Lithos 119 (2010) 34–46

Contents lists available at ScienceDirect

Lithos

j ourna l homepage: www.e lsev ie r.com/ locate / l i thos

Elemental and Sr–Nd isotope geochemistry of microgranular enclaves fromperalkaline A-type granitic plutons of the Emeishan large igneous province,SW China

J.G. Shellnutt a,⁎, B.-M. Jahn a, J. Dostal b

a Academia Sinica, Institute of Earth Science, 128 Academia Road Sec. 2, Nankang Taipei 11529, Taiwanb Saint Mary's University, Department of Geology, 923 Robie Street, Halifax, NS, B3H 3C3, Canada

⁎ Corresponding author. Tel.: +886 2 2783 9910x618E-mail address: jgshelln@earth.sinica.edu.tw (J.G. Sh

0024-4937/$ – see front matter © 2010 Elsevier B.V. Adoi:10.1016/j.lithos.2010.07.011

a b s t r a c t

a r t i c l e i n f o

Article history:Received 3 October 2009Accepted 17 July 2010Available online 24 July 2010

Keywords:PeralkalineA-type granitoidEnclaveEmeishan large igneous provinceAutolithFractional crystallization

Microgranular enclaves are common within intermediate to felsic granitic rocks that have I- and S-typeaffinity however they are rare within alkaline anorogenic granitoids of A-type affinity. The Permian(~260 Ma) Emeishan large igneous province (ELIP) of southwest China contains two peralkaline silicasaturated A-type granitic plutons that host microgranular enclaves. The enclaves from the Baima pluton areintermediate in composition and have lower SiO2 and higher TiO2, CaO and Mg# (SiO2=57.2 to 63.0 wt.%;TiO2=0.8 to 1.8 wt.%; CaO=1.7 to 3.3 wt.%; Mg#=28 to 44) than their host (SiO2=62.6 to 67.8 wt.%;TiO2=0.5 to 1.4 wt.%; CaO=0.4 to 1.8 wt.%; Mg#=15 to 31). The enclaves from the Taihe pluton are morefelsic (SiO2=63.8 to 71.3 wt.%; TiO2=0.3 to 0.6 wt.%; CaO=0.6 to 2.3 wt.%; Mg#=8 to 22) but are still lessevolved than their host (SiO2=69.8 to 75.1 wt.%; TiO2=0.2 to 0.6 wt.%; CaO=0.4 to 0.8 wt.%; Mg#=3to 12). In both cases, the enclaves have very similar εNd(T) values (Baima εNd(T)=+2.8 to +3.2; TaiheεNd(T)=+1.0 to +2.0) as their hosts (Baima εNd(T)=+3.0 to +3.2; Taihe εNd(T)=+1.5 to +1.9). Themajor and trace element trends of the enclave-host pairs suggest that fractional crystallization occurred andthat element diffusion was likely minimal. The enclaves are interpreted as entrained accumulations of earlyformed crystals of a silicic magma which was originally produced by fractional crystallization of a maficmagma.

; fax: +886 2 2783 9871.ellnutt).

ll rights reserved.

© 2010 Elsevier B.V. All rights reserved.

1. Introduction

Microgranular enclaves are dark, fine grained, rounded to elongateinclusions which are common within intermediate to felsic igneousrocks. The study of enclaves offers potentially important informationon the origin and history of the magmatic systems in which theyare found (Didier, 1973, 1987; White and Chappell, 1977; Cantagrelet al., 1984; Vernon, 1984; Bacon, 1986; Vernon et al., 1988; Barbarinand Didier, 1991; Waight et al., 2001; Barbarin, 2005; Garcia-Morenoet al., 2006; Ventura et al., 2006). Enclaves encompass a broad rangeof textures, structures and compositions and are often characterizedaccording to their structural relationship to the host rock. They areoften interpreted as xenoliths, co-mingled magmas, restites orautoliths (Holland, 1900; White and Chappell, 1977; Vernon, 1983;Vernon et al., 1988; Bonin, 1991; Barbarin and Didier, 1991). Xenolithsare foreign lithic fragments, usually country rock, that were incorpo-rated during emplacement and/or crystallization of the host magma(Domenick et al., 1983; Vernon, 1983; Bacon, 1986). Xenoliths are

comparatively easy to identify because they often have magmaticreaction textures. Some enclaves are interpreted as mingled globulesof magma that were entrained while the host was still partially liquid(Cantagrel et al., 1984; Vernon, 1984; Bacon, 1986; Barbarin, 2005). Incontrast, restites are considered to represent pods of the originalrefractorymagma composition of the host granite which did not reacha critical melt fraction during differentiation (White and Chappell,1977; Chappell and White, 1991). Some enclaves or ‘autoliths’ areinterpreted to represent accumulations of early formed genetically-related crystals that were trapped within its own residual liquid(Fershtater and Borodina, 1977; Jones, 1979; Tindle and Pearce, 1983;Ridolfi et al., 2006; Schonenberger et al., 2006). Although enclaves areubiquitouswithin peraluminous andmetaluminous granitic rocks of I-and S-type affinity, they are rarely found within peralkaline graniticrocks of A-type affinity (Bonin, 1991; Barbarin, 1999).

The examination of enclaves from peralkaline A-type granitoidsmayprovide valuable insight into a rare phenomenon and also assist inunderstanding the chemical evolution of their magma systems. TheLate Permian Emeishan large igneous province (ELIP) of southwestChina, contains at least three (i.e. Baima, Taihe and Panzhihua)peralkaline silica saturated A-type granitic plutons that formed byfractional crystallization of mafic magmas (Shellnutt and Zhou, 2007;

35J.G. Shellnutt et al. / Lithos 119 (2010) 34–46

Shellnutt et al., 2009; Shellnutt and Jahn, 2010). The Baima and Taiheperalkaline plutons contain microgranular enclaves whereas thePanzhihua pluton does not. In this paper we present new major andtrace element data and Sr–Nd isotopic data of the enclaves from theBaima and Taihe plutons in order to understand the formation ofenclaves within peralkaline silica saturated A-type granitoids.

2. General geology

The Late Permian ELIP is located in a region covering an area of~0.3×106 km2 in the western part of the Yangtze Block, eastern partof the Songpan–Ganze terrane and northern part of Vietnam (Fig. 1).The ELIP was modified by Mesozoic and Cenozoic faulting associatedwith the eastward extrusion of the Tibetan Plateau during the SouthChina Block–North China Block collision (~220 Ma) and again duringthe Indo-Eurasian collision (~50 Ma). The ELIP consists of floodbasalts, trachytes, felsic plutons and mafic-ultramafic intrusionswhich host either Ni–Cu–(PGE) sulfide deposits or giant Fe–Ti–Voxide deposits (Ali et al., 2005; Zhou et al., 2008). The age of the ELIP isconstrained to the Late Permian with radiometric ages ranging from~260 Ma to ~251 Ma (Zhou et al., 2002; Ali et al., 2005; Shellnutt et al.,2008). Previous interpretations suggest that ELIP-magmatism isrelated to partial melting of a mantle plume source (Chung andJahn, 1995; Chung et al., 1998; Xu et al., 2001; Ali et al., 2005).

Fig. 1. Simplified geological map of the Panxi region showing the lo

Granitic plutons of A-type affinity are located between the cities ofPanzhihua and Xichang, also known as the Panxi region (Shellnutt andZhou, 2007; Zhong et al., 2007) (Fig. 1). The plutons are distributed atintervals of tens of kilometers along a narrow belt ~100 km wide and~200 km long. The regional distribution trend of the plutons changesfrom northeast–southwest to north–south near the town of Miyi. Theplutons are of peralkaline, peraluminous andmetaluminous composi-tions and are spatially and temporally associated with Fe–Ti oxideore-bearing layered gabbroic intrusions (Shellnutt and Zhou, 2007;Zhong et al., 2007). The peralkaline plutons (e.g. Baima, Taihe,Panzhihua) are genetically related to spatially associated layeredgabbroic intrusions and are considered to represent the residualproducts of fractional crystallization (Shellnutt et al., 2009; Shellnuttand Jahn, 2010). The enclaves from the Baima pluton were observedand collected from one locality (i.e. western outcrop) (Fig. 2). Incontrast, enclaves are common throughout the Taihe pluton and werecollected at two localities (Fig. 3). Enclaves were not observed withinthe Panzhihua pluton.

3. Description of the enclaves

Enclaves from the Baima peralkaline pluton were observed at oneoutcrop (26°57′17″N, 102°02′44″) (Fig. 2). The enclaves are abundantbut represent b5% of the total exposure. They vary in size and shape

cations of the Panzhihua, Baima and Taihe peralkaline plutons.

Fig. 2. Simplified geological map for of the Baima area (modified from Wang et al., 1994 and Xiong et al., 1996).

36 J.G. Shellnutt et al. / Lithos 119 (2010) 34–46

but are generally a few centimeters to 10 s of centimeters in diameteror have a 2:1 or 3:1 length to width ratio with a maximum length of~20 cm (Fig. 4a, b). The enclaves are medium to fine grained and tendto be elongate, lobate or rounded rather than angular and are easilyidentified by their colour contrast with their host. The colour contrastgives an appearance of a sharp mineralogical boundary but there aremany minerals which are shared between the host and enclave. It isclear from the hand specimens that the enclaves contain moreamphibole and biotite and less quartz than the host.

Fine grained enclaves are common within the Taihe pluton andwere collected at two localities (Fig. 3). Most enclaves are rounded toelongate and typically ≤7 cm in diameter although there are a fewwhich are ~10 cm (Fig. 4c, d). The enclaves are easily spotted due totheir colour contrast with the host rock although there are manyenclaves which appear to be more leucocratic than melanocratic. It iscommon to observe a few coarse grained crystals of alkali feldsparwithin the enclaves (Fig. 4c).

There is no evidence to suggest that the enclaves in the Baima orTaihe plutons are trapped lithic fragments or were dismembered wall

rock fragments because there is no contact aureole, cross-cuttingveins and the mineralogy is very similar to the host. The enclaveslikely represent mineral accumulations and are best described aseither mafic microgranular enclaves or felsic microgranular enclaves(Didier and Barbarin, 1991).

4. Petrography

4.1. The Baima pluton

The coarse-grained syenites consist of 65–70% perthitic alkalifeldspar, 10% to 15% ferrorichterite, ≤10% quartz, ≤5% aegirine withabout 5% of apatite, titanite, zircon, plagioclase, fluorite, biotite,hematite, magnetite, ilmenite and pyrite combined. The alkali feldsparis typically large (≤1.2 cm), subhedral to anhedral and usuallyseparated from each other by an intergranular mixture of small(b0.3 cm), subhedral grains of K-feldspar, quartz and accessoryplagioclase. Ferrorichterite is euhedral to subhedral, up to 0.5 cmlong, and is commonly associated with subhedral to anhedral crystals

Fig. 3. Simplified geological map of the Taihe gabbro–granite complex (modified from Wang et al., 1993).

37J.G. Shellnutt et al. / Lithos 119 (2010) 34–46

of aegirine. Agglomerates of ferrorichterite, titanite and apatite arecommon. Titanite, although modally minor (≤1%), is commonlyeuhedral but there are many crystals which are equant or subhedral.The euhedral shapes of some titanite suggest that they may havecrystallized early. Other less abundant accessory phases are euhedralbiotite, small (b0.5 mm), euhedral apatite, fluorite, anhedral magne-tite, hematite and ilmenite. The assemblage of magnetite–quartz–titanite is indicative of an oxidized magma (Wones, 1989).

The enclaves from the Baima pluton are fine to medium grainedand are composed of alkali feldspar, biotite, ferrorichterite, titanite,zircon, apatite and accessory amounts of quartz, aegirine, fluorite,plagioclase and Fe–Ti oxideminerals. The alkali feldspars are perthitic,fine to medium grained, equant to subrounded and comprise ~60–70% of the mode. Megacrysts of alkali feldspar are common and are~1.5 cm long and ~0.5 cm wide (Fig. 4e). The megacrysts haveundulating boundaries with the surrounding fine grained mineralsand have biotite, amphibole and apatite inclusions (Fig. 4f). Ferror-ichterite is the most abundant mafic mineral and comprises ~15% ofthe mode. The crystals are anhedral to subhedral and typically finegrained although there are a few coarse grained phenocrysts. Mostamphiboles are interstitial to the feldspars. Commonly the amphi-boles, as well as biotite, titanite, apatite and fluorite, form clustersor agglomerates. Biotite is euhedral to subhedral, typically fine tomedium grained and can comprise up to 10% of the mode (e.g. BM-1)but it is usually less. Zircon inclusions in biotite were not identified.The presence and quantity of biotite in the enclaves is notablydifferent from the host rock as biotite is absent or is an accessorymineral. Anhedral to subhedral titanite is common but comprises lessthan 2% of the mode. The titanites vary in size and appear to be moreabundant than in the host rock. Accessory amounts of fine grainedequant to acicular apatite, zircon, anhedral oxide minerals, anhedralto subhedral quartz, subhedral fluorite and rarely anhedral aegirineare observed.

4.2. The Taihe pluton

Granites from Taihe are granular and coarse grained and consistmostly of alkali feldspar, quartz and amphibole with accessoryamounts of titanite, aenigmatite, apatite, fluorite, ilmenite andaegirine. The alkali feldspar is euhedral to subhedral with fine, linear

exsolution lamellae. The crystals are typically 1–2 cm in length andcomprise ~65–70% of bulk mineralogy. The quartz is ≤0.2 cm and issubhedral and interstitial to the feldspars and comprises 20–25% ofthe mode. The amphibole (e.g. ferrorichterite) are subhedral andinterstitial to the feldspars and contain apatite inclusions. Someferrorichterite are partially altered to riebeckite/arfvedsonite and/orrutile. Small amounts (b1%) of subhedral aegirine, zircon, aenigma-tite, apatite and fluorite are also present. Ilmenite is rare or absent.

The enclaves consist primarily of alkali feldspar, quartz and Fe–Naamphibole (i.e. ferrorichterite and katophorite). The alkali feldsparcomprises ~70% of mode and is typically medium to fine grained withperthitic exsolution lamellae. The crystals are usually euhedral tosubhedral with rounded edges. There are a few megacrysts withinclusions of amphibole but they are generally inclusion-free. Theamphiboles are fine to medium grained with subhedral to anhedralshapes and interstitial to the alkali feldspar. Two types of amphibolesare identified on the basis of their colour and pleochroism andcomprise 10% to 25% of the mode. Ferrorichterite is predominant andgreenish-blue whereas katophorite tend to be larger and reddish-brown. Quartz varies considerably from b5% to 20% of themode and ismedium to fine grained and interstitial to the alkali feldspars andamphiboles. The crystals are typically subhedral to anhedral. The onlyaccessorymineral observedwas apatite. In at least one case (i.e. TH-2)there was a distinct texture of fine grained acicular amphibole withina very fine matrix of alkali feldspar.

5. Analytical methods

5.1. Major and trace elemental analyses

Major elements were analyzed by X-ray fluorescence spectroscopyat the Regional Geochemical Centre, Saint Mary's University, Halifax,Canada, using glass discs. Trace elements were analyzed by induc-tively coupled plasma mass spectrometry (ICP-MS) at NationalTaiwan University, Taipei, Taiwan. Standard reference materials forthe trace element analyses are BHVO-2 and BIR-1. The precision forthe major elements are better than 2% and for trace elements from 2%to 10%, depending on individual elements. The measures standardreference material values are listed in Table 1.

Fig. 4. Photographs of enclaves from Baima (a) and (b) and Taihe (c) and (d). Photomicrographs from BM-1 of the agglomerates of apatite, amphibole, titanite and biotite (e) andalkali feldspar megacrysts with biotite amphibole inclusions (f).

38 J.G. Shellnutt et al. / Lithos 119 (2010) 34–46

5.2. Rb–Sr and Sm–Nd isotopic analyses

Approximately 75–100 mg of whole rock powder was dissolved ina mixture of HF–HClO4 for Sr–Nd isotopic analysis. Strontium andREEs were separated on polyethylene columns with a 5 ml resin bedof AG 50 W-X8, 200–400 mesh. Strontium was further purified bypassing through the same column an addition time. Neodymium wasisolated from other REEs on polyethylene columns using Ln resin as acation exchange medium. For the isotopic measurement, Sr wasloaded on a single W filament with H3PO4 and Nd was loaded with

H3PO4 and measured using a Re-double-filament configuration. 87Sr/86Sr ratios were normalized to 86Sr/88Sr=0.1194 whereas 143Nd/144Nd ratios were normalized to 146Nd/144Nd=0.7219. The 87Sr/86Srisotopic ratios were measured using a Finnigan MAT-262 thermalionization mass spectrometer (TIMS), whereas the 143Nd/144Ndisotopic ratios were measured using a Finnigan Triton TIMS in theMass Spectrometry Laboratory, Institute of Earth Sciences, AcademiaSinica, Taipei. The 2σm values for all samples are less than 0.000018for 87Sr/86Sr and less than 0.000009 for 143Nd/144Nd (Table 2). Thestandard reference materials NBS987 and JMCNd were used and give a

Table 1Whole rock and trace elemental results from enclaves of the Baima and Taihe plutons.

Sample BM-1 BM-2 BM-4 BM-5 TH-2 TH-3 TH-4 TH-6 TH-7 TH-8 BaimaAVG s.d. TaiheAVG s.d. BHVO-2 s.d. BIR-1 s.d.Pluton Baima Baima Baima Baima Taihe Taihe Taihe Taihe Taihe Taihe (11) (24) (3) (2)

SiO2 (%) 58.12 62.94 60.08 57.20 63.77 69.14 67.66 66.44 71.03 67.42 64.99 1.3 72.27 1.4TiO2 1.46 0.77 1.03 1.80 0.58 0.30 0.42 0.49 0.26 0.49 0.69 0.1 0.39 0.1Al2O3 16.19 16.98 16.36 14.18 13.32 13.42 11.76 12.82 12.04 12.92 15.62 0.6 11.10 0.6Fe2O3t 6.05 4.23 5.38 7.99 7.15 3.55 6.13 6.65 4.00 5.52 4.02 0.7 5.26 1.2MnO 0.22 0.19 0.22 0.37 0.22 0.10 0.19 0.22 0.11 0.13 0.17 0.6 0.14 0.0MgO 2.45 0.84 1.79 2.61 0.56 0.19 0.27 0.41 0.18 0.81 0.53 0.1 0.13 0.0CaO 3.28 1.70 3.04 3.17 2.26 0.61 1.36 1.68 0.91 1.44 0.80 0.3 0.55 0.1Na2O 7.17 7.52 7.38 6.59 6.24 5.38 5.79 6.02 5.64 5.33 6.64 0.4 4.50 0.4K2O 3.88 4.66 4.22 4.16 5.34 5.68 5.29 4.57 5.39 5.11 4.99 0.3 4.65 0.2P2O5 0.42 0.20 0.35 0.47 0.14 0.04 0.06 0.11 0.03 0.08 0.12 0.1 0.02 0.0LOI 0.99 0.40 0.69 1.65 0.98 0.60 1.00 0.69 0.39 0.80 0.67 0.2 0.40 0.1TOTAL 100.23 100.43 100.53 100.19 100.55 99.01 99.93 100.10 99.98 100.04 99.24 0.3 99.41 0.5ASI 0.74 0.83 0.74 0.67 0.66 0.72 0.72 0.76 0.72 0.76 0.89 b0.1 0.83 0.1Na+K/Al 0.99 1.03 1.02 1.08 1.20 1.12 1.30 1.16 1.26 1.11 1.05 b0.1 1.12 0.1Sc (ppm) 35 36 36 35 40 40 40 39 40 40 12 1.0 11 4.2 34 0.6 44 1.2V 85 45 74 87 59 38 43 47 38 62 90 51 2 0.7 309 3.1 316 1.4Cr 82 2 39 79 6 3 6 5 4 12 3 3.2 3 1.7 277 2.5 389 3.5Co 9 2 6 8 2 1 1 2 1 4 1 0.5 1 0.1 43 0.1 52 0.1Ni 49 2 21 35 3 2 4 3 2 8 6 3.2 5 7.3 112 1.1 165 0Cu 15 7 15 19 13 7 19 8 6 16 4 1.3 22 30.3 131 0.5 119 1.6Zn 152 160 170 263 282 185 250 300 126 276 143 24.4 220 44.3 105 1.6 72 2.9Ga 30.6 34.3 31.8 30.0 28.5 32.4 34.4 33.3 28.6 29.8 36 3.2 34.0 3.7 21.6 0.3 16.4 0.1Rb 107 136 127 165 212 225 198 181 221 216 126 17.5 131 15.2 9 0.4 0.5 b0.1Sr 756 387 678 416 80 31 47 78 17 53 105 91.1 14 4.1 385 13.7 109 1.5Y 51 59 59 78 72 145 88 252 87 64 47 17.4 100 37.0 26 0.7 16 0.2Zr 601 729 647 834 394 243 462 382 443 462 873 339.7 696 216.1 166 2.6 15 0.1Nb 99 139 105 173 67 36 71 74 42 70 121 27.0 103 14.1 19 0.1 0.6 0Cs 0.72 0.27 0.94 1.27 0.49 0.48 2.08 0.96 0.45 12.4 0.31 0.1 0.56 0.2 0.10 b0.1 0.04 b0.1Ba 1260 834 1189 898 879 510 298 814 248 502 371 179.9 266 93.1 120 4.4La 89 85 90 109 114 335 146 229 147 106 125 112.6 159 120.3 14.8 0.4 0.64 b0.1Ce 193 199 199 248 249 130 256 297 168 146 269 212.5 290 94.2 36.4 0.7 1.91 b0.1Pr 24.6 26.7 25.9 33.3 32.9 88.4 39.2 59.4 36.3 28.9 29.7 17.6 35.2 20.4 5.3 0.1 0.39 b0.1Nd 94.2 102.3 99.9 130.2 127 328 149 237 137 109 109 56.6 130.5 74.6 24.1 0.6 2.46 b0.1Sm 17.2 19.8 19.0 24.9 23.5 59.4 26.9 47.4 24.9 19.2 18.8 7.2 24.8 12.9 6.1 0.2 1.15 b0.1Eu 5.2 4.3 5.1 4.9 4.16 8.46 3.52 8.08 3.28 2.75 2.91 1.0 3.49 1.8 2.0 0.1 0.52 b0.1Gd 15.73 18.0 17.4 22.5 21.8 48.0 25.0 50.5 23.9 17.2 14.1 5.5 22.10 8.8 6.2 0.1 1.69 b0.1Tb 1.92 2.33 2.19 2.90 2.77 6.33 3.19 7.463 3.26 2.18 2.22 0.8 3.47 1.3 0.9 b0.1 0.35 b0.1Dy 10.19 12.56 11.46 15.52 15.0 32.6 17.2 44.1 17.8 11.6 11.0 4.0 19.93 6.1 5.2 0.1 2.6 b0.1Ho 1.92 2.37 2.17 2.97 2.91 5.80 3.40 8.75 3.44 2.34 2.00 0.7 3.95 1.0 1.0 b0.1 0.61 b0.1Er 5.04 6.04 5.59 7.71 7.78 14.3 9.27 21.9 8.49 6.22 5.5 1.9 11.88 2.4 2.5 b0.1 1.73 b0.1Tm 0.66 0.78 0.74 1.04 1.13 1.78 1.33 2.85 1.09 0.87 0.74 0.2 1.67 0.3 0.3 b0.1 0.26 b0.1Yb 4.03 4.74 4.42 6.49 7.69 10.8 8.89 16.3 6.99 5.58 4.95 1.4 10.78 1.8 2.0 b0.1 1.66 b0.1Lu 0.59 0.70 0.66 0.98 1.28 1.63 1.46 2.43 1.19 0.85 0.75 0.2 1.60 0.3 0.3 b0.1 0.26 b0.1Hf 13.1 14.4 13.9 18.5 9.59 6.01 11.2 9.16 11.1 11.1 20.9 8.3 19.4 5.6 4.2 b0.1 4.06 b0.1Ta 6.5 8.7 6.6 9.5 4.0 2.3 4.5 3.9 3.7 4.8 7.4 2.2 6.7 1.1 1.3 b0.1 0.04 b0.1Th 9.0 5.8 6.9 12.2 7.9 4.1 8.5 7.8 11.4 9.7 10.2 6.9 19.0 3.5 1.2 b0.1 0.03 b0.1U 2.5 1.8 1.9 2.0 1.3 1.0 2.5 2.0 2.5 2.1 2.0 0.9 4.3 0.9 0.4 b0.1 0.01 b0.1Eu/Eu* 0.91 0.67 0.82 0.60 0.53 0.45 0.40 0.49 0.39 0.44 0.54 0.1 0.44 0.1(La/Yb)N 16.9 13.5 15.2 12.3 10.2 22.6 11.5 10.4 14.4 13.4 18.8 14.4 10.3 5.9

LOI=loss on ignition; ASI=Al/Ca–1.67P+Na+K; Eu/Eu*=[2*EuN/(SmN+GdN)]; N=chondrite normalized to values of Sun and McDonough (1989). s.d. = standard deviation.BaimaAVG and TaiheAVG are the averaged values of the host plutons. The number in parentheses is the amount of averaged samples.

39J.G. Shellnutt et al. / Lithos 119 (2010) 34–46

mean 87Sr/86Sr value of 0.710248±0.000005 and 143Nd/144Nd valueof 0.511813±0.00001 respectively.

6. Results

6.1. Major and trace elements

Major and trace element results for the enclaves are listed inTable 1. The host rock data can be found in Shellnutt and Zhou (2007)and Shellnutt et al. (2009). The Baima enclaves have aluminumsaturation indices (ASI=Al/Ca–1.67P+Na+K) which are metalumi-nous to weakly peralkaline (ASI=0.69 to 0.84; Na+K/Al=0.99 to1.08), similar to their host (ASI=0.83 to 92; Na+K/Al=1.02 to 1.09).The Fe* (FeOt/MgO+FeOt) values (Fe*=0.69 to 0.82) overlap withtheir host (Fe*=0.85 to 0.91) but generally the enclaves are moremagnesian and are classified as alkalic (Fig. 5). The Baima enclaves are

silica understaturated as indicated by normative nepheline (Ne)whereas the host is quartz-normative. The normative compositionswere calculated assuming an Fe2O3/FeO ratio of 0.5 (Middlemost,1989). Sample BM-2 of the Baima enclaves has the lowest normativenepheline composition (Ne=1.6 wt.%) whereas sample GS04-006has the lowest normative quartz composition (Qz=2.8 wt.%) formthe host. The Taihe enclaves, similar to their host, are ferroan andweakly peralkaline (ASI=0.67 to 0.84; Na+K/Al=1.11 to 1.30;Fe*=0.86 to 0.95) and classify as alkalic to alkali-calcic according tothe modified alkali-line index (MALI) of Frost et al. (2001). All of theTaihe enclaves and host rocks are quartz normative.

The trace element compositions of the Baima and Taihe enclavesare somewhat similar (e.g. V, Cu, Th, U) but their concentration of Sc,Cr, Co, Ni, Sr, Ba, Rb, and Zr are slightly different. The Baima enclaveshave higher Sr (387 to 756 ppm) and Ba (834 to 1260 ppm) but lowerRb (107 to 165 ppm) and Zr (601 to 834 ppm) than the Taihe enclaves

Table 2Whole rock Sr and Nd isotopic data for the Baima and Taihe enclaves.

Sample Pluton Rb(ppm)

Sr(ppm)

87Rb/86Sr

87Sr/86Sr

±2sm

87Sr/86Srinitial(260Ma)

Inducederror inISr

ModelAge (Ma)I=0.704

Sm(ppm)

Nd(ppm)

147Sm/144Nd

143Nd/144Nd

±2sm

εΝδ(0) εΝδ(260 Μα)

f(Sm/Nd)

TDM-1

BM-1 Baima 107 756 0.409 0.70561 7 0.70410 0.00003 277 17.2 89 0.1108 0.512655 8 0.3 +3.2 −0.44 734BM-2 Baima 136 387 1.017 0.70748 16 0.70371 0.00008 240 19.8 85 0.1179 0.512660 7 0.4 +3.0 −0.40 780BM-4 Baima 127 678 0.542 0.70600 13 0.70399 0.00004 259 19 90 0.1155 0.512653 5 0.3 +3.0 −0.41 772BM-5 Baima 165 416 1.148 0.70788 18 0.70364 0.00008 238 24.9 109 0.1156 0.512643 7 0.1 +2.8 −0.41 788TH-2 Taihe 212 80 7.691 0.72718 10 0.69873 0.00057 212 23.5 114 0.1110 0.512574 5 −1.2 +1.6 −0.44 855TH-3 Taihe 225 31 20.959 0.75384 13 0.67632 0.00155 167 59.4 335 0.1097 0.512575 5 −1.2 +1.6 −0.44 843TH-4 Taihe 198 47 12.187 0.77154 15 0.72646 0.00090 389 26.9 149 0.1103 0.512591 6 −0.9 +2.0 −0.44 824TH-6 Taihe 181 78 6.752 0.72710 13 0.70213 0.00050 241 47.4 237 0.1212 0.512560 6 −1.5 +1.0 −0.38 972TH-7 Taihe 221 17 24.9 137 0.1097 0.512572 7 −1.3 +1.6 −0.44 847TH-8 Taihe 216 53 19.2 109 0.1075 0.512588 6 −1.0 +2.0 −0.45 807

Note:(1) Rb, Sr, Sm and Nd concentrations were obtained by ICP-MS and have precision better than±2%.(2) The results of isotopic measurements for Sr and Nd reference materials are: NBS-987 (Sr)=0.710248±3 (2 sm). JMC (Nd)=0.511813±10 (2 sm).(3) Induced error in ISr is:=87Rb/86Sr×(% error assigned)×(el λt−1) of Jahn (2004).(4) f (Sm/Nd) is defined as ((147Sm/144Nd)/0.1967−1).(5) εNd(T) is calculated using an approximate equation of eNd(T)=εNd(0)−Q*f*T; in which Q=25.1 Ga-1, f=f (Sm/Nd), T=age (in Ga).(6) TDM−1=(1/l)*LN(1+((143Nd/144Nd)m−0.51315)/((147Sm/144Nd)m−0.2137))); l=0.00654 Ga-1.

40 J.G. Shellnutt et al. / Lithos 119 (2010) 34–46

(Sr=31 to 80 ppm; Ba=248 to 879 ppm; Rb=198 to 225 ppm;Zr=243 to 462 ppm). The normalized incompatible element patternsof the enclaves are similar as both groups have negative anomalies ofSr, Hf and Zr, however the Taihe enclaves have distinct Ba depletionswhereas the Baima enclaves do not (Fig. 6). The host rock patterns arevery similar to their enclaves. The chondrite normalized rare earthelement patterns of the enclaves (La/YbN=10.3 to 22.6) are alsosimilar to their hosts (La/YbN=7.0 to 52.7) (Fig. 7).

Fig. 5. Classification of the enclaves and their host according to the scheme of Frost et al. (2CaO). (c) Aluminum saturation index (ASI=Al/Ca–1.67P+Na+K) vs. Na+K/Al. (d) K2O/Nand Jahn (2010).

6.2. Radiogenic Sr and Nd isotopes

The Baima enclaves have ISr values ranging from 0.70371 to0.70410 whereas four Taihe enclaves have variable values rangingfrom 0.67632 to 0.72646. Table 2 shows that the four Taihe sampleshave high 87Rb/86Sr ratios (6.7 to 21.0) and their calculated initialratios are either unreasonably low (ISr=0.676 to 0.699) or unreason-ably high (ISr = 0.726). The calculated induced error [ξ=87Rb/

001). (a) Fe* (FeOt/MgO+FeOt) vs. SiO2, (b) modified alkali-line index (Na2O+K2O–a2O vs. SiO2. Data for Panzhihua rocks are from Shellnutt and Zhou (2007) and Shellnutt

Fig. 7. Normalized rare earth element patterns of the Baima and Taihe plutons. a) Baima,

41J.G. Shellnutt et al. / Lithos 119 (2010) 34–46

86Sr×(% error assigned)×(eλt−1)] for the four samples are 0.00057,0.00155, 0.00090 and 0.00050, respectively and when added to the ISrvalues still produce unreasonable values (Jahn, 2004). There are atleast three possibilities which could explain the lower ISr values of theTaihe enclaves which include: 1) reduction of the highly radiogenic Srratios due towater–rock interactions, 2) an increase of the Rb/Sr ratiosdue to Sr loss or Rb gain or 3) a combination of these two possibilities.

In comparison, the initial 143Nd/144Nd ratios for two groups ofenclaves are internally consistent and identicalwithin analytical errorsof their hosts (Table 2). The Baima enclaves have εNd(T) valuesbetween+2.8 and+3.2 whereas the Taihe enclaves range from+1.0to +2.0. The calculated depleted-mantle-based model ages (TDM),regardless of one-stage or two-stage model, are also consistent withineach group (e.g. Baima=~780 MaandTaihe=~850 Ma). The fact thatthe model ages are greater than the emplacement age of 260 Maimplies that the primary magmas were derived from an enrichedmantle source and that crustal contamination is probably insignificantin view of the low initial Sr isotopic.

7. Discussion

7.1. Textural and chemical relationships of the enclaves and their hosts

The enclaves of the Baima and Taihe plutons are not xenoliths nordid they form by stopping because there is no evidence of reactiontextures or cross-cutting relationships (Fig. 4a–d). The mineralassemblages of the enclaves and their hosts are slightly different.The enclaves from the Baima pluton with lower bulk rock SiO2

(b60 wt.%) contain significant quantities (≤10%) of biotite while the

Fig. 6. Primitive mantle normalized incompatible rare earth element normalizedpatterns of the Baima and Taihe plutons. a) Baima, enclaves, host and gabbroic intrusion,b) Baima, fine grained granitic rocks and c) Taihe, enclaves, host and gabbroic intrusion.Data normalized to primitive mantle values of Sun and McDonough (1989).

enclaves, host and gabbroic intrusion and b) Taihe, enclaves, host and gabbroic intrusion.Data normalized to C1 chondrites of Sun and McDonough (1989).

higher bulk rock SiO2 (N60 wt.%) enclaves contain minor amounts(b5%). There are a number of minerals which are common betweenthe hosts and enclaves including titanite, Na–Fe amphibole, zircon,perthitic alkali feldspar, fluorite and aegirine. In both groups ofenclaves (i.e. Baima and Taihe) the amount of quartz, if present, is lessthan their host. The Taihe enclaves are mineralogically very similar totheir host except there is more amphibole and less quartz. In mostcases the colour contrast between some Taihe enclaves (i.e. TH-3, TH-4, TH-7; grey) and the host is not as noticeable as others (i.e. TH-2, TH-6, TH-8; dark). The ellipsoidal shapes and shared minerals indicatethat the enclaves were plastic and in thermal equilibrium with theirhosts implying that the enclaves may be accumulations of earlierformed minerals.

Compositionally, the enclaves are slightly more primitive thantheir host (Fig. 8). The Baima enclaves have 57.2 to 63.0 wt.% SiO2 andMg# between 28 and 44, in comparison, their host has between 63.0and 67.8 wt.% SiO2 and Mg# between 15 and 24. The Taihe enclaveshave 63.8 to 71.0 wt.% SiO2 and Mg# between 8 and 22 compared totheir host which has between 69.8 and 75.1 wt.% SiO2 and Mg#between 3 and 12. The Baima enclaves have higher TiO2, CaO and P2O5

and lower K2O than their host whereas the Taihe enclaves have higherTiO2, Al2O3, CaO, Na2O, K2O and P2O5 than their host. In both cases themajor elements indicate a compositional evolution from the enclaveto the host rock (Fig. 8). The trace element compositions also show acompositional evolution, albeit more obvious within the Taihe pluton(Fig. 9). The primitive mantle normalized patterns of the enclaves arevery similar to their hosts although there is a depletion of Hf–Zr forboth the Baima and Taihe enclaves and a slight depletion of Th–U inthe Baima enclaves (Fig. 6). The Zr depletions, and likely the Hfdepletions, are not due to incomplete sample dissolution as the Zrcontent determined by XRF is within 10% of the ICP-MS results. Thebulk Zr content of the Baima (Zr=601 to 834 ppm) and Taihe(Zr=243 to 462 ppm) enclaves tend to be lower than their hosts

Fig. 8. Major element compositions of the Baima and Taihe host-enclave pairs. Baima host data (GS04-004; -005; -006; -013; -015; -016; -018; -019; -020; -033; GS05-004) fromShellnutt et al. (2009a). Taihe data from Shellnutt and Zhou (2007).

42 J.G. Shellnutt et al. / Lithos 119 (2010) 34–46

(Baima=472 to 1255 ppm; Taihe=421 to 1400 ppm) and suggestspossible compositional evolution. The REE patterns of the enclaves arethe same as their hosts showing negative Eu-anomalies and LREEenriched patterns suggesting a shared evolution (Fig. 7).

The Nd isotopic data of the enclaves are very similar to their hosts.The Baima enclaves have εNd(T) values between +2.8 and +3.2which is indistinguishable from their host (εNd(T)=+3.0 to +3.2).The Taihe enclaves have εNd(T) values between+1.0 and+2.0 whichoverlap with the values of their host (εNd(T)=+1.5 to +1.9). The ISrvalues of the Baima enclaves have a narrow range between 0.7037 and

0.7041whereas the host rock is more variable (ISr=0.7035 to 0.7051)but encompasses the range of the enclaves. The Taihe samples (i.e.host and enclaves) have unreasonably low ISr values (b 0.70) whichcould be due to Sr loss or Rb gain, in either case very little can beconcluded from the Sr-isotopic data of the Taihe pluton.

Themineral textures,major and trace elemental data andwhole rockNd isotopic data indicate that the enclaves and hosts are cogenetic andpart of the same magma system. Since the enclaves are geneticallyrelated to the host, they could be described as ‘autoliths’ (auto=self,lith=stone). The term ‘autolith’ refers to a consanguineous enclave

Fig. 9. Trace elements versus Zr (ppm) diagrams of the Baima and Taihe plutons. Rb, Ba and Sr are considered to be highly diffusive between host and enclaves whereas Y and Zr areless diffusive (Allen, 1991; Holden et al., 1991). Range of gabbroic data for Baima is from Shellnutt et al. (2009). The Taihe gabbroic data are unpublished.

43J.G. Shellnutt et al. / Lithos 119 (2010) 34–46

within its host (Holland, 1900; Fershtater and Borodina, 1977; Didierand Barbarin, 1991). For example, Schonenberger et al. (2006)examined late stage magmatic fluid-autolith interactions from theIlimaussaq intrusion. The Ilimaussaq autoliths are interpreted asoriginating from the same magma as the host but were entrainedwithin their consanguineous magma after roof collapse of the magmachamber. Furthermore, Jones (1979) reported autolithswithin trachytesfrom the Kilombe volcano however, much like the Ilimaussaq enclaves,they did not form in situ. The Baima and Taihe enclaves in contrast to theIlimaussaq and Kilombe enclaves were not entrained, dismemberedlithic fragments. Therefore use of the term ‘autolith’ may be inappro-

priate for the Baima and Taihe enclaves because the suffix ‘-lith’ impliesa solid state which was not likely the case (Fig. 4a–d).

7.2. Chemical diffusion between the enclave and host

It is known that trace elements diffuse between enclaves and theirhosts (Fourcade and Allegre, 1981; Tindle and Pearce, 1983; Holdenet al., 1987, 1991; Eberz et al., 1990; Eberz and Nicholls, 1990; Lesher,1990, 1994; Allen, 1991; Tindle, 1991; Elburg, 1996). Therefore theBaima and Taihe enclaves may have chemically equilibrated withtheir hosts and the measured values may not be their original

Fig. 10. FeOt versus MgOmajor element compositions of ferrorichterite from (a) Baima(BM-1) enclave-host (GS04-004; -015 and 016) pairs and (b) Taihe (TH-2) enclave-host (GS04-054; -057; -059) pairs (unpublished data).

44 J.G. Shellnutt et al. / Lithos 119 (2010) 34–46

compositions. Eberz and Nicholls (1990) and Holden et al. (1991)have shown that K, Rb, Ca, Mg, Mn, Ni, Cr, Zn, V, Cu, Sr, Pb and Ba tendto diffuse faster than Zr, Hf, Nb, Th, REE, Si, Al, Ti, P and Y. If the hostsand enclaves have different compositions, they will equilibrate.Equilibration between the host and enclave may influence theirwhole rock Sr–Nd isotopic compositions because Sr diffuses twice asfast as Nd and therefore it is possible for the isotopic compositions todecouple (Lesher, 1990; Allen, 1991).

The Baima pluton does not show a coherent relationship betweenimmobile andmobile elements versus Zr. The enclaves have higher Baand Sr but there is no correlation with Zr and the host rock is just aslikely to have high Y and Rb concentrations as the enclaves (Fig. 9). Incontrast, the Taihe pluton shows a negative correlation of Rb, Ba andSr with increasing Zr indicating compositional evolution (Fig. 9). It ispossible that diffusion occurred within the Baima pluton but not theTaihe pluton, although the enclaves consistently have higher Mg#, Tiand Ca when compared to their host (Fig. 8). Furthermore the Baimarocks show an increase in total alkalis followed by a decrease withincreasing SiO2 whereas the Taihe rocks show a continuous decreasein alkalis with increasing SiO2. The trends indicate that fractionationof alkali feldspar likely occurred. Although it is possible that diffusionoccurred between the host and enclaves, there is little evidence tosupport this conclusion. Given the fact that the enclaves and host arecogenetic, it is likely that chemical diffusion was minimal.

7.3. Formation of the Baima and Taihe enclaves

The whole rock elemental and isotopic compositions and mor-phology of the enclaves argue against mingling of unrelated magmasor structural dismemberment. It is likely that the enclaves areaccumulations of early formed minerals and not liquid compositions.This could explain why the Baima enclaves are nepheline normativewhereas the host is quartz normative. In the case of the Taihe pluton,the enclaves are host are quartz normative.

Normativenepheline implies the enclaves are silica undersaturatedand could not possibly be the precursor magma to the silica saturatedhost magma. If the enclaves are accumulations of earlier formedcrystals then it is possible that quartz was not an early solidus mineralthereby allowing the accumulation of alkali feldspar, biotite, amphi-bole and titanite and thus having Ne-normative compositions. Thenormative mineralogy and presence of quartz in the enclaves does notpreclude the possibility that the enclaves are magmatic compositionsrather than mineral accumulations but the mineral textures andcompositions would argue against a liquid composition. For example,in the Baima pluton, biotite is present in the host rock as euhedral finegrained laths, although it is very rare. This texture contrasts with theenclaves where biotite is common, medium grained and typicallyassociated with ferrorichterite–apatite–titanite–fluorite agglomerates(Fig. 4e). Similar agglomerate textures are found in the host but biotiteis absent suggesting that it was not a common liquidus mineral whenthe host rocks crystallized. Within the Baima enclaves there are manyphenocrysts of biotite as well as biotite inclusions within alkalifeldspar megacrysts (Fig. 4f). Implicit from the biotite textures isthat they crystallized early because 1) they are present as inclusionswithin megacrysts, 2) their relative rarity in the host suggests theydid not crystallize late and 3) biotite could not have accumulatedalone because it is frequently associated with ferrorichterite–apatite–titanite and fluorite. Furthermore, the ferrorichterites from the Baimaand Taihe enclaves are more magnesian than those within the hostrock (Fig. 10). This evidence suggests that fractional crystallizationwasthe dominant process within the Baima and Taihe plutons.

Curiously, the Panzhihua silica-saturated peralkaline granite doesnot appear to contain enclaves. The Panzhihua granite is composition-ally similar to the Taihe granite and is also considered to be derived byfractional crystallization (Shellnutt and Jahn, 2010). Unlike the Baimaand Taihe gabbro-granitoid complexes, the Panzhihua complex has a

syenodiorite unit between the granite and gabbro. The syenodiorite is ofintermediate composition and is similar in composition to the Baimaand Taihe enclaves (Fig. 5). Large, contiguous, intermediate units are notcurrently known in the Baima and Taihe complexes. The Baima enclavesappear to be isolated tooneoutcrop suggesting their development couldbe a localized phenomenon. In contrast, enclaves are relativelyabundant and found throughout the Taihe pluton suggesting that itwas a magma chamber-scale process.

Considering the major element trends, the whole rock εNd(T)values, the biotite textures and the ferrorichterite compositions, it islikely that the enclaves and host are cogenetic and represent twodifferent stages (e.g. early and later) during progressive fractionalcrystallization of a common parental magma. Therefore the enclavesindicate that there may be large, unmapped intermediate units withinthe Taihe and Baima gabbro–granitoid complexes.

7.4. Enclaves in peralkaline A-type granitoids

The scarcity of enclaves in alkalic, anorogeneic granitoids ispuzzling and as a consequence their formation in not completelyunderstood (Bonin, 1991; Barbarin, 1999). Bonin (1991) discussedthree types of enclaves from alkaline (e.g. metaluminous or peralka-line) anorogenic granitic rocks. The enclaves are broadly described asxenoliths, dismembered magmas (i.e. magma mingling or restite) orhighly evolved (i.e. differentiation of F-rich aqueous fluids). The Baimaand Taihe enclaves do not appear to match the descriptions of Bonin(1991) as they are certainly not xenoliths or highly evolved rocksderived from F-rich aqueous fluids (e.g. rockallite or lindinosite).

The Panxi, peralkaline A-type granitoids likely formed by the sameprocess, yet their mutual evolution created three different expres-sions of intermediate compositions. The differences in major element

Fig. 11. Illustration of the three units of the Panxi gabbro–granitoid and the formation of enclaves. (a) Non-ideal segregation and (b) ideal segregation.

45J.G. Shellnutt et al. / Lithos 119 (2010) 34–46

chemistry between the Baima syenite and the Panzhihua and Taihegranites are attributed to their parental magma bulk composition.Their similar incompatible trace element profiles implies they formedby the same process. Of the Panxi gabbro–granite complexes, thePanzhihua complex may be the best example of ‘ideal’ magmasegregation whereas the Baima and Taihe complexes may beexamples of ‘non-ideal’ magma segregation (Fig. 11). We emphasizethat our use of ‘ideal’ segregation refers to the division of thePanzhihua complex into three distinct units of mafic (i.e. layeredgabbro), intermediate (i.e. syenodiorite) and felsic (i.e. granite andtrachyte) rocks (Fig. 11a). Our use of ‘non-ideal’ segregation refers tothe apparent ‘imperfect’ division or entrainment of the intermediateunit within the felsic unit (i.e. granitic host) and hence the presence ofenclaves within the granitoids (Fig. 11b). The Baima and Taiheenclaves likely represent accumulations of early formed minerals±liquid which were incorporated within their evolved host magma.The reason for ‘ideal’ or ‘non-ideal’ segregation is unknown althoughit may be related to specific internal magma chamber processes (e.g.convection) or physical attributes of the magma (e.g. viscosity).

It is unclear why enclaves are rare in peralkaline A-type granites ingeneral. The critical point may be due to their magmatic origin. Manyperalkaline A-type granitic rocks are derived by fractional crystalliza-tion of mantle derived mafic magmas (e.g. single magma) whereas I-and S-type granites are typically derived frommagmaswhich resultedfrom crust/mantle interactions (e.g. magma mixing, magma minglingand partialmelting of the crust) (Eby, 1990, 1992). Therefore it may beless likely for enclaves to form during differentiation of a single,homogenized magma because of progressive, systematic segregationof crystals (e.g. Panzhihua gabbro–granite complex)whereas enclavesare more likely to form within magma systems which have contribu-tions from multiple sources (e.g. crust+mantle) where magmahomogenization is incomplete.We suggest the occurrence of xenolithswithin I-, S- and A-type granitoids should be about the same.

8. Conclusions

The Baima and Taihe silica saturated peralkaline A-type graniticplutons contain microgranular enclaves. Whole rock and traceelement geochemical data show the enclaves have compositions

that are more primitive than their hosts. The Nd isotopic data of theenclaves and hosts are identical indicating that they are consanguin-eous. Major element trends and amphibole geochemistry suggest thatthe enclaves represent accumulation of early formed minerals duringprogressive fractional crystallization of their parental magmas. Theenclaves are compositionally similar to the intermediate unit of thePanzhihua gabbro–granite complex suggesting that there may beunseen intermediate units within the Baima and Taihe complexes.

Acknowledgements

Wewould like to thank Drs. B. Bonin, T.Waight and P. Robinson fortheir constructive reviews which improved this manuscript. We alsoacknowledge Fu Lung Lin and Wen-yu Hsu of AS and Sun-Lin Chung,Ijhen Lin and Chiyi Lee of NTU for their laboratory assistance andShane Yang and Mei-Fu Zhou for their field support. This work wassupported by Academia Sinica through the post-doctoral fellowship ofJGS. BMJ acknowledges the support of National Science Councilthrough NSC.98-2116-M-001-009 and NSC 97-2116-M-001-011.

References

Ali, J.R., Thompson, G.M., Zhou, M.-F., Song, X.Y., 2005. Emeishan large igneousprovince, SW China. Lithos 79, 475–489.

Allen, C.M., 1991. Local equilibrium of mafic enclaves and granitoids of the Turtlepluton, southeast California: mineral, chemical, and isotopic evidence. AmericanMineralogist 76, 574–588.

Bacon, C.R., 1986. Magmatic inclusions in silicic and intermediate rocks. Journal ofGeophysical Research 81, 6091–6112.

Barbarin, B., 1999. A review of the relationship between granitoid types, their originsand their geodynamic environments. Lithos 46, 605–626.

Barbarin, B., 2005. Mafic magmatic enclaves and mafic rocks associated with somegranitoids of the central Sierra Nevada batholith, California: nature, origin, andrelations with the hosts. Lithos 80, 155–177.

Barbarin, B., Didier, J., 1991. Review of the Main Hypothesis Proposed for the Genesisand Evolution of Mafic Microgranular Enclaves. In: Didier, J., Barbarin, B. (Eds.),Developments in Petrology 13. Elsevier, Amsterdam, pp. 367–373.

Bonin, B., 1991. The Enclaves of the Alkaline Anorogenic Granites: An Overview.Enclaves and Granite Petrology. In: Didier, J., Barbarin, B. (Eds.), Developments inPetrology 13. Amsterdam, Elsevier, pp. 179–189.

Cantagrel, J.M., Didier, J., Gourgaud, A., 1984. Magma mixing: origin of intermediaterocks and “enclaves” from volcanism to plutonism. Physics of the Earth andPlanetary Interiors 35, 63–76.

46 J.G. Shellnutt et al. / Lithos 119 (2010) 34–46

Chappell, B.W., White, A.J.R., 1991. Restite Enclaves and the Restite Model. In: Didier, J.,Barbarin, B. (Eds.), Developments in Petrology 13. Elsevier, Amsterdam, pp. 375–381.

Chung, S.L., Jahn, B.-M., 1995. Plume–lithosphere interaction in generation of theEmeishan flood basalts at the Permian–Triassic boundary. Geology 23, 889–892.

Chung, S.L., Jahn, B.-M., Genyao, W., Lo, C.H., Bolin, C., 1998. The Emeishan Flood Basaltin SW China: A Mantle Plume Initiation Model and its Connection with ContinentalBreakup and Mass Extinction at the Permian–Triassic Boundary. In: Flower, M.F.J.,Chung, S.-L., Lo, C.H., Lee, T.Y. (Eds.), Mantle Dynamics and Plate Interactions in EastAsia: American Geophysical Union Geodynamic Series, vol. 27, pp. 47–58.

Didier, J.D., 1973. Granites and their Enclaves. Developments in Petrology 3. Elsevier,Amsterdam (393 p.).

Didier, J.D., 1987. Contribution of enclave studies to the understanding of origin andevolution of granitic magmas. Geologische Rundschau 76, 41–50.

Didier, J., Barbarin, B. (Eds.), 1991. Developments in Petrology 13. Elsevier, Amsterdam(625 p.).

Domenick, M.A., Kistler, R.E., Dodge, F.C.W., Tatsumoto, M., 1983. Nd and Sr isotopicstudy of crustal and mantle inclusions from the Sierra Nevada and implications forbatholith petrogenesis. Geological Society of America Bulletin 94, 713–719.

Eberz, G.W., Nicholls, I.A., 1990. Chemical modification of enclave magma by post-emplacement crystal fractionation, diffusion and metasomatism. Contributions toMineralogy and Petrology 104, 47–55.

Eberz, G.W., Nicholls, I.A., Maas, R., McColloch, M.T., Whitford, D.J., 1990. The Nd- andSr-isotopic composition of I-typemicrogranitoid enclaves and their host rocks fromthe Swifts Creek Pluton, southeast Australia. Chemical Geology 85, 119–134.

Eby, G.N., 1990. The A-type granitoids: a review of their occurrence and chemicalcharacteristics and speculations on their petrologenesis. Lithos 26, 115–134.

Eby, G.N., 1992. Chemical subdivision of the A-type granitoids: petrogenetic andtectonic implications. Geology 20, 641–644.

Elburg, M.A., 1996. Evidence of isotopic equilibration between microgranitoid enclavesand host granodiorite, Warburton granodiorite, Lachlan fold belt, Australia. Lithos38, 1–22.

Fershtater, G.B., Borodina, N.S., 1977. Petrology of autoliths in granitic rocks.International Geology Review 19, 458–468.

Fourcade, S., Allegre, C.J., 1981. Trace element behavior in granitic gneiss: a case studythe calc-alkaline plutonic association from the Querigut Complex (Pyrenees,France). Contributions to Mineralogy and Petrology 76, 177–195.

Frost, B.R., Barnes, C.G., Collins, W.J., Arculus, R.J., Ellis, D.J., Frost, C.D., 2001. Ageochemical classification for granitic rocks. Journal of Petrology 42, 2033–2048.

Garcia-Moreno, O., Castro, A., Corretge, L.G., El-Hmidi, H., 2006. Dissolution of tonaliticenclaves in ascending hydrous granitic magmas: an experimental study. Lithos 89,245–258.

Holden, P., Halliday, A.N., Stephens, W.E., 1987. Neodymium and strontium isotopecontent of microdiorite enclaves points to mantle input to granitoid production.Nature 330, 53–56.

Holden, P., Halliday, A.N., Stephens, W.E., Henney, P.J., 1991. Chemical and isotopicevidence for major mass transfer between mafic enclaves and felsic magma.Chemical Geology 92, 135–152.

Holland, T.H., 1900. The charnockitic series, a group of Archean hypersthenic rocks inpeninsular India. Geological Survey of India Memoir 28.

Jahn, B.-M., 2004. The Central Asian Orogenic Belt and Growth of the Continental Crustin the Phanerozoic. In: Malpas, J., Fletcher, C.J.N., Ali, J.R., Aitchison, J.C. (Eds.),Aspects of the Tectonic Evolution of China Geological Society of London: SpecialPublication, 226, pp. 73–100.

Jones, W.B., 1979. Syenite boulders associated with Kenyan trachyte volcanoes. Lithos12, 89–97.

Lesher, C.E., 1990. Decoupling of chemical and isotopic exchange during magmamixing. Nature 344, 235–237.

Lesher, C.E., 1994. Kinetics of Sr and Nd exchange in silicate liquids: theory,experiments, and applications to uphill diffusion, isotopic equilibrium, andirreversible mixing of magmas. Journal of Geophysical Research 99, 9585–9604.

Middlemost, E.A.K., 1989. Iron oxidation ratios, norms and the classification of volcanicrocks. Chemical Geology 77, 19–26.

Ridolfi, F., Renzulli, A., Macdonald, R., Upton, B.G.J., 2006. Peralkaline syenite autolithsfrom Kilombe volcano, Kenya rift valley: evidence for subvolcanic interaction withcarbonatitic fluids. Lithos 91, 373–392.

Schonenberger, J., Marks, M., Wagner, T., Markl, G., 2006. Fluid-rock interaction inautoliths of agpaitic nepheline syenites in the Ilimaussaq intrusion, SouthGreenland. Lithos 91, 331–351.

Shellnutt, J.G., Jahn, B.-M., 2010. Formation of the Late Permian Panzhihua plutonic–hypabyssal–volcanic igneous complex: implications for the genesis of Fe–Ti oxidedeposits and A-type granites of SW China. Earth and Planetary Science Letters 289,509–519.

Shellnutt, J.G., Zhou, M.-F., 2007. Permian peralkaline, peraluminous andmetaluminousA-type granites in the Panxi district, SW China: their relationship to the Emeishanmantle plume. Chemical Geology 243, 286–313.

Shellnutt, J.G., Zhou, M.-F., Yan, D.-P., Wang, Y., 2008. Longevity of the PermianEmeishan mantle plume (SW China): 1 Ma, 8 Ma or 18 Ma? Geological Magazine145, 373–388.

Shellnutt, J.G., Zhou,M.-F., Zellmer, G., 2009. The role of Fe–Ti oxide crystallization in theformation of A-type granitoids with implications for the Daly gap: An example fromthe Permian Baima igneous complex, SW China. Chemical Geology 259, 204–217.

Sun, S.S., McDonough, W.F., 1989. Chemical and Isotopic Systematics of Oceanic Basalts:Implications for Mantle Composition and Processes. In: Saunders, A.D., Norry, M.J.(Eds.), Magmatism in the Ocean Basins: Geological Society of London SpecialPublication, vol. 42, pp. 313–435.

Tindle, A.G., 1991. Trace Element Behavior in Microgranular Enclaves from GraniticRocks. In: Didier, J., Barbarin, B. (Eds.), Developments in Petrology 13. Elsevier,Amsterdam, pp. 313–331.

Tindle, A.G., Pearce, J.A., 1983. Assimilation and partial melting of continental crust:evidence from the mineralogy and geochemistry of autoliths and xenoliths. Lithos16, 185–202.

Ventura, G., Gaudio, P.D., Iezzi, G., 2006. Enclaves provide new insights on the dynamicsof magma mingling: a case study from Salina Island (southern Tyrrhenian Sea,Italy). Earth and Planetary Science Letters 243, 128–140.

Vernon, R.H., 1983. Restite, xenoliths and microgranitoid enclaves in granites. Journaland Proceedings. Royal Society of New South Wales 116, 77–109.

Vernon, R.H., 1984. Microgranitoid enclaves in granites — globules of hybrid magmaquenched in a plutonic environment. Nature 309, 438–439.

Vernon, R.H., Etheridge, M.A., Wall, V.J., 1988. Shape and microstructure of themicrogranitoid enclaves: indicators of magma mingling and flow. Lithos 22, 1–11.

Waight, T.E., Maas, R., Nicholls, I.A., 2001. Geochemical investigations of microgranitoidenclaves in the S-type Cowra granodiorite, Lachlan flood belt, SE Australia. Lithos56, 165–186.

Wang, D.R., Xie, Y.M., Hu, Y.J., You, X.X., Cao, J.X., 1993. Geology of Guogailiang area(Map G-48-1-A). Panxi Geological Team, Geology and Mineral Resources Bureau,Sichuan Province, 1: 50000.

Wang, D.R., Xie, Y.M., Wang, Q.J., Tao, Z.D., 1994. Geology of Guabang area (Map G-48-25-C). Panxi Geological Team, Geology and Mineral Resources Bureau, SichuanProvince, 1: 50000.

White, A.J.R., Chappell, B.W., 1977. Ultrametamorphism and granitoid genesis.Tectonophysics 43, 7–22.

Wones, D.R., 1989. Significance of the assemblage titanite+magnetite+quartz ingranitic rocks. American Mineralogist 74, 744–749.

Xiong, X.Y., Xie, Y.M., Nie, K.H., Kan, Z.Z., Ding, L.R., Guo, J.Q., 1996. Geology of Miyi area(Map G-48-37-A). Panxi Geological Team, Geology and Mineral Resources Bureau,Sichuan Province, 1: 50000.

Xu, Y., Chung, S.L., Jahn, B.-M., Wu, G., 2001. Petrologic and geochemical constraints onthe petrogenesis of Permian–Triassic Emeishan flood basalts in southwesternChina. Lithos 58, 145–168.

Zhong, H., Zhu, W.-G., Song, X.-Y., He, D.-F., 2007. SHRIMP U–Pb zircon geochronology,geochemistry, and Nd–Sr isotopic study of contrasting granites in the Emeishanlarge igneous province, SW China. Chemical Geology 236, 112–137.

Zhou, M.-F., Malpas, J., Song, X.Y., Robinson, P.T., Sun, M., Kennedy, A.K., Lesher, C.M.,Keays, R.R., 2002. A temporal link between the Emeishan large igneous province(SW China) and the end-Guadalupian mass extinction. Earth and Planetary ScienceLetters 196, 113–122.

Zhou, M.-F., Arndt, N.T., Malpas, J., Wang, C.Y., Kennedy, A.K., 2008. Two magma seriesand associated ore deposit types in the Permian Emeishan large igneous province,SW China. Lithos 103, 352–368.