic source for rare element pegmatitic dikes is commonly located within anticlinal cores (Černý 1991) from where the magma is capable of propagating several kilometres (commonly 1-3 km; London 2008) from its magmatic source.  2.6.1 Origins of the magma An anatectic or magmatic origin for pegmatites has been debated due, for example, to their composition or lack of progenitor magma chamber (e.g., Stewart 1978, Damm et al. 1992). Pegmatitic textures are recorded from purely anatectic melts after only relatively little fractionation (e.g., Martin and Vito 2005) and anatexis of undepleted metasedimentary rocks can provide many of the appropriate elements. However, it is generally accepted that rare element pegmatitic dikes, such as the LNPG, require the extensive magmatic differentiation of a less evolved magma to achieve their strongly fractionated composition (e.g., Černý 1991) and to saturate the melt, e.g., Nb and Ta oxides (Linnen 1998). The production of Li-rich pegmatitic dikes such as the LNPG is typically related to: • low degrees of partial melting of an undepleted metapelitic protolith, with average REE abundance, where Li is readily incorporated into the melt after muscovite breakdown. • high degrees of magmatic fractionation, typically followed by varying degrees of hydrothermal alteration and replacement (e.g., Walker et al. 1989; Černý 1991; Raimbault et al. 1995; Černý and Ercit 2005).  They are therefore commonly associated with S-type peraluminous granites initially derived from metapelitic protoliths (Patiño Douce and Johnston 1990) at temperatures as low as 750 °C for muscovite-bearing source rocks (Patiño Douce and 56 Harris 1998) corresponding to a depth of ~25-30 km given an average geothermal gradient of 25-30 °C /km).  2.6.2 Geological setting and cooling history Highly evolved, rare element pegmatite dikes contain few minerals suitable for geochronological study, and even fewer with high closure temperatures able to record crystallization ages that are less likely to be reset. However, by employing apatite (U-Pb and Rb/Sr), mica (40Ar/39Ar and Rb/Sr) and feldspar and whole rock (Rb/Sr) in this study we were able to evaluate mineral systems with closure temperatures ranging from 620- 300 °C. The mineral grain sizes, isotopic systems used and relevant closure temperatures of analyses from this study are presented in Table 2.7 with additional information from Mauthner et al. (1995). Although there are still some questions to be answered the new results have helped clarify some aspects of the geochronological history of the LNPG and put them in the context of the deformational history recorded in the host rock lithologies. First-order influences on the thermal history of the area during the Mesozoic include the effect on the geothermal gradient during orogenesis and associated uplift and exhumation that occurred since mid-Jurassic time, heat conduction from the felsic magmatism in the region occurring for 10 m.y. prior to emplacement of the dikes, and the elevated concentration of heat producing elements Th, U and K in the metasedimentary country rocks. All these factors are expected to have elevated the geothermal gradient, moreover, the extended period of orogenesis may have produced close to a thermal steady state (Willett and Brandon 2002). The paleogeothermal gradient for the area has not been constrained; however, in describing other orogenies Burbank et al. (1996) used a range between 35-60 °C/km for the geothermal gradient of the Himalayas, whereas other authors use more conservative values (e.g., 20 °C/km; Marshall et al. 1997) for the Variscan orogeny which incorporated more mafic and carbonate sequences than in the LNPG area. Extremely high rates of erosion would raise the geothermal gradient; however, although the topography is expected to have been elevated by orogeny, only 8 km of overburden has been eroded in the last 90 m.y. This indicates an erosion rate of 0.08 km/ m.y. that must include the significant effects of recent glacial erosion and is low in comparison to average current erosion rates for the mountainous Olympic Peninsula, 57  Location Sample Mineral Dimensions Isotopic system Closure T ° C References LNPG1 EB1305 apatite ~ 10 mm U/Pb 450a, 620b a Chamberlain and Bowring 2000, b Krogstad et al . 1994 LNPG1 EB1305 muscovite, albite, apatite, whole rock ~ 10 mm Rb/Sr 550 (muscovite) Purdy and Jäger 1976 LNPG1 EB02 muscovite 1-5 mm Ar40/Ar39 380-425 Hames and Bowring 1994, Harrison et al . 2009 LNPG1 EB0509 muscovite 10-20 mm Ar40/Ar39 380-425 Hames and Bowring 1994, Harrison et al. 2009 LNPG1 EB1304 lepidolite 1-5 mm Ar40/Ar39 300-350 Smith et al.  2005 LNPG2 various columbite ~ 50 mm U/Pb >lower amphibolite facies Romer and Smeds 1992 LNPG2 235 muscovite 2-4 mm K/Ar 380-425 Hames and Bowring 1994, Harrison et al. 2009 LNPG2 197 lepidolite fine-grained K/Ar 300-350 Smith et al . 2005 LNPG2 230 lepidolite 4-8 mm K/Ar 300-350 Smith et al.  2005 LNPG1 this study LNPG2 Mauthner et al.  1995 Table 2.7 Mineral sizes, isotopic systems, and relevant closure temperatures of geochronological analyses undertaken for this study. Includes data from Mauthner et al.  (1995). 58 Washington State (0.18 to 0.32 km/m.y, reported by Brandon et al. 1998). A conservative range for the geothermal gradient of the LNPG area at ~90 Ma lies between 25-45 °C/km; appropriate for a long-lived orogeny with scattered small-scale, felsic intrusions, and would have produced country rock temperatures between  ~200-360 °C at 8 km depth. However, the presence of a magma chamber at depth would increase the geothermal gradients as illustrated by other peraluminous intrusions such as the extensive, evolved S-type Cornubian Batholith of southwest England with its Sn-W mineralization and minor Li-mica granites (Chappell and Hine 2006; and references therein) where a current geothermal range of 24-76 °C/km has been recorded. Petrographic observations of the metamorphic minerals place constraints on the relative timing of the magmatism which occurred with hornfels porphyroblasts forming between post D1 and late or post D2 at approximately the time the dikes were emplaced along the pre-existing S0 and S1 planes of weakness (Fig. 2.9). It is possible that the emplacement of the progenitor magma for the dikes induced the elevated temperature required to develop the porphyroblasts. Although no intrusions have been identified in the vicinity of the LNPG from fieldwork or airborne radiometric and magnetic data (Charbonneau 2007) it is proposed that the magmatic chamber associated with the formation of the dikes was also the heat source for the contact metamorphism and is expected to be located at depth. The width of the LNPG dikes (~1 m wide average), the small total volume (<1 km3) and relatively low magmatic temperature of the magma (<600 °C) suggests that the LNPG dikes themselves would have probably had little influence on the regional heat budget. Recent thermal modeling studies suggest that sub-vertical pegmatites can solidify within days to weeks of intrusion. Webber et al. (1999) calculated the crystallization of pegmatite-aplite dikes of San Diego County, California, and determined that a 1 m dike at 5 km depth cooled 100 °C from 650 °C within approximately 5 days. London (2008) calculated that a 700 °C magma that was rapidly emplaced to form 1 m wide tabular pegmatitic dikes in cold wall rocks at 6 km depth would have reached 450 °C at its centre (the crystallization temperature calculated from two-feldspar thermometry of the Little Three pegmatitic dike, Ramona, California; Morgan and London 1999) in 107 hours. The implication for the geochronology of the LNPG dikes is that the crystallization of the 59 dikes lies within error of the cooling age from the initial magma and they can be interpreted as the same event. Three distinct groups of ages; ~90 Ma, ~80 Ma and ~65 Ma have been identified using different isotopic systems (this study and Mauthner et al. 1995).  The oldest age was determined from U-Pb analysis of apatite and the grains used were similar in size (~1 cm diameter) to those used by Krogstad and Walker (1994) who proposed a closure temperature of 620 °C (Fig. 2.14A) from the Mn-rich apatite grains from the Li-rich Tin Mountain pegmatites associated with the Harvey Peak Granite of the Black Hills, South Dakota. Chamberlain and Bowring (2000) calculated a lower closure temperature for the U-Pb apatite system (~450 °C) but as fluid inclusion data from the LNPG indicate dike crystallization at <500 °C it is considered that the apatite U-Pb age records the initial cooling of the dikes.  This age coincides with the waning stages of the regional felsic magmatism. The younger ages (~80 and ~65 Ma) recorded by mica and columbite are derived from systems (40Ar/39Ar and U-Pb) with closure temperatures <~400 °C (Fig. 2.14B). Field relations suggest that all of the dikes were emplaced during a single intrusive event (dikes rarely cross-cut one another) at ~90 Ma and are expected to have quickly cooled to the temperature of the country rock, these ages are interpreted to record the overall cooling of the country rock, not the cooling of the dikes. The ages obtained from the dikes lie along a geothermal gradient of 60 °C/km when the average rate of erosion is factored in (Fig. 2.14B). The most likely source for this elevated geothermal gradient is the heat source responsible for the contact metamorphism. The ages around 80 Ma, obtained from the dikes, could be explained if that heat source was still dominating the country rock temperature and had cooled to between 425-380 °C. The ages grouped ~65 Ma may also be explained by further cooling of the country rock from 425-380 °C through the 40Ar/39Ar lepidolite closure temperature at ~300 °C (Hodges 1991) in 15 m.y. One muscovite sample from Mauthner et al. (1995) provided a cooling age of ~ 65 Ma, which does not fit with this proposed cooling history and the reason for this result has not been determined. Notwithstanding, a cooling rate of 8.3-5.3 °C/ m.y (or 83-53 °C/ km given an average erosion rate of < 0.1 mm/yr) can be calculated for the area. 60 61 An alternative explanation for the ~80 Ma and the ~65 Ma ages involves resetting of the mineral ages by separate thermal events. However, there is no supporting evidence for thermal or magmatic events in the area at ~ 80 Ma, and the only thermal event recorded at ~65 Ma lies 450 km away to the northeast (Murphy 1997). If, as proposed, the range of ages determined for the LNPG dikes records the initial emplacement of the dikes and slow cooling of the country rock in the presence of a magma chamber, that heat source would need to be sufficiently large to cool slowly. Exposure of the O’Grady Batholith, 95 km to the north of the LNPG dikes, extends over 270 km2. This batholith is the only other regional example of magmatic fractionation advanced enough to develop a small volume of pegmatite-associated Li mineralization (Ercit et al. 2003) and a pluton of a similar size would be appropriate as the progenitor for the LNPG dikes. In summary, we propose that andalusite+biotite hornfels observed in the LNPG area was likely produced by the progenitor magma for the dikes and was emplaced late to post D2. The magma responsible for the dikes intruded the country rock at a similar time and crystallized at ~8 km depth at ~90 Ma quickly cooling to the ambient country rock temperature. The subsequent rate of cooling of the country rock, as reflected by the ~80 and~65 Ma ages from the dikes, was probably dictated by slow cooling of the underlying pluton.  2.6.3 Rb/Sr results Analyses of sample EB1305e (muscovite, albite, apatite mineral separates and a whole rock sample) define a four-point Rb/Sr errorchron corresponding to a calculated age of 79 ± 11 Ma (Fig. 2.13A,Table 2.6).  The very high MSWD (= 61) suggests a significant amount of geological scatter in the data. Although the calculated Rb/Sr age lies within error of the apatite U/Pb age of ~90 Ma for this same sample, the large degree of scatter in the Rb/Sr data makes the significance of calculated Rb/Sr age dubious. The geological scatter is believed to derive from some degree of open-system behaviour and mobility of Sr in late stage magmatism. Radiogenic 87Sr has been interpreted to be especially mobile in muscovite (Küster 1995) in granitic pegmatites.  Excluding the muscovite analysis and recalculating the age using only the apatite and feldspar analyses 62 yields a two-point isochron age of 90.2 ± 2.9 Ma (Fig. 2.13B) while this age concurs with U-Pb and 40Ar/36Ar results two-point isochrons are statistically meaningless (Ludwig 2003). The initial Sr ratio calculated from the data is very elevated (87Sr/86Sri = 0.742 ± 0.0440 to 0.735116 ± 0.000033 from the two isochrons respectively) in comparison to nearby intrusions of a similar age and is likewise disregarded. The initial Sr signature for the source rock for the magma that evolved to form the pegmatitic dikes is thought to be similar to that determined by Heffernan (2004; 87Sr/86Sr = 0.72048 ± 0.000018) from the nearby mid-Cretaceous Rudi granitic intrusion, indicating a dominantly crustal source.  2.6.4 Dike formation When the fluid pressure of magma exceeds the lithostatic load due to volatile build-up during crystallization of a magma chamber, the outer crystalline carapace can become fractured (Burnham and Ohmoto 1980). Over-pressuring is a well recognized method for the dispersal of evolved magma and can be used to explain the expulsion of the volatile- and flux-rich, buoyant magma of the LNPG dikes into the country rock. Li-bearing rare element pegmatite dikes are among the most highly fractionated and located furthest from their magmatic source according to the zonation of pegmatitic rocks by their geochemistry proposed by Trueman and Černý (1982), and later modified by Černý (1991). A flux component in the magma enable it to disperse long distances (Černý 1991). As anhydrous minerals, such as feldspar and quartz crystallize, incompatible components such as H2O, B, F, P (e.g., Wolf and London 1995; Linnen 1998; Bea and Montero 1999) build up in the remaining magma in the manner of isobaric ‘second boiling’ (Candela 1997). The presence of B and P increase the solubility of H2O and with F they lower the temperature of minimum melting (London 1992). Network modifiers such as the alkali (e.g., Li, Na, K, Rb, Cs) and alkali-earth elements (e.g., Be) can also build up during magmatic fractionation and they are known to restrict polymerization in the silicate melt by disrupting the connectivity of SiO2 tetrahedra and increasing the NBO/T ratio (the number of non-bridging oxygen atoms per tetrahedra), thereby lowering viscosity (Henderson 2005) and enhancing element diffusion.  The resulting magma becomes increasingly buoyant and can be capable of driving dike 63 propagation (Rubin 1995). In addition to lowering the viscosity, Rubin (1995) describes the fundamental role of fluxes in the dike-tip region where the pressure must exceed the hydrostatic pressure difference in order for the dike to continue propagating. Weinberg and Searle (1999) also emphasise the influence of magmatic volatiles in an example of highly mobile volatile-rich magmas that intrude foliated and non-foliated rocks in the Himalayas. They propose that early volatile-rich fluids penetrated the schistose rocks forming pathways used by the parental leucogranitic magma that followed. The LNPG dikes propagated along strength anisotropies with the sub-vertical folded bedding and axial cleavage planes providing channels for the fluid. Fluid inclusion results indicate that the LNPG dikes were emplaced at 7-8 km depth, slightly more than used in the thermal models used by Webber et al. (1999) and London (2008), however they too probably crystallized within days to weeks of emplacement. To estimate the speed of propagation, although it would be expected to be erratic, if the dikes crystallized within 10 days of intrusion and propagated an average distance of 2 km from the progenitor magma chamber during that time, the average rate of propagation would be 10-15 cm/min.  2.6.5 Primary textures reflecting conditions of crystallization Primary, magmatic crystallization textures observed in the pegmatitic dikes are related to the lithostatic and fluid pressure and temperature and the composition of the magma, especially the volatile and flux content. Crystallization of anhydrous minerals, such as in aplite, increases the volatile abundance and even in a cooling system the volatile pressure in the remaining magma is also increased (Lowenstern 1994). Rare miarolitic cavities indicate that some fluid saturation occurred at LNPG (Candela 1997) but the majority of crystallization appears to have occurred in volatile undersaturated conditions. In this environment volatile content and element diffusion rates are high, and melt viscosity and nucleation are low providing conditions conducive to pegmatitic and aplitic crystal growth and the typical primary textures seen in pegmatites (London et al. 1989).  Dike propagation occurs when the hydrostatic pressure difference in the dike tip region exceeds lithostatic pressure (Rubin 1995), and the resulting rapid decrease in fluid 64 pressure in the dike can induce saturation and exsolution of volatiles. Pegmatite-aplite textures are well-studied (e.g., Frindt et al. 2004) with the growth of pegmatitic and aplitic crystals attributed to undercooling of the melt (e.g., Webber et al. 1999; Hammer 2008) in the presence of high concentrations of volatile components (London et al. 1989). Therefore if the system can reseal after a pressure drop caused by dike propagation, the fluid pressure will rebuild, volatile components increase in solubility and the formation of pegmatitic textures can resume (Simmons et al. 2003). Crystallization textures between individual dikes of the LNPG are highly variable and also vary on the cm-scale within dikes. This is fundamentally different from textures observed in well-zoned pegmatite examples, such as the sub-horizontal Tanco pegmatite (Černý 2005) noted for its extensive areas of monomineralic crystallization. And this variation at LNPG suggests that magma composition evolved and physical conditions fluctuated during crystallization. Overall, we envision a dynamic, fluctuating fluid pressure environment occurring during the crystallization of the LNPG dikes with buoyant, mobile volatile-rich melt rapidly propagating upward and crystallising along significant pre-existing strength anisotropies. The primary textures of the LNPG dikes are strongly linked to composition and pressure variations in the melt within the timeframe between intrusion and freezing. Field observations of primary textures provide insight into the physical conditions occurring during crystallization, and interpretations of two examples from the LNPG dikes are given below: 1. Pegmatite-aplite banding is commonly observed in the spodumene-free dikes. Micro-pegmatitic bands of mm- to cm- width are flanked by aplitic bands of similar dimensions (e.g., Fig. 2.10A) with the banding being rarely symmetrical across the dike. These textural bands (pegmatite-aplite) occur parallel to the wallrock contact while the elongate minerals within the bands are perpendicular.  Late-stage aplitic units can crosscut earlier fabrics (Mauthner 1996). Pressure build-up within the dike or the force of the pressure release may have been sufficient to cause rupturing producing the crosscutting aplitic veins. Cyclical pressure fluctuations created during dike propagation are suggested to be the primary cause of these pegmatite-aplite crystallization textures. 65 Rare miarolitic cavities are present and are either planar (between the textural bands) or rounded (towards the centre of the dikes). The formation of miarolitic cavities, would occur at during periods of low fluid pressure (Candela 1997) when ‘bubbles’ of rapidly exsolved volatiles become trapped within the dikes. 2. Elongate minerals such as spodumene, quartz or feldspar crystals tend to have a unidirectional fabric and can be randomly oriented or more commonly occur perpendicular to the wallrock contact (Mauthner 1996). The crystals can be large and extend across the entire width of the dike (e.g., spodumene crystal approximately 80 cm in length; Fig. 2.10B). In these cases the crystals are free of vein parallel textures, such as planes of fluid inclusions or bands of crystal growth within the elongate crystals, and the absence of these textures indicates that they formed during one growth episode. A hiatus in propagation of a few days may provide a period of equilibrium between hydro and lithostatic pressure enabling uninterrupted crystallization to occur. Crystallization of elongate minerals perpendicular to the pegmatite/host-rock contact occurs when the crystal growth rate greatly outstrips the nucleation rate. This texture has been identified elsewhere by Petersen (1985) and described as ‘competitive growth occurring parallel to heat flow’. Grain orientation perpendicular to the host-rock contact may promote more rapid cooling in a similar manner to the convection cells occurring perpendicular to the cooling front in basaltic flows and resulting in columnar jointing (Lee 1988).  2.6.6 Tectonic implications Controls on the orientation of pegmatitic dikes follow the same principles as other dike-like intrusions, and most critically depend on the prevailing regional stress field. In his work on the mechanisms involved in pegmatite intrusion Brisbin (1986) identified the consequences of interplay between changes in lithological and fluid pressures, and the affect of depth (equating to a brittle or ductile environment) and pre-existing strength anisotropies in the host rock.  The lack of cross-cutting relationships between the dikes of the LNPG reflects the strong influence of the host rock foliation on pegmatite emplacement but also suggests that the dikes were emplaced during the same regional 66 stress field conditions and probably contemporaneously.  Generally flat-walled, sub- vertical pegmatitic dikes with a preferred orientation, such as LNPG, are interpreted to have formed by intrusion into brittle rheological conditions (<10 km depth) with minimal stress parallel to the opening direction (Brisbin 1986). The minimal stress field at the time of emplacement was therefore aligned northeast-southwest, although no evidence links dike emplacement to a period of significant extension. The age of the minor brittle deformation of the dikes and the host rock is not known although it must obviously be younger than ~90 Ma. The age limit of deformation in this area was previously determined from regional mid-Cretaceous plutons cutting Mesozoic folds at ~94 Ma (Gordey and Anderson 1993) and a minor revision to that age limit should be made. The LNPG dikes commonly align with pre-existing S1 cleavages and well-defined S0 bedding planes, relating to the asymmetrically folded Fork anticlinorium and these anisotropies were used as conduits by the magma. We propose that the progenitor magma for the dikes intruded late to post D2 and was responsible for the contact metamorphism in that area. Dike propagation occurred at a similar time and was driven by buoyant, volatile-rich magma (Rubin 1995) along steeply inclined strength anisotropies during a hiatus in Mesozoic compression. The majority of the dikes are located on the eastern limb of the Fork anticlinorium where asymmetric folding creates a high concentration of sub- vertical planes (i.e., bedding). In his work on the formation of pegmatitic dikes of the Yellowknife area, Northwest Territories, Kretz (1968) recognized magma migration along the more steeply- dipping pre-existing strength anisotropies. Similarly, we propose that the magma responsible for the LNPG dikes preferentially migrated along steeply dipping, pre-existing S0 and S1 planes rather than the more shallowly dipping bedding planes to the west of the axial surface which would have presented a greater mechanical barrier. 2.7 CONCLUSIONS Formation of the northeast verging, asymmetrically folded Fork anticlinorium during the Mesozoic configured steeply dipping strength anisotropies from the well- defined bedding planes of basinal deposits and pervasive S1 axial-planar cleavage. This 67 was followed by a less significant period of deformation (D2) that formed crenulation cleavage in some areas. An unexposed intrusion at depth beneath the LNPG is postulated to have been the magma source of the pegmatitic dikes, in addition to being the cause of the contact metamorphism and elevation of the overall geothermal gradient in the immediate area of the dikes. Contact metamorphism (assumed to be caused by the presence of the magma chamber) occurred late to post D2, illustrated by S2 fabrics incorporated in porphyroblasts. Intrusion of buoyant, volatile-rich magma occurred ~90 Ma at approximately 7-8 km depth. Sub-vertical strength anisotropies concentrated on the eastern limb of the Fork anticlinorium provided significant conduits for the magma that formed the rare element mid-Cretaceous Little Nahanni Pegmatite Group. The volatile-rich melt cooled rapidly to ambient temperatures and crystallized in situ. Variations in volatile solubility and fluctuating hydrostatic pressure induced by dike propagation are recorded in the primary magmatic textures. Assuming an average erosion rate for the area to the cooling ages of various pegmatitic minerals that were dated suggests an elevated paleogeothermal gradient which is proposed to have been caused by the presence of a progenitor magma chamber for the dikes, located at depth. 2.8 ACKNOWLEDGEMENTS This study was generously supported by War Eagle Mining Co., Ltd. and NSERC Discovery Grants to LG. Thanks also go to the Geological Survey of Canada and Hendrik Falck, in particular, for field support. Excellent field assistance over several seasons was variously provided by Tashia Dzikowski, Allison Brand, Anita Lam, Marie-Eve Caron and Peter Weir. Many thanks go to Bruno Kieffer, Rich Friedman and Tom Ullrich for conducting the Rb-Sr, U-Pb and 40Ar/39Ar analyses, respectively. Archer Cathro & Associates (1981), Ltd is thanked for access to samples and geochemical data and Bill Wengzynowski is thanked for permitting the publication of the notes and geological data on the LNPG whole rock samples collected in preparation for a geological report. This study benefited significantly from discussions with Kelly Russell, Greg Dipple, Ken 68 Hickey, Dan Kontak and Shaun Barker any omissions or errors, however, are the responsibility of the lead author. 69 2.8 REFERENCES BAKKER, R.J. (1999): Adaption of Bowers & Helgeson (1983): Equation of state to isochore and fugacity coefficient calculation in the H2O-CO2-CH4-N2-NaCl fluid system. Chemical Geology 154, 225-236.  BARNES, E.M., GROAT, L.A. & FALCK, H. (2007): A review of the late Cretaceous Little Nahanni Pegmatite Group and associated rare-element mineralization in the Selwyn Basin area, Northwest Territories. In Mineral and Energy Resource Assessment of the Greater Nahanni Ecosystem Under Consideration for the Expansion of the Nahanni National Park Reserve, Northwest Territories (D.F. Wright, D. Lemkow and J.R. Harris, eds.), Geological Survey of Canada Open File 5344, 191–202.  BEA, F. & MONTERO, P. (1999): Behavior of accessory phases and redistribution of Zr, REE, Y, Th, and U during metamorphism and partial melting of metapelites in the lower crust: An example from the Kinzigite Formation of Ivrea-Verbano, NW Italy. Geochimica et Cosmochimica Acta, 63, 1133-1153.  BERANEK, L.P. & MORTENSEN, J.K. (2007): Investigating a Triassic overlap assemblage in Yukon: On-going field studies and preliminary detrital-zircon age data. In Yukon Exploration and Geology 2006 (D. S. Emond, L. L. Lewis and L. H. Weston, eds.). Yukon Geological Survey, 83-92.  BERMAN, R. G. (1988): Internally consistent thermodynamic data for minerals in the system Na2O—K2O—CaO—MgO—FeO—Fe2O3—Al2O3—SiO2—TiO2—H2O—CO2. Journal of Petrology 29, 445–522.  BERMAN, G. (1991): Thermobarometry using multi-equilibrium calculations: a new technique, with petrological applications. Canadian Mineralogist 29, 833-855.  BRANDON, M. T., RODEN-TICE, M. K. & GARVER, J. L. (1998): Late Cenozoic exhumation of the Cascadia accretionary wedge in the Olympic Mountains, northwest Washington State. Geological Society of America Bulletin, 110, 985-1009.  BRISBIN, W.C. (1986): Mechanics of pegmatite intrusion. American Mineralogist 71, 644-651.  BURBANK, D. W., LELAND, J., FIELDING, E., ANDERSON, R.S., BROZOVIC, N., REID, M. R. & DUNCAN, C. (1996): Bedrock incision, rock uplift and threshold hillslopes in the northwestern Himalayas. Nature, 379, 505-510.  BURNHAM, C.W. & OHMOTO, H. (1980): Late-stage processes of felsic magmatism. Mining Geology Special Issue 8, 1-11.  CANDELA, P.A. (1997):  A review of shallow, ore-related granites: textures, volatiles, and ore metals. Journal of Petrology 38, 1619-1633.  70 ČERNÝ, P. (1991): Rare-element granitic pegmatites. Part 2: Regional to global environments and petrogenesis. Geoscience Canada 18, 68-81.  ČERNÝ, P. (2005): The Tanco rare-element pegmatite deposit, Manitoba: Regional context, internal anatomy and global comparisons. In Rare-element geochemistry and mineral deposits (R.L. Linnen and I.M. Samson, eds.). Geological Association of Canada,Short Course Notes 17, 127-158.  ČERNÝ, P. & ERCIT, T.S. (2005): The classification of granitic pegmatites revisited. The Canadian Mineralogist 43, 2005-2026.  CHAMBERLAIN, K.R. & BOWRING, S.A. (2000): Apatite-feldspar U-Pb thermochronometer; a reliable, mid-range ~450 °C diffusion-controlled system. Chemical Geology 172, 173-200.  CHAPPELL, B.W. & HINE, R. (2006): The Cornubian Batholith: an example of magmatic fractionation on a crustal scale. Resource Geology 56, 203-244.  CHARBONNEAU, B.W. (2007): Evaluation of airborne radiometric and magnetic data in the vicinity of the Nahanni National Park Reserve, Northwest Territories, Canada. In Mineral and Energy Resource Assessment of the Greater Nahanni Ecosystem Under Consideration for the Expansion of the Nahanni National Park Reserve, Northwest Territories (D.F. Wright, D. Lemkow and J.R. Harris, eds.). Geological Survey of Canada Open File 5344, 99–124.  DAMM, K.-W., HARMON, R. S., HEPPNER, P.-M. & DORNSIEPEN, U. (1992):  Stable isotope constraints on the origin of the Cabo de Creus garnet-tourmaline pegmatites, Massif des Alberes, Eastern Pyrenees, Spain. Geological Journal 27, 75-86.  DIAMOND, L.W. (1992): Stability of CO2-clathrate-hydrate + CO2 liquid + CO2 vapour + aqueous KCl-NaCl solutions: Experimental determination and application to salinity estimates of fluid inclusions. Geochimica et Cosmochimica Acta, 56, 273-280.  ENGEBRETSON, D.C., GORDON, R.G. & COX, A. (1985): Relative motions between oceanic and continental plates in the Pacific basin. Geological Society of America Special Paper 206, 59.  ERCIT, T.S., GROAT, L.A. & GAULT, R.A. (2003): Granitic pegmatites of the O'Grady Batholith, N.W.T., Canada; a case study of the evolution of the elbaite subtype of rare- element granitic pegmatite. The Canadian Mineralogist 41, 117-138.  FRINDT, S., HAAPALA, I. & PAKKANEN, L. (2004): Anorogenic Gross Spitzkoppe granite stock in central western Namibia: Part I. Petrology and geochemistry. American Mineralogist 89, 841-856.  71 GORDEY, S.P. & ANDERSON, R.G. (1993): Evolution of the northern Cordilleran miogeocline, Nahanni map area [105I], Yukon and Northwest Territories. Geological Survey of Canada Memoir 428.  GROAT, L.A., ERCIT, T.S., RAUDSEPP, M. & MAUTHNER, M.H.F. (1994): Geology and mineralogy of the Little Nahanni Pegmatite Group, part of NTS area 105 I/02. Economic Geological Survey Open File 1994-14. DIAND, NWT.  GROAT, L.A., MULJA, T., MAUTHNER, M.H.F., ERCIT, T.S., RAUDSEPP, M., GAULT, R.A. & ROLLO, H.A. (2003): Geology and mineralogy of the Little Nahanni rare-element granitic pegmatites, Northwest Territories. The Canadian Mineralogist 41, 139-160.  HAMES, W.E. & BOWRING, S.A. (1994): An empirical evaluation of the argon diffusion geometry in muscovite. Earth and Planetary Science Letters 124, 161-169.  HAMMER, J.E. (2008): Experimental studies of the kinetics and energetics of magma crystallization. Reviews in Mineralogy and Geochemistry 69, 9-59.  HARRISON, T.M., CELERIER, J., AIKMAN, A.B., HERMAN, J. & HEIZLER, M.T. (2009): Diffusion of 40Ar in muscovite. Geochimica et Cosmochimica Acta, 73, 1039-1051.  HART, C.J.R. & LEWIS, L.L. (2006): Gold mineralization in the upper Hyland River area: a non-magmatic origin. In Yukon Exploration and Geology 2005 (D.S. Emond, G.D. Bradshaw, L.L. Lewis and L.H. Weston, eds.), Yukon Geological Survey, 109-125.  HART, C.J.R., MAIR, J.L., GOLDFARB, R.J. & GROVES, D.I. (2004): Source and redox controls on metallogenic variations in intrusion-related ore systems, Tombstone-Tungsten Belt, Yukon Territory, Canada. Transactions of the Royal Society of Edinburgh: Earth Sciences 95, 339–356.  HEFFERNAN, R. S. (2004): Temporal, geochemical, isotopic, and metallogenic studies of mid-Cretaceous magmatism in the Tintina Gold Province, southeastern Yukon and southwestern Northwest Territories, Canada. M.Sc. thesis, University of British Columbia, Vancouver, British Columbia.  HENDERSON, G.S. (2005): The structure of silicate melts: A glass perspective. The Canadian Mineralogist 43, 1921-1958.  HODGES, K.V. (1991): Pressure-Temperature-Time Paths. Annual Review of Earth and Planetary Sciences 19, 207-236.  HOLDAWAY, M. J. (1971): Stability of andalusite and the aluminum silicate phase diagram. American Journal of Science 271, 97-131.  72 JAHNS, R.H. (1982): Internal evolution of pegmatite bodies. In Granitic Pegmatites in Science and Industry (Černý, P. ed.). Mineralogical Association of Canada Short Course Handbook 8, 293– 327.  KOHN, M.J. & SPEAR, F.S. (1989): Empirical calibration of geobarometers for the assemblage garnet + hornblende + plagioclase + quartz. American Mineralogist 74, 77- 84.  KOHN, M.J. & SPEAR, F.S. (1990): Two new barometers for garnet amphibolites with applications to southeastern Vermont. American Mineralogist, 75, 89-96.  KREMER, P.D. & LIN, S. (2006): Structural geology of the Bernic Lake area, Bird River greenstone belt, southeastern Manitoba (NTS 52L6): implications for rare element pegmatite emplacement. Manitoba Science, Technology, Energy and Mines, Manitoba Geological Survey, Report of Activities, 206–213.  KRETZ, R. (1968): Study of pegmatite bodies and enclosing rocks, Yellowknife-Beaulieu region, District of Mackenzie. Geological Survey of Canada Bulletin 159.  KROGSTAD, E. J. & WALKER, R. J. (1994): High closure temperatures of the U-Pb system in large apatites from the Tin Mountain pegmatite, Black Hills, South Dakota, USA. Geochimica et Cosmochimica Acta 58, 3845-3853.  KÜSTER, D. (1995): Rb-Sr isotope systematics of muscovite from Pan-African granitic pegmatites of Western and Northeastern Africa. Mineralogy and Petrology 55, 71-83.  LEE, L.J. (1988): Origin of columnar jointing in recent basaltic flows, Garibaldi area, southwest British Columbia. M.Sc. thesis, University of Calgary, Calgary, Alberta.  LINNEN, R.L. (1998): The solubility of Nb-Ta-Zr-Hf-W in granitic melts with Li and Li+F: Constraints for mineralization in rare metal granites and pegmatites. Economic Geology 93, 1013-1025.  LONDON, D. (1992): The application of experimental petrology to the genesis and crystallization of granitic pegmatites. The Canadian Mineralogist 30, 499-540.  LONDON, D. (2008): Pegmatites. Mineralogical Association of Canada, Québec, Canada.  LONDON, D., MORGAN, G.B. & HERVIG, R.L. (1989): Vapor-undersaturated experiments with Macusani glass + H2 0 at 200 MPa, and the internal differentiation of granitic pegmatites. Contributions to Mineralogy and Petrology 102, 1-17.  LOWENSTERN, J.B. (1994): Dissolved volatile concentrations in an ore-forming magma. Geology 22, 893-896.  LUDWIG K.R. (2003): ISOPLOT 3: A geochronological toolkit for Microsoft Excel. 73 Berkeley Geochronology Centre Special Publication 4, 74 p.  MAIR, J.L., HART, C.J.R. & STEPHENS, J.R. (2006): Deformation history of the northwestern Selwyn Basin, Yukon, Canada; implications for orogen evolution and Mid- Cretaceous magmatism. Geological Society of America Bulletin 118, 304-323.  MARSHALL, D., KIRSCHNER, D. & BUSSY, F. (1997): A Variscan pressure-temperature- time path for the N-E Mont Blanc massif. Contributions to Mineralogy and Petrology 126, 416-428.  MARTIN, R.F. & DE VITO, C. (2005): The patterns of enrichment in felsic pegmatites ultimately depend on tectonic setting. The Canadian Mineralogist 43, 2027-2048.  MAUTHNER, M.H.F. (1996): Mineralogy, geochemistry and geochronology of the Little Nahanni Pegmatite Group, Logan Mountains, southwestern Northwest Territories. M.Sc. thesis, University of British Columbia, Vancouver, British Columbia.  MAUTHNER, M.H.F., MORTENSEN, J.K., GROAT, L.A. & ERCIT, T.S. (1995): Geochronology of the Little Nahanni pegmatite group, Selwyn Mountains, southwestern Northwest Territories. Canadian Journal of Earth Sciences 32, 2090-2097.  MORGAN, G.B. & LONDON, D. (1999): Crystallization of the Little Three layered pegmatite-aplite dike, Ramona District, California. Contributions to Mineralogy and Petrology 136, 310-330.  MORTENSEN, J.K., HART, C.J.R., MURPHY, D.C. & HEFFERNAN, S. (2000): Temporal evolution of Early and mid-Cretaceous magmatism in the Tintina Gold Belt. In The Tintina Gold Belt: Concepts, Exploration and Discoveries, J. Jambor (ed.), British Columbia and Yukon Chamber of Mines, Special Volume 2, 49-57.  MURPHY, D.C. (1997): Geology of the McQuesten River Region, Northern McQuesten and Mayo map areas, Yukon Territory (115/P14, 15, 16; 105M/13,14). Indian and Northern Affairs Canada, Exploration and Geological Services Division, Yukon Region, Bulletin. 6.  PASSCHIER, C.W. & TROUW, R.A.J. (2006): Microtectonics. 2nd ed. Springer-Verlag, Berlin, Heidelberg, New York.  PATIÑO DOUCE, A.E. & HARRIS, N. (1998): Experimental constraints on Himalayan anatexis. Journal of Petrology 39, 689-710.  PATIÑO DOUCE, A.E. & JOHNSTON, A.D. (1990): Phase equilibria and melt productivity in the pelitic system: implications for the origin of peraluminous granitoids and aluminous granulites. Contributions to Mineralogy and Petrology 107, 202-218.  74 PETERSEN, J.S. (1985): Columnar-dendritic feldspars in the Lardalite Intrusion, Oslo region, Norway: 1. Implications for unilateral solidification of a stagnant boundary layer. Journal of Petrology 26, 223-252.  RAIMBAULT, L., CUNEY, M., AZENCOTT, C., DUTHOU, J.L. & JOROY, J.L. (1995): Geochemical evidence for a multistage magmatic genesis of Ta-Sn-Li mineralization in the granite at Beauvoir, French Massif Central. Economic Geology 90, 548-576.  RASMUSSEN, K.L., MORTENSEN, J.K., FALCK, H. & ULLRICH, T.D. (2007): The potential for intrusion-related mineralization within the South Nahanni River MERA area, Selwyn and Mackenzie Mountains, Northwest Territories. In Mineral and Energy Resource Assessment of the Greater Nahanni Ecosystem Under Consideration for the Expansion of the Nahanni National Park Reserve, Northwest Territories (D.F. Wright, D. Lemkow and J.R. Harris, eds.). Geological Survey of Canada Open File 5344, 203–278.  RENNE, P.R., DEINO, A.L., WALTER, R.C., TURRIN, B.D., SWISHER III, C.C., BECKER, T.A., CURTIS, G.H., SHARP, W.D. & JAOUNI, A.-R. (1994): Intercalibration of astronomical and radioisotopic time. Geology 22, 783-786.  ROTHERHAM, D.C. (1962): Hyland River Area, Mackenzie - geophysics, geochemistry, regional geology, Canada Tungsten Mining Corporation Ltd., Assessment Report 017568.  RUBIN, A.M. (1995): Propagation of magma-filled cracks. Annual Review of Earth and Planetary Sciences 23, 287-336.  RUDNICK, R.L. & GAO, S. (2003): Composition of the continental crust. In Treatise On Geochemistry, 3, (K.K. Turekian and H.D. Holland, eds.). Elsevier–Pergamon, Oxford, UK.  SIMMONS, W.B., WEBBER, K.L., FALSTER, A.U. &  NIZAMOFF, J.W. (2003): Pegmatology. Rubellite Press, New Orleans, USA.  SMITH, S.R., KELLEY, S.P., TINDLE, A.G. & BREAKS, F.W. (2005): Compositional controls on 40Ar/39Ar ages of zoned mica from a rare-element pegmatite. Contributions to Mineralogy and Petrology 149, 613-626.  SPEAR, F.S. (1995): Metamorphic phase equilibria and pressure-temperature-time paths. Mineralogical Society of America Monograph 1.  TRUEMAN, D.L. & ČERNÝ, P. (1982): Exploration for rare-element granitic pegmatites. Granitic Pegmatites in Science and Industry, Mineralogical Association of Canada Short Course Handbook 8, 463-493.  75 WALKER, R.J., HANSON, G.N. & PAPIKE, J.J. (1989): Trace element constraints on pegmatite genesis: Tin Mountain pegmatite, Black Hills, South Dakota. Contributions to Mineralogy and Petrology 101, 290-300.  WEBBER, K.L., SIMMONS, W.B., FALSTER, A.U. & FOORD, E.E. (1999): Cooling rates and crystallisation dynamics of shallow level pegmatite-aplite dikes, San Diego County, California. American Mineralogist 84, 708-717.  WEINBERG, R.F. & SEARLE, M.P. (1999): Volatile-assisted intrusion and autometasomatism of leucogranites in the Khumbu Himalaya, Nepal. Journal of Geology 107, 27-48.  WENGZYNOWSKI, W.A. (2002): Geological report on the Mac property. War Eagle Mining Company Inc.  WILLETT, S.B. & BRANDON, M.T. (2002): On steady states in mountain belts. Geology 30, 175-178.  WOLF, M.B. & LONDON, D. (1995): Incongruent dissolution of REE-rich and Sr-rich apatite in peraluminous liquids - differential apatite, monazite and xenotime solubilities during anatexis. American Mineralogist 80, 765-775.  YOUNG, I. (2007): Technical report on the Mac Property - 2, War Eagle Mining Company Inc.     76 CHAPTER 3: GEOCHEMICAL EVIDENCE OF LATE-STAGE MAGMATIC FRACTIONATION IN THE DIKES OF THE RARE ELEMENT LITTLE NAHANNI PEGMATITE GROUP, NWT*  3.1 INTRODUCTION Granitic pegmatites are the product of extensive magmatic fractionation (Černý 1991a and b) involving important late-stage igneous processes that result in the redistribution of rare, and technologically useful elements (e.g., Li, Sn, W, Bi, Ta, Nb) within the crust.  Our understanding of the evolution of magma geochemistry is mainly derived from whole rock geochemical analyses, and yet some of the final stages of differentiation occurring in the process of pegmatite formation, are rarely studied in this way. Bulk rock analysis of pegmatitic rock is a challenge; typical coarse grain sizes require the collection of several kilograms of sample material and studies reporting full geochemical analyses of whole rock pegmatites are sparse. Recent data on trace element compositions have been published by Badanina et al. (2004) and Kontak (2006), and in addition, Breaks et al. (2003) and Kontak et al. (2001 and 2002) have provided rare earth element data. More information is still required before a full picture of the whole rock geochemical processes of pegmatite formation is achieved. Currently our understanding is often based on the closest geological equivalents; magmatically evolved igneous rocks with high abundances of fluxing components1 (e.g., Christiansen et al. 1984, Raimbault et al. 1995, Dostal and Chatterjee 1995, Breiter et al. 1999, Förster et al. 1999, Dostal et al. 2004); bulk calculations from combined monomineralic zones (e.g., Tanco pegmatite, Stilling et al. 2006); melt and fluid inclusions (e.g., London 1986, Thomas et al. 2000, Thomas and Webster 2005, Sirbescu and Nabelek 2003, Badanina et al. 2004); or experimental work (e.g., London et al. 1988, London et al. 1989, Webber et al.1999).  * A version of this chapter will be submitted for publication. Barnes, E.M., Weis, D. & Groat, L.A. (2010): Geochemical evidence of late stage magmatic fractionation in the dikes of the rare element Little Nahanni Pegmatite Group, NWT.  1‘Flux’ and ‘fluxing components’ are used as defined in London (2005; and references therein) as components that ‘lower the melting and crystallization temperatures…and enhance miscibility among otherwise less soluble constituents’’.  77 Progressive magmatic differentiation in pegmatites (as in granites) can be observed in the fractionation of element pairs such as K/Rb, Rb/Cs, Li/Cs, Nb/Ta, Zr/Hf (e.g., Černý 1991a, Dostal and Chatterjee 1995, London 2005a and b). Extensive differentiation can concentrate certain minerals to produce economically viable deposits of elements rarely achieved elsewhere. Examples of mines based on rare element enriched pegmatites include the Tanco pegmatite (Li, Ta, Cs, Be, Rb), at Bernic Lake, Manitoba (Černý 2005), the Bikita pegmatite (Li) near Masvingo, Zimbabwe (Norton 1983 and references therein), and the Greenbushes Pegmatite (Sn, Ta, Nb, Li) of Western Australia (Partington et al. 1995). The presence or build-up of some of these elements in the residual magma during crystallisation, especially the alkali metals (Li, Na, K, Rb, Cs) and volatile and fluxing components (e.g., OH, H2O, B, P, F), significantly alter the short-range structure of the melt by reforming its short-range polymerized networks of Si and Al tetrahedral positions and increasing the abundance of non-bridging oxygen (NBO) sites (London et al. 1989, Hannon et al. 1992, Henderson 2005, London 2005b, Soltay and Henderson 2005). This has a fundamental affect on the physical properties of the melt, including lowering the liquidus and solidus temperatures (Manning 1981), lowering viscosity (Thomas and Webster 2000, Giordano et al. 2004) and enhancing diffusivity (Dolejš and Baker 2007). Element partition coefficients are also affected by these significant changes in the structure and change the composition of the silicate melt (Mysen et al. 1982, Linnen 1998, Mysen 2004). Sparse research on silicate melts with high abundances of fluxing components has been undertaken (e.g., Fleet and Pan 1997, Suk 1998 and Veksler et al. 2005) and a thorough understanding of the influences of flux components on partition coefficients of trace elements in these systems has yet to be achieved. This study focuses on dikes of the highly evolved rare element2 Little Nahanni Pegmatite Group (LNPG; Groat et al. 1994 and 2003, Mauthner 1996) in the Northwest Territories. The dike material is primary and relatively homogeneous; grain sizes are typically restricted to <10 cm length, there is an absence of significant monomineralic  2 ‘Rare element’ in terms of the LNPG dikes refers to increased abundance of exotic elements such as Li, Cs, Ta, Nb, Rb, and Sn and should not be confused with ‘rare earth’ elements that are highly depleted in these rocks. 78 zones and few miarolitic cavities, making the collection of sufficiently large samples practical. We contribute an extensive body of geochemical data and make interpretations of the crystallisation history of the LNPG using published partition coefficient values. Trace elements indicate a high degree of magmatic fractionation comparable to units of the evolved Carboniferous granitic intrusions of the Variscan orogeny, the Erzebirge and Ehrenfriedersdorf complex in Germany (e.g., Breiter et al. 1997 and 1999, Webster et al. 1997, Förster et al. 1999, Thomas et al. 2000) and the cupola of the Beauvoir granite, Massif Central, France (Raimbault et al. 1995). In addition, we propose that highly depleted rare earth elements (REE) provide the key to some of the igneous processes including late-stage build-up and saturation of volatile and flux components fundamental to pegmatite formation. The origin of the fractionated REEN patterns from the LNPG dikes is evaluated using appropriate REE partition coefficients and the results illustrate the importance of flux components, F especially, in fractionating REE and otherwise immobile element pairs.  3.2 GEOLOGICAL SETTING 3.2.1 Location Dikes of the Cretaceous Little Nahanni Pegmatite Group (LNPG) crop out on an unnamed ridge formed by the Fork anticlinorium, in the Mackenzie Mountains, Northwest Territories, close to the Yukon border (505000E 6895000N, NTS 105 I/2; Fig. 3.1 and Fig. 3.2A; see also Chapter 2). The >200 rare element-enriched pegmatitic dikes are typically 1-2 m wide (Figs. 3.2B-F; and <10 m wide; Mauthner 1996) rarely cross-cut each other and are concentrated into several steeply-dipping ‘swarms’ consisting of 10s of dikes traceable along strike for 100s m (Wengzynowski 2002). Chapter 2 describes the structural aspects and the geochronology of dike emplacement and metamorphism in more detail, but in outline the host rock foliations (bedding and axial planar cleavage of strongly folded Neoproterozoic sedimentary strata) were utilized as conduits by <1 km3 highly mobile, low viscosity silicate magma that probably rapidly consolidated to form the pegmatites. Cooling models of dikes with similar dimensions as the LNPG (Webber et al. 1999, 79 Figure 3.1 Geologic map of the study area showing areas with the highest density of pegmatites, and numbered locations for the whole rock samples. The elevation of the sample sites is shown to the right of the map. Modified from Groat et al. (2003). Sample numbers on the diagram refer to samples referred to in the text as follows: 1. = P389913 , 2. = P389910, 3. = P389739, 4. = P389875, 5. = P389854, 6. = P389851, 7. = P389726, 8. = P389861, 9. = P389849, 10. = P389870, 11. = P389734, 12. = P389797, 13. = P389760, 14. = P389764, 15. = P389768, 16. = P389774, 17. = P389751, 18. = P389753, 19. = P389711, 20. = P389704, 21. = P389789, 22. = P389791, 23. = P389719, 24. = P389844, 25. = P389817, 26. = P389816, 27. = P389808, 28. = P389809, a. = AL050704, b. = P389859, c. = EB1304, d. EB0509. 80 AB C D E k sp a r k sp a r sp o d u m e n e Orientation of pegmatite/wallrock contact F O ri e n ta ti o n o f p e g m a ti te /w a llr o ck co n ta ct m in e ra l b a n d in g Figure 3.2 Photographs of the study area and examples of the dikes. A) view from cirque 2 to the northeast. B) photograph illustrating typical aspect of dikes splitting and encompassing a raft of the host rock. Outline indicates area of panel D. C) example of megacrystic mineral growth in float block. Pegmatite/country rock boundaries occur at top and bottom of block. Elongate minerals, spodumene and feldspar, typically occur perpendicular to this contact. D) Photograph of dike (indicated in panel B) showing areas of regular compositional banding. E) and F) annotated photographs illustrating crystallisation textures. E) long crystal axes perpendicular to dike/host rock contact and F) mineral banding and pegmatite/aplite banding. White arrows in E) and F) point from host rock contact into pegmatite. 81 London 2008) predict consolidation within days to weeks. The dikes crystallised at ~7-8 km depth at ~90 Ma (Chapter 2; fluid inclusion microthermometry; U-Pb, apatite) and cooled rapidly to the ambient country rock temperature. The subsequent rate of cooling of the country rock (determined by 40Ar/39Ar and U-Pb analyses on mica and columbite respectively) was probably dictated by slow cooling of an underlying pluton. Cretaceous intrusions proximal to the LNPG (Gordey and Anderson 1993) are part of the Tungsten and Tay River-Tungsten suites (100-95 Ma based on U-Pb zircon and monazite dating; Rasmussen et al. 2007). These typically small (< 15 km2 exposure) S-type intrusions are associated with aplitic and pegmatitic dikes and W (Cu-Zn-Mo) and skarn intrusion-related mineralization. The Cretaceous O’Grady Batholith (270 km2) ~95 km further north of the LNPG is a hornblende bearing, alkali-feldspar-rich composite intrusion that includes an area of elbaite subtype of granitic rare element pegmatite (Ercit et al. 2003). The O’Grady Batholith and the LNPG dikes are the only examples of Li mineralization in the region.  3.2.2 Mineralogy and textures The primary mineralogy of the Little Nahanni Pegmatite Group includes: quartz, K-feldspar, plagioclase, spodumene, micas (lepidolite and muscovite), columbite-group (Nb and Ta –rich) minerals, cassiterite, tourmaline, beryl, lithiophilite, garnet (Mauthner 1996; Rollo 1999, Pemberton 2002, Groat et al. 2003) and berlinite. Alteration is minor and secondary minerals such as cookeite and Ca-rich zeolites (Mauthner 1996) are rare. A rare example of system-wide mineralogical or geochemical zonations include a weakly defined zonation in spodumene- (Li-rich clinopyroxene) rich pegmatites occurs at slighter greater depths than spodumene-poor pegmatites (Wengzynowski 2002) and tend to be more prevalent in the centre of the study area. This is in contrast to the more spodumene- poor dikes, which are more common in the north and south of the area, that show some enrichment in Ta and Sn (Wengzynowski 2002). On the outcrop scale, one well-exposed dike towards the north of the study area displayed an increased abundance of quartz in the direction of dike propagation. Overall crystal dimensions rarely exceed ~10 cm length, although single spodumene or feldspar crystals are locally observed to span the width of a dike (Fig. 3.2C 82 and E: ~\<1 m). Several distinctive primary textures are observed in hand sample and at the outcrop scale, including: the crystallization of elongate minerals (feldspar and spodumene) sub-perpendicular to the pegmatite/wallrock contact (Fig. 3.2C and E); mineral banding parallel to the pegmatite/wallrock contact boundary defined by repeated bands of cm-size spodumene, feldspar or quartz crystals (Fig. 3.2D and F); and textural banding in the spodumene-poor pegmatites parallel to the wallrock contact, with bands of equant mm-sized aplite grains (feldspar and quartz) cross-cutting more pegmatitic (coarse-grained) layers (Mauthner 1996). Unbanded, more homogeneous rocks are common (P389859, Fig. 3.3A and P389764, Fig. 3.3B) composed of interlocking grains of the rock-forming minerals. Mica displays diversity in colour, shape and crystal size. Fig. 3.3C and D shows two examples with cleavage normal to the c-axis; Fig. 3.3C shows large books of greenish muscovite intergrown with quartz (EB0509), Fig. 3.3D shows lithian muscovite with a ball-peen habit (EB1304). Rare miarolitic cavities occur, indicating some exsolution of a magmatic volatile phase (Candela 1997). The cavities are planar (planes of euhedral crystal terminations (commonly quartz) across an area of ~15 cm2 between mineral bands) or rounded, <5 cm diameter (Mauthner 1996). 3.2.2.1 Petrography Petrographic observations concentrate on common textures and grain contacts rather than modal compositions due to the coarse grain size of the LNPG samples and they are summarised in Table 3.1. Of especial note are crystallisation textures indicating highly non-equilibrium conditions (Figs. 3.4-3.8). These are typically produced by interplay between liquidus undercooling and rapid crystallisation increasing growth rates and decreasing nucleation densities (Swanson 1977, London 2005a). Examples include: • intergrown plumose mica and feldspar (probably albite) Fig. 3.4E-F, and plumose mica alone Figs. 3.6E-G and 3.7B, displaying the growth of mica along the a- axis in contrast to growth along the c-axis (Figs. 3.6A-D and 3.7A). The plumose habit of mica is indicative of rapid crystallisation (London 2008). D. London (pers. comm, 2010.) cites the Law of Bravais (Friedel 1907) in proposing that these unusual growth directions 83 P389859 A EB0509 C EB1304 D P389764 B Figure 3.3 Examples of hand samples from the LNPG pegmatites. A. and B. Samples (P389859 and P389764) cut across dike widths and cut into sequences of samples for thin section preparation, showing megacrystic, leucocratic minerals. C. and D. LNPG samples EB0509 and EB1304, used to obtain mineral separates. 84 Table 3.1. General petrographic observations from the LNPG samples. Magmatic minerals Quartz - shows dihedral or sutured contacts with surrounding quartz grains, especially in aplitic zones. Single or undulatory extinction. Crystallised throughout the paragenetic sequence. Fig. 3.4A. Potassium feldspar - typically large grains orthoclase (cm across), shows some cloudy alteration and symplectic textures with quartz and plagioclase. Rare microcline. Locally includes adundance of mica chadocrysts. Fig. 3.4. Plagioclase - majority is An20-30 (oligoclase), although minor An10-20. Bimodal grain size, large composite grains (cm across) to small grains (mm across) in aplitic zones. Cross-cutting evidence suggests Na-rich plagioclase crystallised later than Ca-rich. Myrmekitic textures (with K-feldspar) suggests plagioclase crystallised later. Fig. 4B-F. Plumose intergrowths with mica Fig. 3.4Eand F. Spodumene - typically very large grains (commonly >10 cm in length). Commonly show complex symplectic textures with quartz. Fresh/unaltered contact with mica grains. Contains multi-component primary (solids and fluids) and secondary fluid inclusions. Fig. 3.5. Muscovite/lepidolite - several generations and multiple forms. Larger grains (cm across) tend to be earlier than very fine-grained mica. Early grains can also show greater degree of alteration (ragged grain boundaries). Cross-cutting relationships indicated by small (<0.5 cm) mica-filled fractures. Crystal forms include: broad (cm-across) tabular books of muscovite grains stacked in the c-axis direction, plumose muscovite sprays displaying growth along the a-axis (cm in length), ball-peen lithian muscovite (compound mica grains with overall rounded upper surface), skeletal and graphic mica. Figs. 3.4E and F, 3.6, 3.7 and 3.8. Tourmaline - two varieties determined in sample P389765-20 1. In wall zone of dike (normal to contact with host rock); blue (schorl), pleochroic, no cleavage, high relief, uniaxial negative, low birfringence, unzoned to zoned. Contains small fluid inclusions.2. Greenish-brown (uvarite?), stubby, occurring in the metasedimentary host rock, aligned with fabric of the rock. Fluorapatite - blue-green, rare, can contain small primary fluid inclusions Berlinite Al(PO4)- rare, found in sample P389859. Large (several mm across) tabular grain, very similar in appearance to K-feldspar and quartz but cloudy and altered, and lacks twinning; composition confirmed by SEM analysis. Secondary minerals Biotite - associated with small fractures showing Fe-oxidation Hematite - <5%,after pyrite (occasional relict cores) associated with calcite-filled fractures 85 Figure 3.4 Petrographic illustrations of quartz, feldspar and mica. A. Mostly recrystallised quartz in an aplitic zone, showing dihedral, embayed and straight grain boundaries. Minor spodumene middle left. Cross polarized light (XPL). B. Mainly plagioclase, with central grain showing myrmekitic texture. Large K-feldspar grains in lower half of photo. (XPL). C. Twinned spodumene (high birefringence) and K-feldspar (top of image) in matrix of coarsely spherulitic plagioclase grains. (XPL). D. Magnification of area proximal to C. Note fanning of the twinning zones. (XPL). E. Plumose plagioclase (low birefringence) and mica (high birefringence) of aplitic area. Magnified in F. (XPL). F. Plumose plagioclase and mica. (XPL). 86 A 2 mm P389764 B 2 mm P389859 P389799 2 mm C D 0.5 mm P389764 F 0.5 mm P389764 E 1 mm plumose mica growth along a-axis plumose feldspar (albite?) spodumene 87 Figure 3.5 Petrographic illustrations of spodumene. A. Large spodumene grain showing symplectic texture with quartz in recrystallised quartz matrix. (XPL). B. Broken spodumene grain in matrix of quartz grains showing sutured contacts. Plumose mica on right of image. (XPL). C. Spodumene showing symplectic texture with quartz? and vermicular texture between spodumene grains. Quartz grains in matrix show straight or embayed boundaries. (XPL). D. Broken spodumene grains in aplite. (XPL). E. High relief spodumene grain in centre of image in matrix of quartz, K-feldspar and mica. Spodumene has pervasive symplectic texture with quartz? and vermicular texture at the grain boundaries. (XPL). F. Same as E. but in PPL. Showing area of magnified image G. Enlargement of area in image F. Spodumene showing symplectic and vermicular textures in matrix of quartz and plagioclase. (XPL). H. Vermicular texture fringe on spodumene grain boundary (XPL). 88 A 1 mm P389764 C 1 mm P389764 D 2 mm P389764 B 2 mm P389859 E 2 mm P389764 F 2 mm P389764 0.25 mm P389859 H G 1 mm P389764 squi cleavelandite? vermicular spodumene growth 89 Figure 3.6 Petrographic illustrations of mica (I). A. High birefringence of books of mica cut parallel to the c-axis of the grains. Large grains of quartz and K-feldspar also in view, some showing undulose extinction. (XPL). B. Muscovite, parallel to the c-axis with high birefringence, quartz, plagioclase and K-feldspar also in view. K- feldspar shows symplectic, sieve-like texture with quartz? in lower central area of the image. (XPL). C. Radiating books of muscovite cut parallel to the c-axis of the grains plagioclase and K- feldspar also in view. (XPL). D. Muscovite cut parallel to the c-axis, K-feldspar, plagioclase and quartz grains with sutured contacts also in view. (XPL). E. High birefringence of plumose mica on left of image. K-feldspar and plagioclase grain on right of the image showing some sieve-like symplectic texture and additional plumose mica. K-feldspar grain shows slight dusty alteration. Note muscovite cleavage is parallel to c-axis, i.e. along length of grain. (XPL).  F. Same image as F but in PPL. Note muscovite cleavage is parallel to c-axis, i.e. along length of grain, and vermicular texture spodumene (high relief) rims sections of the mica plumes. G. Higher magnification of plumose mica, in quartz matrix. Note muscovite cleavage is parallel to c-axis, i.e. along length of grain. (XPL). H. Commonly broken aligned spodumene grains in aplitic matrix with several plumes of highly birefringent muscovite. Large K-feldspar grain to left of view. (XPL). 90 A 2 mm P389859 B 2 mm P389859 C 2 mm P389859 D 2 mm P389859 E 2 mm P389859 F 2 mm P389859 G 1 mm P389859 H 2 mm P389859 mica growth along a-axis ax a- is mica growth along a-axis a-ax si 91 Figure 3.7 Petrographic illustrations of mica (II). A. Highly birefringent plumose mica associated with a large quartz grain.  Note muscovite cleavage is perpendicular to c-axis, i.e. across width of grain. (XPL). B. Small plumes of muscovite within aplite and larger K-feldspar grain showing minor dusty alteration and symplectic texture. Note muscovite cleavage is parallel to c-axis, i.e. along length of grain. (XPL). C. Skeletal muscovite. Note muscovite cleavage is perpendicular to c-axis, i.e. across width of grain. (XPL). D. Skeletal muscovite. Note muscovite cleavage is perpendicular to c-axis, i.e. across width of grain. (XPL). E. Skeletal muscovite. Grain boundaries show vermicular texture. (XPL). 92 2 mm B P389859 0.5 mm D P389859 P389859 I mm A C 2 mm C P389859 E P389859 0.5 mm m c ce u i a o s  rtin spa g m ica  g ro wt h o g al n c- ax is mica growth along a-axis - x s a a i vermicular mica growth 93 P389859 0.5 mm graphic intergrowth of mica and quartz spherulitic texture Figure 3.8 Petrographic illustration of disequilibrium crystallisation textures. Zone of graphic intergrowth between mica and feldspar surrounding central area containing micaeous and feldspar spherules indicative of rapid, disequilibrium crystallisation. 94 become prominent when conditions requiring rapid crystallisation occur, because less material is required to advance the faces • spherulitic albite (Figs. 3.4C and D and Fig. 3.8; central zone), known to occur in non-equilibrium, typically undercooled, environments (Gránásy et al. 2004, Lofgren 1971) • skeletal mica (Fig. 3.7 C-E), graphic mica and feldspar (Fig. 3.8) (London 2005a) • vermicular or rod-like crystal growth of spodumene (Fig. 3.5H) and mica (Fig. 3.7E) A paragenetic sequence of crystallisation in the major rock-forming minerals is observed in the rock textures and cross-cutting relationships (Fig. 3. 9) that indicate; quartz precipitation occurred throughout the sequence (field observations suggest an increase in quartz abundance); spodumene and K-feldspar precipitation was most abundant in the earlier stages of crystallisation, with K-feldspar less common, and spodumene is typically absent in the later stages; the greatest abundance of plagioclase (albite) and mica crystallisation occurred during the later stages. A groundmass of albite and mica (muscovite and lepidolite) is common as the last stage of crystallisation. Fine- grained lepidolite occurs with fine-grained quartz and albite, and coarse lepidolite occurs more commonly with spodumene; muscovite likewise occurs as fine- and coarse- grained commonly in the wall zone of the dikes (Mauthner 1996). The oxide minerals (tantalite and cassiterite) tend to occur more abundantly in late crystallised quartz and mica rich dikes (Wengzynowski 2002). Textures occurring in spodumene are enigmatic (Fig. 3.5A- C and E-G) but may be due to spodumene breakdown with the formation of squi (spodumene-quartz intergrowth; London 1984) preceding cleavelandite? (Fig. 3.5E-G).  3.3 SAMPLING AND ANALYTICAL METHODS 3.3.1 Sampling Prior to the start of this study, samples from the LNPG were collected as part of an exploration sampling program (Wenzynowski 2002; Appendix 4) and were grouped by the proportions of the dominant rock-forming minerals, quartz, spodumene, feldspar, albite and mica as observed in the field as follows: 95 Quartz Spodumene K-feldspar Plagioclase Mica Figure 3.9 Cartoon interpretation of the crystallisation sequence of the main rock-forming minerals of the LNPG pegmatites from petrographic observations. 96 • SQF (spodumene-quartz-feldspar) • SQFL (spodumene-quartz-feldspar-lepidolite) • QA (quartz-albite) • QFL (quartz-feldspar-lepidolite) • QM (quartz and silver mica)   All sample material was powdered and processed by ALS Chemex, Vancouver, without duplicate analyses. For trace element concentrations samples were prepared using metaborate fusion and analysed by ICP-MS (Appendix 4). The major element geochemistry of a subset of 23 samples were analysed by X-ray fluorescence (XRF; included in Table 3.2). The large sample sizes (1.9-22 kg) are expected to have significantly reduced the effect of mineralogical zonation and grain size heterogeneity on the concentration data. The starting materials for this study were the whole rock powders and the geochemical analyses of the samples. However, it was apparent that the abundance of many of the trace elements was below, or at the limit of, the original analytical techniques (metaborate fusion). The trace elements of the subset of 23 samples were therefore re-analysed at the Pacific Centre of Isotopic and Geochemical Research (PCIGR) using high resolution ICP-MS (HR-ICP-MS; Table 3.3). To facilitate correlation and interpretation of the data by petrographic and further geochemical analysis, five additional powdered rock samples from the original set (P389764, P389768, P389774, P389797, P389870) were and analysed for major (XRF: ALS Chemex) and trace element geochemistry (Quad ICP-MS; Université Joseph Fourier, Grenoble, France; plus sample P389859 (Fig. 3.3A). Hand samples of those five additional samples were collected from the field (e.g., Fig. 3.3B, C and D). Trace element analyses of P389711, P389789 and P389875 (UBC) and P389764 and P389774 (UJF ; included in Table 3.3) were duplicated (re-run on the same aliquot). Mineral grains of spodumene (hand samples P389859 (Fig. 3.3A) and AL050704) and quartz, mica and feldspar (hand samples EB0509 and EB1304; Fig. 3.3D-F) were handpicked and the trace elements analysed by ICP-MS at the Université Joseph Fourier, Grenoble, France. Trace element analyses of spodumene samples (P389859 and AL050704) were duplicated (re-run on the same aliquot). 97 Table 3.2. Major element concentrations of the whole rock LNPG samples. Location LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG UTM E 506643 506770 505975 508291 507970 508558 506824 506824 507179 UTM N 6893518 6893396 6894346 6891110 6892483 6889924 6893279 6893279 6892808 Elevation (m) 1935 1926 1781 1954 1670 2010 1801 1801 1895 Sample  P389704  P389711  P389719  P389726  P389734  P389739  P389751  P389753  P389760 Weight (kg) 21.9 22.0 13.8 1.9 2.4 16.7 1.6 11.8 5.8 Field indentifiers SQF SQF QA SQF QM SQF SQF SQF SQF REE pattern type discontinuous discontinuous straight indeterminate indeterminate straight listric discontinuous listric SiO2 (wt%) 74.32 74.12 74.98 70.67 72.31 71.77 68.89 74.49 72.85 TiO2 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.01 Al2O3 16.71 16.84 15.27 17.24 16.73 17.36 17.33 16.11 16.31 Fe2O3 0.15 0.29 0.33 0.27 0.27 0.28 0.26 0.33 0.34 FeO 0.13 0.26 0.26 0.26 0.26 0.26 0.26 0.32 0.32 MnO 0.14 0.25 0.11 0.21 0.23 0.16 0.17 0.22 0.12 MgO 0.01 <0.01 <0.01 0.04 <0.01 <0.01 0.03 0.03 0.01 CaO 0.42 0.28 0.21 0.63 0.08 0.65 0.37 0.27 0.40 Na2O 3.66 4.38 4.86 4.66 3.24 4.17 5.38 3.80 4.21 K2O 2.32 2.06 3.18 3.22 3.92 2.26 4.43 2.78 3.01 P2O5 0.26 0.23 0.22 0.91 0.45 0.89 0.69 0.25 0.34 LOI 1.07 1.29 0.76 1.73 1.82 1.69 1.23 1.41 1.03 Total 99.2 100.0 100.2 99.8 99.3 99.5 99.0 100.0 99.0 H2O+ 0.71 0.69 0.51 0.91 1.03 0.92 0.50 0.55 0.54 F (ppm) 850 5750 1020 8260 4780 4550 5450 5190 1950 Normative minerals Quartz 44.5 40.7 34.2 31.3 38.9 39.3 20.6 41.7 37.1 Plagioclase 31.9 37.6 41.4 40.2 28.1 36.1 46.5 32.6 36.4 Orthoclase 14.0 12.4 18.9 19.4 23.8 13.7 26.8 16.7 18.1 Corundum 8.2 7.5 3.9 6.2 7.4 8.2 3.8 7.0 6.3 Hypersthene 0.7 1.3 1.1 1.3 1.2 1.1 1.2 1.5 1.2 Magnetite 0.0 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Apatite 0.6 0.5 0.5 2.2 1.1 2.1 1.7 0.6 0.8 Total 100.0 100.0 100.1 100.6 100.5 100.5 100.6 100.0 100.0 98 Table 3.2. Major element concentrations of the whole rock LNPG samples contd. Location LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG UTM E 507267 507267 6892921 6894020 506162 508066 505545 505545 505505 UTM N 6892868 6892868 507384 506162 6894020 6892654 6897443 6897443 6897547 Elevation (m) 1872 1872 1841 1730 1730 1582 1777 1777 1842 Sample P389764 P389768 P389774  P389789  P389791 P389797  P389808  P389809  P389816 Weight (kg) 2.2 3.8 3.0 11.0 12.2 2.0 10.0 5.8 3.7 Field indentifiers SQF SQF SQF SQF SQF SQF QFL QFL QFL REE pattern type discontinuous listric discontinuous discontinuous discontinuous listric straight straight listric SiO2 (wt%) 71.65 70.27 70.68 73.17 72.66 77.65 72.28 71.4 71.56 TiO2 <0.01 0.02 <0.01 0.01 <0.01 0.02 <0.01 <0.01 <0.01 Al2O3 17.74 18.09 17.40 16.11 15.73 13.62 17.26 17.8 16.67 Fe2O3 0.22 0.23 0.22 0.35 0.34 0.32 0.28 0.27 0.27 FeO - - - 0.32 0.32 - 0.26 0.26 0.26 MnO 0.14 0.11 0.2 0.16 0.14 0.08 0.20 0.19 0.16 MgO 0.06 0.09 0.07 0.01 <0.01 0.05 <0.01 0.01 0.03 CaO 0.36 0.22 0.20 0.54 0.22 0.19 0.42 0.57 0.44 Na2O 3.80 3.98 4.57 4.36 4.31 2.68 3.95 4.77 5.32 K2O 2.45 2.60 3.30 2.76 3.43 3.27 2.69 2.38 2.48 P2O5 0.81 0.93 0.39 0.27 0.24 0.20 0.48 0.72 1.10 LOI 1.31 1.61 1.38 1.70 1.21 0.29 1.57 1.42 1.71 Total 98.5 98.2 98.4 99.8 98.6 98.4 99.4 99.8 100.0 H2O+ - - - 1.04 0.84 - 0.80 0.77 0.98 F (ppm) 7260 8170 8310 1970 2190 160 7960 7380 5190 Normative minerals Quartz 41.0 38.2 32.0 37.0 34.8 50.2 39.3 34.6 31.2 Plagioclase 33.1 34.9 39.9 38.6 37.5 23.1 34.2 41.0 45.8 Orthoclase 14.9 15.9 20.1 16.7 20.8 19.7 16.3 14.3 14.9 Corundum 9.1 9.1 6.5 5.7 5.1 5.8 8.0 7.5 5.3 Hypersthene 0.7 0.7 0.9 1.3 1.3 0.7 1.2 1.2 1.2 Magnetite 0.0 0.0 0.0 0.1 0.1 0.0 0.1 0.1 0.1 Apatite 1.9 2.3 0.9 0.7 0.6 0.5 1.1 1.7 2.6 Total 100.7 101.0 100.3 100.0 100.1 100.1 100.2 100.4 101.0 99 Table 3.2. Major element concentrations of the whole rock LNPG samples contd. Location LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG UTM E 505073 505328 507938 508158 508335 508570 508494 508703 509173 509105 UTM N 6897533 6896825 6891309 6891068 6890663 6891150 6891941 6890212 6889545 6889516 Elevation (m) 1927 1550 1780 1938 2012 1781 1595 1755 1657 1673 Sample  P389817  P389844  P389849  P389851  P389854  P389861 P389870  P389875  P389910  P389913 Weight (kg) 6.0 2.6 2.7 8.3 12.0 4.6 7.8 12.8 9.2 4.6 Field indentifiers QM QM QM SQFL SQF SQF SQF SQF QA QM REE pattern type indeterminate listric straight listric straight indeterminate discontinuous discontinuous straight discontinuous SiO2 (wt%) 87.45 78.55 69.6 69.79 77.34 69.71 73.31 74.17 70.76 67.16 TiO2 0.02 <0.01 <0.01 <0.01 0.01 <0.01 0.01 0.02 <0.01 0.01 Al2O3 6.31 12.65 17.31 17.52 13.95 17.15 17.02 16.38 17.56 19.42 Fe2O3 0.33 0.33 0.27 0.27 0.41 0.26 0.32 0.27 0.28 0.46 FeO 0.32 0.32 0.26 0.26 0.32 0.26 - 0.26 0.26 0.32 MnO 0.04 0.09 0.16 0.17 0.12 0.26 0.16 0.10 0.15 0.10 MgO <0.01 <0.01 0.02 <0.01 0.06 <0.01 0.06 <0.01 <0.01 0.10 CaO 0.15 0.02 0.72 0.26 0.60 0.40 0.20 0.20 0.61 1.07 Na2O 1.89 2.46 5.07 5.23 3.43 4.37 3.17 4.07 4.77 5.41 K2O 1.58 2.19 2.81 3.05 2.21 3.88 2.88 3.12 3.58 3.2 P2O5 0.11 0.71 0.95 0.89 0.69 0.70 0.42 0.25 0.64 0.74 LOI 0.47 1.33 1.71 1.56 0.83 1.60 0.99 0.67 1.13 1.59 Total 98.7 98.7 98.9 99.0 100.0 98.6 98.5 99.5 99.7 99.6 H2O+ 0.29 1.07 1.11 0.81 0.52 0.86 - 0.35 0.78 1.30 F (ppm) 180 1420 5550 8620 2620 9940 4110 570 2020 1530 Normative minerals Quartz 71.2 56.9 29.7 27.9 48.7 29.8 44.5 38.6 29.2 23.0 Plagioclase 16.4 21.4 44.2 45.4 29.3 38.2 27.5 34.9 41.0 47.2 Orthoclase 9.5 13.3 17.1 18.5 13.2 23.6 17.4 18.7 21.5 19.3 Corundum 1.5 6.4 6.1 5.8 6.0 5.9 8.9 6.4 5.9 7.1 Hypersthene 1.0 1.2 1.2 1.1 1.5 1.3 0.9 1.0 1.1 1.6 Magnetite 0.1 0.1 0.1 0.1 0.1 0.1 0.0 0.1 0.1 0.1 Apatite 0.3 1.7 2.3 2.1 1.6 1.7 1.0 0.6 1.5 1.8 Total 100.0 100.9 100.6 100.9 100.3 100.6 100.3 100.1 100.3 100.0 100 Table 3.3. Trace element concentrations of the whole rock LNPG samples, the Macusani glass samples and the LNPG mineral separates. Location LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG Sample  P389704  P389711  P389711 duplicate  P389711 average P389711 stdev P389711 %RSD  P389719  P389726  P389734  P389739  P389751  P389753 REE pattern type discontinuous discontinuous discontinuous flat indeterminate indeterminate flat listric discontinuous Li (ppm) 8756 7301 7590 7445 204.4 2.7 4289 4813 11437 14161 4073 7644 Sc <lod 0.019 0.020 0.0196 0.0 4.1 0.138 <lod 0.000 0.044 0.006 <lod Ti na na na na na na na na na na na na V 0.69 0.64 0.68 0.66 0.02 3.8 2.05 0.89 0.438 1.67 0.86 0.65 Cr na na na na na na na na na na na na Co 0.4 0.5 0.4 0.46 0.02 4.8 2.5 0.6 <lod 3.6 0.6 0.5 Ni 2.1 2.6 2.9 2.7 0.2 8.5 2.1 5.6 8.4 8.1 4.3 2.8 Cu 14.0 4.7 6.7 6 1 25.1 4.6 1.8 1.9 6.3 1.8 1.0 Zn 65 115 121 118 4 3.6 29 81 327 85 111 129 As na na na na na na na na na na na na Ga 42 51 49 50 2 3.6 37 52 106 77 30 42 Rb 1229 1558 1587 1572 21 1.3 2587 2074 7441 2155 3566 1715 Sr 156 56 88 72 22 31.0 38 77 35 35 38 17 Y 0.34 0.53 0.52 0.52 0.01 1.1 0.31 0.33 0.51 0.21 0.08 0.29 Zr 21.7 36.9 31.8 34 4 10.4 9.7 19.9 111.8 30.8 33.3 16.7 Nb 71 64 60 62 3 4.4 103 45 461 120 90 75 Mo 0.31 2.86 0.37 2 2 109.2 0.91 0.34 12351 0.94 2.49 0.41 Cd 0.16 1.27 0.11 0.7 0.8 118.8 0.11 0.07 <lod 0.12 0.08 0.11 Sn 79 87 75 81 9 10.9 93 48 11.22 192 227 98 Sb 0.049 0.624 0.043 0.3 0.4 123.2 0.085 0.028 <lod 0.037 0.151 0.049 Cs 98 79 75 77 3 3.5 171 208 497 324 288 98 Ba 8.4 7.17 3.34 5 3 51.5 30.7 7.8 19.0 6.2 8.3 3.1 La 0.250 0.351 0.365 0.36 0.01 2.8 0.580 0.800 0.840 0.420 0.170 0.250 Ce 0.460 0.685 0.621 0.65 0.05 6.9 1.120 0.520 1.120 0.700 0.190 0.470 Pr 0.042 0.061 0.064 0.06 0.00 2.8 0.129 0.102 0.100 0.084 0.016 0.044 Nd 0.130 0.172 0.185 0.18 0.01 5.0 0.450 0.340 0.000 0.290 0.050 0.120 Sm 0.044 0.071 0.074 0.07 0.00 3.8 0.081 0.048 0.041 0.065 0.009 0.048 Eu 0.002 0.003 0.003 0.00 0.00 0.7 0.018 0.010 0.010 0.008 0.002 0.002 Gd 0.041 0.075 0.077 0.08 0.00 1.4 0.064 0.049 0.044 0.051 0.010 0.040 Tb 0.009 0.019 0.019 0.02 0.00 0.7 0.011 0.007 0.000 0.009 0.002 0.011 Dy 0.055 0.093 0.090 0.09 0.00 2.4 0.056 0.041 0.060 0.041 0.010 0.049 Ho 0.005 0.010 0.010 0.01 0.00 0.7 0.011 0.006 0.012 0.007 0.002 0.006 Er 0.016 0.020 0.020 0.02 0.00 0.2 0.029 0.017 0.034 0.017 0.005 0.014 Tm <lod 0.004 0.004 0.00 0.00 0.7 0.004 <lod 0.007 <lod <lod <lod Yb 0.021 0.026 0.027 0.03 0.00 0.7 0.024 0.023 0.061 0.016 0.008 0.062 Lu 0.002 0.003 0.003 0.00 0.00 0.7 0.004 0.003 0.008 0.002 0.002 0.007 Hf 1.89 3.22 2.80 3.0 0.3 9.9 1.50 1.36 8.32 3.55 3.26 1.41 Ta 51.1 23.8 17.6 21 4 21.4 123.5 37.2 469.0 5.3 108.4 47.0 Tl na na na na na na na na na na na na W 1.4 3.3 2.7 3.0 0.4 13.3 2.2 6.4 27.8 7.1 4.4 4.0 Pb 3.7 2.7 2.6 2.68 0.09 3.2 112.6 6.7 11.0 6.5 4.9 3.0 Bi 1.12 5.57 1.29 3 3 88.3 76.14 2.02 <lod 2.16 1.05 0.85 Th 1.26 2.56 2.01 2.3 0.4 16.9 3.27 1.48 6.60 2.86 2.28 1.44 U 2.20 7.32 7.15 7.2 0.1 1.7 3.37 1.23 14.41 7.10 2.26 1.91 ΣREE (ppm) 1.1 1.6 1.6 2.6 2.0 2.4 1.7 0.5 1.1 Eu/Eu* 0.14 0.12 0.12 0.78 0.62 0.70 0.42 0.58 0.14 *# measured at Université Joseph Fourier, Grenoble, France 101 Table 3.3. Trace element concentrations of the whole rock LNPG samples, the Macusani glass samples and the LNPG mineral separates contd. Location LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG Sample  P389760 *P389764 *P389764 duplicate *P389764 average *P389764 stdev *P389764 %RSD *P389768 *P389768 duplicate *P389774 *P389774 duplicate *P389774 average *P389774 stdev *P389774 %RSD REE pattern type listric discontinuous discontinuous listric listric discontinuous discontinuous Li (ppm) 6603 5271 5369 5314 3019 Sc <lod 0.0914 0.0803 0.09 0.01 9.1 0.042 0.0524 0.0717 0.0786 0.0752 0.0049 6.49 Ti na 7.8 7.3 7.6 0.4 4.7 5.6 5.5 6.5 6.2 6.4 0.2 3.3 V 2 0.216 0.212 0.21 0.00 1.3 0.104 0.113 0.182 0.175 0.179 0.005 2.8 Cr na 3.25 3.27 3.26 0.01 0.4 3.37 3.03 2.73 2.63 2.68 0.07 2.6 Co 0.4 0.135 0.134 0.13 0.00 0.5 0.329 0.332 0.404 0.412 0.408 0.006 1.4 Ni 3.2 0.4 0.4 0.40 0.00 0.0 0.5 0.5 0.4 0.4 0.4 0.0 0.0 Cu 1.4 0.504 0.501 0.50 0.00 0.4 0.76 0.736 0.535 0.53 0.53 0.00 0.7 Zn 69 21 20.9 21.0 0.07 0.3 30.5 30.1 51.9 51.7 51.8 0.1 0.3 As na 0.155 0.148 0.15 0.00 3.3 0.142 0.149 0.179 0.177 0.178 0.001 0.8 Ga 38 na na na na na na Rb 1308 2383 na 2467 2451 2826 Sr 46 39.7 39.9 39.8 0.1 0.4 18.9 19 17.7 17.9 17.8 0.1 0.8 Y 0.26 0.377 0.38 0.38 0.00 0.6 0.0554 0.0572 0.691 0.686 0.689 0.004 0.5 Zr 15.3 32.2 32.2 32.2 0.0 0.0 18.8 20.8 29.7 29.7 29.7 0.0 0.0 Nb 59 83.5 77.5 80.5 4.2 5.3 91.4 88.8 66.9 65.7 66.3 0.8 1.3 Mo 0.84 na na na na na na Cd 0.06 0.014 0.013 0.01 0.00 6.3 0.014 0.012 0.016 0.015 0.016 0.001 7.3 Sn 65 na na na na na na Sb <lod na na na na na na Cs 53 257 na 290 290 269 na Ba 8.0 5.1 5.1 5.1 0.0 0.6 2.7 2.9 4.4 4.5 4.4 0.1 1.7 La 0.386 0.292 0.298 0.295 0.004 1.4 0.082 0.097 0.347 0.349 0.348 0.002 0.4 Ce 0.375 0.469 0.454 0.462 0.011 2.3 0.132 0.143 0.426 0.424 0.425 0.002 0.4 Pr 0.047 0.054 0.056 0.055 0.001 1.9 0.010 0.011 0.051 0.049 0.050 0.002 3.6 Nd 0.150 0.178 0.166 0.172 0.008 4.9 0.031 0.034 0.151 0.137 0.144 0.010 7.0 Sm 0.027 0.077 0.066 0.071 0.008 10.9 0.006 0.006 0.047 0.049 0.048 0.001 2.5 Eu 0.007 0.006 0.004 0.005 0.001 18.3 0.001 0.001 0.002 0.003 0.002 0.000 17.9 Gd 0.027 0.072 0.064 0.068 0.006 8.3 0.005 0.005 0.052 0.054 0.053 0.001 2.8 Tb 0.005 0.012 0.014 0.013 0.001 6.0 0.001 0.001 0.014 0.017 0.015 0.002 11.5 Dy 0.032 0.054 0.056 0.055 0.001 2.4 0.006 0.006 0.082 0.086 0.084 0.003 3.1 Ho 0.006 0.007 0.006 0.006 0.001 9.2 0.001 0.001 0.011 0.011 0.011 0.000 0.0 Er 0.018 0.013 0.012 0.013 0.001 5.0 0.003 0.003 0.027 0.028 0.027 0.000 1.2 Tm <lod na na na na na na Yb 0.035 0.015 0.016 0.015 0.001 6.5 0.007 0.006 0.043 0.045 0.044 0.001 2.3 Lu 0.005 0.002 0.002 0.002 0.000 16.8 0.001 0.001 0.005 0.006 0.005 0.000 8.6 Hf 1.20 3.84 3.77 3.81 0.05 1.3 2.03 2.37 2.56 2.53 2.55 0.02 0.8 Ta 23.7 na na 59.9 64.3 29.1 26.4 Tl na 16 16 16 0 0.0 19.3 19.2 22.6 22.5 22.6 0.07 0.3 W 3.4 na na na na na na Pb 5.0 2.63 2.61 2.62 0.01 0.5 4.39 4.44 3.25 3.26 3.26 0.01 0.2 Bi 1.24 na na na na na na Th 1.83 1.42 1.42 1.42 0.00 0.0 2.34 3.03 2.28 2.28 2.28 0.00 0.0 U 3.40 4.31 4.36 4.34 0.04 0.8 3.7 3.78 2.05 2.03 2.04 0.01 0.7 ΣREE (ppm) 1.1 1.2 1.2 0.3 0.3 1.3 1.3 Eu/Eu* 0.85 0.23 0.20 0.59 0.63 0.13 0.16 *# measured at Université Joseph Fourier, Grenoble, France 102 Table 3.3. Trace element concentrations of the whole rock LNPG samples, the Macusani glass samples and the LNPG mineral separates contd. Location LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG Sample  P389789  P389789 duplicate  P389789 average  P389789 stdev  P389789 %RSD  P389791 *P389797  P389808  P389809  P389816  P389817 REE pattern type discontinuous listric flat flat listric indeterminate Li (ppm) 3380 3745 3563 258 7.3 3989 6229 7920 9014 4080 77 Sc 0.10 0.09 0.10 0.01 6.6 0.024 0.0296 0.023 0.047 0.016 0.005 Ti na na na 6.6 na na na na V 1.26 1.16 1.21 0.07 5.9 0.66 0.142 1.82 0.88 0.6 1.21 Cr na na na 4.99 na na na na Co 4.7 4.5 4.6 0.1 2.1 3.9 2.09 4.5 7.4 10.3 16.0 Ni 7.1 7.2 7.2 0.1 1.4 6.0 0.7 3.0 3.9 2.2 5.1 Cu 4.2 4.2 4.2 0.0 0.7 2.9 4.03 7.3 14.4 20.7 27.2 Zn 104 109 106 3 3.3 99 7.01 77 62 39 11 As na na na 0.152 na na na na Ga 34 33 34 1 1.9 28 na 44 33 32 13 Rb 1191 1284 1238 65 5.3 2085 1234 2440 2661 2509 542 Sr 51 54 52 2 3.6 41 32.4 16 46 52 46 Y 0.5 0.5 0.5 0.1 11.0 0.17 0.0439 0.23 0.28 0.14 0.24 Zr 21.0 15.7 18.3 3.7 20.3 8.4 8.71 28.1 31.0 15.7 15.4 Nb 61 63 62 2 2.6 68 45.0 68 77 125 52 Mo 0.27 0.31 0.29 0.03 10.1 0.33 na 1.32 0.4 1.53 0.8 Cd 0.07 0.11 0.09 0.03 31.8 0.13 0.006 0.06 0.14 0.11 0.06 Sn 64 70 67 4 5.8 65 na 107 130 115 19 Sb 0.031 0.019 0.025 0.008 33.3 0.018 na 0.019 0.025 0.023 0.029 Cs 53 55 54 1 2.4 64 310 202 261 47 Ba 9.2 9.3 9 0.1 0.9 4.5 10.3 14.2 8.0 9.0 64.8 La 0.890 0.733 0.812 0.111 13.7 0.260 0.074 0.390 0.380 0.160 0.710 Ce 1.637 1.500 1.569 0.097 6.2 0.430 0.087 0.680 0.660 0.170 0.540 Pr 0.179 0.152 0.165 0.019 11.3 0.041 0.008 0.072 0.075 0.018 0.135 Nd 0.620 0.500 0.560 0.085 15.2 0.100 0.027 0.240 0.240 0.060 0.480 Sm 0.119 0.109 0.114 0.007 6.0 0.029 0.005 0.038 0.043 0.016 0.080 Eu 0.019 0.019 0.019 0.000 0.4 0.003 0.003 0.006 0.011 0.003 0.019 Gd 0.095 0.093 0.094 0.001 1.5 0.022 0.005 0.036 0.044 0.014 0.067 Tb 0.018 0.018 0.018 0.000 2.7 0.005 0.001 0.006 0.008 0.003 0.010 Dy 0.096 0.087 0.091 0.007 7.2 0.028 0.005 0.036 0.046 0.019 0.050 Ho 0.016 0.018 0.017 0.001 5.2 0.004 0.001 0.006 0.009 0.004 0.006 Er 0.041 0.040 0.041 0.001 1.8 0.009 0.003 0.022 0.024 0.010 0.021 Tm 0.007 0.007 0.007 0.001 7.5 <lod na <lod <lod <lod <lod Yb 0.042 0.041 0.042 0.000 0.5 0.015 0.006 0.018 0.020 0.014 0.017 Lu 0.006 0.006 0.006 0.000 1.0 0.002 0.001 0.003 0.003 0.002 <lod Hf 1.12 1.23 1.17 0.07 6.2 0.61 0.856 2.64 3.05 2.37 2.60 Ta 23.5 22.7 23.1 0.6 2.5 21.2 na 78.8 86.8 188.4 88.1 Tl na na na 8.99 na na na na W 6.4 6.1 6.3 0.2 3.3 6.5 na 5.3 4.5 5.3 0.5 Pb 3.3 3.3 3.3 0.0 0.6 3.5 13.5 6.0 5.3 6.7 6.1 Bi 1.01 0.99 1.00 0.02 1.8 1.05 na 1.66 1.10 1.45 0.11 Th 1.42 1.29 1.35 0.09 6.9 1.66 4.61 2.94 5.84 4.01 1.82 U 6.56 6.64 6.60 0.06 0.9 5.15 2.97 3.15 5.24 2.58 1.50 ΣREE (ppm) 3.6 3.3 1.0 0.2 1.6 1.6 0.5 2.1 Eu/Eu* 0.57 0.57 0.37 1.90 0.49 0.76 0.59 0.81 *# measured at Université Joseph Fourier, Grenoble, France 103 Table 3.3. Trace element concentrations of the whole rock LNPG samples, the Macusani glass samples and the LNPG mineral separates contd. Location LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG Sample  P389844  P389849  P389851  P389854  P389861 *P389870  P389875  P389875 duplicate  P389875 duplicate average  P389875 duplicate1* stdev  P389875 duplicate %RSD  P389910 REE pattern type listric flat listric flat indeterminate discontinuous discontinuous flat Li (ppm) 1375 2626 7606 6617 4520 6021 8847 8947 8897 70 0.8 4860 Sc 0.039 0.114 0.013 0.091 0.027 0.0512 0.026 0.029 0.028 0.002 6.6 0.151 Ti na na na na na 11.3 na na na V 0.31 1.81 0.55 2.36 0.72 0.185 0.74 0.76 0.75 0.02 2.1 2.25 Cr na na na na na 4.21 na na na na na na Co 7.0 1.8 12.0 6.5 0.6 0.658 11.7 11.6 11.7 0.1 0.9 3.4 Ni 5.4 3.7 2.4 4.7 3.1 0.8 1.838 1.844 1.841 0.005 0.2 2.2 Cu 13.0 3.6 23.3 11.0 2.0 1.24 21.14 21.03 21.08 0.08 0.4 7.1 Zn 152 39 112 61 102 55.4 51.1 50.0 50.5 0.8 1.5 31 As na na na na na 0.243 na na na na na na Ga 49 49 45 43 60 na 37.64 37.61 37.63 0.02 0.1 41 Rb 1529 1448 3090 2210 2863 1715 2289 2268 2278 15 0.6 2715 Sr 112 95 23 73 90 15.5 16.60 16.64 16.62 0.02 0.1 41 Y 0.20 0.35 0.13 0.64 0.34 0.344 0.25 0.28 0.27 0.02 5.8 0.34 Zr 45.2 18.0 30.7 32.1 36.0 17.2 13.6 14.2 13.9 0.4 3.2 12.4 Nb 368 84 95 76 80 77.6 47 84 65 26 39.6 115 Mo 3 0.46 0.42 0.66 0.4 na 0.40 0.45 0.43 0.04 8.7 1.01 Cd <lod 0.06 0.12 0.10 0.09 0.019 0.07 0.09 0.08 0.02 20.1 0.13 Sn 13882 76 109 87 84 na 73 105 89 23 25.9 117 Sb <lod 0.030 0.043 0.163 0.020 na 0.019 0.017 0.018 0.001 7.9 0.099 Cs 114 170 265 131 241 111 48.7 48.2 48.5 0.4 0.8 204 Ba 5.0 38.3 9.4 16.7 46.3 7.5 12.08 12.15 12.12 0.05 0.4 34.0 La 0.274 0.700 0.140 1.750 0.400 0.244 0.284 0.275 0.280 0.007 2.4 0.600 Ce 0.400 1.060 0.200 3.560 0.460 0.375 0.519 0.508 0.514 0.008 1.5 1.170 Pr 0.039 0.122 0.021 0.374 0.064 0.035 0.051 0.050 0.050 0.001 2.5 0.134 Nd 0.120 0.410 0.070 1.290 0.190 0.098 0.158 0.167 0.162 0.006 3.6 0.450 Sm 0.022 0.064 0.010 0.218 0.045 0.031 0.0558 0.0558 0.0558 0.0000 0.0 0.087 Eu 0.001 0.023 0.003 0.042 0.015 0.002 0.008 0.007 0.008 0.001 13.3 0.019 Gd 0.020 0.057 0.011 0.164 0.049 0.031 0.052 0.049 0.050 0.002 4.6 0.069 Tb 0.003 0.009 0.002 0.022 0.017 0.007 0.0117 0.0131 0.0124 0.0010 7.9 0.012 Dy 0.026 0.055 0.012 0.129 0.048 0.041 0.051 0.053 0.052 0.001 2.1 0.063 Ho 0.005 0.011 0.002 0.020 0.017 0.005 0.0062 0.0058 0.0060 0.0002 3.9 0.013 Er 0.015 0.029 0.006 0.058 0.026 0.014 0.0134 0.0128 0.0131 0.0004 3.1 0.029 Tm <lod 0.005 <lod 0.006 <lod na 0.0021 0.0020 0.0020 0.0001 4.5 0.004 Yb 0.030 0.032 0.012 0.053 0.028 0.022 0.0147 0.0149 0.0148 0.0001 1.0 0.028 Lu 0.004 0.004 0.002 0.007 <lod 0.003 0.00212 0.00200 0.00206 0.0001 3.9 0.004 Hf 4.60 2.52 4.29 3.24 3.67 1.5 1.30 1.34 1.32 0.02 1.8 1.87 Ta 348.0 66.6 87.3 52.0 84.2 na 10 43 27 23 86.9 150.4 Tl na na na na na 13.8 na na na W 4.9 5.8 5.4 2.6 6.0 na 1.0 1.4 1.2 0.3 23.8 2.8 Pb 10.0 9.8 14.6 5.2 14.8 5.2 6.0 6.2 6.1 0.1 1.9 128.3 Bi 1.57 1.90 2.31 1.26 5.23 na 0.92 0.89 0.90 0.02 2.6 82.92 Th 1.90 3.20 5.69 2.73 4.14 1.36 1.60 1.50 1.55 0.07 4.6 3.86 U 9.50 5.28 3.57 3.74 3.06 5.13 1.78 1.88 1.83 0.07 3.8 3.77 ΣREE (ppm) 1.0 2.6 0.5 7.7 1.4 0.9 1.20 1.21 2.7 Eu/Eu* 0.15 1.15 0.90 0.67 0.95 0.16 0.45 0.73 *# measured at Université Joseph Fourier, Grenoble, France 104 Table 3.3. Trace element concentrations of the whole rock LNPG samples, the Macusani glass samples and the LNPG mineral separates contd. Location LNPG Macusani Macusani Macusani Macusani Macusani Macusani Macusani LNPG LNPG LNPG LNPG LNPG Sample  P389913 MAC MAC duplicate MAC average MAC 1*stdev MAC %RSD *MAC *TomMac *AL050704 spodumene *AL050704 spodumene duplicate *AL050704 spodumene average *AL050704 spodumene stdev *AL050704 spodumene %RSD REE pattern type discontinuous Li (ppm) 12105 3984 3746 3865 169 4.4 3419 3808 38112 38387 38249 194 0.5 Sc 0.039 0.85 0.77 0.81 0.06 7.1 1.4 2.9 0.0048 0.0004 0.0026 0.0031 120.5 Ti na na na 364 287 22.7 21.8 22.3 0.6 2.9 V 1.05 1.47 1.38 1.43 0.06 4.5 1.05 0.84 0.10 0.09 0.10 0.00 0.5 Cr na na na na na na 0.72 1.02 5.99 6.01 6.00 0.01 0.2 Co 0.5 0.219 0.209 0.214 0.007 3.2 0.28 0.09 0.18 0.19 0.18 0.01 4.7 Ni 3.8 0.23 0.27 0.25 0.02 9.9 0.3 0.2 27 26.8 26.9 0.1 0.5 Cu 5.0 2.1 1.9 2.0 0.1 6.0 5.4 1.4 2.02 1.98 2.0 0.0 1.4 Zn 66 76 68 72 6 7.9 108 117 27.5 27.9 27.7 0.3 1.0 As na na na 226 393 0.129 0.162 0.146 0.023 16.0 Ga 36 27 25 26 2 6.7 na na na na Rb 1860 1406 1316 1361 64 4.7 1293 1554 na na Sr 38 29 26 27 2 7.3 43.1 2.44 4.8 4.8 4.8 0.0 0.6 Y 0.37 2.3 2.1 2.2 0.1 6.5 5.68 7.01 0.0057 0.0046 0.01 0.00 15.0 Zr 13.5 29 27 28 1.3 4.8 42.9 35.1 1.13 1.12 1.13 0.01 0.6 Nb 42 52 46 48.8 4.03 8.3 54.6 70.3 0.948 0.946 0.947 0.001 0.1 Mo 0.43 1.5 2.1 1.8 0.4 22.0 na na na na Cd 0.09 0.194 0.15 0.17 0.03 16.4 0.05 0.03 0.0069 0.0067 0.0068 0.0001 2.0 Sn 94 482 396 439 61 13.9 na na na na Sb 0.017 5.1 4.6 4.9 0.4 7.4 na na na na Cs 73 911 824 867 61 7.0 1151 761 6 7 6.50 0.71 10.9 Ba 8.6 76 68 72 5 7.2 105 3.82 0.211 0.183 0.197 0.020 10.1 La 0.410 2.71 2.84 2.78 0.09 3.3 6.060 2.750 0.00353 na Ce 0.740 5.7 5.9 5.8 0.1 2.1 12.600 6.460 0.00640 na Pr 0.076 0.67 0.68 0.67 0.01 1.6 1.470 0.822 0.00049 na Nd 0.230 2.27 2.30 2.28 0.03 1.1 5.180 3.070 0.00181 na Sm 0.069 0.537 0.529 0.533 0.005 1.0 1.250 1.090 0.00034 na Eu 0.005 0.083 0.076 0.080 0.005 5.7 0.183 0.025 0.00005 na Gd 0.061 0.44 0.42 0.43 0.02 4.3 1.030 1.110 0.00033 na Tb 0.014 0.075 0.071 0.073 0.003 3.9 0.170 0.206 0.00004 na Dy 0.061 0.43 0.40 0.42 0.02 4.6 0.957 1.170 0.00025 na Ho 0.008 0.074 0.067 0.071 0.005 6.9 0.167 0.188 0.00007 na Er 0.016 0.20 0.18 0.19 0.01 5.7 0.454 0.508 0.00021 na Tm <lod 0.029 0.027 0.028 0.001 3.9 na na na na Yb 0.020 0.20 0.18 0.19 0.01 4.9 0.479 0.540 0.00044 na Lu 0.003 0.030 0.027 0.028 0.003 8.9 0.065 0.067 0.00008 na Hf 0.98 1.13 1.07 1.10 0.04 3.7 1.68 1.83 0.422 0.429 0.426 0.005 1.2 Ta 13.1 59 49 54.2 6.80 12.5 na na 1.22 1.23 Tl na na na 11.5 13.8 0.148 0.155 0.152 0.005 3.3 W 1.7 53 48 51 3.92 7.7 na na na na Pb 2.9 11.7 10.6 11.2 0.84 7.5 17.9 6.57 1.86 1.84 1.85 0.01 0.8 Bi 0.96 2.0 1.8 1.9 0.13 6.5 na na na na Th 1.97 4.6 4.47 4.52 0.07 1.5 5 3.02 0.011 0.012 0.0114 0.001 6.2 U 1.66 10.2 9.2 9.7 0.72 7.5 18.6 22 0.272 0.27 0.27 0.00 0.5 ΣREE (ppm) 1.7 30.1 18.0 0.0 0.0 Eu/Eu* 0.25 0.51 0.07 0.50 *# measured at Université Joseph Fourier, Grenoble, France 105 Table 3.3. Trace element concentrations of the whole rock LNPG samples, the Macusani glass samples and the LNPG mineral separates contd. Location LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG LNPG Sample *P389859 spodumene *P389859 spodumene duplicate *P389859 spodumene average *P389859 spodumene stdev *P389859 spodumene %RSD *EB0509 quartz *EB1304 quartz *EB0509 feldspar *EB1304 feldspar *EB0509 mica *EB1304 mica REE pattern type Li (ppm) na na 4 45 20 871 na 16567 Sc 0.0017 0.0018 0.0018 0.0001 3.6 0.0000 0.0043 0.0000 0.0032 0.0010 0.0466 Ti 16.3 9.3 12.8 4.9 38.7 0.673 9.58 0.819 1.22 32.7 18.8 V 0.17 0.12 0.15 0.03 22.4 0.06 0.09 0.07 0.07 0.03 0.19 Cr 11.5 4.21 7.86 5.15 65.6 0.411 0.135 1.47 0.448 0.321 0.444 Co 0.53 0.55 0.54 0.01 2.1 0.01 0.03 0.01 0.03 0.02 0.05 Ni 26.6 13.5 20.05 9.26 46.2 0.0967 0.254 0.233 0.179 0.201 0.218 Cu 1.88 1.77 1.825 0.078 4.3 0.0285 0.136 0.171 0.082 0.15 0.198 Zn 37.5 38.7 38.1 0.8 2.2 82.6 2.39 12.6 5.2 339 159 As 1.07 0.125 0.598 0.668 111.8 0.0546 0.0486 0.137 0.147 0.11 0.139 Ga na na na na na na na na Rb na na 164 188 82.9 1147 3368 8733 Sr 24.1 22.9 23.5 0.8 3.6 0.3 5.1 9.7 35.8 3.4 19.4 Y 0.0063 0.0059 0.0061 0.0003 4.6 0.0019 0.0430 0.0052 0.0136 0.0046 0.0274 Zr 0.135 0.133 0.134 0.001 1.1 0.16 1.54 0.581 2.67 0.439 14.2 Nb 0.151 0.147 0.149 0.003 1.9 2.05 7.64 1.89 66.7 87.6 167 Mo na na na na na na na na Cd 0.0089 0.0083 0.0086 0.0004 4.8 0.0005 0.0009 0.0031 0.0058 0.0027 0.0060 Sn na na na na na na na na Sb na na na na na na na na Cs 8 13 10.5 3.5 33.7 na na na 105 123 2062 Ba 4.23 4.2 4.2 0.0 0.5 0.0464 2.25 0.0765 3.86 0.315 7.28 La 0.01377 0.01962 0.01670 0.00413 24.8 0.00238 0.11570 0.03980 0.03434 0.00284 0.03422 Ce 0.00883 0.02247 0.01565 0.00964 61.6 0.00366 0.21577 0.02157 0.02408 0.00498 0.03284 Pr 0.00136 0.00315 0.00225 0.00126 56.1 <lod 0.02236 0.00062 0.00163 0.00042 0.00231 Nd 0.00467 0.01138 0.00803 0.00475 59.2 0.00082 0.07764 0.00089 0.00488 0.00146 0.00636 Sm 0.00070 0.00190 0.00130 0.00085 65.3 0.00017 0.01264 0.00013 0.00076 0.00022 0.00099 Eu 0.00020 0.00031 0.00025 0.00008 31.0 0.00004 0.00225 0.00013 0.00041 0.00007 0.00027 Gd 0.00083 0.00167 0.00125 0.00060 47.5 <lod 0.00944 <lod 0.00084 0.00035 0.00115 Tb 0.00011 0.00021 0.00016 0.00007 44.1 0.00003 0.00121 0.00005 0.00018 0.00005 0.00024 Dy 0.00062 0.00107 0.00084 0.00032 37.4 <lod 0.00703 0.00027 0.00098 0.00020 0.00178 Ho 0.00011 0.00017 0.00014 0.00004 30.5 0.00002 0.00138 0.00006 0.00017 0.00005 0.00037 Er 0.00030 0.00048 0.00039 0.00013 33.4 0.00010 0.00375 0.00030 0.00051 0.00016 0.00140 Tm na na na na na na na na Yb 0.00029 0.00037 0.00033 0.00006 17.4 0.00015 0.00330 0.00109 0.00078 0.00017 0.00459 Lu 0.00006 0.00006 0.00006 0.00000 4.3 0.00002 0.00050 0.00016 0.00011 0.00003 0.00067 Hf 0.0378 0.0391 0.0385 0.001 2.4 0.037 0.117 0.0457 0.464 0.0986 1.27 Ta 0.503 0.502 1.72 na na 80.7 na na Tl 0.182 0.198 0.19 0.01 6.0 1.32 1.69 0.667 11.5 25.6 82.2 W na na na na na na na na Pb 2.51 2.94 2.73 0.30 11.2 0.0592 0.632 4.68 14.6 1.27 3 Bi na na na na na na na na Th 0.007 0.013 0.010 0.00 42.5 0.017 0.125 0.122 0.927 0.008 0.692 U 0.101 0.215 0.158 0.081 51.0 0.111 0.429 0.361 3.62 0.156 4.06 ΣREE (ppm) 0.0 0.1 0.0 0.5 0.1 0.1 0.0 0.1 Eu/Eu* 0.79 0.53 0.63 1.57 0.81 0.77 *# measured at Université Joseph Fourier, Grenoble, France 106 Trace element (HR-ICP-MS; PCIGR) analyses was obtained from two samples of Macusani obsidian (Pichavant et al. 1988a and b), a natural peraluminous volcanic glass (1) sample name MAC, obtained for this study from the Smithsonian Institution (Fig. 3.10) and (2) sample name TomMac (received pre-ground) that was previously analysed isotopically by Tomascak (1999). A duplicate trace analysis (re-run on the same aliquot) was undertaken on MAC.  3.3.2 Sample preparation for ICP-MS analysis Approximately 100 mg of sample material to be measured for whole rock trace element content at PCIGR was re-ground to 200 mesh and digested using the method described by Pretorius et al. (2006), in high-pressure PTFE digestion capsules with 5 mL concentrated HF and 0.7 mL 14M HNO3 and 0.7 mL concentrated HClO4. The capsules were sealed in metal assemblies and heated at 190 °C for 5 days. The samples were subsequently dried on a hotplate at ~120 °C and redissolved in 6 mL 6M HCl and returned to the oven for 24 hr at 190 °C. On cooling, if the sample was not completely digested the procedure was repeated after drying down. Fully digested samples were transferred to 15 mL Savillex® beakers and dried on a hotplate at ~120 °C. To remove all traces of HCl prior to analysis 1 mL of 14M HNO3 was added and the vials were sealed and again placed on the hotplate at ~120 °C for 24 hours. Sample solutions were subsequently dried and stored in preparation for dilution and HR-ICP-MS analysis. Mineral separates were handpicked under a binocular microscope and were cleaned for several minutes in an ultrasonic bath using Milli-Q® Element 18.2 MΩ.cm resistivity ultrapure water before being crushed in an agate mortar and pestle. Additional trace element analyses were undertaken on whole rock samples and minerals separates at the Université Joseph Fourier in Grenoble, France. Samples were digested at PCIGR in preparation for the analysis using the method above. On receipt at the Université Joseph Fourier the REE were separated from samples with low REE abundances using chemistry adapted from Barrat et al. (1996). Approximately 50 mg of sample was spiked with a 200 ppb Tm solution and REE were separated using cationic resin AG50W-X8, and eluted with 7M HNO3. 107 Macusani glass stream cobble Figure 3.10 Photograph of the Macusani glass cobble obtained for this study from the Smithsonian Institution. Coin 17 mm diameter. 108  3.3.3 ICP-MS analysis Trace element abundances were measured on whole rock samples (LNPG and Macusani glass) at PCIGR on a double-focusing sector field ELEMENT2 ICP-MS (Thermo Finnigan, Germany) using a Conikal glass concentric nebulizer (Glass Expansion, USA), a Scott-type glass spray chamber, and In as internal standard (mass 115). Samples were calibrated to multi-element calibration curves, typically from 0.1 ppb up to 100 ppb. Several times during analytical runs, measurements of 1% HNO3 were taken to provide baseline values, and of a low concentration sample of the multi-element standard to correct for the change in sensitivity of each element during the sequence. Each batch of analysed samples included duplicates to ensure optimal analytical precision. Additional trace element analyses of LNPG samples (whole rock and mineral separates) and Macusani glass, were undertaken at the Université Joseph Fourier using ICP-MS Agilent 7500CE with the samples diluted with a 2% HNO3: HF mix.  3.3.4 Rietveld analysis Quantitative assessment of the mineralogy of 14 samples (with two samples, P389739 and P389870, run twice) that lacked hand samples was undertaken using the Rietveld method and X-ray powder diffraction data as described in Raudsepp et al. (1999). Preparation of the samples required grinding to a <10 μm powder before being analysed. Measurements were taken using step-scan X-ray powder diffraction between 3- 80 º2θ with CoKα radiation on a standard Siemens (Bruker) D5000 Bragg-Brentano diffractometer equipped with a diffracted-beam with a Fe monochromator foil, 0.6 mm (0.3º) divergence slit, incident and diffracted-beam Sollers slits and a Vantec-1 strip detector. The long fine-focus Co X-ray tube was operated at 35 kV and 40 mV, using a take-off angle of 6º.  109 3.4 RESULTS 3.4.1 Whole rock major elements The major element whole rock analyses of the 28 pegmatite samples from the LNPG analysed by ALS Chemex, Vancouver, are given in Table 3.2. The LNPG samples group relatively tightly together within the granite field (Fig. 3.11; Le Maitre et al. 1989, Total Alkali – Silica diagram), except P389913 and P389751 that fall within the syenite field. All samples are peraluminous with A/CNK values between 1.16 – 1.84 (Fig. 3.12) and normative corundum ranging between 1.5 and 9.1% (Table 3.2). Major element abundances show variable but generally high SiO2 and Al2O3 values except P389817 with extremely high SiO2 and low Al2O3. Abundance of FeO, TiO2, MnO, MgO and CaO are predominantly low. The samples all have a moderate range in Na2O, K2O and variable but typically high P2O5 (0.11-1.1 wt.%) and F (0.02- 0.99 wt.%) levels. The exception again is P389817 with far lower values for P2O5 and F and P389797 that has low F abundance. The mafic oxides (FeO, TiO2 and MgO) show no trend in their abundance variations with increasing SiO2. This is in contrast to the generally comagmatic nature of the trends of CaO, MnO, Na2O, K2O and Al2O3 which all decrease with increasing SiO2 (Fig. 3.13A and B).   Volatile and fluxing components P2O5, F and H2O+ do not show a strong trend with increased SiO2 (Fig. 3.13).  3.4.2 Whole rock minor and trace elements The minor and trace element analyses of whole rock samples from the LNPG and Macusani glass samples are given in Table 3.3. The LNPG trace elements show strong fractionation from upper-crust normalized abundances (Fig. 3.14) with Cs, Rb, Ta and Nb and some U abundances being highly enriched compared to upper crustal composition (>10× ) and Ba and REE showing strong depletion (~100× ). Some of the alkali metals; e.g., Li, Cs, and Rb; High-Field-Strength (HFS) elements Nb, and Ta, and Bi; and high charge elements Sn and W can be moderately to extremely enriched in comparison to upper crustal values (Fig. 3.15).  Other HFS elements (e.g., Zr, Hf and Th) are depleted relative to upper crust values. Samples P389734 and P389844 both have high levels of Sn and high Nb, and Ta possibly caused by grains of cassiterite (given the extremely high levels of Sn) containing 110 65 75 6 8 10 Granodiorite GraniteSyenite 87.5 SiO wt.% 3.47 Na O +K O wt.% 2 2 2 SiO2 N a O + K O 2 2 Figure 3.11 Total Alkali-Silica diagram in wt. %) classifying the LNPG whole rock samples. ( Le Maitre . 1989;et alTAS; 111 0.5 1.0 1.5 0.4 1.4 2.4 (Al O /CaO + Na O + K O)(molar)2 3 2 2 l a  ( r (A O /N O  + K O ) m ol a ) 2 3 2 2 2.0 Peralkaline Metaluminous Peraluminous Figure 3.12 Shand's index diagram (Al O /CaO + Na O + K O)(molar) classifying the LNPG whole rock 2 3 2 2 samples. Displays the peraluminous composition of the whole rock LNPG samples showing A/CNK values of 1.14 to 1.84. 112 .1 1 65 70 80 90 0.1 0.3 0.5 0.7 0.9 1.1 P O2 5 F H O+2 SiO2 t w .%  0 5 10 15 20 w t %  . Al O2 3 Na O2 K O2 CaO MnO w t %  . A B C Figure 3.13 Diagram illustrating the trends of; A. CaO and MnO wt.%, B. Al O, Na O and K O wt.% and C. 2 2 2 P O , F and H O+ wt.%, with increasing SiO  content of the LNPG whole rock samples.2 5 2 2 113 .001 .01 .1 1 10 100 1000 CsRbBaTh U K TaNbLaCeSr NdHf ZrSm Ti Y YbLu s l / pe r c on tin en t cr u t am p e up  al  s Figure 3.14 Upper continental crust-normalized (Taylor and McLennan 1985) extended trace element diagrams of the LNPG whole rock samples. 114 1000 100 10 1 0.1 0.01 0.001 a ve ra ge o fn o rm a ls a m pl e ra n ge va lu es u pp er cr u st Sample values/upper crust in order of decreasing abundance Figure 3.15 Plot of average whole rock LNPG trace element compositions normalized to upper continental crust values (Rudnick and Gao 2003). Shown in order of decreasing abundance. Li Ta Sn Cs Rb Bi Nb Ga W Cd U Zn Mo Hf Pb Cu Th Co Sr Sb Zr Ni Ba Tm La Y Yb Tb Dy Gd Pr Sm Ce Lu V Eu Nd Ho Er Sc 115 Nb- and Ta-rich columbite inclusions, as identified in other LNPG samples by Pemberton (2002) and as occurring the granitic Varuträsk pegmatite, Sweden (Černý et al. 2004). Samples P389719 and P389910 have elevated Bi and Pb, which, in combination with the moderately high Nb and Ta in those samples may indicate the presence of (bismuto?) pyrochlore. Trace elements from two samples of Macusani peraluminous glass (MAC and TomMac) were measured. MAC was analysed at the PCIGR, UBC and both samples were analysed at the Université Joseph Fourier, Grenoble, France. The alkali metals, Ta, Zr, Hf and Th in the Macusani glass are comparable to the levels in the LNPG whole rock samples whereas U and the high charge elements Sn and W are elevated and Cs is almost a level of magnitude higher. The variation in REE abundance between the MAC samples analysed at UBC and Université Joseph Fourier (higher Zn, Pb, Ba, U, Th, Y and REE in the samples measured at UJF) may be the result of micro-phenocrysts. Other rare phenocrysts have been identified in Macusani glass (biotite, monazite and apatite (identified petrographically and by SEM in this study) as well as andalusite, feldspar, quartz, sillimanite and virgilite (Barnes et al. 1970, Noble et al. 1984, French et al. 1978).  3.4.2.1 Rare earth elements Chondrite-normalized (Sun and McDonough 1989) REE patterns of the LNPG samples (Fig. 3.16) display three distinctive LREEN-enriched variations: 1. relatively straight LREEN-enriched patterns weakly negative or more rarely weakly positive Eu anomalies. Similar patterns are observed in a number of granitic and aplitic dikes from the LNPG area (South Nahanni and Rifle Range, respectively; Rasmussen et al. 2007) but at far higher element concentrations. 2.  discontinuous LREEN-enriched patterns with a break in slope at Nd to shallower and positive trends, strongly negative Eu anomalies and upward-convex trends between Gd and Ho. Similar patterns are recognized in highly evolved, commonly P-rich, granitic rocks and pegmatites e.g., Bau 1996, Breiter et al. 1997, Förster 1998 and Kontak et al. 2001). 3. listric LREEN-enriched patterns (concave-up curved pattern; depleted in MREEN, with variable negative Eu anomalies and relatively elevated HREEN). 116 10 .1 1 Sa m pl e/ ch on dr ite La Ce Pr Nd PmSm Eu Gd Tb Dy Ho Er Tm Yb Lu.01 10 .1 1 Sa m pl e/ ch on dr ite ‘flat’ ‘discontinuous’ ‘listric’ LuLa Ce Pr Nd PmSm Eu Gd Tb Dy Ho Er Tm Yb ‘indeterminate’ ‘listric’ ‘flat’ ‘discontinuous’ A D C B Figure 3.16 Rare earth element patterns of the whole rock LNPG samples (normalized to CI chondrite (Sun and McDonough 1989)) separated into 4 groups. A. Grey square: flat REE pattern, B. Open triangle: discontinuous REE pattern, C. Black diamond: listric REE pattern, D. Outlines of A (dotted line), B (dark grey) and C (pale grey) combined to illustrate variation in REE. Inset shows indeterminate REE patterns (black circles). N N N N 117 The remaining four samples are LREEN-enriched but have less distinctive ‘indeterminate’ patterns (P389726, P389734, P389817, P389861). The REEN patterns are fairly evenly distributed across the study area except for the discontinuous patterns that are not found in the north (Fig. 3.17). This could be due to low sample numbers from that area. The samples with straight REEN patterns have a wide range of Eu anomalies (Eu/Eu* ratio calculated by EuN/√[(SmN).(GdN)] (Taylor and McLennan 1985) where N denotes CI-normalized values), up to positive values (0.42 - 1.15); the discontinuous patterns have the largest negative Eu anomalies and a fairly restricted range (0.12 - 0.57); the samples with listric REEN patterns also have large Eu anomalies and a wide range (0.15 - 0.9 plus an outlier at 1.9). The three samples with indeterminate patterns have small Eu anomalies and a relatively restricted range (0.59 to 0.95). The patterns were differentiated by their appearance but they can also be differentiated by calculating the variance from a straight line taken between the elements La to Nd (T1) and Gd to Ho (T3; Fig. 3.18A), using the method described in Monecke et al. (2002; equation 2) to quantify the ‘tetrad effect’ (the splitting of the REEN patterns into 4 rounded segments). For each of the two segments the relative standard deviation was calculated using the following equation:  ( ) ( ) 2 2 CiBi i 2/3 1/3 1/3 2/3 Ai Di Ai Di xx1T * 1 1 2 x *x x *x ⎛ ⎞⎡ ⎤ ⎡ ⎤⎜ ⎟⎢ ⎥ ⎢ ⎥= − + −⎜ ⎟⎢ ⎥ ⎢ ⎥⎣ ⎦ ⎣ ⎦⎝ ⎠                     (1)  where i is one of the two segments of four consecutive rare earth elements used i.e., La to Nd (1) or Gd to Ho (3). xAi refers to the abundance of the first CI-normalized element in the segment, xBi the abundance of the second CI-normalized element etc. The second segment (Pm to Gd) is omitted due to the absence of Pm and the typically anomalous value for Eu in relation to the rest of the REE and in this description segment 4 (Er to Lu) could not be used here as the abundance of Tm was typically below detection levels. Even given these restrictions, this method was able to distinguish the patterns successfully (Fig. 3.18A) using only two segments. The only exceptions were 118 504 E 506 E 508 E 9 N 68 5 6890 N  Geological boundary:approximate, inferred Axial trace of anticlinorium, defined Syncline, extrapolated beneath overburden Thrust fault, teeth indicate upthrust side  Little Nahanni Pegmatite Group - areas of highest density of dikes Cambro-Ordovician Rabbitkettle Fm. Lower Cambrian Vampire Formation Precambrian to Lower Cambrian Narchilla Formation  Precambrian Yusezu Formation Sample site elevation   (m above sea level) 1700 1900 REE  patterns (see text)N Flat Discontinuous Listric Indeterminate Spatial distribution of whole rock REEN patterns Figure 3.17 Spatial distribution of the various LNPG whole rock REE patterns across the study area. Samples N with flat and listric REE patterns are spread throughout the study area, whereas samples with N discontinuous REE patterns occur only in the central and south of the study area. N 119 0 0.2 0.4 0.8 0 0.4 T3 T1 REE patterns flat discontinuous listric indeterminate P389789 P389844 .1 10 P389844 P389789    S am pl /    e ch on dr ite  n rm al ze o i d 0.4 listric flatdiscontinous 1 La/Ce Gd/Tb indeterminate P389844 P389789 3 8 A B Figure 3.18 Two geochemical methods used to differentiate the three groups of distinctive REE  patterns. N A.Variations in the relative standard deviation from a straight line between the first and last elements of two groups of four REE from the whole rock LNPG samples. 0 represents no deviation. T1 refers to elements La to Nd and T3 refers to elements Gd to Ho (Monecke et al. 2002), B. La/Ce against Gd/Tb showing the zonation of the three groups of patterns. 120 samples P389789 and P389844 (see inset in Fig. 3.18A). The indeterminate samples do not correlate with the groups. An alternative method of differentiating the groups uses their La/Ce and Gd/Tb ratios (Fig. 3.18B). The groups split as follows; straight REEN patterns La/Ce 0.5 – 0.7, Gd/Tb 5.6 – 7.5; discontinuous patterns La/Ce0.5 – 0.8, Gd/Tb 3.5 – 5.8; and listric patterns La/Ce 0.6 – 1.0 and Gd/Tb 4.8 – 6.4. This method cannot distinguish the indeterminate patterns well (La/Ce 0.9 – 1.5 and Gd/Tb 2.9 – 6.9), two of which show anomalous Ce abundances, but, in contrast to the previous method, samples P389789 and P389844 are well differentiated by this method. Overall, ΣREE is extremely low (Table 3.3) but correlates with three distinctive normalized REEN patterns. The straight REEN patterns have the highest ΣREE values, between 1.6 and 7.7 ppm, average of 2.1 ppm (omitting outlier P389854 at 7.7 ppm); the discontinuous patterns range between 0.9 and 3.6 ppm, average of 1.2 ppm (omitting outlier P389789 at 3.6 ppm); and the listric patterns have the lowest range between 0.2 and 1.1 ppm, average of 0.6 ppm, the indistinct patterns range between 1.4 and 2.1 ppm, average of 1.8 ppm (see Fig. 3.16).  3.4.2.2 Element ratios The unusual trace element composition of the LNPG limits the application of some commonly used discriminant geochemical diagrams for felsic rocks (e.g., Ta/Yb or Y/Nb; Pearce et al. 1984). Element ratios used to determine the geochemical evolution of pegmatite minerals such as feldspar and muscovite (e.g., K/Rb, K/Cs, Sr/Rb) all of which decrease with magmatic evolution; Černý 1991b), can be used to assess the degree of magmatic fractionation of the whole rock composition due to their predominance in the mineralogy of the pegmatites. At LNPG these ratios are K/Rb 4  – 24; K/Cs 58 – 534; and Sr/Rb 0.005 – 0.127. Rubidium and Cs abundances have strong positive correlations with magmatic evolution; in contrast Li abundance shows a subtle decrease (Fig. 3.19A and B). With magmatic fractionation, volatile and flux components (e.g., F and P2O5) increase in the magma (Fig. 3.19C). Commonly, ratios of elements such as Nb/Ta, Y/Ho and Zr/Hf that are typically fractionate little in most igneous rock compositions, can fractionate significantly in pegmatites. In the LNPG samples these element pairs 121 0 10 200.0 0.5 1.5 2.0 F+ P O 2 5 K/Rb 0 10 200 100 300 400 500 Cs ppm K/Rb 0 20 0 5000 Li ppm K/Rb 0 10 200 2000 6000 8000 Rb ppm K/Rb Nb/Ta 0 10 20 0 1 3 K/Rb 0 1 2 3 40 50 150 200 Li/Cs Nb/Ta 1 10 100 10 100 Zr/Hf Nb/Ta bulk crust A B C D E F Figure 3.19 Assessment of the degree of magmatic fractionation in the whole rock LNPG samples using various element ratios and abundances. Colour coding for the samples reflects their REE patterns as described in Fig.16. Grey arrows are schematic and indicate direction of progressive magmatic fractionation. A-D; K/Rb against Rb ppm, Cs ppm (inset of Li ppm), F+ P O wt.% and Nb/Ta. E. Nb/Ta against Zr/Hf (bulk crust composition, Rudnick and Gao 2003). F. Nb/Ta against Li/Cs. N 2 5 122 fractionate strongly, with decreasing ratios indicating magmatic evolution away from bulk crust values (Rudnick and Gao 2003; Fig. 3.17E), Nb/Ta ranges from 0.6 - 3.2 (except outlier sample P389739 at 22.6); Zr/Hf 6.64 – 15.59; and Y/Ho 20.25 – 67.73. Strong fractionation of Li and Cs (Li/Cs 2 – 183) decreases with evolution of the magma as more incompatible Cs becomes increasingly enriched in the remaining melt (Fig. 3.19F).  3.4.3 Mineral separate minor and trace elements Minor and trace element abundances of the mineral separates; spodumene (AL050704 and P389859), quartz (EB0509 and EB1304), orthoclase (EB0509 and EB1304), muscovite (EB0509) and lithian muscovite (containing 3-4 wt.% Li2O; EB1304) are included in Table 3.3. In these samples fluid inclusions are minimal in the phenocrysts, and the late crystallising groundmass feldspar contains the majority of the small, anhedral inclusions. Spodumene contains relatively high levels of Ti, Cr, Ni, and Zn and very low values of Cs. The quartz and feldspar samples contain noticeable quantities of Li, Zn, Rb and Nb, and feldspar in EB1304 additionally contains Cs, Ta, Tl and Pb. Both mica samples contain high levels of Ti, Zn, Rb, Nb, Cs and Tl. The common occurrence of Zn suggests it may be a component of the fluid inclusions. Total rare earth element abundance in two samples of spodumene (AL050704 and P389859; plus duplicate), and quartz, feldspar and mica from EB0509 and EB1304 were all  <0.09 ppm, except for quartz from EB1304 at 0.47ppm. Chondrite-normalized REE abundances of the mineral separates (Fig. 3.20) show: • Spodumene from both samples had similar listric patterns with elevated LREEN, AL050704 has a slight negative Eu anomaly. • REEN profiles from both quartz samples were fairly straight with elevated LREEN and a minor negative Eu anomaly in sample EB1304. Several elements (Pr, Eu, Gd, Tb, Dy and Ho) were below detection limits in EB0509. The higher concentrations in EB1304 may be recording fluid inclusion contents. 123 .001 .01 .1 1 spodumene La Ce Pr NdPmSmEuGdTb DyHo Er TmYb Lu mica A .0001 .001 .01 .1 1 La Ce Pr NdPmSmEuGdTb DyHo Er TmYb Lu B feldsparC Dquartz Sa m pl e/ ch on dr ite Sa m pl e/ ch on dr ite EB1304 EB0509 AL050704 P389859 EB1304 EB0509 EB1304 EB0509 Figure 3.20 Rare earth element patterns of LNPG mineral separates (normalized to C1 chondrite (Sun and McDonough 1989)). Spodumene from samples AL050704 and P389859 and quartz, feldspar and mica from samples EB0509 and EB1304. 124 • The low abundance of REE in feldspar of EB0509 was below analytical detection for some elements. Where visible both samples show steep negative slopes in the LREEN and slightly negative Eu anomalies. • Micas from sample EB1304 had higher REE abundance and a more defined listric, or MREEN depleted shape than sample EB0509.  3.4.4. Rietveld method results Modal mineral abundance of several selected whole rock LNPG samples was determined using the Rietveld method and X-ray powder diffraction data analyses (Raudsepp et al. 1999). Analysis determined the presence of; quartz, plagioclase, muscovite, ± spodumene, ± K-feldspar, ± lepidolite, ± beryl, ± apatite, ± cassiterite, ± cookeite, ± clinoptilolite and ± calcite (Table 3.4; Fig. 3.21). All samples contain >25% quartz (28.5-42.7%) and >23% plagioclase (23.3- 42.4%). Relative proportions of muscovite (3.3-23%), spodumene (0.8-21.2%), K- feldspar (2.3-18.2%) and lepidolite (0-12.9%) are highly variable. The K-feldspar contents of the duplicate analyses of P389739 differ by ~18%. This is due to overlapping of the microcline peaks on the diffraction pattern by large quartz and albite peaks thereby lowering the precision. The spodumene contents of the duplicate analyses of P389870 also differ by ~18%. This variation may be due to sample inhomogeneity or problems arising from the preferred orientation of spodumene. For the other analyses the results are within the expected reproducibility for this method (Raudsepp et al. 1999). Samples with discontinuous REEN patterns tend to contain a greater proportion of spodumene. In general, K-feldspar and spodumene decrease with increasing muscovite and there is a higher incidence of lepidolite in the samples showing listric and straight REEN patterns, which concurs with the field observations. Of the three less abundant (and commonly absent) minerals, beryl, apatite and cassiterite, beryl has the greatest abundance in one sample (1.3%). Presence of cookeite indicates low temperature late-stage crystallisation or spodumene or lepidolite alteration, and clinoptilolite is a zeolite commonly associated with devitrification. Together with minor calcite these two low temperature minerals occur in very low abundance. There are 125 Table 3.4. Results from Rietveld method and X-ray powder diffraction data analysis of selected whole rock LNPG samples. P 389704 P 389711 P 389719 P 389726 P 389734 P389739 P389739 duplicate mean std %RSD CI-normalised REE pattern discontinuous discontinuous straight indeterminate listric straight straight Mineral Ideal Formula Quartz SiO2 29.4 32.7 32.5 33.5 38.3 32.2 32.9 32.55 0.5 1.5 Plagioclase NaAlSi3O8 – CaAl2Si2O8 30.3 36.1 38.5 38.8 23.4 34.2 35.4 34.8 0.8 2.4 Muscovite KAl2AlSi3O10(OH)2 5.3 13.8 7.5 23 19.4 9.1 9.0 9.05 0.1 0.8 Spodumene LiAlSi2O6 19.7 13.3 5.7 1.2 0.8 14.7 13.5 14.1 0.8 6.0 K-feldspar KAlSi3O8 11.6 2.5 15.1 3 3.3 5.4 4.2 4.8 0.8 17.7 Lepidolite K(Li,Al)3(Si,Al)4O10(F,OH)2 12.9 4 4.4 4.2 0.3 6.7 Beryl Be3Al2Si6O18 0.3 0.4 0.5 0.8 0.5 0.5 0.5 0.0 0.0 Apatite Ca5(PO4)3(OH) 0.3 Cassiterite SnO2 1 Cookeite LiAl4(Si3Al)O10(OH)8 2 1.3 0.4 Clinoptilolite (Ca0.5,Na,K)6[Al6Si30O72]·~20H2O 1.2 Calcite CaCO3 Total% 99.8 100 100.1 100 99.9 100 100 126 Table 3.4. Results from Rietveld method and X-ray powder diffraction data analysis of selected whole rock LNPG samples contd. P 389764 P 389768 P 389774 P 389797 P 389816 P 389870 P 389870 duplicate mean std %RSD P 389910 P 389913 discontinuous listric discontinuous listric listric discontinuous discontinuous flat discontinuous Mineral Quartz 28.5 30.3 29 42.7 33 33.8 32.5 33.15 0.9 2.8 25.4 31.7 Plagioclase 31.1 36.1 40.6 23.3 42.4 28.7 26.7 27.7 1.4 5.1 38.1 25.8 Muscovite 16.6 20.3 22.5 3.3 15.8 11.3 10.4 10.85 0.6 5.9 9.9 4.3 Spodumene 21.2 10.4 1.8 12.4 14.2 18.4 16.3 3.0 18.2 6.8 22 K-feldspar 2.3 2.3 5.5 18.2 11.7 11.7 11.7 0.0 0.0 14.6 11.6 Lepidolite 7 2.9 3.3 Beryl 0.4 0.6 0.5 1.3 0.4 0.3 0.35 0.1 20.2 0.3 Apatite 0.4 0.8 Cassiterite Cookeite 0.8 1.3 Clinoptilolite Calcite 0.4 Total % 100.1 100 99.9 99.9 99.9 100.1 100 100 100 127 Figure 3.21 Pie chart representations of the whole-rock mineralogy of several LNPG samples determined by the Rietveld method. Diagrams are grouped by their REE patterns, and samples with discontinuous patterns tend to comprise a greater portion of spodumene. There is a relatively strong inverse correlation between the abundance of K-Feldspar and muscovite. N 128 P389913 P389910P389816P389768 P389797 P389774 P389764 P389734 P389719 P389711 P389870P389704 Quartz Plagioclase Muscovite K-Feldspar Spodumene Lepidolite Others Modal abundance of minerals grouped by their REE patternsN listric Straight discontinuous P389726 indeterminate duplicate P389739 duplicate 129  130 minor low temperature alteration products and an absence of a clay fraction in all samples.  3.5 DISCUSSION 3.5.1 Comparisons of magmatic fractionation Geochemical proxies for the magmatic evolution of granitoids (including Zr/Hf and Nb/Ta, and Cs and K/Rb) show the LNPG dike whole rock geochemical compositions to be significantly evolved in comparison to >150 regional Cretaceous intrusions from the NWT and Yukon Territory (Fig. 3.22A and B; Groat (unpublished data), Eaton and Armstrong 1980 (unpublished data), Gordey and Anderson 1993, Murphy 1997, Duncan 1999, Lang 2001, Coulson et al. 2002, Heffernan 2004, Rasmussen et al. 2007). This database includes samples from the regional S-type granites proximal to the LNPG dikes and the O’Grady Batholith (the only other local example of Li-rich pegmatite). Only a felsite sample from the O’Grady Batholith (YK97-OG2) and two samples from the pegmatitic, fluorite-rich dikes from the Rau locality approximately 400 km northwest of the LNPG (Groat, unpublished data) show geochemical evolution almost to the extent of the LNPG samples. Within the LNPG array the three distinct REEN patterns reflect the progression of magmatic evolution with the discontinuous REEN patterns typically being the least evolved. To compare the LNPG trace element patterns with the regional intrusions the samples were normalised using: • the highly evolved O’Grady Batholith felsite sample (YK97-OG2) that falls close to the least evolved samples of the LNPG array, • a regional granitic intrusion with Zr/Hf and Nb/Ta compositions relatively similar to bulk continental crust (KR-05-115, Rasmussen et al. 2007; Fig. 3.23A and B). Compared to both the regional granite and the O’Grady felsite, the LNPG samples show stronger fractionation, illustrated by Rb, Ta and Nb enrichment. The LNPG samples are also characterized by extensive depletion in REE and Ba. In a comparison of REEN patterns, the shape of the straight LNPG REEN patterns is similar to that of the regional granite (Fig. 3.23B) although REE are almost two orders .1 1 10 100Nb/Ta 10 Zr/Hf Bulk continental crust 1 10 100 1000 1 10 100 1000 Cs K/Rb Bulk continental crust O’Grady Batholith felsite sample Regional granite used in C McQuesten Suite Tombstone Suite Emerald Lake Tungsten Suite Tay River Suite Anvil Suite Rau dykes O’Grady Batholith LNPG REE patternsN Straight Discontinuous Listric Indeterminate A B Figure 3.22 Comparison of the degree of magmatic fractionation observed in the LNPG with examples of Cretaceous magmatism from the northern Cordillera from Gordey and Anderson 1993, Heffernan 2004, Murphy 1997, Duncan 1999, Coulson 2002, Rasmussen 2007.et al. et al. 131 .0001 .001 .01 .1 1 10 100 1000 Cs Rb Ba Th U K Ta Nb La Ce Sr Nd Hf Zr Sm Ti Y Yb Lu O’Grady Batholith felsite Regional granite LNPG data normalised using: .01 La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu LNPG REE patternsN Straight Discontinuous Listric Indeterminate samples omitted O’Grady Batholith felsite LNPG and O’Grady Batholith samples normalised to regional granite composition .0001 .001 .1 1 B Asa m pl es /e xa m pl es n a m e d in di ag ra m sa m pl es /re gi on al gr an itic in tru si on Figure 3.23. LNPG trace element and REE data compared with samples of regional granite (KR-05- 115) and felsite from the O’Grady Batholith (YK97-OG2).A. Extended trace element diagram comparing the LNPG data normalised by the compositions of the felsite and the granite sample. B. REE diagram of the LNPG and felsite samples normalised by the composition of the granite sample. 132 of magnitude less abundant in the LNPG samples. The felsite sample from the O’Grady Batholith has an REEN pattern similar to the listric patterns from the LNPG although the latter are approximately 10×  less enriched in REE. The LNPG dikes show a broad range of K/Rb values and Cs concentrations that are comparable to other examples of highly fractionated magmas worldwide (whole rock and melt inclusions; Fig. 3.24A and B). Similarities to examples from the Variscan orogeny (Beauvoir granite, Erzgebirge intrusions) and the Tin Mountain pegmatite are noteworthy. In summary, the LNPG shows characteristics of extreme geochemical fractionation. These characteristics are comparable to those of a small number of similar rocks worldwide, and indicate a far greater degree of magmatic fractionation than observed in Cretaceous granitoids from the area around the LNPG.  3.5.2 Rare earth element fractionation Rare earth element patterns at LNPG provide the link between the mineralogy and geochemical indicators of magmatic evolution and may explain processes occurring during late-stage magmatic fractionation. Despite the ΣREE of the LNPG being typically very low (0.2 – ~2 ppm), the three distinct REEN patterns observed at LNPG (straight, discontinuous and listric; Section 3.4.2.1; Fig. 3.18A and B) are consistent. Lanthanide elements typically respond as a cohesive group in a given geochemical environment due to their similar ionic radii and charge, with their atomic configurations only changing across the group with the progressive filling, by electrons, of the 4f orbitals. Discontinuous REEN patterns, similar to those observed in the LNPG, occur in volatile- and flux-rich evolved igneous rocks and accessory minerals (e.g., Jolliff et al. 1989, Bau 1996, Irber 1999, Monecke et al. 2002, Lui and Zhang 2005, Badanina et al. 2006). The origin of these disjointed patterns that can be split into four ‘concave- down’ sections (so-called ‘M’ tetrad patterns) is a matter of ongoing research. Possible processes for fractionation of the lanthanide group focus on (1) crystallisation of accessory minerals and their consequent removal from the melt (e.g., garnet, apatite, monazite; Pan 1997 and Pan and Breaks 1997, Yurimoto et al. 1990), and (2) 133 Figure 3.24 Compilation of K/Rb against Cs ppm and K/Rb against Nb/Ta (used as proxies for magmatic differentiation) for LNPG whole rock samples, bulk crust compositions and various evolved granites and pegmatites. Upper and middle crust, Rudnick and Gao 2003; Khangilay granites, Badanina 2004; Erzgebirge granites, Förster 1999; Keystone pegmatites, Jolliff 1992; Weinebene spodumene deposit, Göd 1989; God's River pegmatites, Chackowsky 1987; Greenbushes pegmatite, Partington 1995; Beauvoir intrusions, Raimbault 1995; Ehrenfriedersdorf melt inclusions, Webster 1997; regional granites, Rasmussen 2007; Superior Province pegmatites, Breaks 2003; Tanco pegmatite, Stilling 2006; Tin Mountain pegmatite, Walker 1986. et al. et al. et al. et al. et al, et al. et al. et al. et al. et al. 134 LNPG upper crust middle crust Beauvoir intrusions Superior Province Ehrenfriedersdorf melt inclusions God’s River Erzgebirge Khangilay Weinebene Keystone Regional granites LNPG REE patternsN straight discontinuous listric indeterminate 1 10 1000 1 10 100 K/Rb Cs 10 1000 1 100 K/Rb Cs Tanco See insets .1 1 10 1 10 100 K/Rb Nb/Ta 1 10 1 100 K/Rb Nb/Ta Tin MountainGreenbushes LNPG samples LNPG samples A B 135 complexation of the elements in a volatile-enriched (especially F) silicate melt (Bau 1996, Irber 1999). The listric REEN patterns occur in more highly evolved samples of the LNPG (as defined by decreased K/Rb, Nb/Ta etc.) than the discontinuous REEN patterns. In the literature the listric REEN pattern is not typically associated with late-stage magmas. In less-evolved rocks it is more commonly associated with hornblende fractional crystallisation or hornblende remaining in the restite during partial melting (Borg and Clynne 1998). The extensive magmatic fractionation experienced by the LNPG pegmatite magma make the retention of the original REE signature from partial melting an unlikely explanation. Fractional crystallisation or element complexation in a volatile-rich melt are possible causes and are discussed below. The straight REEN pattern is associated with some of the most highly evolved dike compositions and yet it is has the highest ΣREE, and the least fractionated REEN pattern. As mentioned previously the slope of these straight patterns is similar to that found in some contemporary regional granites (Rasmussen et al. 2007). Magma of this composition is assumed to have evolved extensively but, unlike the compositions showing discontinuous and listric REEN patterns, did not undergo significant crystal fractionation or phase separation in its late stages, enabling it to retain more REE and producing less REE fractionation. The occurrence of these three distinct REEN patterns from dikes, assuming that they originated from the same magma body, indicates that at least two processes causing REE fractionation occurred. A theory regarding their formation must explain all three and take into account their relative degrees of geochemical evolution. We suggest that an increase in the abundance of flux components in the magma was responsible for the REE fractionation observed at LNPG. Much of the evidence for that lies in the response of ‘element pairs’, such as Zr/Hf and Y/Ho as discussed below.  3.5.3 ‘Immobile’ element pair fractionation The build-up of volatile and flux components (e.g., OH, H2O, B, P, F) and other silicate melt network-modifiers such as the alkali elements (Li, Na, K, Rb, Cs), as a result of magmatic fractionation has a significant influence on the short–range structure of the 136 melt (Hannon et al. 1992, Henderson 2005). Research continues on the role that these components play in the processes occurring in the later stages in pegmatite formation. There is a long standing discussion is whether pegmatite-forming magma is an environment; (1) unsaturated in volatile components and where fractional crystallisation is a key process in driving compositional evolution (e.g., London et al. 1989) or (2) saturated in volatile components causing elemental fractionation between silicate-rich and silicate-poor fluids (Veksler et al. 2005, Badanina et al. 2006). High field strength elements, Zr and Hf, do not fractionate readily in most magmatic conditions and typically retain values of ~33-~36.6 (continental crust, Taylor and McLennan 1985; oceanic basalt, Jochum et al.1986 and references therein, respectively), close to the CI chondrite value of 36.3 (Jochum et al.1986). They have a similar ionic radius (~0.8Å; octahedral co-ordination, Whitaker and Muntos 1970), are quadrivalent and tend to be incompatible in silicate melts until saturation when they form stable compounds. Yttrium and Ho are also regarded as an ‘immobile pair’ of elements, they are trivalent and both have an ionic radius of 0.98Å in octahedral coordination and display similar values in oceanic basalt (~27.7; Jochum et al.1986) and continental crust (26, Taylor and McLennan 1985) to the CI chondrite value (28.1; Jochum et al.1986). Fractionation of these element pairs can occur under specific conditions including; chemical complexation in the presence of fluxing components (e.g., H2O, B, F, P, Li, Cl; Bau 1996) and during crystallisation of evolved peraluminous magmas (Zr/Hf; Linnen and Keppler 2002). Bau (1996) in his study on trace element fractionation and the lanthanide ‘tetrad effect’, identified the CHARAC (Charge and Radius Controlled) field within Y/Ho versus Zr/Hf space that outlines the typical values for these ratios under magmatic conditions (Fig. 3.25A). Outside those limits, at non-chondritic values, Bau (1996) suggested a fluid component becomes increasingly influential to element fractionation. Assessing the variation in these ratios in the LNPG (and some examples introduced earlier) across the three REEN patterns a trend from bulk crustal values (Rudnick and Gao 2003) is observed. Bulk upper and middle crust and the Macusani glass compositions are located within the CHARAC field (Bau 1996), and from that field the Erzebirge melt inclusions (Förster et al. 1999) show decreasing Zr/Hf values. At lower Zr/Hf the LNPG samples diverge into two groups with varying Y/Ho values. The 137 Figure 3.25 Compilation of Y/Ho against Zr/Hf and Hf ppm for LNPG whole rock samples, bulk crust compositions and various evolved granites and pegmatites. The CHARAC field in A) and B) define Y/Ho and Zr/Hf space coherent behaviour of trace elements due to the interelationship between atomic charge and radius, as described in Bau (1996). A) LNPG samples fall outside the limits of the CHARAC field and diverge to high Y/Ho values (samples with discontinuous and listric REE patterns) (indicated with schematic arrows). B) Variation in Hf abundance with decreasing Zr/Hf illustrating the change in Zr solubility with magmatic fractionation and associated F increase (Linnen and Keppler 2002). in the majority of natural igneous rocks resulting from the N 138 2 10 Y/Ho Zr/Hf 80 40 CHARAC field 0 5 15 20 0 10 20 30 40 Zr/Hf Hf ppm 0 10 0 10 20 Zr/Hf Hf LNPG (B only)upper crust middle crust Beauvoir intrusions Superior Province Erzgebirge Khangilay Keystone LNPG REE patternsN flat discontinuous listric indeterminateA and B(inset) Macusani glass 40 20 A B CHARAC field CHARAC field LNPG 139 samples showing discontinuous and listric REEN patterns occur at higher Y/Ho values; samples with straight REEN patterns have lower Y/Ho values that appear to form a trend towards the evolved Khangilay Li-F granitic intrusions (samples O-4 and O-5; Badanina et al. 2006). An increase in Hf concentration is associated with magmatic differentiation (Fig. 3.25B) where an increase in F in the late-stage magma can alter Zr solubility and promote lower Zr/Hf ratios (Linnen and Keppler 2002). The decoupling of Y and Ho seen at LNPG has been observed in a study on the formation of silicate and fluoride immiscible liquids from F-rich silicate melt and two- liquid element partitioning, with Y preferentially partitioning into the fluoride liquid (Veksler et al. 2005). High concentrations of Li in the silicate melt and the stability of Li- F compounds is thought to increase the potential for immiscibility between silicate and fluoride liquids in haplogranite compositions. The generation of separate fluid phases from silicate melt is not uncommon in evolving magmatic systems and fluid-melt element fractionation can become significant in changing the composition of the melt (Hedenquist and Lowenstern 1994, Audétat and Pettke 2003). Similar mechanisms have been suggested for pegmatite systems (e.g., Jahns 1982, Černý et al. 1985, Thomas et al. 2000, Badanina et al. (2004)), and experimentally investigated by Suk (1998), Veksler et al. (2005) and Prowatke and Klemm (2006).  3.5.4 Crystal/melt and fluid/melt fractionation The association between P and F with late-stage peraluminous silicate melts and strong element fractionation is observed elsewhere (e.g., Badanina et al. 2006, Marshall et al. 1998, Thomas et al. 2000 and 2005, Veksler 2004, Veksler et al. 2005) and was assessed in the LNPG dikes. The abundances of P and F are high at LNPG (0.20-1.10 P2O5 wt.% and 160-9940 ppm F) because several P- and F-bearing minerals occur in the dikes; apatite (Ca5(PO4)3(OH,F), lithiophilite (LiMnPO4), montebrasite (LiAl(PO4)(OH,F), rare berlinite (AlPO4), fluorite (CaF2) and micas, including muscovite KAl2(AlSi3O10)(F, OH)2 and lepidolite K(LiAl)2-3(Al,Si3O10)(F,OH)2. Whole rock samples with discontinuous REEN patterns have low, restricted P2O5 content and high CaO/P2O5 (Fig. 3.26) suggesting that these samples originally had somewhat higher concentrations of CaO than the rest. Slightly elevated CaO in these 140 P2O5 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 CaO P O2 5 R =0.932 Figure 3.26 Diagram showing the variation in CaO wt. % against P O wt. % between the three LNPG whole rock REE groups. Samples with flat REE patterns (grey squares) show a positive straight line trend that tends to separate the samples with discontinuous REE patterns at lower P O wt. % from samples with listric REE patterns which show a greater variation in P O wt. % values. 2 5 N N N 2 5 N 2 5 141 samples may have promoted the saturation of apatite and fluorite. Samples with straight REEN patterns have constant CaO/P2O5 ratios (r2 = 0.93), while in contrast samples with listric REEN patterns tend to have lower CaO/P2O5 and a wide range of P2O5. The listric patterns show the lowest CaO/F ratios. To understand the processes that produced the distinct REEN patterns the effect of separating individual components (volatile-rich minerals and fluids) from the magma was investigated. As a starting point for this calculation the average values of each REE from the samples with straight REEN patterns (which have a greater ΣREE than samples with discontinuous or listric REEN  patterns), was taken to represent the REE content of an original silicate magma. Using appropriate partition coefficients (Yurimoto et al. 1980, Fleet and Pan 1997, Suk 1998, Marshall et al. 1998, and Veksler et al. 2005; Table 3.5) the fractional crystallisation of the melt was modeled using a slightly modified Rayleigh fractional crystallisation equation from Arth (1976).  (D-1) L iC C (F' )= ×       (1)  where CL is the concentration of the element in the differentiated liquid to be calculated, Ci is the concentration in the original melt, F’ is the fraction of the remaining liquid and D is the partition coefficient of the phase from the melt. The same method was used to model the separation of P- and F- rich fluids from the silicate melt. As only one phase is separated from the melt at a time in these calculations, the bulk distribution coefficient of Arth (1976) has been replaced by the partition coefficient of the individual phase from melt. It was assumed in the calculations that the system was closed and that the partition coefficients did not change. The straight REEN patterns show the least fractionation within the REE as a group and are similar in shape (although much depleted in concentration) to patterns displayed by a number of regional granitic and aplitic dikes (Rasmussen et al. 2007). However, other indicators of magmatic fractionation (e.g., K/Rb, Li/Cs) show that these samples are in fact highly evolved. This would occur if the minerals influencing the REE concentrations do not incorporate elements used to assess magmatic fractionation. Therefore mica, which does incorporate elements used to quantify magmatic 142 LNPG samples REE pattern type La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Starting composition used for calculations Straight average 0.6886 1.2786 0.1414 0.4814 0.0851 0.0180 0.0693 0.0111 0.0609 0.0110 0.0297 0.0049 0.0273 0.0039 P389797  Listric low 0.074 0.087 0.008 0.027 0.005 0.003 0.005 0.001 0.005 0.001 0.003 <lod 0.006 0.001 P389760 Listric high 0.386 0.375 0.047 0.150 0.027 0.007 0.027 0.005 0.032 0.006 0.018 <lod 0.035 0.005 P389870 Discontinuous low 0.244 0.375 0.035 0.098 0.031 0.002 0.031 0.007 0.041 0.005 0.014 <lod 0.022 0.003 P389789 Discontinuous high 0.810 1.570 0.165 0.560 0.114 0.019 0.094 0.018 0.091 0.017 0.041 0.007 0.042 0.006 Fleet and Pan 1997 KD 4.8 6.2 7 7 5.2 4.7 5.2 na 4 na 2.8 na 1.7 0.9 Separation of: Synthetic fluorapatite F=0.999 0.686 1.272 0.141 0.479 0.085 0.018 0.069 0.061 0.030 0.027 0.004 From: Silicate melt F=0.780 0.268 0.351 0.032 0.108 0.030 0.007 0.024 0.029 0.019 0.023 0.004 Suk 1998 KD 11.330 7.8 Separation of: Phosphate melt F=0.99 0.681 1.270 From: Silicate melt Veksler et al.  2005 KD 134 95.7 96.2 99.9 94.1 79.1 86.7 72 64.7 59.1 52.9 45.7 40.8 35.7 Separation of: Fluoride melt F=0.988 0.138 0.408 0.045 0.146 0.028 0.007 0.025 0.005 0.028 0.005 0.016 0.003 0.017 0.003 From: Silicate melt F=0.970 0.012 0.071 0.008 0.024 0.005 0.002 0.005 0.001 0.009 0.002 0.006 0.001 0.008 0.001 Marshall et al.  1998 KD 4.78 5.81 8.25 12.9 18.1 20.7 26 17.8 14.8 13.12 9.51 na 4.39 3.99 Separation of: Fluorite F=0.99 0.663 1.218 0.131 0.427 0.072 0.015 0.054 0.009 0.053 0.010 0.027 0.026 0.004 From: Peralkaline rhyolite F=0.95 0.567 0.999 0.098 0.261 0.035 0.007 0.019 0.005 0.030 0.006 0.019 0.023 0.003 Yurimoto et al. 1990 KD 3200 3413 3569 3726 2859 228 2144 1786 1429 920 595 395 273 174 Separation of: Monazite F=0.9998 0.363 0.646 0.069 0.229 0.048 0.017 0.045 0.008 0.046 0.009 0.026 0.005 0.026 0.004 From: Granite F=0.9992 0.053 0.083 0.008 0.024 0.009 0.015 0.012 0.003 0.019 0.005 0.018 0.004 0.022 0.003 Table 3.5. Measured LNPG concentrations, published REE partition coefficients of appropriate phases and the calculated REE concentrations after fractionation of the phases. The average REE composition of the samples with straight REEN patterns was used as the starting composition. 143 fractionation, is not thought to influence REE concentrations in the processes considered here. Of the P-rich minerals occurring in the LNPG only REE partition coefficients for apatite (Fleet and Pan 1997) and monazite (Yurimoto et al. 1990) in highly evolved silicate melt were found in the literature. The discontinuous REEN pattern was most successfully modeled using partition coefficients between synthetic fluorapatite (syn. FAp) and a silicate melt (experiment B; Fleet and Pan 1997; Fig. 3.27A) at levels of between F=~0.90 and ~0.78, i.e., fractionation crystallisation of between ~10–22%. The LREEN to MREEN fitted well with the rock compositions. The Eu anomaly is present, but is not as pronounced as in the rock samples. Several partition coefficients for HREE are absent and the inflections in the rocks match poorly, notwithstanding, the modeled abundance of the elements measured is commonly within the range of rock compositions. However, the significant amount of apatite fractionation required to replicate the discontinuous pattern, up to 22%, suggests this model may not be realistic. Using partition coefficients for La and Ce from Suk (1998), derived from the analysis of immiscible silicate and phosphate liquids, experimental values for the separation of 1% P-rich fluid, from 99% silicate melt, fell relatively close to the range observed for the discontinuous REEN pattern (Fig. 3.27A). It is difficult to assess the fit of the modeled immiscible P-rich melt/silicate fractionation to the actual rock compositions as partition coefficients for only two elements La and Ce were available, however, the closest approximation to the rock composition came at <<0.1% separation of a P-rich fluid. Modelling of the listric REEN patterns also used a starting composition of the average REE abundances of the straight REEN patterns. The DREE for P-rich mineral and fluid fractionation (used to replicate the discontinuous REEN patterns) were unsuccessful in replicating the REEN listric patterns. This was expected, as the P2O5 wt.% concentrations in the samples with listric REEN patterns are not depleted, suggesting that these samples had not experienced P fractionation (at least in the late stages observed in the LNPG dikes). Fractionation between immiscible fluoride and silicate melts was modeled very successfully (Veksler et al. 2005; Fig. 3.27B), although the modeled values for La did not fit well, there is no negative Eu anomaly and the enrichment (especially in 144 Starting composition (average value of the straight REE patterns)N .01 .1 1 La Ce Pr Nd PmSm Eu Gd Tb Dy Ho Er Tm Yb Lu F=0.90 F=0.78 Range of discontinuous REE patterns N D (REE) from Exp. B (syn.FAp and melt) Fleet and Pan (1997) D (REE) from immiscible silicate and phosphate melts Suk (1998) La Ce Pr Nd PmSm Eu Gd Tb Dy Ho Er Tm Yb Lu F=0.988 F=0.970 Range of listric REE patternsN D (REE) from immiscible silicate and fluoride melts Veksler (2005)et al. F=0.99 Figure 3.27 Results of modeling crystal fractionation and fluid/melt and fluid/fluid fractionation of LNPG REE abundances. A. Results of modeling crystal fractionation and fluid/melt fractionation using the average REE composition of the group of samples showing flat REE patterns and experimentally derived D (REE) from Fleet and Pan (1997) and Suk (1998)compared to the pattern of the discontinuous REE . B. Results of modeling fluid/melt fractionation using the average REE composition of the group of samples showing flat REE patterns and experimentally derived D (REE) from Veksler (2005) compared to the pattern of the listric REE . N N N N Net al. Starting composition (average value of the straight REE patterns)NA B 145 the lower amounts of fractionation) of the HREEN is missing. The model suggests 1.2 – 3% F-rich fluid separation could fractionate the REE to produce the listric patterns. Crystal fractionation of monazite and fluorite were less successfully modeled from the melt. Extremely low levels of monazite fractionation (Yurimoto et al. 1990; F = 0.9999 to 0.9996) gave LREEN results somewhat above the rock sample fields and the subtle inflection at Nd was missing (Fig. 3.28A).  Modelling of fluorite mineral precipitation was also unsuccessful (Marshall et al. 1998), although the model did fit the upper values of the discontinuous and listric REEN patterns fairly well (Fig. 3.28B). In summary, using the average REE values from samples with straight REEN patterns the modeling results do not rule out the crystal fractionation of fluorapatite to produce the discontinuous REEN patterns. But the modeled values were not able to fully replicate the rock compositions, and the amount of fluorapatite precipitation (10-22%) seems unrealistic when compared to apatite modes measured by Rietveld analysis (<0.8 wt.%). In contrast, the goodness of fit achieved in modeling the listric REEN patterns by the separation of small amounts of F-rich fluid from melt compositions that have straight REEN patterns, suggests that exsolution of a late-stage F-rich fluid may have occurred. The dikes at LNPG probably originated from the same magma, and while the straight patterns can be related to the listric REEN patterns by the separation of small volumes of a late-stage F-rich fluid, the magmatic history of the discontinuous REEN pattern is less straightforward. The formation of the discontinuous REEN patterns (often described as showing a ‘tetrad effect’ in the literature) has been assigned to the increasing influence or separation of a volatile-rich fluid (Irber 1999, Badanina et al. 2006), fluorite (Monecke et al. 2002) or apatite (Lui and Zhang 2005) precipitation. These patterns frequently occur in the residual melt and the mineral precipitates, or melt inclusions (Irber 1996, Lui and Zhang 2005, Badanina et al. 2006). In comparison to the other LNPG compositions, samples with discontinuous REEN patterns show a low and restricted range of geochemical differentiation, and are associated with the presence of spodumene in the dikes with typically higher Ca2O/P2O5 ratios. We suggest that the discontinuous and listic REEN patterns evolved from the same silicate melt (bearing a straight REEN pattern). The discontinuous REEN patterns resulted from REE fractionation (due to slightly elevated Ca concentration) either by crystal 146 .01 .1 1 10 .01 .1 La Ce Pr Nd PmSm Eu Gd Tb Dy Ho Er Tm Yb Lu 1 LNPG discontinuous REE patternsN D (REE) fluoride/ peralkaline rhyolite fractionation Marshall (1998)et al. D (REE) monazite/granite fractionation Yurimoto (1990)et al. Yurimoto (1990) F=0.9998 to F=0.9992 et al. Marshall et al. (1998) F=0.99 to F=0.95 LNPG listric REE patternsN D (REE) fluoride/ peralkaline rhyolite fractionation Marshall (1998)et al. D (REE) monazite/granite fractionation Yurimoto (1990)et al. A B Figure 3.28 Results of modeling LNPG REE abundances with crystal fractionation of fluoride and monazite. A. Comparison of results modeling D (REE) of fluoride from peralkaline rhyolite from Marshall (1998) and monazite from granite from Yurimoto (1990) (using the average REE composition of the group of samples showing flat REE patterns) to the discontinuous REE patterns, B. Comparison of the same results as illustrated in A. to the listric REE patterns. N N N N et al. et al. 147 fractionation or by the separation of small amounts of fluid (that probably includes P) at an earlier stage than the process that created the REEN listric pattern.  3.6 CONCLUSIONS The LNPG dikes originate from a single magmatic source with each dike essentially behaving as a closed system during propagation of the magma through to its eventual consolidation. The dikes show a wide range of magmatic fractionation up to levels comparable with highly evolved and mineralised granites and pegmatites from the Variscan orogeny in France and Germany (e.g., Raimbault et al. 1995 and Webster et al. 1997) the Khangilay Li-F granites of Transbaikalia, Russia (Badanina et al. 2006), and the Tin Mountain pegmatite in South Dakota (Walker et al. 1986). Geochemical evolution of the LNPG magma is associated with the formation of: • three distinctive REEN patterns and the • strong fractionation of element pairs Y/Ho and Zr/Hf The REEN patterns and the Y/Ho ratios of some of the most evolved samples (with straight REEN patterns) show no evidence of volatile saturation. Crystal and fluid fractionation of the REEN was modeled from the average composition of the samples with the straight REEN patterns using available crystal/melt and fluid/melt partition coefficients. Results from the modeling suggest the saturation of volatile-rich phases (P and F; crystal or melt) is responsible for the unusually strong fractionation of otherwise typically immobile groups or pairs of elements. Strong Zr/Hf fractionation (that can result from crystal fractionation or fluid exsolution) appears in all three groups, but only the samples with listric and discontinuous REEN patterns (i.e. the two groups which appear to show late-stage element fractionation) also show strong Y/Ho fractionation from typical magmatic values. This may indicate that element fractionation by fluid separation is the process occurring in both these groups. We suggest that extensive magmatic fractionation of the original magma occurred during dike propagation. The dikes behaved essentially as closed systems evolving flux- 148  149 or volatile-saturated and undersaturated conditions, that influenced the trace element composition of each dike.  3.7 ACKNOWLEDGEMENTS This project was made possible by NSERC Discovery Grants to Lee Groat and Dominique Weis, and funding from War Eagle Mining Company Inc.. Archer Cathro & Associates (1981) Ltd. is thanked for access to samples and geochemical data. From UBC: Wilma Pretorius, Bert Mueller and Vivian Lai from PCIGR, and Mati Raudsepp, Elisabetta Pani and Jenny Lai are thanked for their professional help and advice. Many thanks go to Catherine Chauvel and her team, Sarah Bureau and Christéle Poggi, at the Université Joseph Fourier in Grenoble, France for their considerate assistance. Thought processes were encouraged by discussions with James Scoates, Greg Dipple and Kelly Russell and they are thanked for sharing their time and knowledge.  3.8 REFERENCES ARTH, J.G. (1976): Behavior of trace elements during magmatic processes: a summary of theoretical models and their applications. Journal of Research, U.S. Geological Survey 4, 41-47.  AUDÉTAT, A. & PETTKE, T. (2003): The magmatic-hydrothermal evolution of two barren granites: A melt and fluid inclusion study of the Rito del Medio and Cañada Pinabete plutons in northern New Mexico (USA). Geochimica et Cosmochimica Acta 67, 97-121.  BADANINA, E.V., VEKSLER, I.V., THOMAS, R., SYRITSO, L.F. & TRUMBULL, R. B. (2004): Magmatic evolution of Li–F, rare-metal granites: a case study of melt inclusions in the Khangilay complex, Eastern Transbaikalia (Russia). Chemical Geology 210, 113-133. BADANINA, E.V., TRUMBULL, R.B., DULSKI, P., WIEDENBECK, M., VEKSLER, I.V. & SYRITSO, L.F. (2006): The behavior of rare-earth and lithophile trace elements in rare- metal granites: a study of fluorite, melt inclusions and host rocks from the Khangilay Complex, Transbaikalia, Russia. The Canadian Mineralogist 44, 667-692.  BARNES, V.E., EDWARDS, G., MCLAUGHLIN, W.A., FRIEDMAN, I. &  JOENSUU, O. (1970): Macusanite occurrence, age, and composition, Macusani, Peru. Geological Society of America Bulletin 81, 1539-1546.  BARRAT, J.A., KELLER, F., AMOSS, J., TAYLOR, R.N., NESBITT, K.W. & HIRATA, T. (1996): Determination of rare-earth elements in sixteen silicate references samples by ICP-MS after TM addition and ion exchange separation. Geostandards Newsletter 20, 133-139.  BAU, M. (1996): Controls on the fractionation of isovalent trace elements in magmatic and aqueous systems: evidence from Y/Ho, Zr/Hf, and lanthanide tetrad effect. Contributions to Mineralogy and Petrology 123, 323-333. BORG, L.E. & CLYNNE, M.A. (1998): The petrogenesis of felsic calc-alkaline magmas from the Southernmost Cascades, California: origin by partial melting of basaltic lower crust. Journal of Petrology 39,1197-1222.  BREAKS, F.W., SELWAY, J.B. & TINDLE, A.G. (2003): Fertile peraluminous granites and related rare element mineralization in pegmatites, Superior Province, northwest and northeast Ontario: Operation Treasure Hunt. Ontario Geological Survey, Open File Report 6099.  BREITER, K., FRYDA, J., SELTMANN, R. & THOMAS, R. (1997): Mineralogical evidence for two magmatic stages in the evolution of an extremely fractionated P-rich rare-metal granite: the Podlesí Stock, Krušné Hory, Czech Republic. Journal of Petrology 38, 1723- 1739.  150 BREITER, K., FÖRSTER, H.-J. & SELTMANN, R. (1999): Variscan silicic magmatism and related tin-tungsten mineralization in the Erzgebirge-Slavkovský les metallogenic province. Mineralium Deposita 34, 505-521.  CANDELA, P.A. (1997):  A review of shallow, ore-related granites: textures, volatiles, and ore metals. Journal of Petrology 38, 1619-1633.  ČERNÝ, P. (1991a): Rare-element granitic pegmatites, Part I: Anatomy and internal evolution. Geoscience Canada 18, 49-67.  ČERNÝ, P. (1991b): Rare-element granitic pegmatites, Part II: Regional to global environments and petrogenesis. Geoscience Canada 18, 68-80.  ČERNÝ, P. (2005): The Tanco rare-element pegmatite deposit, Manitoba: regional context, internal anatomy, and global comparisons. In Rare-element Geochemistry and Mineral Deposits (R.L. Linnen and I.M. Samson, eds.), Geological Association of Canada, Short Course Notes 17, 127-158.  ČERNÝ, P., MEINTZNER, R.E. & ANDERSON, A.J. (1985): Extreme fractionation in rare element granitic pegmatites: selected examples of data and mechanisms. The Canadian Mmineralogist 23, 381-421. ČERNÝ, P., CHAPMAN, R., FERREIRA, K. & SMEDS, S.-A. (2004): Geochemistry of oxide minerals of Nb, Ta, Sn, and Sb in the Varuträsk granitic pegmatite, Sweden: The case of an "anomalous" columbite-tantalite trend. American Mineralogist 89, 505-518.  CHACKOWSKY, L.E. (1987): Mineralogy, geochemistry and petrology of pegmatitic granites and pegmatites at Red Sucker Lake and Gods Lake, Northeastern Manitoba. M.Sc. thesis, University of Manitoba, Winnipeg, Manitoba.  CHRISTIANSEN, E.H., BIKUM, J.V., SHERIDAN, M.F. & BURT, D.M. (1984): Geochemical evolution of topaz rhyolites from the Thomas Range and Spor Mountain, Utah. American Mineralogist 69, 223-236.  COULSON I. M., VILLENEUVE, M. E., DIPPLE G. M., DUNCAN R. A., RUSSELL J. K., & MORTENSEN, J. K. (2002): Time-scales of assembly and thermal history of a composite felsic pluton: constraints from the Emerald Lake area, northern Canadian Cordillera, Yukon. Journal of Volcanology and Geothermal Research 114, 331-356.  DOLEJŠ, D. & BAKER, D.R. (2007): Liquidus equilibria in the system K2O-Na2O-Al2O3- SiO2-F2O-1 - H2O to 100MPa: II. Differentiation paths of fluorosilicic magmas in hydrous systems. Journal of Petrology 48, 807-828  DOSTAL, J. & CHATTERJEE, A. K. (1995): Origin of topaz-bearing and related peraluminous granites of the Late Devonian Davis Lake pluton, Nova Scotia, Canada: crystal versus fluid fractionation. Chemical Geology 123, 67-88.  151 DOSTAL, J., CHATTERJEE, A. K. & KONTAK, D. J. (2004): Chemical and isotopic (Pb, Sr) zonation in a peraluminous granite pluton: role of fluid fractionation. Contributions to Mineralogy and Petrology 147, 74-90.  DUNCAN, R.A. (1999); Physical and chemical zonation in the Emerald Lake pluton, Yukon Territory. M.Sc. thesis: Vancouver, University of British Columbia, Vancouver, British Columbia.  ERCIT, T.S., GROAT, L.A. & GAULT, R.A. (2003): Granitic pegmatites of the O'Grady Batholith, N.W.T., Canada: a case study of the evolution of the elbaite subtype of rare- element granitic pegmatite. The Canadian Mineralogist 41, 117-138.  FLEET, M.E. & PAN, Y. (1997): Rare earth elements in apatite: Uptake from H2O-bearing phosphate-fluoride melts and the role of volatile components. Geochimica et Cosmochimica Acta 61, 4745-4760.  FÖRSTER, H.-J. (1998): The chemical composition of REE-Y-Th-U-rich accessory minerals in peraluminous granites of the Erzgebirge-Fichtelgebirge region, Germany, Part I: The monazite-(Ce)-brabantite solid solution series. American Mineralogist 83, 259-272.  FÖRSTER, H.-J., TISCHENDORF, G., TRUMBULL, R.E. &  GOTTESMANN, B. (1999): Late collisional granites in the Variscan Erzgebirge, Germany. Journal of Petrology 40, 1613- 1645.  FRENCH, B.M., JEZEK, P.A. & APPLEMAN, D.E. (1978): Virgilite: a new lithium aluminum silicate mineral from the Macusani glass, Peru. American Mineralogist  63, 461-465.  FRIEDEL, G. (1907) : Etudes sur la loi de Bravais. Bulletin de la Societe Francaise Mineralogique 30, 326-455.  GIORDANO, D., ROMANO, C., DINGWELL, D.B., POE, B. & BEHRENS, H. (2004): The combined effects of water and fluorine on the viscosity of silicic magmas. Geochimica et Cosmochimica Acta 68, 5159-5168.  GÖD, R. (1989): The spodumene deposit at "Weinebene", Koralpe. Mineralium Deposita 24, 270-278.  GORDEY, S.P. & ANDERSON, R.G. (1993): Evolution of the northern Cordilleran miogeocline, Nahanni map area [105I], Yukon and Northwest Territories. Geological Survey of Canada Memoir 428.  GRÁNÁSY, L., PUSZTA, T., BÖRZSÖNY, T., WARREN, J. A. & DOUGLAS, J.F. (2004): A general mechanism of polycrystalline growth. Nature Materials 3, 645-650.  152 GROAT, L.A., ERCIT, T.S., RAUDSEPP, M. & MAUTHNER, M.H.F. (1994): Geology and mineralogy of the Little Nahanni Pegmatite Group, part of NTS area 105 I/02. Economic Geology Survey Open File 1994 14. DIAND, NWT.  GROAT, L.A., MULJA, T., MAUTHNER, M.H.F., ERCIT, T.S., RAUDSEPP, M., GAULT, R.A. & ROLLO, H.A. (2003): Geology and mineralogy of the Little Nahanni rare-element granitic pegmatites, Northwest Territories. The Canadian Mineralogist 41, 139-160.  HANNON, A.C., VESSAL B. & PARKER, J.M. (1992): The structure of alkali silicate glasses. Journal of non-crystalline solids 150, 97-102.  HEDENQUIST, J.W. & LOWENSTERN, J.B. (1994) : The role of magmas in the formation of hydrothermal ore deposits. Nature 370, 519-527.  HEFFERNAN, R. S. (2004): Temporal, Geochemical, Isotopic and Metallogenic Studies of Mid-cretaceous Magmatism in the Tintina Gold Province, Southeastern Yukon and Southwestern Northwest Territories, Canada. M.Sc. thesis: Vancouver, University of British Columbia, Vancouver, British Columbia.  HENDERSON, G.S. (2005): The structure of silicate melts: a glass perspective. The Canadian Mineralogist 43, 1921-1958.  IRBER W. (1999): The lanthanide tetrad effect and its correlation with K/Rb, Eu/Eu*, Sr/Eu, Y/Ho, and Zr/Hf of evolving peraluminous granite suites. Geochimica et Cosmochimica Acta 63, 489-508.  JAHNS, R.H. (1982): Internal evolution of pegmatite bodies. In Granitic Pegmatites in Science and Industry (P. Černý, ed.), Mineralogical Association of Canada Short Course Handbook 8, 293-327.  JOCHUM, K.P., SEUFERT, H.M., SPETTEL, B. & PALME, H. (1986): The solar-system abundances of Nb, Ta and Y, and the relative abundances of refractory lithophile elements in differentiated planetary bodies. Geochimica et Cosmochimica Acta 50, 1173- 1183.  JOLLIFF, B.L., PAPIKE, J.J., SHEARER, C.K. & SHIMIZU, N. (1989): Inter- and intra-crystal REE variations in apatite from the Bob Ingersoll pegmatite, Black Hills, South Dakota. Geochimica et Cosmochimica Acta 53, 429-441.  JOLLIFF, B. L., PAPIKE, J. J. & SHEARER, C.K. (1992): Petrogenetic relationships between pegmatite and granite based on geochemistry of muscovite in pegmatite wall zones, Black Hills, South Dakota, USA. Geochimica et Cosmochimica Acta 56, 1915-1939.  KONTAK, D.J. (2006): Nature and origin of an LCT-suite pegmatite with late-stage sodium enrichment, Brazil Lake, Yarmouth County, Nova Scotia. I. Geological setting and petrology. The Canadian Mineralogist 44, 563-598. 153  154 KONTAK, D.J., ANSDELL, K., DOSTAL, J., HALTER, W., MARTIN, R. & WILLIAMS-JONES, A.E. (2001): The nature and origin of pegmatites in a fluorine-rich leucogranite, East Kemptville Tin Deposit, Nova Scotia, Canada. Transactions of the Royal Society of Edinburgh: Earth Sciences 92, 173-200.  KONTAK, D.J., DOSTAL, J., KYSER, T.K. & ARCHIBALD, D.A. (2002): A petrological, geochemical, isotopic and fluid-inclusion study of 370 Ma pegmatite-aplite sheets, Peggys Cove, Nova Scotia, Canada. The Canadian Mineralogist 40, 1249-1286.  LANG, J. R. (ed.) (2001): Regional and system-scale controls on the formation of copper and/or gold magmatic-hydrothermal mineralization. Mineral Deposit Research Unit Spec. Publ. 2, 115 p.  LE MAITRE, R.W., BATEMAN, P., DUDEK, A., KELLER, J., LAMEYRE, J., LE BAS, M. J., SABINE, P.A., SCHMID, R., SORENSEN, H., STRECKEISEN, A., WOOLLEY, A.R. & ZANETTIN, B. (1989): A classification of igneous rocks and glossary of terms: Recommendations of the International Union of Geological Sciences Subcommission on the Systematics of Igneous Rocks. In Blackwell Scientific Publications, Oxford, UK.  LINNEN, R.L. (1998): The solubility of Nb-Ta-Zr-Hf-W in granitic melts with Li and Li + F: Constraints for mineralization in rare metal granites and pegmatites. Economic Geology 93, 1013-1025.  LINNEN, R.L. & KEPPLER, H. (2002): Melt composition control of Zr/Hf fractionation in magmatic processes. Geochimica et Cosmochimica Acta 66, 3293-3301.  LOFGREN, G. (1971): Spherulitic textures in glassy and crystalline rocks. Journal of Geophysical Research 76, 5635-5648.  LONDON, D. (1984): Experimental phase equilibria in the system LiAlSiO4 -SiO2 -H2O; a petrogenetic grid for lithium-rich pegmatites. American Mineralogist 69, 995-1004.  LONDON, D. (1986): Magmatic-hydrothermal transition in the Tanco rare-element pegmatite: Evidence from fluid inclusions and phase-equilibrium experiments. American Mineralogist 71, 376-395.  LONDON, D. (2005a): Geochemistry of alkali and alkaline earth elements in ore-forming granites, pegmatites and rhyolites. In Rare-element Geochemistry and Mineral Deposits (R.L. Linnen and I.M. Samson, eds.), Geological Association of Canada, Short Course Notes 17, 17-43.  LONDON, D. (2005b): Granitic pegmatites: an assessment of current concepts and directions for the future. Lithos 80, 281-303.  LONDON, D. (2008): Pegmatites. Mineralogical Association of Canada, Québec, Canada.  LONDON, D., HERVIG, R.L. & MORGAN VI, G.B. (1988): Melt-vapor solubilities and elemental partitioning in peraluminous granite-pegmatite systems: experimental results with Macusani glass at 200 Mpa. Contributions to Mineralogy and Petrology 99, 360- 373.  LONDON, D., MORGAN IV, G.B. & HERVIG, R.L. (1989): Vapor-undersaturated experiments with Macusani glass + H20 at 200 MPa, and the internal differentiation of granitic pegmatites. Contributions to Mineralogy and Petrology 102, 1-17.   LUI, C.-Q. & ZHANG, H. (2005): The lanthanide tetrad effect in apatite from the Altay No. 3 pegmatite, Xingjiang, China: an intrinsic feature of the pegmatite magma. Chemical Geology 214, 61-77.  MANNING, D.A.C. (1981): The effect of fluorine on liquidus phase relationships in the system Qz-Ab-Or with excess water at 1 kb. Contributions to Mineralogy and Petrology 76, 206-215.  MARSHALL, A.S., HINTON, R.W. & MACDONALD, R. (1998): Phenocrystic fluorite in peralkaline rhyolites, Olkaria, Kenya Rift Valley. Mineralogical Magazine 62, 477-486.  MAUTHNER, M.H.F. (1996): Mineralogy, geochemistry and geochronology of the Little Nahanni Pegmatite Group, Logan Mountains, southwestern Northwest Territories. M.Sc. thesis, University of British Columbia, Vancouver, British Columbia. MONECKE, T., KEMPE, U., MONECKE, J., SALA, M. & WOLF, D. (2002): Tetrad effect in rare earth element distribution patterns: a method of quantification with application to rock and mineral samples from granite-related rare metal deposits. Geochimica et Cosmochimica Acta 66, 1185-1196. MURPHY, D.C. (1997): Geology of the McQuesten River Region, Northern McQuesten and Mayo map areas, Yukon Territory (115/P14, 15, 16; 105M/13,14). Indian and Northern Affairs Canada, Exploration and Geological Services Division, Yukon Region, Bulletin. 6. MYSEN, B.O. (2004): Element partitioning between minerals and melt, melt composition, and melt structure. Chemical Geology 213, 1-16.  MYSEN, B.O., VIRGO, D. & SEIFERT, F.A. (1982): The structure of silicate melts: Implications for chemical and physical properties of natural magma. Reviews of Geophysics and Space Physics 20, 353-383.  NOBLE, D.C., VOGEL, T.A., PETERSON, P.S., LANDIS, G.P., GRANT, N.K., JEZEK, P.A. & MCKEE, E.H. (1984): Rare-element-enriched, S-type ash-flow tuffs containing phenocrysts of muscovite, andalusite, and sillimanite, southeastern Peru. Geology 12, 35- 39.  NORTON, J.J. (1983): Sequence of mineral assemblages in differentiated granitic pegmatites. Economic Geology 78, 854-874.  155 PAN, Y. (1997): Controls on the fractionation of isovalent trace elements in magmatic and aqueous systems: evidence from Y/Ho, Zr/Hf, and lanthanide tetrad effect - a discussion of the article by M. Bau (1996). Contributions to Mineralogy and Petrology 128, 405-408.  PAN, Y. & BREAKS, F.W. (1997): Rare earth elements in fluorapatites, Separation Lake Area, Ontario: Evidence of S-type granite – rare-element pegmatite linkage. The Canadian Mineralogist 35, 659-671.  PARTINGTON, G.A., MCNAUGHTON, J. & WILLIAMS, I.S. (1995): A review of the geology, mineralization and geochronology of the Greenbushes Pegmatite Western Australia. Economic Geology 90, 616-635.  PEARCE, J.A., HARRIS, N.B.W. & TINDLE, A.G. (1984): Trace element discrimination diagrams for the tectonic interpretation of granitic rocks. Journal of Petrology 25, 956- 983.  PEMBERTON, B.C. (2002): The origin of cassiterite at the Little Nahanni Pegmatite Group, Northwest Territories. B.Sc. thesis, University of British Columbia, Vancouver, British Columbia.  PICHAVANT, M., KONTAK, D.J., HERRERA, J.V. & CLARK, A.H. (1988a): The Miocene- Pliocene Macusani Volcanics, SE Peru I. Mineralogy and magmatic evolution of a two- mica aluminosilicate-bearing ignimbrite suite. Contributions to Mineralogy and Petrology 100, 300-324.  PICHAVANT, M., KONTAK, D.J., HERRERA, J.V. & CLARK, A.H. (1988b): The Miocene- Pliocene Macusani Volcanics, SE Peru II. Geochemistry and origin of a felsic peraluminous magma. Contributions to Mineralogy and Petrology 100, 325-338.  PRETORIUS, W., WEIS, D., WILLIAMS, G., HANANO, D., KIEFFER, B. & SCOATES, J. (2006): Complete trace elemental characterization of granitoids (USGS G-2, GSP-2) reference materials by high resolution inductively coupled plasma-mass spectrometry. Geostandards and Geoanalytical Research 30, 39-54.  PROWATKE, S. & KLEMME, S. (2006): Trace element partitioning between apatite and silicate melts. Geochimica et Cosmochimica Acta 70, 4513-4527.  RAIMBAULT, L., CUNEY, M., AZENCOTT, C., DUTHOU, J.L. & JOROY, J.L. (1995): Geochemical evidence for a multistage magmatic genesis of Ta-Sn-Li mineralization in the granite at Beauvoir, French Massif Central. Economic Geology 90, 548-576.  RASMUSSEN, K.L., MORTENSEN, J.K., FALCK, H. & ULLRICH, T.D. (2007): The potential for intrusion-related mineralization within the South Nahanni River MERA area, Selwyn and Mackenzie Mountains, Northwest Territories. In Mineral and Energy Resource Assessment of the Greater Nahanni Ecosystem Under Consideration for the Expansion of 156 the Nahanni National Park Reserve, Northwest Territories (D.F. Wright, D. Lemkow and J.R. Harris, eds.). Geological Survey of Canada Open File 5344, 203–278.  RAUDSEPP, M., PANI, E. & DIPPLE, G. M. (1999): Measuring mineral abundance in skarn. I. The Rietveld method using X-ray powder-diffraction data. The Canadian Mineralogist 37, 1-15.  ROLLO, H. A. (1999): Mineralogy and geochemistry of mica minerals from the Little Nahanni Pegmatite Group, Northwest Territories. B.Sc. thesis, University of British Columbia, Vancouver, British Columbia.  RUDNICK, R.L. & GAO, S. (2003): Composition of the continental crust. In Treatise on Geochemistry, 3, (K.K. Turekian, and H.D. Holland, eds.), Elsevier–Pergamon, Oxford, UK.  SIRBESCU, M.-L. & NABELEK, P.I. (2003): Crystallization conditions and evolution of magmatic fluids in the Harney Peak Granite and associated pegmatites, Black Hills, South Dakota—Evidence from fluid inclusions. Geochimica et Cosmochimica Acta 67, 2443-2465.  SOLTAY, L.G. & HENDERSON, G.S. (2005): The structure of lithium-containing silicate and germinate glasses. The Canadian Mineralogist 43, 1643-1651. STILLING, A., ČERNÝ, P. & VANSTONE, P.J. (2006): The Tanco pegmatite at Bernic Lake, Manitoba. XVI. Zonal and bulk compositions and their petrogenetic significance. The Canadian Mineralogist 44, 599-623.  SUK, N.I. (1998): Distribution of ore elements between immiscible liquids in silicate- phosphate systems (experimental investigation).  Acta Universitatis Carolinae, Geologica 42, 138-140.  SUN, S.-S. & MCDONOUGH, W.F. (1989): Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In Magmatism in the Ocean Basins (A.D. Saunders and M.J. Norry, eds.), Geological Society Special Publication 42, 313-345.  SWANSON, S.E. (1977): Relation of nucleation and crystal-growth rate to the development of granitic textures. American Mineralogist 62, 966-978.  TAYLOR, S.R. & MCLENNAN, S.M. (1985): The continental crust: Its composition and evolution. Blackwell Scientific Publications, Palo Alto, CA.  THOMAS, R. & WEBSTER, J.D. (2000): Strong tin enrichment in a pegmatite-forming melt. Mineralium Deposita 35, 570-582.  157 THOMAS, R., WEBSTER, J.D. & HEINRICH, W. (2000): Melt inclusions in pegmatite quartz: complete miscibility between silicate melts and hydrous fluids at low pressure. Contributions to Mineralogy and Petrology 139, 394-401.  THOMAS, R., FÖRSTER , H.-J., RICKERS , K. & WEBSTER , J.D. (2005): Formation of extremely F-rich hydrous melt fractions and hydrothermal fluids during differentiation of highly-evolved tin-granite magmas: a melt/fluid inclusion study. Contributions to Mineralogy and Petrology 148, 582-601.  TOMASCAK, P.B.,CARLSON, R.W. & SHIREY, S.B. (1999): Accurate and precise determination of Li isotopic compositions by multi-collector sector ICP-MS. Chemical Geology 158, 145-154.  VEKSLER, I.V. (2004): Liquid immiscibility and its role at the magmatic–hydrothermal transition: a summary of experimental studies. Chem. Geol. 210, 7-31.  VEKSLER I.V., DORFMAN, A.M., KAMENETSKY, M., DULSKI, P. & DINGWELL, D.B. (2005): Partitioning of lanthanides and Y between immiscible silicate and fluoride melts, fluorite and cryolite and the origin of the lanthanide tetrad effect in igneous rocks. Geochimica et Cosmochimica Acta 69, 2847-2860.  WALKER, R.J., HANSON, G.N., PAPIKE, J.J., O’NEIL, J.R. & LAUL, J.C. (1986): Internal evolution of the Tin Mountain pegmatite, Black Hills, South Dakota. American Mineralogist 71, 440-459.  WEBBER, K.L., SIMMONS, W.B., FALSTER, A.U. & FOORD, E.E. (1999): Cooling rates and crystallisation dynamics of shallow level pegmatite-aplite dikes, San Diego County, California. American Mineralogist 84, 708-717.  WEBSTER, J.D., THOMAS, R., RHEDE, D., FÖRSTER, H.-J. & SELTMANN, R. (1997): Melt inclusions in quartz from an evolved peraluminous pegmatite: Geochemical evidence for strong tin enrichment in fluorine-rich and phosphorus-rich residual liquids. Geochimica et Cosmochimica Acta 61, 2589-2604.  WENGZYNOWSKI, W.A. (2002): Geological report on the Mac property. War Eagle Mining Company Inc. WHITAKER, E.J.W. & MUNTUS, R. (1970): Ionic radii for use in geochemistry. Geochimica et Cosmochimica Acta 34, 945-956. YURIMOTO, H., DUKE, E.F., PAPIKE, J.J. & SHEARER, C.K. (1990): Are discontinuous chondrite-normalized REE patterns in pegmatitic granite systems the results of monazite fractionation? Geochimica et Cosmochimica Acta 54, 2141-2145.    158  CHAPTER 4: STRONG LI ISOTOPE FRACTIONATION IN THE HIGHLY EVOLVED LITTLE NAHANNI PEGMATITE GROUP, NWT* 4.1 INTRODUCTION Lithium isotope analysis is being increasingly used in the Earth Sciences especially in the study of processes where solid and fluid phase interaction can produce strong Li isotopic fractionation, such as hydrothermal, alteration and dehydration processes (e.g., Chan et al. 1992 and Chan and Kastner 2000, Tomascak et al. 2000, and Coogan et al. 2005) and magmatic fractionation (e.g., Foustoukos et al. 2004, Teng et al. 2006, Rudnick and Ionov 2007). In these natural settings the large relative mass difference between the two stable isotopes of Li (6Li and 7Li; ~17%) can produce isotopic fractionation of up to δ7Li 60‰ (Hoefs 1997). Equilibrium and kinetic processes are both significant in Li isotope fractionation due to the the large relative mass difference. In crystallising silicate melts the preferred tetrahedral bonding of Li with O (Wenger and Armbruster 1991) tends to make elemental Li somewhat incompatible, and it builds up in the residual melt. At chemical equilibrium if Li becomes saturated in the melt, the mass difference influences the location of the isotopes (Chacko et al. 2001) during the crystallization of Li- minerals. As a result the ‘lighter’ isotope (6Li) will be preferentially incorporated in more highly co-ordinated (commonly octahedral) mineral structures (Wunder et al. 2007).  Therefore the increasingly Li-rich residual melt is expected to become more 7Li-rich during magmatic fractionation (e.g., Teng et al. 2006 and references therein). Recent research into the kinetic fractionation of Li isotopes has identified an exceptionally high diffusion rate of elemental Li and a 5% increase in the diffusion rate of 6Li over 7Li (Coogan et al. 2005, Richter et al. 2003 and 2009). A combination of these processes can have a significant influence on Li isotopic signatures if retained, for example, by rapid crystallization (Maloney et al. 2008).  * A version of this chapter will be submitted for publication. Barnes, E.M., Weis, D. & Groat, L.A. (2010): Strong Li isotope fractionation in the highly evolved Little Nahanni Pegmatite Group, NWT. 159  Pegmatites commonly have an overall granitic composition (Černý 1991a), and a high flux component1 (especially H2O, F, B and P) that increases with magmatic fractionation. The flux components (in addition to elements such as Li, Na and K) disrupt the short-range structure of the silicate melt (consisting of tetrahedral groups, Si and Al with O) by increasing the number of non-bridging oxygen atoms (NBO; London 2005b, Henderson 2005). This has the effect of reducing melt viscosity (Thomas and Webster (2000) and increasing diffusivity (Dingwell et al. 1996). In addition, the liquidus and solidus temperatures of the silicate melt are greatly reduced enabling the melt to remain liquid to as low as ~400ºC (Sirbescu and Nabelek 2003).  Flux components are also integral to the promotion of extremely rapid crystallization and the formation of typical pegmatite characteristics including the growth of a wide range of crystal sizes (even up to m’s across). The processes that form pegmatites can also concentrate rare elements to levels several orders of magnitude above typical granite compositions e.g., Li, Cs, Bi, Ta, Nb, Sn and W, all of which are becoming increasingly useful to industry and modern technology (e.g., Černý 1991a, Raimbault et al. 1995). Lithium isotopic data on pegmatites is somewhat limited, however, Tomascak et al. (1995) identified fractionated isotopic signatures from mineral samples (albite, muscovite and quartz) collected from various compositional zones of the Tin Mountain pegmatite, South Dakota. The preference of 7Li for the mobile (melt) phase during magmatic fractionation is reflected in the increased δ7Li value recorded in the bulk (whole rock; and wall zone) samples from the same pegmatite (Teng et al. 2006). Gordienko et al. (2007) associated increased δ7Li signatures of pegmatitic minerals from high- and low-F intrusions in Russia with magma evolution, suggesting that F accumulation in evolved melts is a mechanism that promotes Li isotope fractionation. Maloney et al. (2008) suggested that kinetic isotopic fractionation could result in slower- diffusing 7Li becoming preferentially incorporated in rapidly crystallizing tourmaline.  1‘Flux’ is used as defined in London (2005b; and references therein) as components that ‘lower the melting and crystallization temperatures…and enhance miscibility among otherwise less soluble constituents’. 160  We focus on the Li isotopic signature of the rare element2 Little Nahanni Pegmatite Group. This swarm of highly evolved, Li-rich pegmatitic dikes displays evidence of increasing magmatic fractionation and fluctuating non-equilibrium crystallization conditions induced by rapid dike propagation and consolidation (see Chapter 2). Lithium isotope measurements were undertaken at the Pacific Centre for Isotopic and Geochemical Research (PCIGR) at the University of British Columbia, and at the University of Maryland, MD (US) of whole rock pegmatite samples, mineral separates and whole rock samples from regional granitic intrusions. The Li isotopic values are correlated with trace element geochemistry, mineralogy and observations of the primary textures from the pegmatites to assess the degree and the mechanisms of Li isotopic fractionation during pegmatite formation. We relate the very broad range of δ7Li from the whole rock LNPG samples (–0.94‰ to +11.36‰) with the consolidation of the final ~15% melt fraction of this peraluminous magma and associate it with F build up in the late stages of magmatic fractionation. Rock-forming minerals (quartz, albite, spodumene and mica) display δ7Li values that corroborate consolidation of the dikes under non-equilibrium conditions. We propose that Li isotopic signatures of highly evolved peraluminous magmas reflect the build up of F and can provide a qualitative assessment of the state of mineral/melt chemical equilibrium. 4.2 GEOLOGY The Late Cretaceous rare element pegmatitic dikes of the LNPG are located in the Mackenzie Mountains, Northwest Territories (505000E 6895000N; NTS 105 I/2; Fig. 4.1). In a regional study, Gordey and Anderson (1993) describe the host rocks, Proterozoic Yusezu and Narchilla formations of the Hyland Group, as probably upper fan facies deposited on the eastward migrating margin of the Selwyn Basin. These basinal deposits comprise coarse- to fine-grained siliciclastic beds that vary between millimetres to metres in thickness with variable amounts of carbonate matrix. A prolonged period of compression during the Mesozoic period formed a regional northwest-southeast tectonic fabric including the northeast verging asymmetric Fork anticlinorium of the study area  2 The term ‘rare element’ as used here for the LNPG is an accepted pegmatite classification denoting an increased abundance of rare elements such as Li, Rb, Cs, Be, Sn, Nb and Ta (Černý 1991). It should be not confused with ‘rare earth elements’ (REEs) that are highly depleted in the LNPG samples. 161 Figure 4.1 Geologic map of the study area showing areas with the highest density of pegmatites, and numbered locations for the whole rock samples. The elevation of the sample sites is shown to the right of the map. Modified from Groat et al. (2003). Sample numbers on the diagram refer to samples referred to in the text as follows: 1. = P389913 , 2. = P389910, 3. = P389739, 4. = P389875, 5. = P389854, 6. = P389851, 7. = P389726, 8. = P389861, 9. = P389849, 10. = P389870, 11. = P389734, 12. = P389797, 13. = P389760, 14. = P389764, 15. = P389768, 16. = P389774, 17. = P389751, 18. = P389753, 19. = P389711, 20. = P389704, 21. = P389789, 22. = P389791, 23. = P389719, 24. = P389844, 25. = P389817, 26. = P389816, 27. = P389808, 28. = P389809, a. = AL050704, b. = P389859, c. = EB1304, d. EB0509. 162  (see Chapter 2). The dikes average 1-2 m in width (up to 10 m; Mauthner 1996), and are traceable across the high relief topography for several 100’s m along strike. Sharp dike contacts are typical, although decimetre-wide aureoles of silver mica or schorl (Groat et al. 1994) occur within the pelitic host rock, and within the carbonate-rich host rock the dikes can be cut by calcite veins (Mauthner 1996), indicating some fluid interaction between the dikes and the country rock. The mineralogy of the dikes includes: quartz, K-feldspar, albite, spodumene, various micas (lepidolite, muscovite and minor cookeite), columbite-group minerals, cassiterite, tourmaline, beryl, lithiophilite, garnet (Mauthner 1996; Groat et al. 2003) and berlinite. Few mineralogical zonations are observed across the LNPG, and where present they are relatively diffuse. Zonations include spodumene-rich dikes exposed at slightly greater depths than slightly Ta and Sn enriched spodumene-poor dikes that are more common in the north and south of the study area (Wengzynowski 2002), and at one well- exposed location the quartz content of the dikes increases with elevation. The lack of cross-cutting relationships between the dikes suggests they were emplaced during a single event (~90 Ma; apatite, U-Pb, see Chapter 2), that correlates with the waning of regional magmatism (Rasmussen et al. 2006 and 2007). Nearby intrusions (within ~50 km) tend to be small volume S-type granitic plutons (~15 km2 exposure) associated with pegmatitic and aplitic units and W-Cu-Zn-Mo exoskarn and veining. The hornblende bearing, alkali-feldspar-rich composite O’Grady batholith (~270 km2; Gordey and Anderson 1993), approximately 95 km to the north, is a rare example of the elbaite subtype of granitic rare element pegmatite (Ercit et al. 2003) and the only other example of Li mineralization in the area. The spatial distribution of distinctive primary textures (e.g., aligned mineral growth, pegmatitic and aplitic zones and mineral banding consolidated sub-parallel to country rock contacts) vary over relatively short distances within the dikes (decimetres) and between nearby individual dikes. Small, rare miarolitic cavities provide evidence of magmatic fluid exsolution (Candela 1997) and occur either as planar areas of euhedral crystal terminations (commonly quartz) approximately 15 ×  15 ×  0.5 cm  (~110 cm3) between pegmatite-aplite bands or rounded cavities approximately 5 cm diameter (Mauthner 1996) located towards the centre of the dikes. 163  4.3 SAMPLES AND ANALYTICAL METHODS The Li isotopic value was measured on twenty-eight whole rock samples and several mineral separates from LNPG dikes, regional granitic intrusions coeval with the LNPG, and examples of peraluminous natural volcanic glass. The same samples were geochemically analysed and reported in Chapter 3. Twenty-three of the whole rock samples were collected from individual dikes from across the LNPG as part of an exploration program (>200 samples; Wenzynowski 2002: Appendix 4) and were received for this study only as rock powders. The rocks had been grouped in the field by the abundance of the main rock-forming minerals, into the groups: • SQF (spodumene-quartz-feldspar), • SQFL (spodumene-quartz-feldspar-lepidolite), • QA (quartz-albite), QFL (quartz-feldspar-lepidolite) • and QM (quartz and silver mica). Hand samples from 5 dikes (for which we already had powdered samples; P389764, P389768, P389774, P389797, P389870) were collected by the author from areas proximal or adjacent to the original sampling, and were used for petrographic and further trace element analysis to facilitate correlation and interpretation of the data. Because many of the samples were only available as powders the modal mineralogy of a number of samples was determined using the Rietveld method and X-ray powder-diffraction data (Raudsepp et al. 1999; Table 4.1). The following section is an outline of the geochemical results (reported in full in Chapter 3), which pertain to this chapter. The samples are granitic (Le Maitre et al. 1989) and strongly peraluminous (A/CNK = 1.16 – 1.88) with normative corundum ranging between 1.5 and 9.1 %. Lithium, Rb, Cs, Nb, Ta, Sn and W all show moderate to extreme enrichment in the whole rock compositions compared to upper crustal values (Rudnick and Gao 2003). In contrast, several high-field strength elements (e.g., Zr, Hf, Th) are strongly depleted, and the REE have a low total abundance of 0.25 to 7.69 ppm. The chondrite-normalized REE (REEN) patterns of the whole rock LNPG samples display three distinctive groups (Fig. 4.2); (1) straight patterns with minor Eu anomalies, 164 P 389704 P 389711 P 389719 P 389726 P 389734 P 389764 P 389768 P 389774 δ7Li -0.73 0.12 -0.75 4.35 1.08 2.12 2.5 1.96 Mineral Ideal Formula Quartz SiO2 29.4 32.7 32.5 33.5 38.3 28.5 30.3 29 Plagioclase NaAlSi3O8 – CaAl2Si2O8 30.3 36.1 38.5 38.8 23.4 31.1 36.1 40.6 Muscovite KAl2AlSi3O10(OH)2 5.3 13.8 7.5 23 19.4 16.6 20.3 22.5 Spodumene LiAlSi2O6 19.7 13.3 5.7 1.2 0.8 21.2 10.4 1.8 K-feldspar KAlSi3O8 11.6 2.5 15.1 3 3.3 2.3 2.3 5.5 Lepidolite K(Li,Al)3(Si,Al)4O10(F,OH)2 12.9 Beryl Be3Al2Si6O18 0.3 0.4 0.5 0.8 0.4 0.6 0.5 Apatite Ca5(PO4)3(OH) 0.3 Cassiterite SnO2 1 Cookeite LiAl4(Si3Al)O10(OH)8 2 1.3 0.4 Clinoptilolite (Ca0.5,Na,K)6[Al6Si30O72]·~20H2O 1.2 Calcite CaCO3 Table 4.1. Table of modal mineral abundance as determined from several LNPG whole rock samples correlated with their δ7Li values and predominant mineralogy. 165 P 389797 P 389816 P 389870 P 389870 duplicate mean std %RSD P 389910 P 389913 δ7Li 1.26 8.01 2.87 4.3 0.78 Mineral Quartz 42.7 33 33.8 32.5 33.15 0.9 2.8 25.4 31.7 Plagioclase 23.3 42.4 28.7 26.7 27.7 1.4 5.1 38.1 25.8 Muscovite 3.3 15.8 11.3 10.4 10.85 0.6 5.9 9.9 4.3 Spodumene 12.4 14.2 18.4 16.3 3.0 18.2 6.8 22 K-feldspar 18.2 11.7 11.7 11.7 0.0 0.0 14.6 11.6 Lepidolite 7 2.9 3.3 Beryl 1.3 0.4 0.3 0.35 0.1 20.2 0.3 Apatite 0.4 0.8 Cassiterite Cookeite 0.8 1.3 Clinoptilolite Calcite 0.4 Table 4.1. Table of modal mineral abundance as determined from several LNPG whole rock samples correlated with their δ7Li values and predominant mineralogy contd. 166 .01 10 .1 1 Sa m pl e/ ch on dr ite La Ce Pr Nd PmSm Eu Gd Tb Dy Ho Er Tm Yb Lu La Ce Pr Nd PmSm Eu Gd Tb Dy Ho Er Tm Yb LuLa Ce Pr Nd PmSm Eu Gd Tb Dy Ho Er Tm Yb Lu LuLa Ce Pr Nd PmSm Eu Gd Tb Dy Ho Er Tm Yb.01 10 .1 1 Sa m pl e/ ch on dr ite ‘straight’ ‘discontinuous’ ‘listric’ ‘indeterminate’ ‘listric’ ‘discontinuous’ Figure 4.2 Chondrite normalized REE patterns of the LNPG whole rock samples illustrating three groups of distinctive patterns; straight, discontinuous and listric, and examples with indeterminate patterns. ‘straight’ 167  (2) ‘discontinuous’ patterns with a break in slope to shallower or positive trends at Nd (recognized elsewhere in P-rich intrusions e.g., Breiter et al. 1997 and Kontak et al. 2001), strongly negative Eu anomalies and upward-convex trends between Gd and Ho (3) curved concave-up, ‘listric’ patterns. Three of the samples are LREEN enriched but have less distinctive, ‘indeterminate’ patterns with variable Eu anomalies. Elemental ratios indicate strong magmatic fractionation with K/Rb=4-24; K/Cs=58-876; Sr/Rb=0.005-0.127; Nb/Ta=0.6-3.2 (with one outlier at 22.6); Y/Ho=25.6- 67.7; and Zr/Hf=5.9-15.6. Generally the samples show variable but high P2O5 and F levels (P2O5=0.20-1.10, F=0.016-0.994 wt. %) that decrease with increased SiO2; B was not measured. Mineral separates of spodumene, quartz, mica and feldspar were manually picked from hand samples (spodumene: AL050704 and P389859; quartz, mica and feldspar: EB0509 and EB1304). The five whole rock samples from regional Cretaceous granitic intrusions (Gordey and Anderson 1993, Rasmussen et al. 2006) and two natural felsic volcanic glass samples (Macusani, Peru; Pichavant et al. 1988a and b) were also measured. One sample of Macusani obsidian used in this study was a stream cobble (see Chapter 3) obtained from the Smithsonian Institution and is similar to sample JV1 described by Pichavant et al. (1988a). A separate pre-ground sample of Macusani glass (Tomascak et al. 1999) was also analysed for this study. United States Geological Survey (USGS) reference materials, G-2 and RGM-1 and international Li standard IRMM-015 (L-SVEC), the University of Maryland (UM) in-house standard UMD-1 and a Li carbonate compound (99.998%) manufactured by Puratronic® were also analysed.  4.3.1 Sample preparation The whole rock LNPG and regional granitic samples were ground by ALS Chemex, Vancouver. Sample preparation at UBC commenced with coning and quartering the powdered samples to ensure homogeneity and each were reground to < 200 µm by hand in an agate pestle and mortar. The Macusani glass cobble was sonicated in Milli-Q® 168  Element 18.2 MΩ.cm resistivity ultrapure water (Milli-Q 18Ω water) for 30 minutes to remove particulate matter, before sample material was chipped off and ground in the agate pestle. The mineral separates were sonicated for 10 minutes in Milli-Q 18Ω water and dried down before being ground. All samples analysed at the University of British Columbia (except the Macusani glass) were initially digested in high-pressure, individual PTFE capsules using the method described in Pretorius et al. (2006). All acids used in contact with sample material (and the methanol (MeOH) used in the column chemistry) were sub-boiled and only Milli-Q® Element 18.2 MΩ.cm resistivity ultrapure water was used. Digestion of 0.05-0.1 g of sample material required 6.4 mL of concentrated HF/ 14M HNO3/ HClO4 (8:1:1), at 190° C for 5 days. The samples were subsequently dried on a hotplate at 180° C, before 5 mL 6M sub-boiled HCl was added and the samples returned to the oven for 24 hours at 190° C. On cooling, if the sample was not completely digested the procedure from the addition of the acid mix was repeated after drying down, typically digestion was successful on the first attempt. Fully digested samples were transferred to 15 mL Savillex® beakers and dried on a hotplate at ~120 °C. To remove all traces of HCl prior to analysis, 1 mL of concentrated HNO3 was added and the vials were sealed and again placed on the hotplate at ~120 °C for 24 hours. Sample solutions were subsequently dried in preparation for dilution and analysis. Samples sent for isotopic analysis to the University of Maryland were digested and prepared using ion exchange chemistry to separate Li (see Section 4.3.2) at PCIGR, UBC.  Digestion followed standard hotplate dissolution procedures in 15 mL screw-top Savillex® vials with 5 mL concentrated HF and 0.5 mL 14M HNO3 (Weis et al. 2006). The covered vials were sonicated for 5 minutes and kept at 120°C for 2 days on a hotplate, after which the lids were removed and the contents dried down. When dry, 5 mL of 6M HCl was added, the vials were re-covered, sonicated for 5 minutes and then kept at 120°C for a further 2 days. Prior to the HCl being finally dried down, the samples were transferred to Savillex® vials and aliquots drawn off for trace element analyses. Although the standard hotplate method is more typically used for mafic or volcanic rocks, these digestions produced clear liquids with no visible residue due to the 169  typically low abundance of refractory minerals in the pegmatites. Only two samples (P389734 and P389844) retained several miniscule black grains, probably cassiterite given the high Sn concentrations in these samples (12351 and 13882 ppm respectively; see Chapter 3).  4.3.2 Column chemistry The column chemistry followed the two-column method of Jeffcoate et al. (2004). After digestion, samples (containing up to 1400 μg of Li), were diluted so that each 5 mL sample aliquot delivered 100-375 ng Li onto the first column. Given the average rock composition of the pegmatites, the calculated exchange capacity of the columns used was at least 6× that required to effectively separate the cations loaded onto the columns. Conditioning of the resin prior to the addition of the sample to the column, used 6M HCl (quartz-distilled). The first column procedure used a 20 mL plastic Econo column (Bio-Rad®) with 6 mL cationic resin (50W×  8, 200-400 mesh, Bio-Rad®) and a 1M HNO3: 80% v/v MeOH mix to separate Li from most other cations. This is a 5-hour procedure and the flow rates of the various eluents are as follows: conditioning 6M HCl ~86 mL/hr, H2O ~52 mL/hr and acid: MeOH mix, 60 mL/hr. The second column separates Li from Na (the major remaining impurity; see Jeffcoate et al. 2004). It uses a custom-made, shrink- fit Teflon column (2 mL reservoir) with 0.5 mL cationic resin (50W ×  12, 200-400 mesh, Bio-Rad®) and 0.5M HNO3: 80% v/v MeOH mix and a 1M HNO3:80% v/v MeOH mix as eluents. This is an 11-hour procedure and the flow rates are all between 5-6 mL/hr. The required eluent volumes per sample and information on eluent preparation are outlined in Table 4.2. To ensure the method was delivering  ~100% yield of Li the eluent fractions collected before and after the Li fraction from the second column were analyzed on a high-resolution single collector ICP-MS (HR-ICP-MS: Finnigan Element 2) or by graphite-furnace atomic absorption spectrometry (Varian Zeeman Graphite Tube Atomizer and SpectrAA-300 Atomic Absorption Spectrometer; GFAAS). For measurement by GFAAS (detection limit ~0.1 ppb Li) the column cuts from before and after the Li fraction from the second column were dried and brought up in 1 mL 170 Column 1 Eluent compositions and volume per sample: Eluent 1 122 mL of 6M HCl Eluent 2 102 mL of 1M HNO3:80% v/v methanol (see below) Column 2 Eluent compositions and volume per sample: Eluent 1 24 mL of 6M HCl Eluent 2 12.75 mL of 1M HCl:80% v/v methanol (see below) Eluent 3 7.25 mL of 0.5M HCl:80% v/v methanol (see below) Eluent formulae 1M HNO3:80% v/v methanol 1L requires: 800 mL sub-boiled methanol + 200 mL x 5M HNO3 (200 mL x 5M HNO3 = 71.4 mL 14M HNO3 and 128.6 mL H2O) 1M HCl:80% v/v methanol 200 mL requires: 160 mL sub-boiled methanol + 40 mL x 5M HCl (40 mL x 5M HCl = 33.3 mL 6M HCl and  6.7mL H2O) 0.5M HCl:80% v/v methanol 200 mL requires: 160 mL sub-boiled methanol + 40 mL x 2.5M HCl (40 mL x 2.5M HCl = 16.7 mL 6M HCl and 23.3 mL H2O) Table 4.2. List of eluent compositions for cationic exchange columns 1 and 2 for the separation of Li, and per sample volumes. 171  1%HNO3. A measurement of 1 ppb Li in these aliquots would correspond to a 1% Li loss in processing a sample with an initial Li content of 100 ng. Measurements of these aliquots by GFAAS were consistently <1ppb Li and rarely achieved the detection limit of 0.1 ppb Li. The elemental composition of the reagents and eluents was also measured  (Table 4.3). Distilled methanol (MeOH) was determined to be the most significant cation contributor to the chemistry (especially Na and Mg) but these contaminants are removed from the eluate by the column chemistry.  4.3.3 Instrumental set-up Simultaneous measurement of 6Li and 7Li was undertaken on the Nu 021 Plasma MC-ICP-MS at UBC (Nu Instruments, UK) using a DSN-100 membrane desolvator (Nu Instruments, UK) to introduce the sample material to the system. Typical instrumental settings for Li (Table 4.4) differ significantly from those used for other commonly measured isotopes (e.g. Pb) and to avoid signal instability from frequent setting changes, Li analyses were undertaken in several periods of 3-5 days when the MC-ICP-MS could be dedicated solely to Li isotope analysis. The instrument was tuned daily during the analytical period. Once a stable signal of around 6V (at 30-40 ppb Li) was achieved the instrumental settings were not modified during the day to maintain a reproducible mass bias, except minor adjustments to the gas membrane to maintain a steady gas flow.  4.3.4 Analyses 4.3.4.1 Notation Analytical results are normalised to the international standard L-SVEC and Li isotopic ratios are reported using the accepted Li del notation (δ7Li) suggested by Coplen et al. (1996).  { }7 6 sample7 6 L-SVEC Li/ Li7 Li/ Li Li 1 1000δ = − ×                    (1) 172 Table 4.3. Compositions of eluents and their constituents measured by HR-ICP-MS. Isotope Average ppb Stdev ppb Average ppb Stdev ppb Average ppb Stdev ppb Average ppb Stdev ppb Average ppb Stdev ppb Average ppb Stdev ppb 7Li (LR) 0.016 0.001 0.014 0.002 0.018 0.005 0.02 0.003 0.011 0.002 0.011 0.003 9Be (LR) 0.001 0.001 0.002 0 0.001 0 0.001 0 0.001 0.001 0.001 0.001 23Na (MR) 4.794 0.144 1.824 0.14 2.351 0.14 24.873 0.256 4.761 0.046 2.66 0.078 24Mg (MR) 0.664 0.039 1.382 0.074 0.388 0.078 32.862 0.117 5.987 0.126 3.268 0.229 27Al (LR) 2.851 0.079 1.429 0.083 1.052 0.036 4.276 0.067 8.839 0.275 1.774 0.024 44Ca (MR) 3.96 0.883 2.295 0.681 1.317 0.939 7.667 0.444 7.43 0.386 3.503 1.095 47Ti (MR) 0.004 0.006 0.021 0.017 n.d. 0.073 0.002 0.507 0.017 0.01 0.008 51V (MR) 0.063 0.002 0.097 0.002 0.022 0 0.042 0.002 0.059 0.003 0.025 0.001 52Cr (MR) 0.274 0.004 0.117 0 0.092 0.002 0.315 0.01 0.199 0.003 0.138 0.003 55Mn (MR) 0.097 0 0.029 0 0.028 0.001 0.098 0.002 0.116 0.003 0.05 0.002 56Fe (MR) 0.823 0.005 0.66 0.01 0.317 0.01 2.962 0.104 6.85 0.123 0.899 0.011 59Co (MR) 0.024 0 0.037 0.001 0.025 0.001 0.035 0.001 0.123 0.004 0.03 0 60Ni (MR) 0.189 0.003 0.15 0.005 0.127 0.004 0.44 0.018 0.226 0.006 0.492 0.01 63Cu (MR) 0.132 0.007 0.065 0.003 0.055 0.002 0.215 0.009 0.118 0.003 0.241 0.005 66Zn (MR) 0.375 0.031 0.199 0.009 0.368 0.038 2.119 0.023 0.358 0.007 0.286 0.025 85Rb (MR) 0.012 0.002 0.013 0.001 0.014 0.001 0.02 0 0.018 0.001 0.012 0 88Sr (LR) n.d. n.d. n.d. n.d. 0.017 0.001 n.d. 95Mo (LR) n.d. 0.005 0.001 n.d. 0.025 0.002 n.d. 0.003 0 111Cd (LR) 0.014 0 0.014 0 0.014 0 0.023 0 0.013 0 0.013 0 114Cd (LR) 0.023 0.001 0.023 0 0.023 0 0.031 0.002 0.023 0.001 0.023 0.001 118Sn (LR) 0.041 0.002 0.045 0.001 0.023 0 0.048 0.003 0.083 0.007 0.04 0.001 118Sn (MR) 0.022 0.003 0.023 0.003 0.004 0.003 0.03 0.004 0.068 0.003 0.022 0.002 138Ba (LR) n.d. n.d. n.d. 0.041 0.001 0.107 0.002 0.582 0.002 208Pb (LR) n.d. n.d. n.d. 0.028 0.001 n.d. n.d. n.d. not detectable LR=measured on low resolution MR=measured on medium resolution  HNO3 HCl 1 mol l-1 HCl:80% v/v MeOH 1 mol l-1 HNO3:80% v/v MeOH Milli-Q 18Ω H2O  MeOH 173 Nu 021 Plasma MC-ICP-MS Typical settings DSN Ar gas flow L/min 2.95 Deflectors HV1 3989 HV2 3065 HV3 2870 HV4 na HV5 2010 HV6 1180 SV1 75.5 SH1 48 SH2 6.6 TH1 8.5 TV1 168 TV2 192 Torch position B/F 2.65 I/O 5.15 U/D 5.41 Lenses Q1 -24.5 Q2 30.5 Lin1 261.1 Lin2 260.9 Table 4.4. Typical settings achieved after tuning the Nu 021 Plasma MC-ICP-MS in preparation for Li isotope analysis. 174  4.3.4.2 Method protocol A rapid analytical protocol is necessary to mitigate the effect of instrumental mass bias drift. A sequence of blank, standard, sample, standard, blank measurements of 10 ratios each at 5 s per ratio, completed within 16-19 minutes was developed following Jeffcoate et al. (2004). An ‘on-peak zero’ (OPZ), measured on a blank aliquot (0.05 M HNO3) at the start and end of each analysis was used to calculate a blank correction.  The 7Li/6Li of the first standard (std1) was blank (B1 and B2) corrected by:  ( ) ( ) 7 7 7 std1 B1 B27 6 std1 6 6 6 std1 B1 B2 Li 0.75 Li 0.25 Li Li / Li Li 0.75 Li 0.25  Li ⎛ ⎞− × + × ⎜ ⎟= ⎜ ⎟− × + ×⎝ ⎠                                        (2)  and second standard (std2) by:  ( ) ( ) 7 7 7 std2 B1 B27 6 std2 6 6 6 std2 B1 B2 Li 0.75 Li 0.25 Li Li / Li Li 0.75 Li 0.25  Li ⎛ ⎞− × + × ⎜ ⎟= ⎜ ⎟− × + ×⎝ ⎠            (3) and of the sample (smp) by:  ( ) ( ) 7 7 7 smp B1 B27 6 smp 6 6 6 smp B1 B2 Li 0.5 Li 0.5 Li Li / Li Li 0.5 Li 0.5 Li ⎛ ⎞− × + ×⎜ ⎟= ⎜ ⎟− × + ×⎝ ⎠           (4)   The values used are the measured intensities and; B1 = on peak zero at the start of the sequence B2 = on peak zero at the end of the sequence  175  The del notation was then calculated by:  ( )7 Eqn. 4Li 1 1000 Eqn. 2  Eqn. 3 2 ⎛ ⎞⎜ ⎟⎜ ⎟δ = − ×⎜ ⎟+⎛ ⎞⎜ ⎟⎜ ⎟⎝ ⎠⎝ ⎠                          (5)  The analytical protocol was not automated and in order to speed up the process, (critical for accurate Li isotopic measurement and to control instrument drift) peak centring was done manually before each measurement. Occasionally the 3% HNO3 wash, performed after each sample and standard analysis, was extended to ensure the signal returned to blank levels. This caused some variation in the length of time the protocol took. The Li concentration in LNPG samples that lacked trace element data was measured using GFAAS, in order for the Li content of sample and standard to be matched for analysis on the MC-ICP-MS. The Li concentration of other samples was calculated on average literature values (e.g., granite 15-26 ppm, Gao et al. 1998). Sample concentrations were commonly matched within 25% of the standard (e.g., 40 ppb standard, 30-50 ppb sample). Sample dilution of more than ~60% from the standard concentration was found to compromise the accuracy of the ‘blank corrected data’ due to the greater influence of the intensity blank correction on the 7Li/6Li ratio of dilute samples. The whole rock and many of the mineral samples from the LNPG have extremely high original Li concentrations; whole rock up to ~14 000 ppm Li, and ideal mineral stochiometric compositions of spodumene and lepidolite of ~37 300 and ~35 000 ppm Li, respectively. To produce peak signal voltages of ~6V, all samples were diluted to ~30-50 ppb Li, often requiring dilution factors of  > 1 ×  106. Despite these high dilution factors, the potential of sample components other than Li, to cause matrix effects compromising the accuracy of the data was a concern. Therefore the effect of Al  (the most abundant cation in the sample material apart from Si) on the accuracy on Puratronic® and L-SVEC standard solution measurements was investigated in a series of doping experiments. The 176  Puratronic® and L-SVEC (~40 ppb and ~50 ppb Li respectively) were doped with 0.5 ppm and 1 ppm Al (Table 4.5). The 7Li/6Li ratio results from the doped L-SVEC fell within the range of undoped material and although a slight decrease in 7Li/6Li ratio was determined by doping the Puratronic® compound solution with 1 ppm Al it was within error of the 7Li/6Li ratio (<0.14‰) of undoped samples of the same material. 4.4 RESULTS Results from the MC-ICP-MS analyses at UBC and the University of Maryland (UMD) are shown in Table 4.6.  4.4.1 δ7Li results from whole rock pegmatite, mineral separate and regional granitic intrusions The 28 samples of whole rock pegmatitic material from the LNPG measured in this study show a wide range of δ7Li values (δ7Li= –0.94‰ to +11.36‰). In comparison, whole rock pegmatite data from the Black Hills, South Dakota (Teng et al. 2006) shows (1) a range of δ7Li =+1.4‰ to +7.3‰ from simple unzoned pegmatites and (2) a range of +7.5‰ to +11.1‰ from the wall zone of the evolved Tin Mountain pegmatite (Fig. 4.3.A, Appendix 5). The spatial distribution of the δ7Li values from the LNPG area are shown in Fig. 4.4. The highest δ7Li values (+6‰ to +12‰) occur in restricted areas in the north and towards to the south of the study area. Of the mineral separates from the LNPG; two spodumene samples (AL050704 and P389859) have comparable δ7Li values of +3.5‰ and +3.7‰. In contrast, the δ7Li value of mineral separates from co-precipitated mineral assemblages can vary from sample to sample. Sample EB1304 has very uniform δ7Li values for co-precipitated minerals with muscovite at +7.9‰, plagioclase at +7.9‰ and quartz at +8.7‰, whereas sample EB0509 shows very different δ7Li values for the same mineral assemblage with muscovite at +2.2‰, plagioclase at +3.4‰ and quartz at +15.7‰. Averaged δ7Li values of minerals from the Tin Mountain pegmatite show spodumene at +8.1‰, muscovite at +10.4‰, plagioclase at +9.2‰ and quartz at +17.8‰ (Teng et al. 2006) and a range for 177 Sample δ7Li range 2SD L-SVEC ~40 ppb Li (n=9 ) -0.2 -0.93 to 0.15 0.8 L-SVEC ~40 ppb Li + 1 ppm Al (n=2 ) -0.5 0.3 L-SVEC ~40 ppb Li + 0.5 ppm Al (n=2 ) -0.6 0.1 Puratronic® ~50 ppb Li (n=158 ) 80.9 79.84 to 83.49 0.9 Puratronic® ~50 ppb Li + 1 ppm Al (n=2 ) 79.7 0.1 Table 4.5.Results from matrix effect tests, determined by doping L-SVEC and Puratronic® with varying amounts of Al. 178 Table 4.6. Results of δ7Li analysis from PCIGR (UBC) and the University of Maryland. Sample location Sample Field id. Li (ppm) REE pattern Average δ7Li n ‰2σ δ7Li Average δ7Li n ‰2σ δ7Li Total average δ7Li Total n Total ‰2σ δ7Li LNPG  P389704 SQF 8756 1 -0.7 1 -0.7 1 LNPG  P389711 SQF 7445 1 0.8 2 1.8 0.8 2 1.8 LNPG  P389719 QA 4289 2 -0.3 2 0.2 -0.8 1 -0.5 3 0.5 LNPG  P389726 SQF 4813 4 4.8 2 0.9 4.5 3 0.4 4.6 5 0.6 LNPG  P389734 QM 11437 4 1.1 1 1.1 1 LNPG  P389739 SQF 14161 2 2.1 2 1.7 2.1 2 1.7 LNPG  P389751 SQF 4073 3 2.4 2 0.6 2.4 2 0.6 LNPG  P389753 SQF 7644 1 0.1 3 2.0 0.1 2 1.8 0.1 5 1.0 LNPG  P389760 SQF 6603 3 -0.1 2 0.8 0.1 1 0.0 1 0.2 LNPG  P389789 SQF 3563 1 -0.9 1 -0.9 1 LNPG  P389791 SQF 3989 1 0.5 1 0.5 1 LNPG  P389808 QFL 7920 2 3.4 2 0.3 3.4 2 0.3 LNPG  P389809 QFL 9014 2 3.7 2 0.5 3.8 1 3.7 3 0.4 LNPG  P389816 QFL 4080 3 7.9 2 0.1 7.9 2 0.2 7.9 4 0.1 LNPG  P389817 QM 77 4 11.4 2 0.1 11.4 2 0.1 LNPG  P389844 QM 1375 3 3.2 2 0.2 4.4 2 1.1 3.8 4 1.5 LNPG  P389849 QM 2626 2 9.1 3 0.9 9.7 3 0.5 9.3 6 1.0 LNPG  P389851 SQFL 7606 3 6.4 2 0.2 6.1 2 0.4 6.2 4 0.4 LNPG  P389854 SQF 6617 2 -0.7 1 -0.7 1 LNPG  P389861 SQF 4520 4 5.8 2 0.8 5.8 2 0.4 LNPG  P389875 SQF 8897 1 0.0 2 0.3 0.0 2 0.3 LNPG  P389910 QA 4860 2 3.4 2 0.5 3.4 2 0.5 LNPG  P389913 QM 12105 1 0.6 2 0.5 0.6 2 0.5 LNPG P389870 SQF 6021 1 2.9 2 0.7 2.9 2 0.7 LNPG P389764 SQF 5271 1 2.1 3 0.2 2.1 3 0.2 LNPG P389768 SQF 5342 3 2.5 4 0.8 2.5 4 0.8 LNPG P389774 SQF 3019 1 2.0 2 1.4 2.0 2 1.4 LNPG P389797 SQF 6229 1 1.3 3 0.1 1.3 3 0.1 δ 7Li values UBC δ 7Li values UMD Combined UBC and UMD data Whole rock samples 179 Table 4.6. Results of δ7Li analysis from PCIGR (UBC) and the University of Maryland contd. Sample location Sample Li (ppm) REE pattern Average δ7Li n ‰2σ δ7Li Average δ7Li n ‰2σ δ7Li Total average δ7Li Total n Total ‰2σ δ7Li LNPG AL050704b spod 38250 3 3.5 4 0.6 3.5 4 0.6 LNPG P389859b spod 32398 2 3.7 2 0.0 3.7 2 0.0 LNPG EB0509e quartz 3.6 4 15.7 2 0.5 15.7 2 0.5 LNPG EB0509e mica 1362 4 2.2 3 0.8 2.2 3 0.8 LNPG EB0509e albite 20.2 4 3.4 2 1.0 3.4 2 1.0 LNPG EB1304e albite 871 3 7.9 2 1.8 7.9 2 1.8 LNPG EB1304e mica 16567 3 7.9 2 0.9 3.5 2 0.6 LNPG EB1304e quartz 44.7 2 8.7 3 1.1 8.7 3 1.1 NWT KR-05-110 na na -0.4 2 0.5 -0.4 2 0.5 NWT KR-05-130 na na -0.1 2 0.3 -0.1 2 0.3 NWT KR-05-175 na na 0.8 2 0.2 0.8 2 0.2 NWT KR-05-215 na na 0.9 3 0.9 0.9 3 0.9 NWT KR-05-97b na na 2.2 4 0.9 2.2 4 0.9 Peru My Mac 3865 na -0.7 6 0.8 -0.7 6 0.8 Peru Tom Mac na na -2.4 3 0.9 0.3 1 -1.7 4 2.7 Standard RGM-1 na na 5.7 1 5.7 1 Standard BCR-2 na na 7.1 1 7.1 1 Standard G-2 na na 0.3 8 1.7 0.3 8 1.7 Standard Puratronic na na 80.9 59 0.9 80.9 59 0.9 Standard UMD-1 na na 54.9 3 0.1 54.2 9 1.0 54.4 12 1.0 Minerals separates δ 7Li values UBC δ 7Li values UMD Combined UBC and UMD data 180 REE patterns discontinuous straight listric -2 0 2 4 6 8 10 12 K/Rb  7Li B indeterminate 0 5000 10000 15000 -2 0 2 4 6 8 10 12 Li ppm  7Li AGrey area denotes range of values from Tin Mountain pegmatite (Teng et al. 2006) 0 50 100 150 200 -2 0 2 4 6 8 10 12 Li/Cs  7Li C 0 1 2 3 -2 0 2 4 6 8 10 12 P O + F + H O wt%2 5 2 D  7Li 1 2 3 -2 0 2 4 6 8 10 12 Nb/Ta 24 Zr/Hf5 15  7Li G  7Li 0.0 0.5 1.0 -2 0 2 4 6 8 10 12 P O2 5 wt.% E local granites Ca O w t.% 0 1.0 0 1.0 P O wt.%2 5 see text for details  7Li D 0.0 0.2 0.4 0.6 0.8 1.0 -2 0 2 4 6 8 10 12 F wt. % F  7Li 20 30 40 50 60 70 -2 0 2 4 6 8 10 12 Y/Ho H  7Li Figure 4.3 Whole rock Li values of the Little Nahanni Pegmatite Group plotted against selected trace element concentrations and ratios. A. Li ppm B. K/Rb C. Li/Cs D. P O + F + H O wt.% E. P O wt.% (with and inset of CaO wt.% and P O wt.% (see text for details). F. F wt% G. Nb/Ta with inset of Zr/Hf. H.Y/Ho. The symbol coding corresponds to the groups based on the LNPG REE patterns identified in Fig. 4.2. Regional Cretaceous granitic intrusions (E. only) = black dots. Ellipses in E. indicates group of LNPG samples with similar P O wt.% values.  7 2 5 2 2 5 2 5 2 5 181 182  albite of  +7.1‰ to +11.1‰ and for muscovite and quartz of +16‰ to +19‰ from different zones (Tomascak et al. 1995). The five granitic samples from the region around the LNPG show a range of δ7Li= –0.3‰ to +2.2‰.  4.4.2 Accuracy and precision To establish satisfactory precision and accuracy, each analytical session typically started with 1 or 2 measurements of our δ zero standard (L-SVEC) and 2 measurements of our in-house secondary standard (Puratronic®). When these fell within 2σ of our laboratory averages, sample measurements were undertaken. Additional analyses of the in-house secondary standard (Puratronic®) were made periodically during and at the end of each session. In addition to these analyses, internationally accepted rock standards were also run periodically (Table 4.6).  4.4.2.1 UBC analyses From analyses undertaken over a 3-month period L-SVEC had a δ7Li value of – 0.22‰ (2SD 0.85‰, n=12); our in-house secondary standard, a lithium carbonate compound (Puratronic®; 99.998%) had a value of δ7Li =80.87‰ (2SD 0.88‰, n=59). The δ7Li value of the 0.05 M HNO3 used as a blank was 12.9 (2SD 0.80‰, n=158) over an 8-month period. Reference material G-2 (granite) had a δ7Li value of 0.06‰ (2SD 1.8‰, n=4) (Table 4.6) that falls between values obtained by James and Palmer (2000; 1.2‰) and Pistiner and Henderson (2003; –0.3‰). The in-house University of Maryland standard, UMD-1, had a δ7Li value of 54.9‰(2SD 0.01‰, n=3). The sample of Macusani glass previously analysed by Tomascak (1999) (TomMac) showed a δ7Li value of -2.4‰ (2SD 0.9‰, n=3). Replicate measurements of 15 whole rock pegmatite samples produced a range of 2SD values between 0.2‰ and 1.42‰ (mean 0.6‰ 2SD), measurements of 8 mineral separates produced a range of 2SD values between 0.0‰ and 1.8‰ (mean 0.8‰ 2SD), and measurements of the 5 regional granitic intrusions produced a range of 2SD values between 0.2‰ and 0.9‰ (mean 0.6‰ 2SD). 183  4.4.2.2 UMD analyses The in-house University of Maryland standard, UMD-1 had a δ7Li value of 54.2‰ (2SD 0.1‰, n=9) and rock standard RGM-1 had a δ7Li value of +5.7‰, higher than the value obtained by Schuessler et al. (2009; +2.6‰). Analysis of the TomMac sample undertaken at UMD recorded δ7Li =+0.3‰ (n=1), and the other sample of Macusani glass sample (MAC) showed a δ7Li value of –0.7‰ (2SD 0.8‰, n=6). Replicate measurements of 14 whole rock pegmatite samples produced a range of 2SD values between 0.1‰ and 1.8‰ (mean 0.7‰ 2SD).  Nine whole rock LNPG pegmatite samples were analysed at both UBC and UMD, eight of which agree within error (Table 4.6).  4.5 DISCUSSION 4.5.1 Whole rock analyses: evidence of magmatic and Li isotopic fractionation Processes involving fluid and solid phases, such as magmatic crystallization or hydrothermal alteration, can result in strong Li isotope fractionation (Foustoukos et al. 2004). It appears that the LNPG whole rock samples and minerals were only minimally affected by hydrothermal activity; rare secondary minerals include minor cookeite (identified by Rietveld analysis; Table 4.1) and Ca-bearing zeolites (Mauthner 1996). Minor calcite veins also provide evidence of a small amount of late-stage calcite-rich fluid, probably influenced by the host-rock (Mauthner 1996). The clay fraction segment of the Rietveld analytical spectra is insignificant (see Chapter 3) and petrographic inspection (Figs. 3.5-3.9 and Appendices 6 and 7) shows the pegmatite dike rock to be relatively fresh. Magmatic and isotopic fractionation values are therefore assumed to derive predominantly from igneous processes. The majority of the 208 dikes sampled from across the LNPG (Appendix 4; of which a subset formed part of this study), are rich in spodumene. An 85/15 ratio of spodumene-rich to spodumene-poor material was identified in the field i.e., those that 184  include spodumene in the mineralogical description, SQF, SQFL etc. and those that do not (Wengzynowski 2002; see Chapter 3). This is interpreted to represent the ratio of the melt volumes that consolidated to form the spodumene-rich and spodumene-poor dike compositions. Spodumene-rich samples identified thus, tend to have distinctive discontinuous REEN patterns and strongly negative Eu anomalies, and correspond to the least evolved of the LNPG compositions (Fig. 4.2 and see Chapter 3). The samples with discontinuous REEN patterns also have a small range of low δ7Li values (–0.94‰ to +2.9‰), similar to those of the regional granites (–0.4‰to +2.2‰) and other examples of S-type magmatism (Lachlan Fold Belt, Bryant et al. 2004; –0.4‰ to +2.1‰) agreeing with a relative lack of Li isotope fractionation identified in granites elsewhere (Teng et al. 2009). In contrast, the more differentiated, spodumene-poor LNPG samples that constitute the remaining ~15% of the melt volume (determined from the 85/15 ratio of spodumene-rich to spodumene–poor dikes) are more geochemically evolved, display LREEN -enriched, relatively straight or listric patterns (Fig. 4.2), and have δ7Li values up to ~12‰. By grouping the LNPG whole rock samples according to their REEN patterns (see Fig. 4.2) and plotting δ7Li against trace element content or element ratios, several distinctive trends are observed (Fig. 4.3A-H). Element ratios, such as K/Rb (Fig. 4.3B) are used to assess magmatic fractionation (Černý 1991a and b; K/Rb decreasing as the magma evolves). The δ7Li value of the LNPG samples increases as the K/Rb ratio decreases and there is a minor corresponding decrease in the Li content (Fig. 4.3A). The reduction in Li concentration is probably due to the precipitation of Li-rich minerals (e.g., less evolved samples have modal mineral abundances of up to ~22% spodumene and up to 13% lepidolite; see Table 4.1). The highest δ7Li value from the LNPG samples at +11.36‰, occurs in a sample with 87.45 SiO2 wt.%  (71.2% quartz CIPW normative; P389817). The preference of 7Li for the less coordinated crystal sites in quartz over the octahedral sites available in most other minerals is also observed in increased δ7Li value quartz signature from the pegmatites of the Black Hills, South Dakota (Teng et al. 2006; Fig. 4.3A). Of the three distinctive LNPG REEN patterns the ‘discontinuous’ group have high K/Rb values indicating they are the least fractionated. 185  Typically the elemental ratio of Li/Cs in granitic pegmatite magma is driven by mica composition (London 2005a). In the LNPG samples low Li/Cs values accompany the significant increase in δ7Li values in the more evolved compositions (Fig. 4.3C), whereas the least-evolved pegmatites fractionate little over a broad range of Li/Cs. Caesium is highly incompatible but can substitute for K in mica, feldspar and beryl, and is thought to increase with F content (and form complexes with F) in late-stage magmas (Černý et al. 1985 and references therein). Late enrichment of Cs in the melt is illustrated by lepidolite compositions and Cs-rich rims on late beryl (Mauthner 1996) and explains the progressive decreasing Li/Cs ratio in the LNPG system. The δ7Li values increase with volatile and flux content (F, P and H2O) in the melt during magmatic fractionation (Fig. 4.3D). Phosphorus content of the pegmatites shows a positive correlation with δ7Li, increasing from values similar to those in the regional granites (Fig. 4.3E).  Several dikes with discontinuous REEN patterns have very similar low P2O5 contents (circled on Fig. 4.3E) that may have been caused by precipitation and removal of a mineral phase (see Chapter 3) at melt content of ~0.25 P2O5 wt.%. The inset in Fig. 4.3E shows that these samples tend to have higher CaO/P2O5 ratios than more evolved samples and may be due to mineral precipitation of Ca-rich minerals such as apatite and plagioclase (London et al. 1990 and references therein). There is a positive correlation between F abundance and δ7Li values (Fig. 4.3F). The rare earth elements and element ratios (e.g., Nb/Ta, Zr/Hf and Y/Ho) fractionate strongly in the LNPG samples (Figs. 4.2, 4.3G and H). Similar fractionation trends have been recognised in other evolved melts associated with a high concentration of volatile components(not necessarily pegmatites; e.g., Bau 1996, Dolgopolova et al. 2004). The Zr/Hf and Y/Ho ratios from the LNPG samples are non-chondritic and fall well outside the ‘CHArge and Radius Controlled (‘CHARAC’) field’ of Bau (1996; see Chapter 3) who suggested that the non-chondritic ratios were due to the influence of components such as F, B and P as ligands for high field strength elements in the magma. The spread of Y/Ho (Fig. 4.3H) separates the most evolved samples (grey squares; low Y/Ho) from the least evolved samples (open triangles; high Y/Ho) with samples with the listric pattern (black triangles) generally having mid-value Y/Ho values and higher δ7Li. 186   To evaluate the influence of the build-up of volatile and flux content in the LNPG magma on the trace elements, modeling of REE abundances was undertaken (for more detail please refer to Chapter 3). Using partition coefficients derived from evolved F- and P-rich silicate melts (Fleet and Pan 1997, Veksler et al. 2005, and Suk 1998) the effect on REE abundances of the precipitation and removal of increasing amounts of volatile rich phases (by fractional crystallisation and fluid exsolution) from a melt was determined. As the straight LREEN patterns have the highest ΣREE the average values for each of the REE from these patterns was used as the starting composition for the modeling. The samples displaying discontinuous REEN patterns (and the least fractionated Li isotopic signature) were modeled relatively successfully by the fractional crystallisation of fluorapatite or the possible removal of small amounts of a P-rich fluid. By modeling the separation of small amounts of an F-rich fluid the listric pattern was well matched. The suitability of using geochemical modeling of REE on these types of rocks can be debated (see Chapter 3) however, a high F content in the magma (possibly close to saturation) appears to be associated with the more highly evolved compositions; the same compositions that show strong Li isotope fractionation. We suggest that increased levels of F in the magma can result in conditions that promotes fractionation of Li isotopes (Gordienko et al. 2007), ‘immobile’ element ratios (Bau 1996, Dolgopolova et al. 2004) and REEN patterns (Irber 1999, Liu and Zhang 2005, Veksler et al. 2005). Teng et al. (2006) suggested fractionation of Li isotopes could occur with fluid exsolution from a granitic magma with 7Li preferentially partitioned into the exsolved fluid. Although this may be the case it is important to note that fluid exsolution does not appear to be a prerequisite of Li isotope fractionation at LNPG. Similarly high Li isotopic values were determined from potentially F saturated and F undersaturated compositions (listric and straight REEN patterns respectively). In conclusion, the LNPG dikes are assumed to have originated from a single magmatic source and show some system-wide lateral zonation in δ7Li values (Fig. 4.4) and additional subtle vertical increase in the predominance of spodumene-poor dikes (associated with Sn and Ta mineralization; Wengzynowski 2002) and quartz abundance. However, we envisage that each dike developed individually (as closed systems) during 187  extensive magmatic fractionation, creating the range of compositions and Li isotopic values due to minor variations in magma composition between the dikes. The sequence of events forming the distinctive REEN patterns in the LNPG samples can be summarized as follows; (1) a volatile-rich phase with relatively high CaO/P2O5 ratios (F- and P-rich; possibly fluorapatite) is assumed to have precipitated  and produced discontinuous REEN patterns. The precipitation of Ca-rich plagioclase in these dikes may also explain the strongly negative Eu anomalies common in the discontinuous patterns. (2) Compositions with lower CaO/P2O5 ratios retained F in the melt for longer gaining a listric REEN pattern by precipitating a later F-rich phase. (3) The compositions with straight REEN patterns had increased F content but did not reach saturation. Furthermore, we observe strong Li isotope fractionation only in association with increased F concentration. The process governing this late stage increase in Li isotope fractionation to heavier signatures has not been determined. However, in a system such as a silicate melt, that potentially has a variety of atomic sites available to Li, the incorporation of the heavier isotope into structures with lower coordination numbers will increase the entropy of the system (Chacko et al. 2001). The late-stage build up of F in the pegmatite melt may promote a significant increase in the formation of stable Li-F bonds. If so this would influence the Li isotopic signature as a simple compound such as Li-F would be expected to attract 7Li more strongly than the common Li-O tetrahedra (Henderson 2005). This is envisaged as a potentially highly efficient mechanism for the concentration of 7Li in the melt in the late stages of geochemical evolution.  4.5.1.1 Macusani glass An explanation for the variation in the δ7Li values for Macusani glass is not yet confirmed however, isotopic heterogeneity may reflect the presence of rare phenocrysts; biotite, monazite and apatite (identified petrographically and by SEM in this study) andalusite, feldspar, quartz, sillimanite, spinel and virgilite (Barnes et al. 1970, French et al. 1978, Noble et al. 1984 and Pichavant et al. 1988a). 188  4.5.2 Comparison of LNPG δ7Li values with literature The range of δ7Li values (12.3‰) obtained on the whole rock pegmatite samples of the LNPG is slightly broader than that seen in the Tin Mountain pegmatite, Black Hills, South Dakota (Teng et al. 2006; δ7Li range of 9.7‰; Appendix 5). The lowest LNPG δ7Li values are more akin to the S-type regional granitic intrusions between – 0.4‰ and +2.2‰, S-type granites of the Lachlan Fold Belt (LFB) between –0.4‰ and +2.1‰ (Bryant et al. 2004), and other LFB two-mica intrusions –1.1‰ (Teng et al. 2004). All the whole rock pegmatite Li isotope data determined so far (this study, Teng et al. 2006) have been collected from pegmatites that are thought to be magmatically derived from S-type granitic intrusions (Fig. 4.5). The δ7Li signature is known to increase during magmatic fractionation (Teng et al. 2006) and this is reflected in the lower δ7Li values for granite and the typically higher average values for pegmatites.  4.5.3 Li isotope fractionation in minerals 4.5.3.1 Calculation of whole rock δ7Li signatures assuming equilibrium conditions If magmatic fractionation is the first-order control on whole rock δ7Li signature of the LNPG dikes, a second-order influence is the state of equilibrium of the silicate melt during crystallisation. Whole rock Li isotopic values represent the combined values of the minerals (plus fluid or melt inclusions). At given equilibrium conditions the δ7Li values of minerals are specific to that mineral under those conditions (Wunder et al. 2006 and 2007) and in those circumstances the δ7Li value of the whole rock can be calculated from the modal abundances of the minerals. To assess whether the mineral δ7Li values obtained from several samples (EB0509, EB1304, AL050704 and P389859) were typical and possibly representative of equilibrium conditions, the δ7Li values of whole rock samples were calculated (Table 4.7) using the measured mineral δ7Li values and the modal mineral abundance of the whole 189 ‰7Li Pelite and schist Arc basalts Whole rock(or wall zone) pegmatite material Granite Pegmatite minerals 0 10 0 5 0 10 0 0 10 5 -5 -3 -1 +1 +3 +5 +7 +9 +11 +13 +15 +17 +19 +21 +23 o n ly e N o. fa a s s 20 Quartz Plagioclase Spodumene Mica Tourmaline Holmquistite Legend for the pegmatite minerals LNPG Tin Mountain Regional granite Macusani glass Undifferentiated All boxes with thick outlines represent data measured in this study Figure 4.5 7 Li histograms comparing results from this study (highlighted with thick outlines) with literature values. Pegmatite: Teng (2006); arc basalt: Bryant (2004); pelite and schist:et al. et al. Teng (2006) and Bryant (2004); granite: Bryant (2004), Teng (2006),et al. et al. et al. et al. Pistiner and Henderson (2003), James and Palmer (2000); rhyolite: James and Palmer (2000) and Macusani obsidian: Tomascak (1999a). The mineral colour code is in the figure.et al.  190 Average Rietveld data from samples (modal abundance) See Table 3.3 for sample details sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 Quartz 0.29 0.33 0.33 0.34 0.38 0.29 0.30 Spodumene 0.20 0.13 0.06 0.01 0.01 0.21 0.10 Mica 0.05 0.14 0.08 0.23 0.32 0.17 0.20 Feldspar 0.42 0.39 0.54 0.42 0.27 0.33 0.38 Actual δ7Li of the sample -0.73 0.12 -0.75 4.35 1.08 2.12 2.5 δ7Li Li content Quartz (EB0509) 15.7 0.01 0.05 0.05 0.05 0.05 0.06 0.04 0.05 Quartz (EB1304) 8.7 0.01 0.03 0.03 0.03 0.03 0.03 0.02 0.03 Spodumene (av. AL050704+P389859) 3.6 3.5 2.48 1.68 0.72 0.15 0.10 2.67 1.31 Mica (EB0509) 2.2 1.6 0.19 0.49 0.26 0.81 1.14 0.58 0.71 Mica (EB1304) 7.9 1.6 0.67 1.74 0.95 2.91 4.08 2.10 2.57 Feldspar (EB0509) 3.4 0.1 0.14 0.13 0.18 0.14 0.09 0.11 0.13 Feldspar (EB1304) 7.9 0.1 0.33 0.30 0.42 0.33 0.21 0.26 0.30 calculated δ7Li using average spodumene and values from EB0509 3.5 3.2 3.2 2.5 2.4 3.3 3.0 range = 2.3-3.5 2std from measured δ7Li value 6.0 4.4 5.6 2.6 1.9 1.6 0.7 calculated δ7Li using average spodumene and values from EB1304 4.3 5.2 5.6 7.5 7.7 4.8 5.8 range = 4.3-7.9 2std from measured δ7Li value 7.1 7.1 9.0 4.5 9.4 3.9 4.6 Table 4.7.Whole rock δ7Li values calculated from the modal mineralogy of the LNPG whole rock samples determined by the Rietveld method and the δ7Li values of rock-forming minerals. 191 Average Rietveld data from samples (modal abundance) See Table 3.3 for sample details sample 8 sample 9 sample 10 sample 11 sample 12 sample 13 sample 14 Quartz 0.29 0.43 0.33 0.34 0.33 0.25 0.32 Spodumene 0.02 0.12 0.00 0.14 0.18 0.07 0.22 Mica 0.23 0.03 0.23 0.11 0.10 0.13 0.08 Feldspar 0.46 0.42 0.42 0.40 0.38 0.53 0.37 Actual δ7Li of the sample 1.96 1.26 8.01 2.87 na 4.3 0.78 δ7Li Li content Quartz (EB0509) 15.7 0.01 0.05 0.07 0.05 0.05 0.05 0.04 0.05 Quartz (EB1304) 8.7 0.01 0.03 0.04 0.03 0.03 0.03 0.02 0.03 Spodumene (av. AL050704+P389859) 3.6 3.5 0.23 1.56 0.00 1.79 2.32 0.86 2.77 Mica (EB0509) 2.2 1.6 0.79 0.12 0.80 0.40 0.37 0.45 0.27 Mica (EB1304) 7.9 1.6 2.84 0.42 2.88 1.43 1.31 1.62 0.96 Feldspar (EB0509) 3.4 0.1 0.16 0.14 0.14 0.14 0.13 0.18 0.13 Feldspar (EB1304) 7.9 0.1 0.36 0.33 0.33 0.32 0.30 0.42 0.30 calculated δ7Li using average spodumene and values from EB0509 2.6 3.5 2.4 3.3 3.4 3.1 3.5 range=2.3-3.5 2std from measured δ7Li value 0.9 3.2 7.9 0.6 1.7 3.8 calculated δ7Li using average spodumene and values from EB1304 7.3 4.4 7.9 4.9 4.7 5.8 4.4 range = 4.3-7.9 2std from measured δ7Li value 7.6 4.4 0.1 2.9 2.2 5.0 Table 4.7.Whole rock δ7Li values calculated from the modal mineralogy of the LNPG whole rock samples determined by the Rietveld method and the δ7Li values of rock-forming minerals contd. 192  rock samples obtained using the Rietveld method and X-ray powder diffraction data (Raudsepp et al. 1999). The mass balance equation:  [ ] [ ] [ ] [ ] 7 7 min1 min1 min 2 min 27 min1 min2 WR min1 min 2min1 min 2 Li Li a Li Li a Li etc. Li a Li a δ δδ ⎛ ⎞ ⎛ ⎞× × × ×= +⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎜ ⎟× ×⎝ ⎠ ⎝ ⎠        (6)  was used, where the δ7Li of whole rock samples (δ7LiWR) is the product of the measured δ7Li value of each mineral sample (min1, min2), the concentration of Li [Li] specific to that mineral phase, and the proportion (a) of the mineral (quartz, spodumene, mica and feldspar) in the whole rock, normalized to the concentration of Li in the mineral and its proportion in the rock. The Li content of quartz  (4-45 ppm) and feldspar (2-871 ppm; Table 3.3) was rounded up to 100 ppm and 1000 ppm respectively for the calculation. The modal proportions of muscovite and lepidolite were combined as ‘mica’ and given a nominal Li content of 16 000 ppm (Table 3.3), this is the typical Li content of lithian muscovite and provides an average value between lepidolite (35 000 ppm Li) and Li-free muscovite. As K-feldspar was not analysed for trace elements the Li concentration for plagioclase was used in the calculation as a proxy for all feldspars. Using the δ7Li mineral values measured in EB0509, plus an average value for spodumene, and the modal mineralogy of the 14 samples analysed by the Rietveld method the range of calculated whole rock δ7Li values of those samples is ~2.4-3.5‰; by using the EB1304 δ7Li mineral values, plus an average value for spodumene, and the same modal abundances the range for those samples is 4.3-7.9‰. The majority of the δ7Li values of the whole rock samples calculated in this way do not relate to the measured values, indicating that the measured mineral δ7Li values, or the combination of values from the different minerals, are not standard across the LNPG samples (Table 4.7) as expected under non-equilibrium conditions. The overall lack of correlation between the mineralogy and the measured δ7Li is also observed in Fig 4.6. The closest match between the calculated and measured δ7Li values (7.9 and 8.01‰, respectively) is found in sample P389816, using the δ7Li mineral values of EB1304. The similarity between the calculated and measured δ7Li values may indicate 193 Figure 4.6 Pie chart representations of the whole-rock mineralogy of several LNPG samples determined by the Rietveld method and X-ray powder diffraction data as described in Raudsepp et al. (1999). The del7Li value for the sample is indicated within each chart. 194  P389910  P389816 P389768 P389797 P389774 P389764  P389734  P389726  P389719  P389711 P389870  P389704 Quartz Plagioclase Muscovite K-Feldspar Spodumene Lepidolite Others Modal abundance of minerals -0.7 1.1 1.3 0.8 2.1 7.9 -0.5 2.5 2.9 4.6 2.0 3.4 0.6 7Including whole rock d Li 195  that this rock consolidated under near-equilibrium conditions. This does not infer that sample EB1304 was consolidated under similar conditions. Even from these simple calculations it is obvious that Li-rich mica with a relatively high δ7Li value would make the most significant contribution to higher whole rock δ7Li values in the LNPG dikes if they achieved equilibrium during consolidation. The samples comprising >20% mica (given a nominal Li concentration of 16,000 ppm) have significantly higher whole rock δ7Li values (samples 4, 5 7, 8 and 10; Table 4.7) than other samples. Values for mica from other highly evolved igneous rocks are variable but consistently even higher than those occurring at LNPG (~δ7Li =10-20‰; Tomascak et al. 1995, Teng et al. 2006, Gordienko et al. 2007). Mineral/silicate melt equilibrium constants for Li have not yet been determined and a quantitative evaluation of the relationship between mica δ7Li values and equilibrium conditions in the LNPG dikes cannot be made at this time. The quartz and mucovite/aqueous fluid fractionation factors of Lynton et al. 2005 (Δquartz-aqueous =  +8 to +12‰, and Δmuscovite-aqueous =+18 to +20‰) do not concur with these conclusions and while the reason is uncertain, it may be due to the difference between the composition of the experimental starting fluid (chlorine- bearing aqueous fluid, Lynton et al. 2005) and the natural peraluminous silicate melt. 4.5.3.2 Effect of non-equilibrium conditions on mineral δ7Li values The measured δ7Li values of quartz, mica and feldspar vary between samples and, as described above, the majority of the whole rock measured δ7Li values do not match the values calculated from the modal mineralogy. Under non-equilibrium conditions crystallizing minerals will not gain δ7Li values specific to that mineral under set conditions (Watson and Müller 2009), which may explain the variation observed in the LNPG samples and would concur with textural evidence of rapid crystallisation from the primary minerals in the dikes (Chapter 3). Chemical gradients in magma are constantly changing during mineral growth as elements and isotopes diffuse towards or away from sites of mineral precipitation. Incompatible components (including fluxes) that are not incorporated in a precipitating mineral will build up in the boundary layer at the growing crystal face, resulting in a low viscosity boundary layer enriched in incompatible components. The growth of a mineral 196  from a silicate melt that incorporates only trace amounts of Li in its structure (e.g., 7Li in the Si position in the Si-O tetrahedra), will build up a boundary layer of excluded Li (Fig. 4.7A). The difference in the rate of diffusion between 6Li and 7Li (in magma 3.35% Richter et al. 2003, Parkinson et al. 2007) will result in the melt closest to the growth surface being anomalously high in 7Li as the faster 6Li diffuses away from this area of high Li abundance. Crystallization could incorporate the 7Li-rich zone (as fluid inclusions or as a minor structural constituent) thereby increasing the δ7Li of the whole mineral. The boundary layer is also expected to be 7Li-enriched when the mineral does include Li in its structure (e.g., muscovite, spodumene). Most Li-bearing silicate minerals incorporate Li in the octahedral structural sites (Wunder et al. 2007) that will preferentially incorporate 6Li because the lighter isotope is more thermodynamically stable in more coordinated (typically octahedral) sites (Chacko et al. 2001). Lithium in the silicate melt will tend to be in the form of 7Li in co-ordination with 4 oxygen atoms in tetrahedral structures (Henderson 2005, and references therein) or in different (Soltay and Henderson 2005) or lower coordination configurations (Hannon et al. 1992).  During crystallization 7Li will preferentially be retained by the melt but rapid crystallization may incorporate this 7Li –rich zone of the boundary layer increasing its overall δ7Li signature of the mineral (Fig. 4.7A and B). In summary, if crystallisation is rapid enough the mineral may not achieve the isotopic signature specific to equilibrium conditions.  Rapid crystallization could reduce the typically high Li isotopic value of quartz, which includes trace amounts of 7Li, by incorporating 6Li diffusing away from the mineral growth sites. The 6Li could possibly be incorporated within the quartz structure, and certainly within fluid inclusions, thereby reducing the crystals’ δ7Li signature. Minerals that incorporate Li into their structure (typically 6Li into octahedral structural sites) may incorporate 7Li built up in the boundary layer during rapid crystallisation, thereby increasing the typically lower δ7Li signatures of these minerals. If the Li isotopic signature of co-precipitated minerals reflects the relative rate of crystallization and thereby a relative degree of equilibrium in the system, we may use the mineral signatures to interpret aspects of pegmatite formation. At LNPG sample EB0509 has low δ7Li values for mica that incorporates 6Li in its structure, and a high δ7Li value 197 quartz trace Li, mostly in fluid inclusions 6Li 6Li 7Li 6Li gro wth fronts layer che mic al melt 7Li 7Li 6Li ~3% faster diffusion than Li7 7Li boundary A mica 6Li 7Li 6Li7Li 6Li 7Li grow th fronts layer che mic al melt 7Li 7Li 6Li Li in crystal structure boundary B Figure 4.7 Cartoon of the different distributions of Li and Li during mineral growth from a silicate melt. A. mineral contains no structural Li (e.g., quartz). B. mineral (e.g., mica) that includes Li in octahedral structural sites. 6 7 198  for quartz, which preferentially incorporates 7Li. This suggests crystallization of this sample may have occurred closer to equilibrium and was less rapidly consolidated than EB1304 that shows more uniform isotopic signatures. The increased uniformity of the EB1304 signature suggests diffusion gradients may have been overtaken by rapid crystallisation. Primary textures in EB0509 show large mica books stacked in the direction of the c-axis; EB1304 shows plumose mica (growth in the direction of the a- axis) and spherulitic growth of small elongate (locally skeletal) feldspar crystals, both the textures in EB1304 are characteristic of rapid growth (London 2008) (Appendices 6 and 7). The petrographic observations therefore corroborate that the Li isotopic signature may reflect the crystallisation rates and by implication the degree of equilibrium in the system.  Re-equilibration of an original δ7Li signature of the minerals may have been caused by the presence of a Li-rich fluid. Use of in situ analytical techniques by other researchers has identified Li isotopic zonation in minerals that is interpreted as post- crystallization migration of a Li-rich fluid through the minerals with a wave of faster 6Li followed and re-equilibrated by a wave of slower 7Li (Parkinson et al. 2007). However, petrographic observations of the LNPG dikes (Chapters 2 and 3) indicate rapid crystallisation and show no evidence of the continuing presence of a Li-rich fluid.  4.5.4 Timescale of pegmatite consolidation and Li isotope diffusion In comparison to minerals from the Tin Mountain pegmatite (Tomascak et al. 1995 and Teng et al. 2006; Appendix 5) and lepidolite from the high-F intrusions of Voron’ya Tundra (Gordienko et al. 2007), spodumene, Li-rich muscovite, plagioclase and quartz from the LNPG dikes tend to have significantly ‘lighter’ isotopic signatures which we suggest may be due to more rapid consolidation of the narrow, sub-vertical LNPG dikes producing non-equilibrium conditions. Preservation of non-equilibrium mineral-melt Li isotopic signatures requires crystallization rates to exceed the rate of Li diffusion. Pegmatite crystallization appears to occur on a similar timescale as Li diffusion (Webber et al. 1999 (line rock calculation), London 2008, Richter et al. 2003), and could potentially preserve non-equilibrium Li isotopic signatures in pegmatitic minerals (Watson and Müller 2009). Lithium is mobile in fluids and in silicate melts and diffuses up to 3 orders of magnitude faster than any 199  other cation or molecule except for dissolved H2O or He (in basaltic magma Li diffuses 6 ×  10-5 cm2/s; Richter et al. 2003). Parameters for pegmatite melts which are necessary for accurate calculations are as yet unavailable and this value for basalt is taken as an approximation of the rate of Li diffusion in pegmatite melt. Assuming Li becomes enriched in a boundary layer along mineral crystallisation fronts (London 2005a) a difference in the diffusion rates between 7Li and 6Li of ~3% (Richter et al. 2003, Parkinson et al. 2007) would result in 6Li diffusing ~17mm further from the crystallising front of a mineral than 7Li, in 24 hours. The mobility of Li may presumably be even faster in the low viscosity fluids (Thomas and Webster 2000) of a Li-rich boundary layer. Webber et al. (1999) and London (2008) used heat-flow models to calculate the timescale of pegmatite consolidation (line rock portion in Webber et al. 1999) , using parameters that are similar in dimension, depth and composition to the LNPG dikes. Total solidification was calculated by both authors to be complete within days to weeks of pegmatite emplacement (see Chapter 2). Given the similarity in the rate of these two processes, highly variable, non- equilibrium Li isotopic signatures may be expected to occur in minerals under conditions of rapid pegmatite crystallisation. This concurs with conclusions from Maloney et al. 2008 on the heterogeneity of the Li isotope signature in rapidly crystallised tourmaline crystals. 4.6 CONCLUSIONS The whole rock Li isotopic signatures of the highly evolved, rare element Little Nahanni Pegmatite Group are strongly fractionated (δ7Li range of 12.3‰).  The correlation of mineralogy and geochemistry of the dikes with the Li isotopic data illustrates the evolution of the original magma during magmatic fractionation and indicates the significance of the build up and influence of volatiles and flux components. Less evolved dikes do not display Li isotope fractionation. These dikes can be recognized by: • a greater abundance of spodumene • less fractionated element ratios indicating geochemical fractionation (e.g., K/Rb, Li/Cs, Zr/Hf, Nb/Ta, Y/Ho) 200  • discontinuous normalized REEN patterns that can be derived by modelling P- and F-rich phase saturation using experimental partition coefficients  Strong Li isotope fractionation is recorded only in more evolved dike compositions represented by the final 15% of the melt to consolidate. These dikes are characterized by: • lower spodumene abundance • highly fractionated element ratios indicating geochemical fractionation • typically higher abundances of volatile and flux components • straight, or listric (that can be derived by modelling F-rich phase saturation using experimental partition coefficients) REEN patterns  The increase in F during magmatic fractionation appears to be critical in Li isotope and elemental fractionation in this highly evolved peraluminous system. The strong Li isotope fractionation observed is proposed to be due to the formation of Li-F bonds in the magma at high concentrations of F. These bonds could attract 7Li more than the common Li-O tetrahedral bonding (lower coordination numbers), and be an efficient method to concentrate 7Li in the late-stage melt. The rapid rate of consolidation of the LNPG dikes is comparable to the rate of diffusion of Li in a magma. This allows evidence of kinetic fractionation, produced by the significant variation in the diffusion rate between 6Li and 7Li, to be preserved in the dikes. A co-precipitated mineral assemblage that consolidated exceptionally rapidly shows a relatively homogeneous Li isotopic signature suggesting that the δ7Li values of mineral assemblages may indicate the relative degree of chemical equilibrium in crystallising magma. The overall lower δ7Li signature of the LNPG dikes may result from their relatively more rapid consolidation than examples of similar rocks reported in the literature. 4.7 ACKNOWLEDGEMENTS The authors wish to acknowledge the financial support provided by the Natural Sciences and Engineering Research Council of Canada (Research Grants to DW and LG) and by 201  War Eagle Mining Company Inc., especially Terry Schorn. Archer Cathro & Associates (1981) Ltd is thanked for access to samples and geochemical data. Logistical and analytical acknowledgements include those mentioned in Chapters 2 and 3, in addition we are indebted to Roberta Rudnick for the isotopic analyses undertaken at the University of Maryland and Jane Barling (PCIGR) is thanked for her professional help and advice during the isotopic analysis at UBC.  Much appreciation also goes to Bert Mueller (PCIGR) for assistance in measuring trace element concentrations and Bruno Kieffer (PCIGR) for sharing his clean-lab expertise. Paul Tomascak is thanked for providing a pre-ground sample of Macusani glass.                       202  4.8 REFERENCES BARNES, V.E., EDWARDS, G., MCLAUGHLIN, W.A., FRIEDMAN, I. &  JOENSUU, O. (1970): Macusanite occurrence, age, and composition, Macusani, Peru. Geological Society of America Bulletin 81, 1539–1546.  BAU, M. (1996): Controls on the fractionation of isovalent trace elements in magmatic and aqueous systems: evidence from Y/Ho, Zr/Hf, and lanthanide tetrad effect. Contributions to Mineralogy and Petrology 123, 323-333.  BREITER, K., FRYDA, J., SELTMANN, R., & THOMAS, R. (1997): Mineralogical evidence for two magmatic stages in the evolution of an extremely fractionated P-rich rare-metal granite: the Podlesí Stock, Krušné Hory, Czech Republic. Journal of Petrology 38, 1723- 1739.  BRYANT, C.J., CHAPPELL, B.W., BENNETT, V.C., & MCCULLOCH, M.T. (2004): Lithium isotopic compositions of the New England Batholith: correlations with inferred source rock compositions. Transactions of the Royal Society of Edinburgh: Earth Sciences 95, 199-214.  CANDELA, P.A. (1997): A review of shallow, ore-related granites: textures, volatiles and ore metals. Journal of Petrology 38, 1619-1633.  ČERNÝ, P. (1991a): Rare-element granitic pegmatites, Part I: Anatomy and internal evolution. Geoscience Canada 18, 49-67.  ČERNÝ, P. (1991b): Rare-element granitic pegmatites, Part II: Regional to global environments and petrogenesis. Geoscience Canada 18, 68-80. ČERNÝ, P., MEINTZER, R.E. & ANDERSON, A.J. (1985): Extreme fractionation in rare- element granitic pegmatites; selected examples of data and mechanisms. The Canadian Mineralogist  23, 381-421.  CHACKO, T., COLE, D.R. & HORITA, J. (2001): Equilibrium oxygen, hydrogen and carbon isotope fractionation factors applicable to geologic systems. In Stable Isotope Geochemistry (J.W. Valley and D.R. Cole, eds.), Reviews in Mineralogy and Geochemistry 43, 1-62.  CHAN, L.H. & KASTNER, M. (2000): Lithium isotopic compositions of pore fluids and sediments in the Costa Rica subduction zone: implications for fluid processes and sediment contribution to the arc volcanoes. Earth and Planetary Science Letters 183, 275-290. 203  CHAN, L.H., EDMOND, J.M., THOMPSON, G. & GILLIS, K. (1992): Lithium isotopic composition of submarine basalts: implications for the lithium cycle in the oceans. Earth and Planetary Science Letters 108, 151-160.  COOGAN, L.A., KASEMANN, S.A. & CHAKRABORTY, S. (2005): Rates of hydrothermal cooling of new oceanic upper crust derived from lithium-geospeedometry. Earth and Planetary Science Letters 241, 415-424.  COPLEN, T.B. (1996): Atomic weights of the elements (1995): Pure Applied Chemistry 68, 2339-2359.  DINGWELL, D.B., ROMANO, C. & HESS, K.-U. (1996): The effect of water on the viscosity of a haplogranitic melt under P-T-X conditions relevant to silicic volcanism. Contributions to Mineralogy and Petrology 124, 19-28.  DOLGOPOLOVA, A., SELTMANN, R., KOBER, B., WEISS, D., STANLEY, C. & DULSKI, P. (2004): Geochemical characteristics and lead isotope systematics of highly fractionated Li-F-enriched amazonite granites and related host rocks of the Orlovka-Spokoinoe mining district, Eastern Transbaikalia (Russia). Applied Earth Science: IMM Transactions section B 113, 83-99.  ERCIT, T.S., GROAT, L.A. & GAULT, R.A. (2003): Granitic pegmatites of the O'Grady Batholith, N.W.T., Canada; a case study of the evolution of the elbaite subtype of rare- element granitic pegmatite. The Canadian Mineralogist 41, 117-138.  FLEET, M.E. & PAN, Y. (1997): Rare earth elements in apatite: Uptake from H2O-bearing phosphate-fluoride melts and the role of volatile components. Geochimica et Cosmochimica Acta 61, 4745-4760.  FOUSTOUKOS, D.I., JAMES, R.H., BERNDT, M.E. & SEYFRIED, J.W.E. (2004): Lithium isotopic systematics of hydrothermal vent fluids at the Main Endeavour Field, Northern Juan de Fuca Ridge. Chemical Geology 212, 17–26.  FRENCH, B.M., JEZEK, P.A. & APPLEMAN, D.E. (1978): Virgilite: a new lithium aluminum silicate mineral from the Macusani glass, Peru. American Mineralogist  63, 461-465.  GAO, S., LUO, T.-C., ZHANG, B.-R., ZHANG, H.-F., HAN, Y.-W., ZHAO, Z.-D. & HU, Y.- K. (1998): Chemical composition of the continental crust as revealed by studies in East China. Geochimica et Cosmochimica Acta 62, 1959-1975.  GORDEY, S.P. & ANDERSON, R.G. (1993): Evolution of the northern Cordilleran miogeocline, Nahanni map area [105I], Yukon and Northwest Territories. Geological Survey of Canada Memoir 428.  204  GORDIENKO, V.V., GORDIENKO, VL.VL.,  SERGEEV, A.S., LEVSKII , L.K., LOKHOV, K.I., KAPITONOV, I.N. & SERGEEV, S. A. (2007): First data in favor of the crystallization model of lithium isotope fractionation in the pegmatitic process. Doklady Earth Sciences 413, 441-443.  GROAT, L.A., ERCIT, T.S., RAUDSEPP, M. & MAUTHNER, M.H.F. (1994): Geology and mineralogy of the Little Nahanni Pegmatite Group, part of NTS area 105 I/02. Economic Geology Survey Open File 1994-14. DIAND, NWT.  GROAT, L.A., MULJA, T., MAUTHNER, M.H.F., ERCIT, T.S., RAUDSEPP, M., GAULT, R.A. & ROLLO, H.A. (2003): Geology and mineralogy of the Little Nahanni rare-element granitic pegmatites, Northwest Territories. The Canadian Mineralogist 41, 139-160.  HANNON, A.C., VESSAL B. & PARKER, J.M. (1992): The structure of alkali silicate glasses. Journal of non-crystalline solids 150, 97-102.  HENDERSON, G.S. (2005): The structure of silicate melts: a glass perspective. The Canadian Mineralogist 43, 1921-1958.  HOEFS, J. (1997): Stable Isotope Geochemistry. Springer-Verlag, Berlin, Germany.  IRBER, W. (1999): The lanthanide tetrad effect and its correlation with K/Rb, Eu/Eu*, Sr/Eu, Y/Ho, and Zr/Hf of evolving peraluminous granite suites. Geochimica et Cosmochimica Acta, 63, 489-508.  JAMES, R.H. & PALMER, M.R. (2000): The lithium isotope composition of international rock standards. Chemical Geology 166, 319-326.  JEFFCOATE, A.B., ELLIOTT, T., THOMAS, A. & BOUMAN, C. (2004): Precise, small sample size determinations of lithium isotopic compositions of geological reference materials and modern seawater by MC-ICP-MS. Geostandards and Geoanalytical Research 28, 161-172.  KONTAK, D.J., ANSDELL, K., DOSTAL, J., HALTER, W., MARTIN, R. & WILLIAMS-JONES, A.E. (2001): The nature and origin of pegmatites in a fluorine-rich leucogranite, East Kemptville Tin Deposit, Nova Scotia, Canada. Transactions of the Royal Society of Edinburgh: Earth Sciences 92, 173-200.  LE MAITRE, R.W., BATEMAN, P., DUDEK, A., KELLER, J., LAMEYRE, J., LE BAS, M. J., SABINE, P.A., SCHMID, R., SORENSEN, H., STRECKEISEN, A., WOOLLEY, A.R. & ZANETTIN, B. (1989): A classification of igneous rocks and glossary of terms: Recommendations of the International Union of Geological Sciences Subcommission on the Systematics of Igneous Rocks. In Blackwell Scientific Publications, Oxford, UK. LONDON, D. (2005a): Geochemistry of alkali and alkaline earth elements in ore-forming granites, pegmatites and rhyolites. In Rare-Element Geochemistry and Mineral Deposits 205  (R.L. Linnen and I.M. Samson eds.), Geological Association of Canada Short Course Notes 17, Geological Association of Canada, Ottawa, Ontario, Canada. 17-43.  LONDON, D. (2005b): Granitic pegmatites: an assessment of current concepts and directions for the future. Lithos 80, 281-303.  LONDON, D. (2008): Pegmatites. Mineralogical Association of Canada, Québec, Canada.  LONDON, D., ČERNÝ, P., LOMIS, J.L.& PAN, J.J. (1990): Phosphorus in alkali feldspars of rare-element granitic pegmatites. The Canadian Mineralogist 28, 77l-786.  LUI C.-Q. & ZHANG, H. (2005): The lanthanide tetrad effect in apatite from the Altay No. 3 pegmatite, Xingjiang, China: an intrinsic feature of the pegmatite magma. Chemical Geology 214, 61-77.  LYNTON, S.J., WALKER, R.J. & CANDELA, P.A. (2005): Lithium isotopes in the system Qz-Ms-fluid: An experimental study. Geochimica et Cosmochimica Acta 69, 3337–3347. MALONEY J.S., NABELEK, P.I., SIRBESCU, M.-L.C. & HALAMA, R. (2008): Lithium and its isotopes in tourmaline as indicators of the crystallization process in the San Diego County pegmatites, California, USA. European Journal of Mineralogy 20, 905-916.  MAUTHNER, M.H.F. (1996): Mineralogy, geochemistry and geochronology of the Little Nahanni Pegmatite Group, Logan Mountains, southwestern Northwest Territories. M.Sc. thesis, University of British Columbia, Vancouver, British Columbia.  NOBLE, D.C., VOGEL, T.A., PETERSON, P.S., LANDIS, G.P., GRANT, N.K., JEZEK, P.A. & MCKEE, E.H. (1984): Rare-element–enriched, S-type ash-flow tuffs containing phenocrysts of muscovite, andalusite, and sillimanite, southeastern Peru. Geology 12, 35- 39.  PARKINSON, I.J., HAMMOND, S., JAMES, R.H. & ROGERS, N.W. (2007): High-temperature lithium isotope fractionation: insights from lithium isotope diffusion in magmatic systems. Earth and Planetary Science Letters 257, 609-621.  PICHAVANT, M., KONTAK, D.J., HERRERA, J.V. & CLARK, A.H. (1988a): The Miocene- Pliocene Macusani Volcanics, SE Peru I. Mineralogy and magmatic evolution of a two- mica aluminosilicate-bearing ignimbrite suite. Contributions to Mineralogy and Petrology 100, 300-324.  PICHAVANT, M., KONTAK, D.J., HERRERA, J.V. & CLARK, A.H. (1988b): The Miocene- Pliocene Macusani Volcanics, SE Peru II. Geochemistry and origin of a felsic peraluminous magma. Contributions to Mineralogy and Petrology 100, 325-338.  206  PISTINER, J.S. & HENDERSON, G.M. (2003):  Lithium-isotope fractionation during continental weathering processes. Earth and Planetary Science Letters 214, 327-339.  PRETORIUS, W., WEIS, D., WILLIAMS, G., HANANO, D., KIEFFER, B. & SCOATES, J. (2006): Complete trace elemental characterization of granitoids (USGS G-2, GSP-2) reference materials by high resolution inductively coupled plasma-mass spectrometry. Geostandards and Geoanalytical Research 30, 39-54.  RAIMBAULT, L., CUNEY, M., AZENCOTT, C., DUTHOU, J.L., & JOROY, J.L. (1995): Geochemical evidence for a multistage magmatic genesis of Ta-Sn-Li mineralization in the granite at Beauvoir, French Massif Central. Economic Geology 90, 548-576.  RASMUSSEN, K.L., MORTENSEN, J.K. & FALCK, H. (2006): Geochronological and lithogeochemical studies of intrusive rocks in the Nahanni region, southwestern Northwest Territories and southeastern Yukon. In Yukon Exploration and Geology 2005 (D.S. Emond, G.D. Bradshaw, L.L. Lewis and L.H. Weston eds.), Yukon Geological Survey, 287-298.  RASMUSSEN, K.L., MORTENSEN, J.K., FALCK, H. & ULLRICH, T.D. (2007): The potential for intrusion-related mineralization within the South Nahanni River MERA area, Selwyn and Mackenzie Mountains, Northwest Territories. In Mineral and Energy Resource Assessment of the Greater Nahanni Ecosystem Under Consideration for the Expansion of the Nahanni National Park Reserve, Northwest Territories (D.F. Wright, D. Lemkow and J.R. Harris, eds.). Geological Survey of Canada Open File 5344, 203–278.  RAUDSEPP, M., PANI, E. & DIPPLE, G. M. (1999): Measuring mineral abundance in skarn. I. The Rietveld method using X-ray powder-diffraction data. The Canadian Mineralogist 37, 1-15.  RICHTER F.M., DAVIS, A.M., DEPAOLO, D.J. & WATSON, E.B. (2003): Isotope fractionation by chemical diffusion between molten basalt and rhyolite. Geochimica et Cosmochimica Acta 67, 3905-3923.  RICHTER, F.M., WATSON, E.B., MENDYBAEV, R., DAUPHAS, N., GEORG, B., WATKINS, J. & VALLEY, J. (2009): Isotopic fractionation of the major elements of molten basalt by chemical and thermal diffusion. Geochimica et Cosmochimica Acta 73, 4250-4263.  RUDNICK, R.L. & GAO, S. (2003): Composition of the continental crust. In Treatise on Geochemistry, 3, (K.K. Turekian and H.D. Holland, eds.). Elsevier–Pergamon, Oxford, UK.  RUDNICK, R.L. & IONOV, D.A. (2007): Lithium elemental and isotopic disequilibrium in minerals from peridotite xenoliths from far-east Russia: product of recent melt/fluid–rock reaction. Earth and Planetary Science Letters 256, 278-293.  207  SCHUESSLER, J. A., SCHOENBERG, R. & SIGMARSSON, O. (2009): Iron and lithium isotope systematics of the Hekla volcano, Iceland — Evidence for Fe isotope fractionation during magma differentiation. Chemical Geology 258, 78-91.  SIRBESCU, M.-L. & NABELEK, P.I. (2003): Crustal melts below 400 °C. Geology 31, 685- 688.  SOLTAY, L.G. & HENDERSON, G.S. (2005): The structure of lithium-containing silicate and germinate glasses. The Canadian Mineralogist 43, 1643-1651.  SUK, N.I. (1998): Distribution of ore elements between immiscible liquids in silicate- phosphate systems (experimental investigation).  Acta Universitatis Carolinae, Geologica 42, 138-140.  TENG, F.-Z., MCDONOUGH, W.F., RUDNICK, R.L., DALPÉ, C., TOMASCAK, P.B., CHAPPELL, B.W. & GAO, S. (2004): Lithium isotopic composition and concentration of the upper continental crust. Geochimica et. Cosmochimica Acta 68, 4167-4178. TENG, F.-Z., MCDONOUGH, W.F., RUDNICK R.L., WALKER, R.J. & SIRBESCU, M.-L.C. (2006): Lithium isotopic systematics of granites and pegmatites from the Black Hills, South Dakota. American Mineralogist 91, 1488-1498.  TENG, F.-Z., RUDNICK, R.L.,  MCDONOUGH, W.F. & WU, F.-Y. (2009): Lithium isotopic systematics of A-type granites and their mafic enclaves: Further constraints on the Li isotopic composition of the continental crust. Chemical Geology 262, 370-379.  THOMAS, R. & WEBSTER, J.D. (2000): Strong tin enrichment in a pegmatite-forming melt. Mineralium Deposita 35, 570-582.  TOMASCAK, P.B., LYNTON, S.J., WALKER, R.J. & KROGSTAD, E.J. (1995): Li isotope geochemistry of the Tin Mountain Pegmatite, Black Hills, South Dakota. In The origin of granites and related rocks (M. Brown and P.M. Piccoli, eds.), US Geological Survey Circular 1129, 151-152.  TOMASCAK, P.B.,CARLSON, R.W. & SHIREY, S.B. (1999): Accurate and precise determination of Li isotopic compositions by multi-collector sector ICP-MS. Chemical Geology 158, 145-154.  TOMASCAK, P.B., RYAN, J.G. AND DEFANT, M.J. (2000): Lithium isotope evidence for light element decoupling in the Panama subarc mantle. Geology 28, 507-510.  VEKSLER I.V., DORFMAN, A.M., KAMENETSKY, M., DULSKI, P. & DINGWELL, D.B. (2005): Partitioning of lanthanides and Y between immiscible silicate and fluoride melts, fluorite and cryolite and the origin of the lanthanide tetrad effect in igneous rocks. Geochimica et Cosmochimica Acta 69, 2847-2860.  208  WATSON, E.B. & MÜLLER, T. (2009): Non-equilibrium isotopic and elemental fractionation during diffusion-controlled crystal growth under static and dynamic conditions. Chemical Geology 267, 111-124.  WEBBER, K.L., SIMMONS, W.B., FALSTER, A.U. & FOORD, E.E. (1999): Cooling rates and crystallisation dynamics of shallow level pegmatite-aplite dikes, San Diego County, California. American Mineralogist 84, 708-717. WEIS, D., KIEFFER, B., MAERSCHALK, C., BARLING, C., DE JONG, J., WILLIAMS, G.A., HANANO, D., PRETORIUS, W., MATTIELLI, N., SCOATES, J.S., GOOLAERTS, A., FRIEDMAN, R.M. & MAHONEY, J.B. (2006): High-precision isotopic characterization of USGS reference materials by TIMS and MC-ICP-MS. Geochemistry, Geophysics, Geosystems 7, Q08006. WENGER, M. & ARMBRUSTER, T. (1991): Crystal chemistry of lithium: oxygen coordination and bonding. European Journal of Mineralogy 3, 387-399.  WENGZYNOWSKI, W.A. (2002): Geological report on the Mac property. War Eagle Mining Company Inc.  WUNDER, B., MEIXNER, A., ROMER, R.L. & HEINRICH, W. (2006): Temperature- dependent isotopic fractionation of lithium between clinopyroxene and high-pressure hydrous fluids. Contributions to Mineralogy and Petrology 151, 112–120.  WUNDER, B., MEIXNER, A., ROMER, R.L., FEENSTRA, A., SCHETTLER, G. & HEINRICH, H. (2007): Lithium isotope fractionation between Li-bearing staurolite, Li-mica and aqueous fluids: An experimental study. Chemical Geology 238, 277-290.  209  CHAPTER 5: CONCLUSIONS The aim of this study was to gain a better understanding of the magmatic evolution, emplacement and consolidation of rare element pegmatites by studying the dikes of the Little Nahanni Pegmatite Group. An integrated approach using several lines of investigation was successfully applied in this study to determine various aspects of pegmatite formation. For example; • a detailed structural study related the influence of the regional deformation on pegmatite emplacement • fluid inclusion microthermometry constrained the pressure and temperature of dike emplacement • the application of several geochronological systems to dating the emplacement of the dikes allowed us to interpret the regional cooling history • whole rock geochemical analyses enabled the study of the trace element geochemistry and in the REEN patterns and allowed us to assess the pegmatites in the context of the regional peraluminous intrusions • the application of the Rietveld diffraction method to powdered bulk rock samples provided quantification of the pegmatite mineralogy that would otherwise have been unavailable • Li isotope measurements gave insight into the influence of flux components in the magma on isotopic fractionation and appears to provide a method to assess disequilibrium conditions of crystallisation At its conclusion it has identified key associations between host rock structures and dike orientation, confirmed the age of the dikes and provided an interpretation of the geothermal history of the study area during the mid-Cretaceous period. In addition, geochemical indicators have been used to detect processes of rare element and rare earth element fractionation occurring in highly evolved late-stage peraluminous magma and important steps have been made towards a better understanding of Li isotope behaviour in these environments. In the process, it has provided a substantial database of structural, whole rock geochemical and isotopic results relating to the formation of rare element pegmatites that will benefit future research. While there are undoubtedly aspects that 210  remain unresolved, this study has made a significant contribution in several areas of pegmatite petrogenesis and increased our understanding of the consolidation of the LNPG dikes in particular. The significant findings of each research chapter (Chapters 2, 3 and 4) are outlined below.  Chapter 2: Upper crustal dike emplacement in an orogenic region This chapter focussed on the structural and geochronological aspects of the emplacement of the Little Nahanni Pegmatite Group dikes. The general absence of cross- cutting relationships in the LNPG imply that the magma (total volume < 1 km3) that consolidated to form the dikes was emplaced in a single event, ~90Ma. Furthermore, the minor syn-post emplacement deformation of the dikes indicates that this was a relatively quiescent tectonic period in this area of the northern Canadian Cordillera. Strong fractionation of partial melt originating from a metapelitic source is the typical origin of rare element pegmatites (Černý 1991a) and this is reflected in the evolved peraluminous composition of the LNPG dikes. The thick wedge of basinal sedimentary rock and the crystalline basement rocks of the underlying margin ancestral North America would both provide suitable source lithologies. An upper level progenitor magma chamber for the dikes is, so far undetected in the area, but was probably the heat source responsible for the zone of contact metamorphism. The chamber is expected to be located within the core of the Fork anticlinorium (Fig. 5.1), such structures are typical traps for the ‘fertile’ granite magma associated with rare element pegmatites (Černý 1991b). An elevated geothermal gradient (~60 °C/km), interpreted to have been caused by the emplacement of the magma chamber, is implied from the variation in 40Ar/39Ar cooling dates from different minerals, and this regional increase in temperature will have assisted propagation of the magma into the country rock. Dike propagation was driven by buoyancy, derived from the high volatile content of the melt, to a depth of ~7-8 km depth (determined by fluid inclusion microthermometry). The passage of the magma was focused by pre-existing planar, sub- vertical strength anisotropies in the deformed host rock (bedding planes and axial planar foliations). In the absence of steeply dipping foliations the upward force of the magma 211 S0 Precambrian sedimentary basin strata S1     M e zo c  so i co r ss io n m p e Little Nahanni Pegmatite Group Age: ~90 Ma Depth: ~7-8 km Temperature: ~400-500 °C Location: S  and S planes0 1 magma chamber within      anticlinorium core not to scale peraluminous   partial melt SW NE p at ite eg m s n m Fork an ticli or iu undepleted metasedimentary source rock                    ~25-30 km depth 7-8 km depth msihpromatem tcatnoc fo ti mil Figure 5.1 Cartoon of the geological setting of the Little Nahanni Pegmatite Group.  The approximate location of the study area is indicated by the pale grey box. 212  was sufficient to cut across planes. Dike emplacement is not expected to have strongly influenced the regional geothermal gradient due to their relatively low temperature (~400-500 °C) and rapid rate of cooling; their consolidation was probably complete within days to weeks (suggested by cooling model calculations of similar dikes by Webber et al. 1999 and London 2008). The interplay between flux and fluid pressure loss and build up during sporadic magma propagation and a dynamic crystallization environment is recorded in the igneous textures.  Chapter 3: Evidence of late-stage magmatic fractionation A broad range of magmatic fractionation is recorded in the LNPG geochemistry and mineralogy, and the distinctive whole rock trace element compositions of the dikes and their interpretation using geochemical modelling of REE fractionation, form the basis for this chapter. Trace element ratios and abundances indicate that the dikes experienced different histories of magmatic fractionation. The implication being that they were closed systems during dike propagation and crystallisation, as the influence of external fluids or later magma pulses would have created homogeneity in the geochemistry. These observations broadly correlate with the mineralogy and indicate a diffuse zone of elevated magmatic fractionation towards the centre of the study area. The late stage build up of fluxes in the silicate melt is proposed to have been influential in the fractionation of ‘immobile’ element ratios (especially Y/Ho) and the formation of the distinctive whole rock REEN patterns. Replication of the REEN patterns was achieved by geochemical modeling the saturation of two flux-rich phases. Several well-known examples of typically peraluminous, commonly mineralised intrusions worldwide exhibit a comparable degree of geochemical evolution to the LNPG. Some of the closest comparisons include selected intrusions associated with the European Variscan orogeny (e.g., Raimbault et al. 1995 and Webster et al. 1997) the Khangilay Li-F granites of Transbaikalia, Russia (Badanina et al. 2006), and the Tin Mountain pegmatite, South Dakota (Walker et al. 1986).  213  Chapter 4: Strong Li isotope fractionation in highly evolved peraluminous melt  The development of the experimental and analytical set-up required to measure Li isotopes was a significant aspect of this final research chapter. Lithium isotopic signatures of whole rock LNPG dike samples are strongly fractionated  (up to δ7Li  = 11.4‰) and show a broad range of values (δ7Li range of 12.3‰), a slightly greater range than from the Tin Mountain pegmatite (Teng et al. 2006b) that shows a similar degree of advanced magmatic fractionation. Interpretations of the Li isotope results build on the geochemical data reported in Chapter 3; showing that the strongly fractionated Li isotopic results relate to the geochemical signatures and mineralogy of the dikes. This important correlation shows that no Li isotope fractionation occurred in dikes with the least-evolved magma compositions; and only the more geochemically evolved samples show significant Li isotope fractionation (Fig. 5.2). As observed in the fractionation and modeling of REEN patterns and some trace element ratios, flux build up (probably F) in the more evolved dike compositions is associated with Li isotope fractionation. However, the saturation of flux-rich phases is not required to cause Li isotope fractionation. An increased homogeneity in the δ7Li signature of co-precipitated minerals in the dikes is proposed to be the result of kinetic Li isotopic fractionation. Rapid crystallisation (as petrographically observed) may retain the Li isotopic fractionation caused by diffusion and provide us with a method to assess the degree of chemical equilibrium in the system.  The following section compares evidence from the LNPG dikes with typical features found in rare element pegmatites and looks at them from an exploration viewpoint. 214 012 7δ Li Saturation of F-rich phase Saturation of P-rich phase? No late-stage saturation d k  p pa at io n i e ro g 7δ Li Figure 5.2 Representation of the relationship between the three distinct REE  whole rock patterns, and their N 7approximate del Li values. The arrows represent pegmatitic dikes and the direction of propagation. The symbols used throughout the manuscript to signify the specific REE  patterns N are incorporated in the arrow heads. 215  5.1 Exploration tools Rare element pegmatites are uncommon, the ‘perfect storm’ of igneous intrusions, requiring specific components in their initial composition and a geological setting permissive of extended magmatic fractionation. The potentially high concentrations of valuable rare elements in these intrusions can make them attractive exploration targets. The ability to classify them by composition, texture and setting etc. (e.g., Černý and Ercit 2005) illustrates their commonality this has been applied in reviews by Černý (1991b) and Selway et al. (2005) that provide advise on rare element pegmatite exploration. Successful exploration for rare element pegmatites requires an understanding of their formation and the influences on their location; recognizing regional indicators of their presence, and mineralogical and geochemical indicators in the pegmatites of the degree of fractionation or mineralization. Factors important in the formation of the Little Nahanni Pegmatite Group (Table 5.1) that can be used during exploration are compared (and added to) to those outlined by Černý (1991b) and expanded on by Selway et al. (2005), who focused on factors more specific to rare element pegmatites of the Superior Province and in the exploration of Ta-rich pegmatites.  Source lithology The fundamental requirement in the formation of rare element pegmatites is an appropriate source rock lithology to provide the requisite components, e.g., Li, Cs, Rb. Muscovite in undepleted metapelite is commonly accepted to be the typical source for these elements (Černý 1991b), however extremely fractionated I-type magmas as a source, may also be appropriate (London 2008). Abundant P and Cs are indicative of an S-type rather than I-type origin. LNPG: Examples of potential muscovite-bearing source rocks are common in the study area as the LNPG occurs within Precambrian sedimentary strata of the Selwyn Basin overlying the ancient rift margin and crystalline basement of ancestral North America. There are also nearby examples (<90 km) of fractionated I-type magmas e.g. the O’Grady Batholith of the Tombstone Suite.  216 Table 5.1. Illustrations of the rare element pegmatite model from the LNPG. Aspect of formation Level LNPG Source lithology Required Precambrian Selwyn Basin strata overlying ancestral North America Tectonic setting Emplaced at 7-8 km, towards the end of an extensive orogenic period. Currently exposed at surface Pathway Typical March Fault, lithological boundary between Selwyn Basin and ancestral North America (potential) regional deformation fabric (confirmed) Trap Fork anticlinorium Greenschist to amphibolite host rock facies Regional greenschist facies plus contact metamorphic aureole Regional magmatism and timing Emplaced at ~90 Ma toward the end of period of regional magmatism. Nearby contemporaneous intrusions commonly include pegmatitic and aplitic phases. Dike emplacement post-deformation and peak metamorphism, indicated by relative timing of Geochemical and mineralogical anomalies in the country rock Indicator Mineralogical and geochemical haloes; tourmaline and muscovite plus Li and Ta (Young 2007) Mineralogy indications Spodumene, lepidolite, cookeite,  cassiterite, columbite, tantalite, tourmaline and beryl Pegmatite geochemistry Nb/Ta typically < 3.2, K/Rb <24, Cs typically 50-500 ppm, Li typically 1 000-14 000 ppm, average P2O5 of 0.6 wt.% Whole rock Li isotope signature High, up to δ7Li = 11.4‰ Geophysical anomalies Not necessary No radiometric or magnetic anomalies 217  Tectonic setting Rare element pegmatites are associated with orogenic terrains and are typically emplaced relatively high in the crust. The age of pegmatite emplacement and the degree of erosion since that time are fundamental to their survival and accessibility. LNPG: Emplaced at 7-8 km depth, ~90 Ma the LNPG is currently exposed at surface. The rugged terrain of the area improves access to the sub-vertical dikes.  Pathways, traps and host rock As with any fluid movement, pathways and traps influence the final location and structure of the pegmatites. Rare element pegmatites can occur within or proximal to the progenitor granite, but due to the high flux content in the melt they commonly propagate several km’s into the typically greenschist to amphibolite facies country rock. Pathways for the magma include significant crustal faults and deformational fabric and lithological boundaries while anticlinal cores commonly act as traps (Černý 1991b). LNPG: The March Fault, the strong deformational fabric and the buried lithological boundary between the Selwyn Basin strata and ancestral North America may all have influenced the movement of the magma. The dikes are located in the Fork anticlinorium where the progenitor granite is proposed to have formed a contact metamorphic aureole in the regional greenschist facies country rock.  Regional magmatism Rare element pegmatites tend to be emplaced towards the end of an active period of magmatism, in an area with an elevated geothermal gradient, post-peak deformation and metamorphism (Černý 1991b). Therefore areas with a history of typically peraluminous magmatism, have a higher potential for rare element pegmatites. LNPG: The LNPG was emplaced slightly later than nearby exposures of Tungsten Suite intrusions (94-99 Ma, Rasmussen et al. 2006) and the other example of Li mineralization in the area, the more mafic Tombstone Suite O’Grady Batholith (92.5 +/- 1.8 Ma; Mortensen, unpublished data). Cross-cutting relationships and evidence from the contact metamorphic zone provide relative timing of the emplacement of the LNPG dikes 218  post-peak deformation and metamorphism. Geochronological results suggest the area had an elevated geothermal gradient.  Geochemical and mineralogical anomalies in the country rock The distinctive mineralogy and geochemistry of rare element pegmatites can create geochemical anomalies and exomorphic mineral assemblages in the country rock (due to fluid interaction from the dikes), and in the overlying soil horizons. LNPG: A soil-sampling program in the LNPG study area detected Li, Cs, Ta and Sn anomalies beyond the southern extent of the dikes, although the presence of dikes at depth has not yet been confirmed (Young 2007). Tourmaline and mica aureoles occur in the country rock, proximal (cm’s) to the dikes.  Mineralogical indications The mineralogy of pegmatites can provide an initial indication of the potential for mineralisation and what elements may be abundant. LNPG: Spodumene, lepidolite and cookeite in the LNPG dikes illustrate the high Li abundance. Cassiterite, columbite, tantalite, tourmaline and beryl are significant indicators of highly evolved magma composition.  Pegmatite geochemistry  Geochemical analysis (especially of trace elements) of whole rock and mineral samples provides a more accurate and precise evaluation of their economic potential. Magmatic fractionation improves rare element concentration (Li, Cs, Ta) and is indicated by various trace element ratios and abundances, e.g., low Nb/Ta, low K/Rb, high Li and in the case of an S-type magma derivation high Cs and P. LNPG: The whole rock geochemical signatures of the LNPG dikes typically show Nb/Ta <3.2, K/Rb <24, Li between 1,000 to 14,000 ppm, Cs ranges between 50- 500 ppm and the average value of P2O5 is 0.6 wt.%. 219  Lithium isotopic signature LNPG: δ7Li values of LNPG whole rock samples range up to 11.4‰. Increasing values for Li isotopic signatures of whole rock samples provide an indication of the degree of magmatic fractionation, as higher δ7Li values occur with geochemical evolution and flux component build up.  Geophysical evidence LNPG: An airborne geophysical survey of the study area (Charbonneau 2007) did not detect the presence of LNPG dikes or strong anomalies distinguishing the area. The granitic composition of rare element pegmatites can be similar to country rock composition, which, in combination with their lack of magnetic minerals may mean that geophysical radiometric (measuring U, Th and K surface distribution) and magnetic data collection may not determine the presence of rare element pegmatites.  As outlined above, many aspects of rare element pegmatites and their formation suggested by Černý (1991b) and Selway et al. (2005), that can be useful to exploration, correspond to features observed in the LNPG. In addition, results from this thesis, from the geophysical survey, and especially from the Li isotopic study, increase our understanding of these systems and improve the opportunity for successful exploration. 5.2 NOTABLE PETROGENETIC ASPECTS OF THE LNPG The previous section brought together the features common to rare element pegmatites that may be useful to exploration. The LNPG however, also displays several unusual or exceptional features that are note-worthy; • the range of magmatic fractionation occurring within the ~200 contemporaneous dikes (<1 km3 total magma) •  the distinct variation in the trace element chemistry between the dikes • the wide variation in igneous textures and the excellent examples of crystallisation disequilibrium 220  •  the correlation between the geochemistry, the mineralogy and the Li isotopic fractionation. We propose that the majority of these features can be attributed to the propagation of the magma into a well-stratified and deformed region of the brittle upper crust. Pervasive, sub-vertical planar foliations focused the magma during dike propagation. The significant strength anisotropies between the foliations will have limited the interaction between individual dikes, which therefore retained a heterogeneous geochemical signature. The propagation and consolidation of the magma is expected to have been rapid, but somewhat intermittent, leading to dynamic variations in, for example, fluid pressure and composition during crystallisation, and creating variations in equilibrium, preserved in textures and recorded in the Li isotopic signature of co- precipitated minerals. The final feature mentioned, the correlation between geochemistry, mineralogy and Li isotopic fractionation, is not attributed to the physical setting of the dikes but the chemical process involved in their formation. And in this respect the LNPG is remarkable; it is the exposure of a swarm of rare element pegmatites at a crucial and dynamic point when the influence of fluxes in the magma have a fundamental influence on its magmatic and isotopic fractionation. In these respects the LNPG provides an exceptional opportunity for further studies on rare element pegmatites. 5.3 IMPLICATIONS FOR LI STUDIES A fundamental influence on our ability to make interpretations from the Li isotopic data in this study is the wide range of magmatic fractionation recorded in the dikes of the LNPG. This range allows us to observe the transition from less evolved magmas with no appreciable Li isotope fractionation, to the highly evolved, flux-rich compositions associated with strong isotopic fractionation. This integrated study is able to confirm the findings of Teng et al. (2006b and 2009) identifying a lack of Li isotope fractionation in granites and elevated δ7Li signatures from the highly evolved Tin Mountain pegmatite. It also provides a significant body of results that indicates strong Li isotopic fractionation only occurring in highly evolved magmas apparently in association 221  with a build up in F as suggested by Gordienko et al. (2007). From our study we propose that the higher δ7Li values observed in the late stages of magmatic fractionation may be partly due to the formation of stable Li-F bonds as the F concentration increases; and that these bonds attract 7Li more strongly than the Li-O tetrahedra in the melt. As the magma continues to fractionate this process increases the efficiency of concentrating 7Li. An indication of the degree of disequilibrium in the system appears to be provided by the Li isotopic signature from co-precipitated minerals. Samples that show textures of disequilibrium during crystallisation; e.g., spherulitic crystals, and mica growth along the a-axis, display increased Li isotopic homogeneity. It is proposed that the rapid consolidation of the dikes captures the signature of kinetic isotopic fractionation produced by the different diffusion rates of 6Li and 7Li from crystallisation sites. Maloney et al. (2008) suggested this process in their study on the isotopic signature of tourmaline.  Ongoing research into Li isotopes and the role of Li in silicate melt is steadily increasing our understanding of the isotopic and elemental processes. From this study it is clear that well exposed, rare element pegmatite systems such as the LNPG can provide remarkable opportunities to study natural examples. 5.4 FUTURE RESEARCH The LNPG is an excellent example of typical REL-Li pegmatite classification (Černý and Ercit 2005) in accordance to its location, host-rock and mineralogy, and provides a wealth of opportunities for future studies. Suggestions for future research fall into two categories, 1) those which relate directly to the LNPG and 2) experiments which would expand our understanding of some of the approaches used in this study. Although the dikes are expected to have consolidated extremely rapidly there is evidence that the melt evolved significantly before consolidation. Studies related to the compositional evolution of the magma and the variable igneous textures, would be valuable contributions to our interpretation of pegmatite consolidation: 1.     Crystallisation textures, paragenetic sequences and compositional variations The dikes would provide excellent material for detailed studies on compositional variations during consolidation display examples of crystallisation textures, including rare disequilibrium textures, and unusual minerals (e.g., berlinite). 222  223 Probe analyses to determine the compositional variation of individual phenocrysts through their growth sequence and between early and late examples of the same minerals (e.g., lepidolite, albite, fluorite). The study could also include primary and secondary textures and the formation of a paragenetic sequence of mineral crystallisation, alteration and replacement. 2. Melt and fluid inclusions. The influence of flux components, especially F, on the physical parameters of the melt, and the potential for liquid immiscibility make inclusion studies of immense interest. The volatile content and composition of melt inclusions may identify immiscible phases co-existing during crystallisation. Studies on similar rocks recognized a similarity with the composition of whole rock samples (Badanina et al. 2006) which may be useful in pegmatite studies that otherwise require large bulk samples. In addition, further microthermometric and compositional analysis of dilute fluid inclusions would increase our understanding of the range of P – T conditions of crystallisation. 3.  Lithium isotope studies. The wide range of δ7Li values occurring in the LNPG system make it suitable for a follow-up study relating the degree of disequilibrium in the system with the δ7Li values of co-precipitated minerals. A study of Li isotope diffusion in the country rock similar to that undertaken by Teng et al. (2006a) would improve our understanding of fluid interaction from the dikes into the country rock. 4.  Berlinite study. A detailed characterisation of the occurrence of berlinite in the LNPG would fill a gap in our knowledge about this rare alumino-phosphate mineral. The presence of berlinite in the LNPG dikes has implications on the build up and solubility of phosphorous in the magma. 5.  Date the latest deformation of greenschist country rock. The age of the most recent deformation in the area is unknown and may be related to emplacement of the dikes. Dating the formation of white micas associated with faults using Rb/Sr has been successfully undertaken elsewhere (e.g., Carrapa et al. 2009 and Inger and Cliff 1994). 6.  Experimental research into the combined effect of flux components on the structure of evolved silicate melts. Research in this field is ongoing (Henderson  2005, Mysen 2004). There remain however, gaps in our knowledge about the influence of combined fluxes on element partition coefficients and Li isotopic fractionation in highly evolved magmas, and a more comprehensive understanding is required if we are to obtain full benefit of geochemical studies of these systems.                         224  5.5 REFERENCES BADANINA, E.V., TRUMBULL, R.B., DULSKI, P., WIEDENBECK, M., VEKSLER, I.V. & SYRITSO, L.F. (2006): The behavior of rare-earth and lithophile trace elements in rare- metal granites: a study of fluorite, melt inclusions and host rocks from the Khangilay Complex, Transbaikalia, Russia. The Canadian Mineralogist 44, 667-692.  CARRAPA, B., DESCELLES, P.G., REINERS, P.W., GEHRELS, G.E. & SUDO, M. (2009): Apatite triple dating and white mica 40Ar/39Ar thermochronology of syntectonic detritus in the Central Andes: A multiphase tectonothermal history. Geology 37, 407-410.  ČERNÝ, P. (1991a): Rare-element granitic pegmatites, Part I: Anatomy and internal evolution. Geoscience Canada 18, 49-67.  ČERNÝ, P. (1991b): Rare-element granitic pegmatites, Part II: Regional to global environments and petrogenesis. Geoscience Canada 18, 68-80.  ČERNÝ, P. & ERCIT, T.S. (2005): The classification of granitic pegmatites revisited. The Canadian Mineralogist 43, 2005-2026.  CHARBONNEAU, B.W. (2007): Evaluation of airborne radiometric and magnetic data in the vicinity of the Nahanni National Park Reserve, Northwest Territories, Canada. In Mineral and Energy Resource Assessment of the Greater Nahanni Ecosystem Under Consideration for the Expansion of the Nahanni National Park Reserve, Northwest Territories (D.F. Wright, D. Lemkow and J.R. Harris, eds.). Geological Survey of Canada Open File 5344, 99–124.  GORDIENKO, V.V., GORDIENKO, VL.VL.,  SERGEEV, A.S., LEVSKII , L.K., LOKHOV, K.I., KAPITONOV, I.N. & SERGEEV, S. A. (2007): First data in favor of the crystallization model of lithium isotope fractionation in the pegmatitic process. Doklady Earth Sciences 413, 441-443.  HENDERSON, G.S. (2005): The structure of silicate melts: a glass perspective. The Canadian Mineralogist 43, 1921-1958.  INGER, S. & CLIFF, R.A. (1994): Timing of metamorphism in the Tauern Window, Eastern Alps: Rb-Sr ages and fabric formation. Journal of Metamorphic Geology 12, 695-707.  LONDON, D. (2008): Pegmatites. Mineralogical Association of Canada, Québec, Canada.  LYNTON, S.J., WALKER, R.J. & CANDELA, P.A. (2005): Lithium isotopes in the system Qz-Ms-fluid:An experimental study. Geochimica et Cosmochimica Acta 69, 3337–3347.  225  MALONEY J.S., NABELEK, P.I., SIRBESCU, M.-L.C. & HALAMA, R. (2008): Lithium and its isotopes in tourmaline as indicators of the crystallization process in the San Diego County pegmatites, California, USA. European Journal of Mineralogy 20, 905-916.  MYSEN, B.O. (2004): Element partitioning between minerals and melt, melt composition, and melt structure. Chemical Geology 213, 1-16.  RAIMBAULT, L., CUNEY, M., AZENCOTT, C., DUTHOU, J.L. & JOROY, J.L. (1995): Geochemical evidence for a multistage magmatic genesis of Ta-Sn-Li mineralization in the granite at Beauvoir, French Massif Central. Economic Geology 90, 548-576.  RASMUSSEN, K.L., MORTENSEN, J.K. & FALCK, H. (2006): Geochronological and lithogeochemical studies of intrusive rocks in the Nahanni region, southwestern Northwest Territories and southeastern Yukon. In Yukon Exploration and Geology 2005 (D.S. Emond, G.D. Bradshaw, L.L. Lewis and L.H. Weston eds.), Yukon Geological Survey, 287-298.  SELWAY, J.B., BREAKS, F.W. & TINDLE, A.G. (2005): A Review of Rare-Element (Li-Cs- Ta) Pegmatite Exploration Techniques for the Superior Province, Canada, and Large Worldwide Tantalum Deposits. Exploration and Mining Geology 14, 1-30.  TENG, F.-H., MCDONOUGH, W.F., RUDNICK, R.L. & WALKER, R.J. (2006a): Diffusion- driven extreme lithium isotopic fractionation in country rocks of the Tin Mountain pegmatite. Earth and Planetary Science Letters 243, 701–710.  TENG, F.-Z., MCDONOUGH, W.F., RUDNICK R.L., WALKER, R.J. & SIRBESCU, M.-L.C. (2006b): Lithium isotopic systematics of granites and pegmatites from the Black Hills, South Dakota. American Mineralogist 91, 1488-1498.  TENG, F.-Z., RUDNICK, R.L.,  MCDONOUGH, W.F. & WU, F.-Y. (2009): Lithium isotopic systematics of A-type granites and their mafic enclaves: Further constraints on the Li isotopic composition of the continental crust. Chemical Geology 262, 370-379.  WALKER, R.J., HANSON, G.N., PAPIKE, J.J., O’NEIL, J.R. & LAUL, J.C. (1986): Internal evolution of the Tin Mountain pegmatite, Black Hills, South Dakota. American Mineralogist 71, 440-459.  WEBBER, K.L., SIMMONS, W.B., FALSTER, A.U. & FOORD, E.E. (1999): Cooling rates and crystallisation dynamics of shallow level pegmatite-aplite dikes, San Diego County, California. American Mineralogist 84, 708-717.  WEBSTER, J.D., THOMAS, R., RHEDE, D., FÖRSTER, H.-J. & SELTMANN, R. (1997): Melt inclusions in quartz from an evolved peraluminous pegmatite: Geochemical evidence for strong tin enrichment in fluorine-rich and phosphorus-rich residual liquids. Geochimica et Cosmochimica Acta 61, 2589-2604.  226  YOUNG, I. (2007): Technical report on the Mac Property - 2, War Eagle Mining Company Inc.       227     APPENDICES 228 APPENDIX 1 Fluid inclusion microthermometry The Linkam THMS-G 600 heating and freezing stage was calibrated with two synthetic fluid inclusions with four observable phase transitions at -56.6, 0.0, 10.0 and 374.1 ºC. Analytical precision was determined at < ±0.1 ºC for temperatures < 50 ºC and 0.5 ºC for temperatures >50 ºC. The average of the melting points of the carbonic phase (TMCO2) from 49 fluid inclusions is –57.9 º ± 0.9 ºC (ranging from –59.7 ºC to –56.0 ºC), close to the triple point for pure CO2. The minor depression of the CO2 triple point is consistent with the presence of some other compressible gas within the carbonic phase of these inclusions. The most probably gases are N2 or CH4.  Eutectic melting (TE) in the aqueous phase was observed in only 2 fluid inclusions and occurred at –42.8º ± 2.4 ºC (ranging from –44.2 ºC to –40.0 ºC). The depression of the H2O-NaCl eutectic temperature is consistent with the presence of other chloride species such as KCl and CaCl2. The pegmatites also have Li bearing minerals and the presence of LiCl may also be responsible for the depression of the H2O- NaCl eutectic temperature. Ice melted (TMICE) at an average temperature of –6.5º ± 1.8 ºC (ranging from –9.0 ºC to –4.3 ºC, n=5) and clathrate melted (TMCLATH) at an average temperature of 9.3 º ± 0.7 ºC (ranging from 7.8 ºC to 10.6 ºC, n=29).  Homogenization of the carbonic phase (THCO2) occurred between 23.5 ºC and 27.3 ºC, with homogenization to liquid in the 5 cases where this was visible. Final homogenization (THtotal) occurred between 295 ºC and 340 ºC with homogenization to vapor occurring in 6 out of the 8 occasions when this was visible. On two occasions critical behavior was observed at this stage suggesting conditions near the critical point. Nucleation of clathrate occurred at an average temperature of –30.5º ± 2.0ºC (ranging from –28.6 ºC to –33.1 ºC, n=10), nucleation of ice was only visible in 5 fluid inclusions and occurred at an average temperature of –47.0º ± 2.4ºC (ranging from –50.0 ºC to –44.7 ºC, n=5) and finally the nucleation of solid CO2 occurred at an average temperature of – 97.2º ± 1.7ºC (ranging from –100.6 ºC to –95.5 ºC, n=36).  229  230 APPENDIX 2 U-Pb analysis Dry apatite grains were weighed accurately into 3.5 mL screwtop PFA Teflon beakers after having been ultrasonicated for 5 minutes in 2 mL 0.5 M HNO3 and rinsed twice, once with ultra-pure water then sub-boiled acetone. Grains were completely dissolved in the capped beakers using 1 mL of sub-boiled 6N HCl (a weighed drop of 233- 235U-205Pb isotopic tracer was also added) on a hotplate at 130 ºC for a minimum of 48 hours.  The digested samples were then prepared for column chemistry by drying down at 130 ºC and bringing up in1 mL of sub-boiled 3.1 N HCl before being placed on a hotplate at 130 ºC for 24 hours.  Anion exchange column procedures were slightly modified from that used for zircon, described in Parrish et al. (1987).  After elution into 7 mL PFA beakers 2 mL of 0.5 N H3PO4 was added to each.  Samples were then loaded on single, degassed zone refined Re filaments in 5 mL of a silicic acid emitter (Gerstenberger and Haase, 1997). Isotopic ratios were measured using a modified single collector VG-54R thermal ionization mass spectrometer equipped with an analogue Daly photomultiplier. Measurements were done in peak-switching mode on the Daly detector.  Analytical blanks during the course of this study were 1.0 pg for U and for Pb 5.5 pg.  U fractionation was determined directly on individual runs using the 233-235U tracer, and Pb isotopic ratios were corrected for fractionation of 0.28%/amu, based on replicate analyses of the NBS-982 Pb standard and the values recommended by Thirlwall (2000).  Reported precisions for isochron data were determined using the algorithms of Roddick (1987). Isochron diagrams were constructed and ages calculated with Isoplot 3.00 (Ludwig 2003).  Unless otherwise noted, all errors are quoted at the 2s level.       231 REFERENCES GERSTENBERGER, H. & HAASE, G. (1997): A highly effective emitter substance for mass spectrometric Pb isotope ratio determinations. Chemical Geology 136, 309-312.  LUDWIG, K. R. (2003): Users manual for Isoplot/Ex 3.00, a geochronological toolkit for Microsoft Excel. Berkeley Geochronology Center, Special Publication 4.  PARRISH, R., RODDICK, J.C., LOVERIDGE, W.D., & SULLIVAN, R.W. (1987): Uranium- lead analytical techniques at the geochronology laboratory, Geological Survey of Canada. In Radiogenic age and isotopic studies, Report 1, Geological Survey of Canada, Paper 87-2, 3-7.  RODDICK, J.C. (1987): Generalised numerical error analysis with application to geochronology and thermodynamics. Geochimica et Cosmochimica Acta 51, 2129-2135.  THIRLWALL, M.F. (2000): Inter-laboratory and other errors in Pb isotope analyses investigated using a 207Pb–204Pb double spike. Chemical Geology 163, 299-322.   APPENDIX  3 Rb/Sr analysis The mineral separates were separated under a binocular microscope before being rinsed in ultra-pure water and then sub-boiled acetone; ~ 0.5-0.8 g of each mineral and 0.14 g of whole rock powder were then weighed out accurately into custom-made high- pressure Teflon digestion capsules, and dissolved by the same method. Five mL of sub- boiled HF (29M) + 0.7 mL sub-boiled HNO3 (14M) + 0.7 mL HClO4 (12M) were added to each capsule before they were secured in a custom-made metal container and heated in an oven at 190 ºC for 5 days. On cooling the samples were dried down at 180 ºC on a hotplate in a perchloric fume hood before 5 mL sub-boiled HCl (~6M) was added, the high pressure assemblage reconstructed and returned to the oven for one day at 190 ºC. The samples were by this time fully dissolved and were quantitatively transferred into 15 mL capped Savillex beakers before being dried down at 130 ºC. The samples were prepared for column chemistry by being brought up in 1 mL sub-boiled HNO3 (14M) and heated, capped, for at least 12 hours on a hotplate at 130 ºC before being dried down.  The apatite grains were rinsed in ultra-pure water and then sub-boiled acetone, dried and dissolved in 15 mL Savillex beakers in 1 mL of sub-boiled 6N HCl on a hotplate at 130º C for a minimum of 48 hours.  The digested apatite samples were then prepared for column chemistry by drying down at 130º C. Column chemistry (method described in Weis et al. 2006) was undertaken using Teflon PFA custom-made columns (produced by Savillex) and BioRad AG50W-X8 resin (100-200 mesh). The isotopic measurements are ratioed to SRM 987 87Sr/86Sr = 0.7102325 ± 0.0000015 (n = 2). Isotopic compositions were measured in static mode with relay matrix rotation on a single Ta filament. The data were corrected for mass fractionation by normalizing to 86Sr/88Sr = 0.1194 using an exponential law. 232 Appendix 4. List of whole rock samples of pegmatitic material from the LNPG collected by Wengzynowski (2002). Sample # Northing Easting Elevation (m) Kg Ba (ppm) Cs (ppm) Nb (ppm) Rb (ppm) Sn (ppm) Sr (ppm) Ta (ppm) U (ppm) Zr (ppm) Rb/Sr Nb/Ta P389914 6889555 509228 1627 9.94 17 64.6 102 1690 2360 19.7 103 1 4 86 0.99 P389913 6889516 509105 1673 4.58 45 62.7 117 1220 266 62.8 142 4 4.5 19 0.82 P389912 6889530 509113 1691 8.84 42.5 54.4 75 925 341 327 86 4.5 16.5 3 0.87 P389911 6889545 509173 1657 8.84 24.5 119 70 1435 228 28.5 110 3 16.5 50 0.64 P389910 6889545 509173 1657 9.16 31.5 125.5 106 1665 548 33.5 166 5 21 50 0.64 P389909 6887092 509779 1122 na 86.5 121.5 109 1815 311 103 170.5 10.5 60.5 18 0.64 P389908 6890011 509087 1734 na 12.5 47.2 50 1050 201 19.2 45 3 12 55 1.11 P389907 6890011 509087 1734 na 11 52.9 95 1060 439 19.7 73 4.5 37.5 54 1.30 P389906 6890054 509074 1644 6.7 19 31.8 74 1375 402 11.5 50 2 17 120 1.48 P389905 6890054 509074 1644 6.48 24 29.9 118 858 545 23.9 79.5 3 26.5 36 1.48 P389904 6890054 509074 1644 10 8 39.3 66 1315 250 9.9 46 2 8.5 133 1.43 P389903 6890054 509074 1644 9.16 8.5 39.2 49 963 125 15.6 25.5 4 8 62 1.92 P389902 6890054 509074 1644 9.68 16.5 27.3 74 1390 417 15.2 50 2.5 18 91 1.48 P389901 6890054 509074 1644 6.4 19.5 31.1 79 1325 507 10 62.5 2 19 133 1.26 P389875 6890212 508703 1755 12.8 11.5 34.9 62 1410 361 15.6 32.5 2 10.5 90 1.91 P389874 6890212 508703 1755 12.1 14 46.4 79 1635 340 60.4 38.5 1 12 27 2.05 P389873 6890212 508703 1755 10.92 7.5 25.8 67 748 320 8.7 27.5 1.5 9.5 86 2.44 P389872 6890212 508703 1755 10.84 8 47 62 1675 438 10.8 39 3 19 155 1.59 P389871 6890212 508703 1755 13.16 6.5 48.3 71 1555 406 6.8 40.5 2.5 20 229 1.75 P389870 6891941 508494 1595 14.8 7.5 72.4 58 1450 156 9.5 58.5 3.5 15 153 0.99 P389869 6891941 508494 1595 12.88 8.5 97.1 61 2120 149 15.8 63 3.5 20.5 134 0.97 P389868 6891941 508494 1595 13.46 7.5 134.5 50 1435 167 22.5 75.5 6 31 64 0.66 P389867 6891765 508554 1638 3.56 50.5 29.4 61 663 181 79.5 57.5 1.5 28 8 1.06 P389866 6891753 508521 1650 2.96 103.5 29.6 77 635 120 89.1 94 3.5 35.5 7 0.82 P389865 6891211 508758 1675 3.42 52 41.3 84 1090 86 79.2 75.5 5 13 14 1.11 P389864 6891210 508721 1633 2.74 61 25.5 79 633 201 30.8 94 4.5 23.5 21 0.84 P389863 6891156 508501 1851 3.74 13.5 35.1 69 1115 269 63.2 65.5 2.5 19.5 18 1.05 P389862 6891156 508501 1851 2.38 21.5 17.6 80 797 209 62.8 82.5 3 32.5 13 0.97 P389861 6891150 508570 1781 4.56 44.5 154 91 2430 299 65.8 145 4 35 37 0.63 P389860 6891150 508570 1781 1.12 62.5 73.8 63 1390 276 34.2 80.5 2 25 41 0.78 P389859 6892608 507845 1605 11.88 11.5 71.2 72 1220 358 29.3 111.5 8 16 42 0.65 P389858 6892515 507870 1630 10 23 51.1 63 1630 55 32.3 69 9 9.5 50 0.91 P389857 6892510 507958 1639 6.84 17.5 87.7 142 1375 184 23.3 277 2.5 22.5 59 0.51 P389856 6892510 507958 1639 4.14 28.5 79 27 2240 185 33.9 25 3 6.5 66 1.08 P389855 6892510 507958 1639 11.84 21.5 77 75 1250 360 54.4 81 4 21 23 0.93 P389854 6890663 508335 2012 12.02 21.5 88.6 73 1945 294 60.4 91.5 4.5 45.5 32 0.80 P389853 6890663 508335 2012 4.4 17 110.5 95 2840 446 60.9 112 5.5 45.5 47 0.85 P389852 6891060 508099 1960 8.58 41.5 125.5 57 1430 395 92.6 64.5 6 61 15 0.88 233 Appendix 4. List of whole rock samples of pegmatitic material from the LNPG collected by Wengzynowski (2002). Sample # Northing Easting Elevation (m) Kg Ba (ppm) Cs (ppm) Nb (ppm) Rb (ppm) Sn (ppm) Sr (ppm) Ta (ppm) U (ppm) Zr (ppm) Rb/Sr Nb/Ta P389851 6891068 508158 1938 8.26 8.5 165.5 109 1790 209 18.5 199 4 28 97 0.55 P389851 6891068 508158 1938 8.26 8.5 165.5 109 1790 209 18.5 199 4 28 97 0.55 P389849 6891309 507938 1780 2.7 43 117.5 84 1250 288 91.5 96.5 9 17.5 14 0.87 P389848 6891359 508007 1749 18.46 20 122 118 1795 2030 23.3 132.5 9 39 77 0.89 P389847 6891359 508007 1749 4.92 6.5 115.5 91 2000 534 29.5 75 8.5 25.5 68 1.21 P389846 6891359 508007 1749 12.38 4 130 84 2190 453 20 62 11.5 35.5 110 1.35 P389845 6896822 505255 1552 7.22 19.5 86 68 1050 532 93.5 86.5 10.5 61 11 0.79 P389844 6896825 505328 1550 2.58 6 69.2 316 846 >10000 81.6 454 7.5 51 10 0.70 P389843 6896825 505334 1550 11.62 64.5 92 85 866 269 87.1 152 11.5 56.5 10 0.56 P389842 6896825 505338 1550 4.84 25.5 137.5 223 1050 195 108 461 18.5 50 10 0.48 P389841 6896825 505346 1550 9.6 10 67.5 111 947 615 32.4 193 20.5 47.5 29 0.58 P389840 6896825 505346 1550 12.54 8 86.8 61 1815 305 40.8 70.5 9 19.5 44 0.87 P389839 6896846 505389 1540 10.32 16 106 33 2590 161 80.1 30 9.5 28.5 32 1.10 P389838 6896846 505389 1540 2.46 85.5 171 60 1410 228 181.5 96.5 12 94.5 8 0.62 P389837 6896846 505389 1540 3 134.5 177 68 1620 198 319 176 6.5 45 5 0.39 P389836 6896846 505389 1540 9.2 17 134 46 2320 91 57.2 63.5 6.5 22.5 41 0.72 P389835 6896854 506089 1542 6.16 11 29 44 778 76 39.9 29 8.5 18 19 1.52 P389834 6896798 506152 1552 1.92 57.5 50 88 829 115 133 147 8 32 6 0.60 P389833 6896833 506160 1554 2.02 29.5 75.2 72 1300 192 74.4 58 15.5 28.5 17 1.24 P389832 6896833 506209 1544 9.4 17 42.6 79 1150 676 58.8 42 12 22 20 1.88 P389831 6895328 505487 1563 13.34 50 51.7 62 924 208 87.4 61 10.5 43 11 1.02 P389830 6895328 505487 1563 15.2 8 42.2 59 980 227 48.8 41.5 3 24.5 20 1.42 P389829 6895328 505487 1563 11.86 10 49.8 84 1330 330 57.4 54 7.5 28.5 23 1.56 P389828 6895328 505487 1563 11.96 6.5 48.3 73 1030 484 40.5 84 6 29 25 0.87 P389827 6896106 505593 1656 2.7 15 83.5 60 1375 240 39.4 41 10.5 25.5 35 1.46 P389826 6896325 505665 1754 2.54 73.5 94.8 75 1210 123 92.6 168.5 6 53 13 0.45 P389825 6896285 505577 1759 5.08 28 105.5 67 1365 413 28.8 68 10 57 47 0.99 P389824 6896263 505516 1774 5.1 64.5 127 76 1175 646 53.2 74 10.5 56.5 22 1.03 P389823 6896263 505516 1774 12.98 42 108 64 1685 398 63.6 65.5 2 29 26 0.98 P389822 6896026 505289 1709 7.62 9 78.6 75 1215 251 38.8 52 6.5 32.5 31 1.44 P389821 6896026 505289 1709 12.12 5 94.9 67 1695 294 35.5 65.5 6 31 48 1.02 P389820 6896026 505289 1709 8.96 8 57.9 49 698 272 37.7 35.5 4 28 19 1.38 P389819 6896026 505289 1709 8.66 4 39.2 76 646 364 13.8 29.5 4 19.5 47 2.58 P389818 6897533 505073 1927 3.78 61 116.5 58 1550 183 74.4 69 1.5 36.5 21 0.84 P389817 6897533 505073 1927 6 53.5 37 59 475 32 41.1 148.5 1.5 31.5 12 0.40 P389816 6897547 505505 1842 3.7 9 186.5 97 1655 352 57.8 182.5 3 23 29 0.53 P389815 6897533 505692 1786 10.3 5 204 78 2040 359 17.1 76.5 6.5 37 119 1.02 P389814 6897533 505692 1786 7.24 4 144.5 79 1785 274 14.2 59 6 31 126 1.34 234 Appendix 4. List of whole rock samples of pegmatitic material from the LNPG collected by Wengzynowski (2002). Sample # Northing Easting Elevation (m) Kg Ba (ppm) Cs (ppm) Nb (ppm) Rb (ppm) Sn (ppm) Sr (ppm) Ta (ppm) U (ppm) Zr (ppm) Rb/Sr Nb/Ta P389813 6897443 505545 1777 13.6 19.5 227 107 2080 478 34.9 152 4.5 50 60 0.70 P389812 6897443 505545 1777 3.1 34 195 64 3180 490 69.1 115 6.5 23 46 0.56 P389811 6897443 505545 1777 2.76 32 127.5 92 1550 489 104.5 105 4.5 31 15 0.88 P389810 6897443 505545 1777 5.38 35 198 84 2020 570 55.8 144 4.5 39 36 0.58 P389809 6897443 505545 1777 5.8 11.5 164 57 1665 355 51.2 80.5 6.5 46.5 33 0.71 P389808 6897443 505545 1777 10.04 18.5 219 62 2230 272 16.5 85 4.5 59.5 135 0.73 P389807 6893328 507545 1603 21.76 11 118 69 2170 368 156.5 62.5 10.5 33.5 14 1.10 P389806 6893314 507510 1608 9.86 46 51.9 60 703 370 254 33 6 35 3 1.82 P389805 6893312 507488 1602 7.26 18.5 215 71 2740 303 33.9 86 8 24.5 81 0.83 P389804 6893301 507478 1596 7.06 7.5 45.7 59 790 417 38.2 25.5 4.5 21 21 2.31 P389803 6893229 507354 1616 12.76 8 48.3 64 923 433 31.7 28 7 17.5 29 2.29 P389802 6893207 507299 1620 10.7 6.5 184.5 60 2410 474 13.2 51.5 2 22.5 183 1.17 P389801 6893307 507131 1641 14.42 12 94.4 59 1260 388 39.6 62.5 4 21 32 0.94 P389800 6892976 508313 1520 12.34 46.5 77.3 56 1050 280 23.4 42 10.5 30.5 45 1.33 P389799 6892657 508132 1597 4.16 18.5 24.8 70 841 70 54.4 31.5 9 17 15 2.22 P389798 6892660 508073 1591 11.96 13 19.7 55 761 47 23.2 21 3.5 12 33 2.62 P389797 6892654 508066 1582 8.6 12 26.4 81 815 61 29.4 27 3.5 12.5 28 3.00 P389796 6892904 508025 1577 20.66 6 47.7 84 1080 225 12.4 38 14 28 87 2.21 P389795 6892716 507691 1633 10.94 10.5 36.3 67 798 140 309 33 6 13 3 2.03 P389794 6892842 507753 1664 17.32 208 50.3 61 950 195 87.9 38 0.5 18 11 1.61 P389793 6892842 507753 1664 12.88 13 38.5 59 810 275 35.1 37.5 5 13 23 1.57 P389792 6892842 507753 1664 8.98 14.5 41.3 62 933 165 33.7 29 4.5 27.5 28 2.14 P389791 6894020 506162 1730 12.2 9.5 49.9 62 1330 97 35 19.5 6.5 14 38 3.18 P389790 6894020 506162 1730 8.92 6 61.8 78 2050 181 35 43.5 12.5 17 59 1.79 P389789 6894020 506162 1730 10.96 12.5 35.3 58 1070 161 51 19 6.5 19 21 3.05 P389788 6894368 506024 1779 12.44 14 57.4 108 1060 2960 61.5 76.5 3.5 37 17 1.41 P389787 6894368 506024 1779 12.76 15.5 23.4 73 463 115 60.5 22 1.5 12 8 3.32 P389786 6894368 506024 1779 12.64 12.5 43.7 70 1030 175 15.9 30.5 1 11 65 2.30 P389785 6893002 507514 1816 1.5 820 152 66 3870 356 20.7 46 0.5 14.5 187 1.43 P389784 6893002 507512 1816 11.18 1390 149 71 3360 377 28.6 56.5 1 24.5 117 1.26 P389783 6892996 507502 1818 8.14 10 106 58 3120 404 52.1 46.5 2 38 60 1.25 P389782 6892993 507492 1821 3.26 17 77.2 86 1945 310 46 55.5 2 38 42 1.55 P389781 6892980 507473 1828 4.78 15 58.6 73 1040 373 45.5 39.5 1 42 23 1.85 P389780 6892980 507473 1828 8.72 27.5 38.3 71 1895 209 56.3 26 1 16 34 2.73 P389779 6892958 507446 1827 16.54 31.5 84.6 58 2040 297 29.9 46 3 26 68 1.26 P389778 6892958 507446 1827 10.22 10.5 47.9 40 869 185 13.2 15.5 0.5 13 66 2.58 P389777 6892958 507446 1827 7.9 8.5 89.3 63 1160 375 12.1 56.5 1 51.5 96 1.12 P389776 6892947 507426 1830 13.84 17.5 70.4 56 2520 247 16.5 23.5 3 15.5 153 2.38 235 Appendix 4. List of whole rock samples of pegmatitic material from the LNPG collected by Wengzynowski (2002). Sample # Northing Easting Elevation (m) Kg Ba (ppm) Cs (ppm) Nb (ppm) Rb (ppm) Sn (ppm) Sr (ppm) Ta (ppm) U (ppm) Zr (ppm) Rb/Sr Nb/Ta P389775 6892926 507403 1840 5.06 10.5 126.5 62 2730 395 47.7 71 2 19.5 57 0.87 P389774 6892921 507384 1841 21.26 13.5 185 66 3140 396 38 49.5 2 34.5 83 1.33 P389773 6892921 507384 1841 16.76 30 142.5 60 2600 432 57.7 63.5 2 33 45 0.94 P389772 6892893 507319 1863 8.04 9 252 82 3040 385 23.2 96.5 3.5 25 131 0.85 P389771 6892851 507249 1874 10.04 7 210 70 2650 497 33.6 65 4 20.5 79 1.08 P389770 6892868 507267 1872 26.34 21.5 211 73 3160 485 119.5 81.5 2.5 29.5 26 0.90 P389769 6892868 507267 1872 4.74 13 227 91 3020 416 70.5 92 2 22 43 0.99 P389768 6892868 507267 1872 5.82 6 220 83 3110 357 19.2 76.5 2 30 162 1.08 P389767 6892868 507267 1872 14.04 8 206 86 2890 361 25.6 78 2 26 113 1.10 P389766 6892868 507267 1872 4.14 17.5 102.5 87 2340 462 81.1 104.5 3 32.5 29 0.83 P389765 6892868 507267 1872 6.06 14 198 75 3530 403 73.5 100 3.5 33 48 0.75 P389764 6892868 507267 1872 6.18 15 151.5 71 2520 445 66.5 82 3 33.5 38 0.87 P389763 6892868 507267 1872 3.38 22 194 59 2990 305 74 50.5 1.5 75 40 1.17 P389761 6894330 505957 1773 3.42 8.5 312 69 2970 518 46.2 153.5 5 28 64 0.45 P389760 6892808 507179 1895 5.82 15.5 48.1 68 1205 280 45.4 50 3.5 21 27 1.36 P389759 6894299 506005 1788 10.54 6.5 59.9 62 1265 126 14.4 29 8 20 88 2.14 P389758 6894299 506005 1788 8.5 8.5 52.1 75 963 263 8 44.5 9.5 23.5 120 1.69 P389757 6894299 506005 1788 6.62 10.5 45.3 45 1075 244 4.8 24 3.5 18.5 224 1.88 P389756 6894299 506005 1788 11.34 7.5 45.6 68 839 159 7.5 37 9.5 21.5 112 1.84 P389755 6893279 506824 1801 5.32 11 77.3 65 1405 187 45.7 46 10 44 31 1.41 P389754 6893279 506824 1801 12.62 9.5 68.9 85 1835 263 29.8 62.5 5.5 41 62 1.36 P389753 6893279 506824 1801 11.78 5 67.6 61 1450 203 15.2 39 3 26 95 1.56 P389752 6893279 506824 1801 1.38 36 109.5 101 2730 620 67.1 160.5 8.5 30 41 0.63 P389751 6893279 506824 1801 1.6 14 219 70 4220 522 38.9 107 4 33 108 0.65 P389750 6890054 509074 1664 9.56 26 54.8 75 1630 345 12.7 47.5 4 22.5 128 1.58 P389749 6890054 509074 1664 8.48 7 22.2 78 833 327 13.7 44 2.5 21 61 1.77 P389748 6890054 509074 1664 12.22 8 34 74 1145 326 8.1 40.5 1.5 10.5 141 1.83 P389747 6891891 508143 1665 7.22 9 158.5 69 2100 321 12.3 83.5 3.5 21 171 0.83 P389746 6891891 508143 1665 4.64 17 156.5 91 1860 391 22.6 124 1.5 33 82 0.73 P389745 6891891 508143 1665 6.44 2370 166.5 71 1485 291 20.8 80 1.5 18 71 0.89 P389744 6891917 508162 1670 16.02 8.5 107 76 1430 336 38.8 74.5 2 25.5 37 1.02 P389743 6891917 508162 1670 12.12 12 109 56 1560 355 21.9 60 1.5 12.5 71 0.93 P389742 6890050 508567 2019 11.84 13 124 92 2310 486 68.5 117 3 40 34 0.79 P389741 6890050 508567 2019 10.12 17 106 79 2000 534 85.4 136.5 2 28 23 0.58 P389740 6890050 508567 2019 7.18 18.5 95.2 66 2150 458 70.3 107.5 2 40.5 31 0.61 P389739 6889924 508558 2010 16.66 6.5 112.5 64 1755 397 27.5 114 6.5 24 64 0.56 P389738 6889924 508558 2010 10.64 11 97.6 125 1970 1525 55.5 281 6.5 39.5 35 0.44 P389737 6889924 508558 2010 9.76 3 184 59 2190 431 27.6 131 10.5 33 79 0.45 236 Appendix 4. List of whole rock samples of pegmatitic material from the LNPG collected by Wengzynowski (2002). Sample # Northing Easting Elevation (m) Kg Ba (ppm) Cs (ppm) Nb (ppm) Rb (ppm) Sn (ppm) Sr (ppm) Ta (ppm) U (ppm) Zr (ppm) Rb/Sr Nb/Ta P389736 6889924 508558 2010 11.86 10.5 114 71 1645 599 34.1 171.5 7.5 33.5 48 0.41 P389734 6892483 507970 1670 2.38 6 120.5 184 1800 >10000 12 329 6 52.5 150 0.56 P389732 6892573 508506 1690 2.28 13 224 51 1250 459 77.1 90.5 2 20.5 16 0.56 P389731 6889721 508491 2011 1.66 14.5 83.8 117 900 2680 295 185.5 1.5 22.5 3 0.63 P389729 6890757 508209 2008 2 4.5 140.5 71 2030 404 21.1 103.5 2.5 32 96 0.69 P389727 6891001 508017 2004 1.44 54 81.7 69 986 226 106 134 3 51.5 9 0.51 P389726 6891110 508291 1954 1.94 7 120.5 67 1595 189 51.6 72 1.5 41.5 31 0.93 P389725 6891420 508409 1860 1.58 10 9 46 293 259 57.3 51.5 1.5 25 5 0.89 P389724 6893617 506450 1910 15.66 11 77.2 74 1450 378 17.5 43 3.5 44 83 1.72 P389723 6894261 506757 1613 8.1 50.5 66.8 73 1395 271 109.5 53 8 27.5 13 1.38 P389722 6894203 506640 1624 10.62 13 37.1 74 801 726 16.3 31.5 7.5 16.5 49 2.35 P389721 6894203 506640 1624 15.58 6.5 40.7 63 990 312 23.4 49 10 35.5 42 1.29 P389720 6894346 505975 1781 13.58 14 46.7 60 1165 293 25.6 28.5 6.5 23.5 46 2.11 P389719 6894346 505975 1781 13.8 8 38.8 64 1090 270 19 31 7 19.5 57 2.06 P389718 6894346 505975 1781 13.82 7 35.3 61 989 154 18.8 27.5 8 21 53 2.22 P389717 6894346 505975 1781 12.88 8.5 56 67 1235 218 33.8 60 9 21 37 1.12 P389716 6893774 506200 1870 13.58 30.5 86.7 77 2010 336 80.5 43.5 2.5 36.5 25 1.77 P389715 6893774 506200 1870 13.8 21 57.8 53 1375 249 79 25.5 2.5 23 17 2.08 P389714 6893774 506200 1870 13.44 30 54.4 63 1160 222 134.5 38.5 2 22.5 9 1.64 P389713 6893396 506770 1926 6.16 9 51.6 69 1260 211 33.2 33 2 23.5 38 2.09 P389712 6893396 506770 1926 3.52 8.5 63.2 60 1175 144 26 48 3 14.5 45 1.25 P389711 6893396 506770 1926 22 9 50.9 73 1275 381 43.9 34.5 8.5 54 29 2.12 P389710 6893396 506770 1926 6.7 19 62.9 83 954 256 50.6 50.5 3.5 47 19 1.64 P389709 6893396 506770 1926 5.38 53.5 80.4 80 1360 393 62 70.5 6 53.5 22 1.13 P389708 6893396 506770 1926 3 119.5 46.7 76 893 450 105.5 56 4.5 59.5 8 1.36 P389707 6893396 506770 1926 6.56 14 42.7 77 921 470 236 68.5 2 52.5 4 1.12 P389706 6893396 506770 1930 13.16 12.5 84.5 58 2690 431 31.3 37 3 51.5 86 1.57 P389705 6893518 506643 1935 16.4 12.5 66.3 87 1185 459 66.6 66 3.5 65.5 18 1.32 P389704 6893518 506643 1935 21.9 8.5 69.5 63 1035 327 130 50 3 66 8 1.26 P389703 6893518 506643 1935 15.44 9.5 53.3 86 929 380 27.5 53.5 3 67 34 1.61 P389702 6893518 506643 1935 15.9 17.5 48.9 66 1340 294 43.6 54 1.5 75.5 31 1.22 P389701 6893518 506643 1935 13.74 16.5 83.4 91 1420 477 54.8 69.5 4.5 120 26 1.31 M452700 6899016 505211 1697 2.22 10.5 14 73 376 181 43.4 161 4 21 9 0.45 M452699 6898932 505016 1677 1.86 23 17 64 495 670 73.8 153 6.5 23 7 0.42 M452694 6895397 505527 1527 4.08 2.5 22.1 66 647 97 93.6 22 5 12 7 3.00 M452693 6895397 505527 1527 4.12 3 33.7 74 826 94 622 31 6 14 1 2.39 M452692 6895397 505527 1527 3.12 3.5 35.2 50 950 136 1085 25 3.5 22.5 1 2.00 M452691 6895397 505527 1527 3.22 1.5 20.3 60 538 95 28.1 27 6 20.5 19 2.22 237 Appendix 4. List of whole rock samples of pegmatitic material from the LNPG collected by Wengzynowski (2002). Sample # Northing Easting Elevation (m) Kg Ba (ppm) Cs (ppm) Nb (ppm) Rb (ppm) Sn (ppm) Sr (ppm) Ta (ppm) U (ppm) Zr (ppm) Rb/Sr Nb/Ta M452690 6895444 505588 1503 5.22 11 20.1 47 551 96 23.8 25.5 9 25 23 1.84 M452689 6895444 505588 1503 4.54 2.5 41.9 35 1245 101 14.7 21 11 17 85 1.67 M452688 6895444 505588 1503 4.68 3.5 53.3 17 1545 79 16.4 8.5 8 15.5 94 2.00 M452687 6895444 505588 1503 3 15 53.4 37 1160 138 21.5 32 7 13.5 54 1.16 M452686 6895444 505588 1503 4.18 5.5 57.7 35 1330 156 19.5 18 8.5 10 68 1.94 M452685 6895444 505588 1503 3.98 19 48.8 68 828 200 48.5 41 5.5 12 17 1.66 M452684 6895444 505588 1503 5.44 6 39 75 715 142 24.6 36.5 18 13.5 29 2.05 M452683 6895451 505665 1515 3.34 17 83.2 40 1260 180 627 37.5 7 17 2 1.07 M452682 6895451 505665 1515 3.54 9 52.9 41 1085 369 111.5 24.5 5.5 14.5 10 1.67 M452680 6896356 505921 1617 1.9 41.5 87.5 89 1055 672 107 230 4 44.5 10 0.39 M452679 6896356 505921 1617 2 12 109 57 1560 339 112.5 105 8.5 31.5 14 0.54 M452678 6896356 505921 1617 2.86 7.5 78.2 48 978 420 127 85 6 36.5 8 0.56 M452677 6895851 505006 1676 2.46 9 220 70 2370 412 117.5 133 6.5 36 20 0.53 M452675 6896724 506107 1603 7.28 24.5 72.6 52 2050 65 45.6 42 9 16 45 1.24 M452674 6896568 506035 1685 2.06 23 46.7 70 1125 90 51.5 91 4 23 22 0.77 M452673 6896168 505447 1737 2.62 20 33.7 70 975 217 145 28.5 2.5 14 7 2.46 M452672 6896102 505391 1715 1.82 15 138 79 1590 393 81 99 5.5 41.5 20 0.80 M452671 6896223 505272 1857 1.44 28 109 80 1285 998 175 91.5 6 37 7 0.87 Little Nahanni Pegmatite Group sample set collected and documentedby Wengzynowski (2002). Samples analysed at ALS Chemex, Vancouver. Shaded rows indicate groups of samples from same location Bold denotes subset  used for further analyses in this study 237 Appendix 4. List of whole rock samples of pegmatitic material from the LNPG collected by Wengzynowski (2002). Sample # P389914 P389913 P389912 P389911 P389910 P389909 P389908 P389907 P389906 P389905 P389904 P389903 P389902 P389901 P389875 P389874 P389873 P389872 P389871 P389870 P389869 P389868 P389867 P389866 P389865 P389864 P389863 P389862 P389861 P389860 P389859 P389858 P389857 P389856 P389855 P389854 P389853 P389852 Rock type and comments QM-no S, large segregations of quartz and a few prismatic black oxides QM-no S and a few prismatic black oxides SQ-very thin aplite margins QAS-minor opaques QAS-minor opaques SQF-quartz rich, silver mica at selvage and black minerals disseminated at contact SQF-several aplite zones and 20cm aplite selvage with abundant balck opaques SQF-10cm aplite selvage with clusters of block oxides SQF SQF SQF SQF SQF SQF SQF-coarse grained S rich (white) SQF-coarse grained S rich (white) SQF-coarse grained S rich (white) SQF-coarse grained S rich (white) SQF-coarse grained S rich (white) SQF SQF SQF SQF-with quartz rich zones QA-A mod silver mica in sections SQF-mica in selvage SQF-mica and black minerals at selvage SQFL-quartz banding with adjacent black minerals and 5cm aplite band SQF-some mica and quartz banding SQF-some mica throughout and 10cm L rich selvage SQF-some mica throughout SQFL-black oxides at contact SQFL-soft black oxides throughout SQF-Qm with 5cm aplite on east selvage with black oxides plus minor black oxides in matrix SQF-minor black oxides in center band, no aplite selvages SQF-with 10cm aplite on west selvage containing black oxide SQF-no aplite SQF-no aplite SQF-no aplite 233 Appendix 4. List of whole rock samples of pegmatitic material from the LNPG collected by Wengzynowski (2002). Sample # P389851 P389851 P389849 P389848 P389847 P389846 P389845 P389844 P389843 P389842 P389841 P389840 P389839 P389838 P389837 P389836 P389835 P389834 P389833 P389832 P389831 P389830 P389829 P389828 P389827 P389826 P389825 P389824 P389823 P389822 P389821 P389820 P389819 P389818 P389817 P389816 P389815 P389814 Rock type and comments SQFL-no aplite selvage QFLS-no aplite selvage QML-3cm aplite margin SLQF SLQF SLQF-black oxides at selvage SQF-no aplite QM-abundant black oxides SQF-mod apatite SLQF SQF-mod apatite-no aplite SQF-no aplite SQF-no aplite SQF-no aplite SQF-no aplite SQF-no aplite SQF-no aplite SQF A-mica rich SQF-aplite margins with minor black oxide SQF SQF SQF SQF-20cm aplite zone SQF SQF SQF SQF-no aplite SQF-no aplite SQF SQF SQF SQF QV-sucrosic with mica selvages QV-sucrosic with mica selvages LQF SLQF SLQF 234 Appendix 4. List of whole rock samples of pegmatitic material from the LNPG collected by Wengzynowski (2002). Sample # P389813 P389812 P389811 P389810 P389809 P389808 P389807 P389806 P389805 P389804 P389803 P389802 P389801 P389800 P389799 P389798 P389797 P389796 P389795 P389794 P389793 P389792 P389791 P389790 P389789 P389788 P389787 P389786 P389785 P389784 P389783 P389782 P389781 P389780 P389779 P389778 P389777 P389776 Rock type and comments SLQF SQF SQF SQFL LQF LQF SQF-coarse grained, 2cm aplite developed on both selvages but no oxides SQF-fine grained bluish apatite and no reported aplite selvage SQF SQF-banded aplite zones SQF-S rich with several 2-3 cm aplite bands SQF-vague banding and 20cm wallrock zone QA-banded with several aplite zones and S rich zones A-fine to medium grained with minor spodumene SQF-medium grained, brownish color and no black oxide SQF-S rich SQF-3cm aplite selvages SQF-S rich SQF-1cm clots of lithiophyllite QA-fine grained, S altered to clay SQF-S rich with A ellipsoids, fault gouge and quartz-albite zone SQF-S rich with A ellipsoids SQF-coarse Kspar and S SQF-coarse Kspar and S SQF-coarse Kspar and S SQF-coarse grained with 10cm aplite margin (west) and no black oxides SQF-coarse grained and no black oxides SQF-coarse grained and no black oxides SQF-medium grained with narrow aplite margin SQF-half banded aplite and half coarse S SQF-banded plus one aplite selvage SQF SQF-medium grained SQF-coarse banded with one aplite selvage containing minor black oxide SQF-very coarse S and F across half of dyke SQF-no black oxides SQF-S rich and trace black oxide SQF-S rich and trace black oxide 235 Appendix 4. List of whole rock samples of pegmatitic material from the LNPG collected by Wengzynowski (2002). Sample # P389775 P389774 P389773 P389772 P389771 P389770 P389769 P389768 P389767 P389766 P389765 P389764 P389763 P389761 P389760 P389759 P389758 P389757 P389756 P389755 P389754 P389753 P389752 P389751 P389750 P389749 P389748 P389747 P389746 P389745 P389744 P389743 P389742 P389741 P389740 P389739 P389738 P389737 Rock type and comments SQF-granular to med grained -no black opaques SQF-Coarse quartz -feldspar and minor pink-white mica SQF-banded with 20cm aplite selvage SQF-minor white mica and black opauqes SQF-quartz rich and minor black opaques SQF-quartz rich selvage and alternating aplite bands throughout. Also mod white mica in sections SQF-quartz-mica rich selvages SQF-quartz-mica rich selvages SQF-quartz -mica rich selvages SQF-quartz rich and no black opaqes SQF-abundant banded white mica and no black opaques SQF-mod white mica and no black opaques SQF-mod white mica and minor black opaques SQF-coarse S SQF with banded L zones and narrow aplite margins SQF with 65cm aplite and minor black oxides SQF SQF A-black opaques speckled throughout SQF SQF-multiple narrow aplite zones SQF-white to pink mica common and fine black opaques scattered SQF SQF with 10cm aplite-white to purple mica abundant SQF SQF SQF SQF-5cm mica rich aplite selvage SQF-disseminated siver mica and feldspar clots up to 10cm long SQF-30cm aplitic selvage with siver mica SQF-S up to 10cm long, coarse aplite selvage and silvery mica throughout SQF-S up to 10cm long, coarse aplite selvage and silvery mica throughout SQF-lower portion is lepidolite rich and contains large feldspar clots SQF-mod apatite plus Mn staining. 1cm aplite selvage SQFL-well developed banding-Mn present SQF-abundant lepidolite SQF-abundant lepidolite and lesser apatite. 10cm aplite selvage with mod abundant fine black oxides SQF-abundant lepidolite and lesser apatite 236 Appendix 4. List of whole rock samples of pegmatitic material from the LNPG collected by Wengzynowski (2002). Sample # P389736 P389734 P389732 P389731 P389729 P389727 P389726 P389725 P389724 P389723 P389722 P389721 P389720 P389719 P389718 P389717 P389716 P389715 P389714 P389713 P389712 P389711 P389710 P389709 P389708 P389707 P389706 P389705 P389704 P389703 P389702 P389701 M452700 M452699 M452694 M452693 M452692 M452691 Rock type and comments SQF-abundant lepidolite QM-minor lepidolite in core and mod abundant large black oxide grains throughout SQFL-mod apatite and 3.5cm aplite selvage with pepper size oxide grains QA-with white quartz clots and lesser white S. Mod abundant fine black oxide grains throughout SQF-minor apatite-20cm aplite selvage with 1% fine disseminated black oxide SQF-mod apatite and 2cm aplite QFL-abundant apatite and lesser S. 10cm aplite selvage but no oxides A-selvages marked by silver-white mica. Matrix contains minor S SQF-aplite zones within main body containing grey quartz and minor black flecks-no aplite selvage SQF-abundant silver-white mica and quartz SQF-quartz rich, mod apatite with Mn staining and patches of lithiophyllite. SQF-quartz rich, mod apatite with Mn staining and patches of lithiophyllite. 12cm aplitie selvage with dull black specs QA-minor S, apatite and some lithiophyllite stringers QA-minor S and apatite SQF-strongly aplitic and containing abundant grey quartz. Some areas are peppered with fine black flecks SQF-strongly aplitic and containing abundant dull black lithiophyllite stringers SQF-moderate Mn staining and 10cm aplite at 5.0m containing fine black flecks SQF-moderate Mn staining and 10cm aplite at 4.0m containing fine black flecks QA-A-minor S SQF-S rich and mod apatite SQF-quartz+S rich SQF-interbanded aplite with fine felted tourmaline SQF-sucrosic aplite selvages A-80%, SQF-20% moderate black specks A-minor S ,trace L and abundant dark specks QA-minor S-1mm black fleck SQF-35% S and mod apatite -10cm aplite on both selvages containing trace to 2% fine black flecks SQF with a 34cm and 28cm alitic selvage contianing minor to moderate oxides SQF with 15cm aplite zone containing pepper size black oxide SQF SQF with 10cm aplite selvage, oxides present in aplite SQF-coarse S-no black minerals SQF-mod silver mica and Mn films throughout-no pronounced aplite selvage SQF-20cm aplite selvage with minor black flecks SQF-S rich with abundant Mn films SQF-S rich with abundant Mn films SQF-S rich with abundant Mn films SQF-S rich with abundant Mn films 237 Appendix 4. List of whole rock samples of pegmatitic material from the LNPG collected by Wengzynowski (2002). Sample # M452690 M452689 M452688 M452687 M452686 M452685 M452684 M452683 M452682 M452680 M452679 M452678 M452677 M452675 M452674 M452673 M452672 M452671 Little Nahanni P Shaded rows ind Bold denotes su Rock type and comments SQF-S rich with 20% brown-gold aplite bands SQF-S rich with 20% brown-gold aplite bands SQF-S rich with 20% brown-gold aplite bands SQF-S rich mod abundant siver mica throughout and a 5cm shale parting SQF-S rich SQF-S rich with lithiophyllite scattered in some sections SQF-S rich SQF-S rich SQF-S rich SQF-minor apatite-no aplite selvage SQF-minor apatite-no aplite selvage SQF-minor apatite-no aplite selvage SQFL-sivery mica booklets throughout and 6cm aplite selvage marked by white and glassy quartz SQF-10cm aplite develped on footwall portion of dyke containing silver-gold mica SQF-coarse grained with white and clear S QA-mostly coarse grained with strongly aplitic sections SQF SQF-selvage aplite not well developed Little Nahanni Pegmatite Group sample set collected and documentedby Wengzynowski (2002). Samples analysed at ALS Chemex, Vancouver. Shaded rows indicate groups of samples from same location Bold denotes subset  used for further analyses in this study 238 δ 7Li values average low high n References Whole rock samples and reference materials LNPG pegmatites2 -0.9 11.4 28 This study Simple pegmatites (Black Hills)2 1.4 7.3 4 Teng et al . 2006 Tin Mountain pegmatite (wall zone) 7.5 11.1 3 Teng et al . 2006 S-type Regional Cretaceous intrusions -0.4 2.2 5 This study S-type granite (Lachlan Fold Belt, Mount Flakney) -1.1 Teng et al . 2004 S-type granites -0.4 2.1 9 Bryant et al . 2004 S-type granite (Harney Peak) -3.1 6.6 25 Teng et al . 2006 G-2 granite 0.1 -0.5 0.5 6 This study G-2 granite 1.2 0.9 1.5 James & Palmer 2000 G-2 granite -0.3 3 Pistiner & Henderson 2003 G-2 granite -1.2 Teng et al . 2004 Macusani peraluminous glass (Mac) -0.7 -1.2 -0.1 5 This study Macusani peraluminous glass (TomMac) -1.7 -2.5 0.3 4 This study Macusani peraluminous glass 1.2 2.3 Tomascak et al . 1999 RGM-1 5.7 1 This study RGM-1 2.6 6 Schuessler et al.  2009 BCR-2 7.1 1 This study BCR-2G1 4.1 10 Kasemann et al. 2005 Pegmatitic minerals Spodumene (P389859, LNPG) 3.5 3.1 3.8 4 This study Spodumene (AL050704, LNPG) 3.7 3.7 3.7 2 This study Spodumene (Tin Mountain)3 8.1 7.9 8.3 4 Teng et al . 2006 Spodumene (Low-F Kolmozero) 6.7 6.9 Gordienko et al.  2007 Spodumene (High-F Voron'ya Tundra) 12.7 15.1 Gordienko et al . 2007 Li-rich muscovite (EB0509, LNPG) 2.2 1.8 2.6 3 This study Li-rich muscovite (EB1304e, LNPG) 7.9 7.6 8.3 2 This study Li-rich muscovite (Tin Mountain)3 10.4 8.5 11.4 11 Teng et al . 2006 Muscovite (Tin Mountain) 16 Tomascak et al . 1995 Lepidolite (High-F Voron'ya Tundra) 20.3 1 Gordienko et al.  2007 Plagioclase, albite (EB0509, LNPG) 3.4 3.7 3.0 2 This study Plagioclase, albite (EB1304e, LNPG) 7.9 7.2 8.5 2 This study Plagioclase (Tin Mountain)3 9.2 8.8 9.9 7 Teng et al . 2006 Plagioclase, albite (Tin Mountain)4 8.9 7.1 11.1 5 Tomascak et al . 1995 Quartz (EB0509, LNPG) 15.7 15.5 15.8 2 This study Quartz (EB1304e, LNPG) 8.7 8.2 9.3 3 This study Quartz (Tin Mountain)3 17.8 14.7 21.3 12 Teng et al . 2006 Quartz (Tin Mountain) 19 2 Tomascak et al . 1995 Holmquistite (Low-F Kolmozero) 5.3 Gordienko et al . 2007 Holmquistite (High-F Voron'ya Tundra) 8.8 Gordienko et al.  2007 Tourmaline (Cryo-Genie, Little Three and Himalaya)2 11.2 22.9 16 Maloney et al.  2008 1 glass material 2 samples taken from several pegmatites 3 minerals sampled from several locations within the Tin Mountain pegmatite (e.g. core vs.  wall) 4 originally reported as δ 6Li values. Converted to δ 7Li values by sign change ( - to +), therefore values are approximate Appendix 5. Compilation of δ7Li values from this study and from similar rock types, minerals and standards reported in the literature. 240 P389765, dike width ~ 80 cm (507267E, 6892868N, 1872 m elev.) a l  W l ro ck a co nt ct all rock W  c ntact o 1 2 3 4 5 6 7 8 10 11 12 13 14 15 17 18 19 20 1 2 3 4 5 6 7 8 10 11 12 13 14 15 17 18 19 20 9 16 9 2 cm Grey=feldspar megacrysts   Purple=increased mica content, commonly plumose  Orange=increased spodumene content Aligned crystal orientation Plumose mica Bimodal grain size Spodumene Plumose mica Spodumene Spherulitic crystal growth -mica and feldspar Feldspar megacrysts Aligned crystal orientation 16 Bimodal grain size Appendix 6.Composite annotated image of thin sections (cross-polarised) cut across the width of a ~80 cm width dike (using cross-polarized light), sample P389765. 241 EB04 EB0509a EB1007 EB1304e EB1305a-21 EB1305c EB1306 P389764 P389768 P389768a P389774 P389774b P389797 P389799-1 P389799-2 P389870 Approximate 10% enlargement spherulitic crystal growth plumose mica plumose mica wall rock sutured quartz grain contacts axc is megacrystic mica bimodal crystal sizes spodumene plumose mica plumose mica aligned mineral growth bimodal grain size spodumene spherulitic crystal growth Appendix 7. Annotated images of thin sections (photographed using cross-polarized light) from selected LNPG samples displaying various textural and mineralogical details. 242

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
China 23 33
Canada 14 3
United States 6 0
Austria 1 0
City Views Downloads
Beijing 23 0
Vancouver 8 1
Seattle 4 0
Greater Sudbury 2 0
North Vancouver 2 0
Ottawa 2 2
San Bruno 1 0
Villach 1 0
Ashburn 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items