Mesozoic Adakites in the Lingqiu Basin of the Central North China Craton: Partial Melting of Underplated Basaltic Lower Crust

Geochemical Journal, Vol. 40, pp. 447 to 461, 2006

Mesozoic adakites in the Lingqiu Basin of the central North China Craton: Partial melting of underplated basaltic lower crust

XUAN-CE WANG,1,2,3,4* YONG-SHENG LIU2 and XIAO-MING LIU3

1Key Laboratory of Isotope Geochronology and Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou 510640, China 2State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences, Wuhan 430074, China 3Key Laboratory of Continental Dynamics, Department of Geology, Northwest University, Xi’an 710069, China 4Graduate School of the Chinese Academy of Sciences, Beijing 100039, China

(Received June 14, 2005; Accepted February 16, 2006)

Intermediate to felsic volcanic rocks of the Baiqi formation from the Lingqiu basin in the central part of North China Craton were studied. Single zircon U-Pb dating indicates that these volcanics formed at 125.8 ± 3.0 Ma. Their Sr and Nd isotopic compositions (143Nd/144Nd = 0.51180–0.51182, 87Sr/86Sr = 0.7062–0.7063) fall in the range of the nearby late- Mesozoic basaltic rocks. These volcanics share geochemical affinities to the adakites formed in the modern arcs, e.g.,

high Na2O (>4.06%), Al2O3 (>15.4%) and Sr (645–1389 ppm) contents and Sr/Y ratios (55~103), and thus being termed as adakitic rocks. However, the Baiqi adakitic rocks were not temporospatially associated with active subduction. Further- more, their low Cr (2.19–47.4 ppm, with average of 25) and Ni (1.57–20.7 ppm, with average of 12) contents and Mg# (22–47, with average of 32) argue against interaction with the lithospheric mantle. Combined with the geological setting, we suggest that the Baiqi adakitic rocks resulted from partial melting of a thickened lower continental crust associated two episodes of basaltic underplating events. We propose that enormous conductive heating from 80–140 Ma basaltic underplating resulted in partial melting of pre-existing mafic lower crust formed by ~150–160 Ma basaltic underplating. This study provides a case for partial melting of the thickened lower continental crust in association with basaltic underplating events.

Keywords: adakitic rocks, basaltic underplating, lower crust, Lingqiu, North China

Rapp et al., 1999), and thicken continental crust (Wang, INTRODUCTION Q. et al., 2005). Adakites in subduction zones have been Adakite is characterized by low K, Y, HREE contents widely investigated (Defant and Drummond, 1990; Mar- ≤ ≤ (Y 18, Yb 1.9 ppm), and high Al, Na (Al2O3 >15% at tin, 1999; Martin et al., 2005; Peacock et al., 1994; Stern the 70% SiO2; Na2O/K2O >1), Sr (>400 ppm) contents, and Kilian, 1996), whereas the adakitic rocks within the and high Sr/Y and La/Yb ratios (Defant and Drummond, continent are yet to be fully understood (Gao et al., 2004; 1990). Partial melting of basaltic rocks can produce Liu et al., 2005; Wang, Q. et al., 2005). adakitic melt at pressures equivalent to a crustal thick- Many adakitic rocks in the Eastern China were found ness of >40 km (Rapp and Watson, 1995; Rapp et al., and studied in the last five years (e.g., Gao et al., 2004; 1991). This rock attracts widespread attention due to its Xiao et al., 2004; Xu et al., 2002; Zhang et al., 2001a, b). significance in revealing deep geodynamic processes, e.g., These adakitic rocks are predominately formed in late oceanic crust subduction (Defant and Drummond, 1990; Jurassic to early Cretaceous (160–120 Ma, peak at 137– Martin et al., 2005; Martin, 1999; Peacock et al., 1994; 130 Ma). Zhang et al. (2001b) classified adakitic rocks Stern and Kilian, 1996), basalt underplating (Atherton and into two types: O-type adakites (typical adakites, related Petford, 1993; Petford and Atherton, 1996; Rapp and to the slab subduction) and C-type adakites (produced Watson, 1995; Xiong et al., 2003), recycling of lower within intracontinent). Xiao et al. (2004) classified the continental crust (Gao et al., 2004; Kay and Kay, 1991; C-adakites from North China Craton into type A and type Xu et al., 2002), melt-peridotite reaction (Liu et al., 2005; B. The type A rocks (most of C-type adakites, not all) have lower Mg# number and higher K contents than typical subduction-related adakites, which could be produced *Corresponding author (e-mail: [email protected]) by partial melting of underplated basaltic rocks (Xiao et Copyright © 2006 by The Geochemical Society of Japan. al., 2004; Zhang et al., 2001a, b), whereas the type B

447 N 15km Archean genesis Division of the North China craton 5

Mesozoic volcanics 1 6 Western Block 2 Intrusive rocks Hunyuan 3 4 Sediment rocks Eastern Block

Trans-North China Orogen 39 30 (B)

1100 Xiguayuan formation 1000 Dabiegou formation Baiqi adakitic A 900 sample site Zhangjiakou

Lingqiu formation Thickness(m) 200 Baiqi formation

114 (C ) (A)

Fig. 1. (a) Simplified geological map of the study area. (b) Tectonic division of the North China Craton (After Zhao et al., 2001). Locations of the Mesozoic gabbros and lower crustal xenoliths in the Trans-North China Orogen were also marked. 1 = Location of the Baiqi adakitic rocks, 2 = gabbro from Laiyuan (Zhang et al., 2004), 3 = gabbro from Hanxing (Zhang et al., 2004), 4 = gabbro from Linxian (Wang, Y. J. et al., 2005), 5 = lower crustal xenolths from the Hannuoba, 6 = Xinglonggou high-Mg adakites From Liaoxi (Gao et al., 2004). (c) Profile of the Mesozoic volcanics in Hunyuan-Guangling-Lingqiu basin.

rocks featured by high Mg# were interpreted as foundering Mesozoic magmatism (e.g., Davis et al., 1998; Chen et lower continental crust-derived melt interacted with al., 2002, 2003, 2004; Chen and Zhai, 2003; Zhang et al., peridotites in the mantle (Gao et al., 2004; Xiao et al., 2004). 2004). Despite these investigations, some questions re- The adakitic rocks in this study were collected from main: how were these adakitic rocks related to the the Lingqiu volcanic basin in the northern part of the foundering lower continental crust or basalt underplating? Trans-North China Orogen. The late Mesozoic volcanics And what is the mechanism triggered the surge of the in the Hunyuan-Guangling-Lingqiu basins (Figs. 1A and Mesozoic adakites in the North China Craton? To address C) are composed of intermediate to felsic volcanics with these questions, typical adakitic volcanics from the Trans- interlayers of sandstones and mudstones. These volcanics North China Orogen, central North China Craton were were classified into four formations (Fig. 1C): from bot- studied in this paper. Except for some adakitic intrusive tom to top, Baiqi, Zhangjiakou, Dabeigou and Xiguayuan. rocks, typical adakitic volcanic rocks have not been re- The Baiqi formation was systematically studied in this ported in the central of North China Craton. work. The emplacement age of diorite-granite-rhyolite and tuff lava of the Trans-North China Orogen has been dated at 127–138 Ma using U-Pb zircon and Rb-Sr whole- GEOLOGICAL SETTING rock isochron methods (Cai et al., 2003; Davis et al., The North China Craton is one of the oldest continen- 1998; Peng et al., 2004). Several episodes of Mesozoic tal nuclei in the world, with basement of mainly Archean mafic igneous rocks in the Trans-North China Orogen to Early Proterozoic gneisses (Jahn et al., 1988). Based have also been identified at 150–160 Ma (e.g., on isotopic age, lithological assemblage, tectonic evolu- Yunmengshan gabbro-diorite complex, Davis et al., 1998; tion and P-T-t paths, the North China Craton can be di- Hanxing gabbro, Zhang et al., 2004), 135–145 Ma (e.g., vided into the Eastern Block, the Western Block and the Laiyuan gabbro, Zhang et al., 2004) and 120–130 Ma intervening Trans-North China Orogen (Zhao et al., 2000, (e.g., Linxian gabbro, Wang, Y. J. et al., 2005). Laiyuan 2001) (Fig. 1B). This craton underwent a dramatic change locate in the eastern of the Lingqiu basin with a distance from a Paleozoic cratonic mantle to a Cenozoic “oceanic” of ~80 km, and Linxian in the southeastern with a dis- lithospheric mantle, accompanied by lithospheric thinning tance of 300 km. Lower crustal xenoliths found in the (Gao et al., 2002; Griffin et al., 1998; Menzies et al., Neogene Hannuoba basalts adjacent to the studied area 1993; Rudnick et al., 2004; Wu et al., 2003; Xu, 2001), are dated at two age intervals of ~160–140 Ma and ~140– lower crustal recycling (Gao et al., 2004) and widespread 80 Ma (Liu et al., 2004; Wilde et al., 2003), and they

448 X.-C. Wang et al. were interpreted as products of ~160–140 Ma basaltic TEMORA 1 as an unknown in this LA-ICPMS over a underplating and subsequent ~140–80 Ma granulite-facies period of 16 months yielded a weighted mean 206Pb/238U metamorphism (Liu et al., 2004). The xenoliths with con- age of 415 ± 4 Ma (MSWD = 0.112) (Gao et al., 2004; vex-upward REE patterns had a cumulate origin (gabbro Yuan et al., 2004), which is in good agreement with the or pyroxenite) (Liu et al., 2001). The Mesozoic Hannuoba recommended ID-TIMS age of 416.75 ± 0.24 Ma (Black granulite xenoliths was products of basaltic underplating et al., 2003). has been widely accepted (e.g., Chen et al., 2001; Fan et Fresh chips of whole rock samples were powdered to al., 1998, 2001; Liu et al., 2001, 2004; Zhang et al., 1998; 200 meshes using a tungsten carbide ball mill. Major and Zhou et al., 2002). trace elements were analyzed using XRF (Rikagu RIX The mafic igneous rocks from the Trans-North China 2100) and ICP-MS (PE 6100 DRC), respectively at the Orogen were derived from a special mantle metasomatised Key Laboratory of Continental Dynamics, Northwest by a SiO2-rich melt (Chen et al., 2004; Wang, Y. J. et al., University in Xi’an, China. Analyses of USGS and Chi- 2005; Zhang et al., 2004). Metasomatism lower the nese national rock standards (BCR-2, GSR-1 and solidus significantly, triggered intensely melting of the GSR-3) indicate that both analytical precision and accu- enriched sub-continental lithospheric mantle. The SiO2- racy for major elements are generally better than 5% (Ap- rich melt may be derived from the paleo-Pacific subducted pendix 1). For trace element analysis, sample powders slab (Chen et al., 2004), or the Paleoproterozoic subduc- were digested using HF + HNO3 mixture in high-pres- tion slab during the collision between the Eastern and sure Teflon bombs at 190°C for 48 hours. Analytical pre- Western Blocks of the North China Craton along the cision and accuracy are better than 2% and 10%, respec- Trans-North China Orogen (Wang, Y. J. et al., 2005), or tively, for most of the trace elements except for transi- basaltic layers that were previously subducted (a fossil tion metals (Appendix 2). oceanic slab) or underplated into the base of the Sr-Nd isotopes were determined at the China Univer- lithospheric mantle (Liu et al., 2005). sity of Geosciences (Wuhan), following the procedures described by (Ling et al., 2003). A mixture solution of 84Sr, 85Rb, and 145Nd and 149Sm isotope spikes was SAMPLES AND ANALYTICAL METHODS weighted and added to each sample aliquots. Rb, Sr and Samples REE were separated using cation columns; Sm and Nd Adakitic rocks from the Baiqi formation were system- fractions were further separated by HDEHP-coated Kef atically sampled at the Ganhegou-Matou Mountain pro- columns. The measured 143Nd/144Nd and 87Sr/86Sr ratios file in the Lingqiu basin (Fig. 1A). This profile is com- were normalized to 146Nd/144Nd = 0.7219 and 88Sr/86Sr = posed of minor andesites and abundant dacites with mi- 8.375209, respectively. The La Jolla standard measured nor amount of rhyolites. Andesite contains phenocrysts during the course of this study gives an average 143Nd/ of plagioclase, brown amphibole and pyroxene. In most 144Nd = 0.511862 ± 5 (2σ, n = 15), the BCR-2 gives 143Nd/ dacite specimens, plagioclase is the main phenocrysts and 144Nd = 0.512635 ± 4 (2σ, n = 6), Nd = 29.10 ppm and accompanied by brown amphibole and mica set in a Sm = 6.591 ppm; and the NBS-987 gives 87Sr/86Sr = groundmass. The rhyolites contain quartz and plagioclase 0.710236 ± 16 (2σ, n = 6). microphenocryst within perlitic groundmass. Fresh samples were only selected for whole rock analyses based on RESULTS detailed petrological studies. U-Pb Zircon age Analytical methods Zircons from a dacite sample GHG02 from the Baiqi Hand picked zircon grains were mounted in epoxy formation were separated for U-Pb dating. Zircon grains, blocks, polished to obtain an even surface, and cleaned together with a zircon U-Pb standard (TEMORA 1), were in an acid bath prior to LA-ICPMS analysis. Zircon U- cast in an epoxy mount, and then were documented with Th-Pb measurements were made on 30–40 µm diameter Cathodoluminescence (CL) images. In contrast to the pre- spots of single grain using the Laser Ablation-Inductively dominantly inherited zircons from those adakites derived Coupled Plasma Mass Spectrometry (LA-ICPMS) at the from melting of recycling lower continental crust (Gao Key Laboratory of Continental Dynamics, Northwest et al., 2004), most zircons from the Baiqi formation are University in Xi’an, and the analytical procedures are euhedral and show typical igneous oscillatory zonation similar to those described by Yuan et al. (2004). The data (Fig. 6B). Thirteen zircons were analyzed, and they have were processed and plotted using Isoplot 3.0 (Ludwig, relatively high Th/U ratios of 0.65 to 2.5 (Table 3). The 2003). Common Pb were corrected following the method measured 206Pb/238U ratios are in good agreement within of Andersen (2002). Age uncertainties are quoted at the analytical precision, yielding a mean 206Pb/238U age of 95% confidence level. Measurements of zircon standard 125.8 ± 3.0 Ma (95% confidence interval).

Mesozoic adakites: Partial melting of underplated basalts 449 Table 1. Chemical composition of the Baiqi adakitic rocks

Sample GHG16 GHG15 GHG03 GHG05 GHG14 GHG13 GHG06 GHG09

Major elements (%)

SiO 2 64.60 63.30 65.00 64.20 59.70 57.30 60.30 70.50 TiO2 0.79 0.71 0.71 0.86 0.90 0.97 0.84 0.23 Al2O3 16.00 16.50 15.40 14.60 15.90 16.70 16.10 15.40 to tal Fe 2O3 4.78 5.23 5.14 6.09 6.97 7.13 6.41 2.23 MnO 0.03 0.05 0.07 0.06 0.07 0.10 0.07 0.04 MgO 0.70 0.90 1.06 0.98 2.20 3.24 2.45 0.33 CaO 3.02 2.97 3.21 3.45 3.76 4.30 3.72 1.43

Na2O 5.63 4.97 5.78 5.68 4.12 6.89 4.06 5.63 K2O 2.59 3.26 1.60 2.77 3.89 1.30 3.80 3.00 P2O5 0.42 0.38 0.38 0.41 0.44 0.43 0.41 0.10 Total 99.60 99.80 99.90 100.00 99.90 100.00 99.90 99.90

Na2O/K2O 2.17 1.52 3.61 2.05 1.06 5.3 1.07 1.88 Mg#(a) 22 25 29 24 38 47 43 23

Trace elements (ppm) Cr 39.9 15.8 11.0 27.4 34.6 47.4 20.1 2.19 Co 49.0 30.2 44.3 60.7 40.3 38.1 32.3 34.5 Ni 14.0 8.85 7.05 12.5 18.6 20.7 12.1 1.57 Rb 51.3 69.2 39.8 47.4 72.7 40.9 80.2 48.3 Sr 1389 1129 1140 1205 1198 1077 1225 645 Y 13.5 11.9 13.0 13.4 14.7 16.2 14.5 11.2 Zr 178 163 157 156 157 155 164 176 Nb 11.4 9.69 9.53 9.27 9.63 9.41 10.3 12.6 Ba 1681 1699 770 1791 2015 470 2056 1488 La 49.3 41.0 42.1 39.7 40.4 42.0 45.0 53.2 Ce 87.3 70.6 76.4 74.8 74.8 77.2 80.6 92.4 Pr 10.1 8.35 8.71 8.77 8.97 9.19 9.54 9.91 Nd 38.2 32.4 33.2 34.6 35.5 37.3 37.4 34.2 Sm 5.43 4.72 5.01 5.23 5.42 5.68 5.46 4.40 Eu 1.71 1.53 1.50 1.70 1.82 1.71 1.80 1.24 Gd 4.87 4.31 4.54 4.78 4.96 5.25 5.04 4.12 Tb 0.52 0.48 0.51 0.53 0.55 0.60 0.55 0.42 Dy 2.40 2.32 2.43 2.50 2.68 2.89 2.59 2.00 Ho 0.39 0.40 0.42 0.43 0.47 0.51 0.44 0.34 Er 1.02 1.05 1.12 1.11 1.23 1.34 1.16 0.98 Tm 0.14 0.15 0.16 0.15 0.18 0.19 0.16 0.14 Yb 0.88 1.02 1.00 0.96 1.14 1.25 1.04 0.98 Lu 0.13 0.15 0.16 0.14 0.18 0.20 0.16 0.16 Hf 4.29 3.93 3.99 3.73 3.76 3.99 4.04 4.60 Ta 0.65 0.60 0.63 0.55 0.54 0.56 0.59 0.80 Pb 18.6 15.6 16.3 8.44 13.1 11.5 14.1 19.5 Th 5.51 5.00 4.84 4.44 4.33 4.27 5.08 6.84 U 1.21 1.18 1.25 0.89 0.85 1.98 1.27 1.36 Rb/Sr 0.04 0.06 0.03 0.04 0.06 0.04 0.07 0.07 (b) (La/Yb)N 38 27 29 28 24 23 30 37 Sr/Y 103 95 88 90 81 66 85 58

(a)Mg# = 100∗Mg/(Mg + Fe), in atomic number. (b)Subscript N denotes chondrite normalization.

Major and trace elements morphic basaltic rocks (Fig. 2). SiO2 correlates positively The Baiqi volcanic rocks range from andesite to with MgO and FeOT with correlation coefficients of 0.80 rhyolite with SiO2 = 57–71% (Table 1). Their major ele- and 0.92, respectively (Figs. 2B and C). ment compositions are characterized by high Na2O (4.06– The Baiqi volcanic rocks have high Sr contents (645– ≥ 6.89%) and Al2O3 ( 15.4%) and high Na2O/K2O ratio 1389 ppm), and high Sr/Y (55–103), LaN/YbN ratios (17– (1.06–5.30), similar to those of the partial melts of meta- 38; where subscript N denotes chondrite normalization),

450 X.-C. Wang et al. 26 7 A B 6 24 WW 5

20 Adakite 4 23 MgO(wt%) Al O (wt%) R 3 18 1

2 WW RSW2-4 16 1 RSW Adakite 2-4 R1

10 10 C D 9 9 RSW2-4 R1 8 8

7 7 RSW2-4 6 6

5 WW 5 FeO(wt%) 4 Adakite 4 R 2 Adakite 3 1 3 Na O(wt%)

2 2 WW 1 1 10 7 E F 9 6 8 WW RSW2-4 7 5 Adakite Adakite 6 4 R1 5 CaO(wt%) RSW2-4 3

4 2 KO(wt%) 3 2 WW 2 R1 1 1 0 0 50 60 70 80 50 60 70 80 SiO (wt%) 2 SiO2 (wt%)

Fig. 2. Geochemical comparison of the Baiqi adakitic rocks with partial melts of basalts. R1 = Partial melts of alkali basalt at 12~38 kbar (Rapp, 1995; Rapp et al., 1991, 1999; Rapp and Watson, 1995); R2–4SW = Partial melts of basalts compositionally close to N-MORB (Rapp and Watson, 1995; Rapp et al., 1991; Sen and Dunn, 1994; Winther, 1996); WW = Partial melts of low- K and Na, high-Mg and Ca basalt (Wolf and Wyllie, 1994). The arc-related adakites (grey field) following Xiong et al. (2003).

and low heavy rare-earth element (HREE; Yb ≤ 1.8 ppm), and insignificant Eu anomaly (Fig. 3A). Their very low Y contents (≤18 ppm) (Table 1), resembling those of the Rb/Sr ratios (0.03 to 0.07) suggest little, if any, fractional typical adakite (Defant and Drummond, 1990). In the crystallization of plagioclase (Xiong et al., 2003). In the primitive mantle-normalized trace element spidergram, Sr/Y-Y diagram, the Baiqi volcanic rocks fall into the field they display clear depletion in Nb-Ta and pronounced of typical adakites and TTG, contrasting the arc andesite enrichment in Pb relative to neighboring elements. In and dacites (Fig. 4B). All these indicate that the Baiqi addition, they exhibit clear positive Sr anomaly (Fig. 3B). volcanics belong to adakite in terms of geochemical fea- Their rare earth element (REE) patterns are characterized tures (Defant and Drummond, 1990). Thus, we term the by LREE enrichment, relatively flat HREE distribution Baiqi volcanics as adakitc rocks in the following discus-

Mesozoic adakites: Partial melting of underplated basalts 451 1000 100 Xinglonggou A Baiqi Baiqi B Adakite(n=221) 20 80

100 60 N

) 30 20 Adakites Yb /

a 40 40 30

10 (L

Sample/Chondrite 50 40 20 Classic A island arc (A) 1 0 La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu 0 3 6 9 12 15

YbN 1000 120 B

100 A 100 30 30 80

Y Adakites

10 Sr/ 40 60 40

50 40 50 1

Sample/primitive mantle Sample/primitive 20 Classic B (B) island arc 0.1 0 Ba U Ta Ce Pr Nd Hf Eu Tb Y Er Yb 51015202530354045 Rb Th Nb La Pb Sr Zr Sm Gd Dy Ho Tm Lu Fig. 3. Chondrite-normalized REE patterns and Primitive man- Y(ppm) tle (PM)-normalized spider diagrams for the Baiqi adakitic Fig. 4. (La/Yb) -Yb and Sr/Y-Y plots (following Defant and rocks. The mean values of adakites are from Condie (2005). N N Drummond, 1990) of the Baiqi adakitic rocks. The solid lines The Xinglonggou high-Mg adakites (Gao et al., 2004) were also represent batch melting model assuming 150–160 Ma gabbros shown for comparison. Chondrite and PM values are from Sun from Hanxing (Zhang, H. F., 2005, unpublished data) as start- and McDonough (1989). ing material, and numbers show the partial melting degrees (%). Parameters used for calculations are listed in Table 4. A and B represent melts in equilibrium with 50%Cpx + 50%Grt and 20%Cpx + 80%Grt, respectively. sion in order to distinguish them from the typical adakites formed by slab melting in modern arc settings.

Whole rock Sr-Nd isotopic compositions DISCUSSIONS The Baiqi adakitic rocks are characterized by constant The case of thickened lower continental crust-derived initial Nd and Sr isotopic compositions at T = 126 Ma: adakitic rocks ε 87 86 Nd (T) = –14.4 to –14.8 and ( Sr/ Sr)i = 0.7060~0.7061 The significances of adakitic rocks rest with its (Table 2). We noticed that their Nd and Sr isotopic com- geochemical indicator for high-pressure processes at sta- positions are consistent with those of the Mesozoic bility field of garnet, most probably under eclogite-facies 87 86 ε Hannuoba granulites ( Sr/ Sr = 0.706–0.707, Nd (130 conditions. Various hypotheses have been put forward to Ma) = –18 to –12, Liu et al., 2004) and the late-Mesozoic explain the origin of the adakitic magmas. These include 87 86 basaltic magmatism from North China Craton (( Sr/ Sr)i formation by partial melting of subducted oceanic slab ε = 0.704~0.708, Nd (T) = –17 to –12) (Cai et al., 2003; (e.g., Defant and Drummond, 1990; Martin, 1999; Mar- Chen et al., 2004; Wang, Y. J. et al., 2005; Zhang et al., tin et al., 2005), partial melting of thickened lower conti- 2004) (Fig. 5), and clearly differ from the MORB-like nental crust (Atherton and Petford, 1993; Petford and isotopic compositions of the modern arc-related adakites Atherton, 1996; Rapp and Watson, 1995; Xiong et al., (e.g., Kay and Kay, 1993; Stern and Kilian, 1996). 2003; Wang, Q. et al., 2005) or partial melting of recy-

452 X.-C. Wang et al. Table 2. Sr-Nd isotopic compositions of the Baiqi adakitic rocks

143 144 143 144 ± σ 147 144 ε Sample Nd Sm Nd/ Nd Nd/ Nd 2 Sm/ Nd Nd (ppm) (ppm) (126 Ma) (126 Ma)

GHG03 34.8 5.48 0.511799 0.51172 5 0.09517 –14.6 GHG03_Dupl. 34.5 5.43 0.511794 0.51171 5 0.09508 –14.7 GHG16 38.5 5.69 0.511796 0.51172 5 0.09519 –14.7 GHG07 43.8 7.54 0.511817 0.51173 5 0.10396 –14.4 GHG13 39.4 6.34 0.511794 0.51171 5 0.09831 –14.8

Sample Sr Rb 87Sr/86Sr 87Sr/86Sr ±2σ 87Rb/86Sr (ppm) (ppm) (126 Ma)

GHG03 1194 37.9 0.70619 0.70603 11 0.0918 GHG03_Dupl. 1192 37.9 0.70620 0.70603 9 0.0920 GHG16 1390 50.9 0.70620 0.70603 12 0.0916 GHG07 1299 34.6 0.70620 0.70606 8 0.0769 GHG13 1090 36.5 0.70631 0.70613 11 0.0967

143Nd/144Nd ratios have been adjusted relative to the La Jolla standard = 0.511860. Dupl. = duplicate sample.

Table 3. U-Pb isotopic compositions of zircons from the Baiqi formation

Spots Th/U *Pb 207Pb/206Pb 207Pb/235U 206Pb/238U 208Pb/232Th 207Pb/235U 206Pb/238U 208Pb/232Th ±1σ ±1σ ±1σ ±1σ ±1σ (%) age (Ma) age (Ma) age (Ma) 10# 0.65 1 0.052 ± 0.004 0.144 ± 0.011 0.020 ± 0.00031 0.006 ± 0.00019 137 ± 10 129 ± 2 128 ± 1 16# 0.71 0.047 ± 0.006 0.129 ± 0.015 0.020 ± 0.00043 0.006 ± 0.00028 124 ± 14 128 ± 3 128 ± 6 17# 1.22 0.051 ± 0.003 0.133 ± 0.007 0.019 ± 0.00027 0.007 ± 0.00011 127 ± 6 121 ± 2 131 ± 2 18# 1.27 0.052 ± 0.002 0.143 ± 0.006 0.020 ± 0.00026 0.006 ± 0.00010 135 ± 5 127 ± 2 125 ± 2 21# 1.85 0.049 ± 0.001 0.137 ± 0.004 0.020 ± 0.00022 0.006 ± 0.00006 131 ± 3 130 ± 1 123 ± 1 22# 1.23 0.17 0.053 ± 0.002 0.149 ± 0.006 0.018 ± 0.00027 0.006 ± 0.00011 141 ± 6 130 ± 2 131 ± 2 27# 2.5 0.046 ± 0.001 0.129 ± 0.003 0.019 ± 0.00021 0.006 ± 0.00005 123 ± 3 130 ± 1 121 ± 1 29# 1.85 0.047 ± 0.001 0.128 ± 0.004 0.020 ± 0.00022 0.006 ± 0.00006 122 ± 3 126 ± 1 111 ± 1 34# 1.47 0.047 ± 0.002 0.128 ± 0.006 0.020 ± 0.00027 0.006 ± 0.00009 122 ± 6 125 ± 2 114 ± 2 37# 2.08 0.047 ± 0.001 0.134 ± 0.004 0.021 ± 0.00023 0.006 ± 0.00006 127 ± 4 131 ± 1 119 ± 1 41# 1.36 0.05 ± 0.003 0.133 ± 0.007 0.019 ± 0.00030 0.005 ± 0.00015 127 ± 7 124 ± 2 109 ± 3 42# 1.11 0.051 ± 0.003 0.131 ± 0.008 0.019 ± 0.00030 0.005 ± 0.00013 125 ± 7 120 ± 2 110 ± 3 43# 1.45 0.047 ± 0.002 0.126 ± 0.005 0.019 ± 0.00024 0.006 ± 0.00007 120 ± 4 123 ± 2 112 ± 1

*Pb indicate common Pb (corrected by ComPbCorr#3_151, Andersen, 2002).

cling lower continental crust in the mantle (e.g., Gao et formation locates in the interior of the North China al., 2004; Xu et al., 2002), and/or assimilation- Craton, >1500 km away from the subduction zone of Pa- fractionation-crystallization (AFC) process (e.g., Feeley cific oceanic crust, and thus no significant contribution and Hacker, 1995). Either partial melting of subducted from the subducted Pacific oceanic slab. Although geo- oceanic slab or AFC of a basaltic magma is unlikely for physical data implies a possible old large flat subducted the Baiqi adakitic rocks based on the following observa- slab beneath the studied area (Chen et al., 2005; Pei et tions. al., 2005), these adakitic rocks present no melt-peridotite First, zircon U-Pb dating indicates that the Baiqi interaction feature (e.g., high Mg#) of slab-derived melts adakitic rocks formed at about 126 Ma (Fig. 6A), when traversed part of the mantle peridotite. Second, fractional there was no active subduction zone at the northern mar- crystallziation of basaltic magma is an inefficient mecha- gin of the North China Craton. Furthermore, the Baiqi nism for felsic rocks (Green and Ringwood, 1968). Frac-

Mesozoic adakites: Partial melting of underplated basalts 453 Gabbro from study area 10 138 Pyroxenite-mafic granulite 0.0215 Intermediate-felsic granulite Nushan lower crustal xenolith 134 0 Peridotite from Hannuoba Archean TTG 0.0205 130 Baiqi adakitic rocks -10 126 Pb/ U

2060.0195 238 -20 122 Epsilon Nd(125Ma) Epsilon

0.0185 -30 206Pb/ 238 U age=125.8 3.0 114 (n=13,MSWD=1.7)

-40 0.0175 0.70.710.720.730.740.75 0.08 0.10 0.12 0.14 0.16 0.18

87 86 207 235 ( Sr/ Sr)i Pb/ U 87 86 ε (A) Fig. 5. Sr/ Sr (125 Ma) – Nd(125 Ma) plot for the Baiqi adakitic rocks. Gabbros from Laiyuan, Hanxing, Linxian and Xishu (Cai et al., 2003; Chen et al., 2004; Wang, Y. J. et al., 2005; Zhang et al., 2004), lower crustal xenoliths from Nushan (Huang et al., 2004; Yu et al., 2003), granulite xenoliths from Hannuoba (Liu et al., 2004; Zhang et al., 1998; Zhou et al., 2002), Archean TTG (Jahn et al., 1988; Jahn and Zhang, 1984; Sun et al., 1992), peridotite xenoliths from Hannuoba (Rudnick et al., 2004) are shown for comparison.

tional crystallization degree of >70% is required to produce the trondhjemitic liquids (Spulber and Rutherford, 1983) or rhyolite (Meijer, 1983). Furthermore, fractional crystallization of garnet will produce liquid with not only (B) high Sr/Y and La/Yb ratios, but also negative Al O -SiO 2 3 2 Fig. 6. (a) Cathodoluminescence (CL) images of representa- correlations, differing from the trends of the Baiqi adakitic tive zircon grains. (b) Concordia plots of zircon U-Pb isotopic rocks (Fig. 2A). Third, although the Baiqi adakitic rocks results. have various SiO2 contents from 57 to 71%, their Sr and Nd isotopic compositions are remarkably homogeneous (Table 2). This is the most important, because the homogeneous Sr and Nd isotopic compositions suggest that the petrology demonstrate that partial melting of dry Baiqi formation derived from one magma chamber, and peridotite produces basaltic melt (e.g., Falloon and Green, were not contaminated significantly by evolved old base- 1987; Falloon et al., 1988; Hirose and Kushiro, 1993; ment rocks such as 1.9 Ga or 2.5 Ga lower crustal Kushiro, 2001; Walter, 1998), and partial melting of hy- xenoliths from Hannuoba (Liu et al., 2004) and Nushan drous peridotite can produce high Mg# andesitic melt (Huang et al., 2004; Yu et al., 2003) or Archean terrane (Hirose, 1997). On the other hand, dehydration melting granulites (Jahn et al., 1988; Jahn and Zhang, 1984; Liu of mafic lower crust is known to be capable of giving et al., 2004; Sun et al., 1992) (Fig. 5). rise to felsic melts. It remains to resolve whether or not Early Cretaceous magmatism is also developed exten- adakitic rocks can be produced by this kind of anatexis in sively along the Dabie-Sulu orogenic belt that formed by post-orogenic tectonic settings. Nevertheless, the remark- Triassic collision between the North China and Yangtze ably homogeneous Sr-Nd isotopic compositions for the Cratons (Cong, 1996; Jahn et al., 2003; Zheng et al., Baiqi adakitic rocks preclude the possibility that they were 2003). Anatexis of subducted continent and thus result- produced by the melting of orogenic lithospheric keel that ant orogenic lithospheric keel has been suggested to in- has sandwich-layers composed of felsic and mafic- terpret the petrogenesis of post-collisional intrusives (in- ultramafic rocks like those in the Dabie orogen (Zhao et cluding adakitic ones) in the Dabie orogen (Zhang et al., al., 2004, 2005). 2002; Zhao et al., 2004, 2005). Studies of experimental As suggested by experiments (Rapp and Watson,

454 X.-C. Wang et al. 1000 1000 DMP-72 DMP-72

e 0.1 0.1 0.2 0.2 100 0.3 100 0.3 0.4 0.4 Baiqi Baiqi 10 10

1 1 Sample/primitive mantl Sample/primitive Grt:Cpx=1:1 A Grt:Cpx=4:1 B 0.1 0.1 Th Nb La Pr Nd Sm Gd Dy Ho Yb Th Nb La Pr Nd Sm Gd Dy Ho Yb Ba U Ta Ce Sr Zr Eu Tb Y Er Lu Ba U Ta Ce Sr Zr Eu Tb Y Er Lu

1000 1000 Gabbro Gabbro 0.1 0.1 0.2 0.2 100 0.3 100 0.3 0.4 0.4 Baiqi Baiqi 10 10

1 1 Sample/primitive mantle Sample/primitive Grt:Cpx=4:1 C Grt:Cpx=1:1 D 0.1 0.1 Th Nb La Pr Nd Sm Gd Dy Ho Yb Th Nb La Pr Nd Sm Gd Dy Ho Yb Ba U Ta Ce Sr Zr Eu Tb Y Er Lu Ba U Ta Ce Sr Zr Eu Tb Y Er Lu

Fig. 7. Primitive mantle-normalized spider diagrams of melts equilibrated with garnet-rich pyroxenite/eclogite. Based on batch melting model (Shaw, 1970), the melts were calculated assuming the Mesozoic gabbros from Hanxing (Zhang, H. F., 2005, unpublished data) and mafic granulite xenolith (DMP-72) from Hannuoba (Liu et al., 2001) as starting material. The Parameters used for calculations are listed in Table 4. The results indicate that gabbro-derived melts agree well with the Baiqi adakitic rocks.

1995), the Baiqi adakitic rocks could have been formed basaltic underplating) triggering partial melting of the by partial melting of mafic lower crust. Melts derived from preexisting lower crust in the Trans-North China Orogen. partial melting of recycled lower continental crust in the If the lower crust would be thickened by collisional mantle will obtain mantle signature (e.g., high Mg# num- orogeny, the petrogenetic model for the Dabie post- bers and Cr, Ni contents) due to reaction with the mantle collisional magmatism (Zhao et al., 2004, 2005) could peridotite (Liu et al., 2005; Rapp et al., 1999). However, be also applicable to the Baiqi adakitic rocks. In the next low Mg# number and Cr and Ni contents of the Baiqi section, we discuss possible sources of the Baiqi adakitic adakitic rocks (Table 1), differing from the Mesozoic rocks and processes triggering the partial melting of thick- volcanics from Xinglonggou (Gao et al., 2004), are at ened lower crust. variance with their derivation from partial melting of recycled lower continental crust in the mantle, but consist- Links between partial melting of thickened lower crust ent with the conclusion that no significant Mesozoic and basaltic underplating lithospheric thinning took place within the Trans-North If the Baiqi adakitic rocks were produced by partial China Orogenic belt based on Re-Os isotopic system of melting of thickened lower crust, what were their parent the peridotite from the Hannuoba basalt (Gao et al., 2002). rocks and what did drive the partial melting of the thick- Thus, partial melting of a thickened lower continental ened lower crust? The Baiqi adakitic rocks have remark- 87 86 ε crust could be responsible for the formation of the Baiqi ably different Sr/ Sr ratios and Nd(t) values from that adakitic rocks, as suggested by Wang, Q. et al. (2005) for of the Precambrian lower crustal xenoliths (Huang et al., those from Tibet. This recognition is very important, be- 2004; Liu et al., 2004; Yu et al., 2003) and granulite-TTG- cause (1) it suggests that partial melting of the crustal khondalite exposed on the Earth’s surface (Jahn et al., rocks were not attributed to recycling of lower continen- 1988; Jahn and Zhang, 1984; Liu et al., 2004; Sun et al., tal crust into asthenospheric mantle as demonstrated by 1992) at 125 Ma (Fig. 5), which demonstrate clearly that those adakitic volcanics from Xinglonggou (Gao et al., the Baiqi adakitc rocks were not formed by anatexis of 2004); and (2) it implies a crust with thickness >40 km the Precambrian lower continental crust. (>1.2 GPa) (Petford and Atherton, 1996; Rapp and However, the Baiqi adakitic rocks have Sr-Nd isotopic Watson, 1995) and a strong Mesozoic thermal event (e.g., compositions overlapping with the adjacent Mesozoic

Mesozoic adakites: Partial melting of underplated basalts 455 Table 4. Parameters used for modeling calculations

Partition coefficient Initial material Grt:Cpx = 1:1 Grt:Cpx = 4:1 Baiqi adakite

Dcpx/melt Dgrt/melts Gabbro average 1 average 2 average(n = 8) (ppm) (ppm) (ppm) (ppm)

K 0.007 0.0003 9841 35164 35368 23047 Sr 0.08 0.005 300 1454 1629 1126 Y 0.58 3.10 22.0 13.9 10.4 13.6 Nb 0.021 0.008 4.58 15.9 16.0 10.2 Ta 0.012 0.004 0.26 0.92 0.92 0.62 Rb 0.0035 0.0007 31.3 112 113 56.2 La 0.054 0.01 11.2 36.2 37.5 44.1 Ce 0.098 0.021 27.5 84.9 89.9 79.3 Pr 0.15 0.054 3.86 10.8 11.5 9.19 Nd 0.21 0.087 19.1 48.6 52.4 35.3 Sm 0.26 0.217 4.63 10.1 10.3 5.17 Eu 0.31 0.32 1.42 2.77 2.76 1.63 Gd 0.3 0.498 4.05 7.04 6.56 4.73 Tb 0.31 0.75 0.57 0.85 0.75 0.52 Dy 0.33 1.06 3.21 4.08 3.41 2.48 Ho 0.31 1.53 0.6 0.63 0.50 0.43 Er 0.30 2.00 1.55 1.43 1.06 1.13 Tm 0.29 3.00 0.2 0.14 0.13 0.16 Yb 0.28 4.03 1.25 0.69 0.48 1.03 Lu 0.28 5.50 0.17 0.07 0.05 0.16 Th 0.007 0.0015 1.03 3.67 3.69 5.04 U 0.008 0.006 0.35 1.23 1.23 1.25 Pb 0.13 0.18 6.78 17.1 16.6 14.7 Ba 0.0019 0 429 1545 1547 1496 Zr 0.093 0.40 55.1 119 104 163 Hf 0.17 0.31 2.06 4.35 4.08 4.04 Sr/Y 13.6 105 156 83.2 La/Yb 8.76 52.3 77.5 43.3

The initial material is a gabbro sample from Hanxing area (Zhang, H. F., 2005, unpublished data). The 1 and 2 are average calculation melts who are derived from batch melting of Hanxing gabbro with partial melting degree F = 0.2, 0.3, 0.4. The REE partition coefficients data are from McKenzie and O’Nions (1991) and the other partition coefficients from Klemme et al. (2002).

lower crustal xenoliths from the Hannuoba basalts (Liu 238U age of 125.8 ± 3.0 Ma should record the formation et al., 2004; Zhou et al., 2002) and gabbros from the age of the Baiqi adakitic rocks. Furthermore, the ages of Trans-North China Orogen (Zhang et al., 2004) (Fig. 5), intermediate-felsic igneous rocks around the study area which implies that the Baiqi adakitic rocks could be ge- cluster in the range of 138–125 Ma (Cai et al., 2003; Chen netically related with the Mesozoic lower crustal xenoliths et al., 2004; Davis et al., 1998; Peng et al., 2004). All and/or gabbros. Zircon cathodoluminescence (CL) these ages make it reasonable to speculate that the ~150– imaging and U-Pb dating for granulite and olivine 160 Ma gabbro or granulite xenoliths formed by basaltic pyroxenite demonstrate two episode of basaltic underplating could be the parent rocks of the Baiqi adakitc underplating and subsequent granulite-facies metamor- volcanics, consistent with their similar Sr-Nd isotopic phism at 160–140 Ma and 140–80 Ma, respectively. The compositions (Fig. 5). Heating of the subsequent ~140– most important is that the overlapping timing for basaltic 120 Ma basaltic underplating caused granulite-facies underplating and granulite-facies metamorphism indicates metamorphism and partial melting of the mafic lower crust that the latter was induced by the former (Liu et al., 2004). formed at ~150–160 Ma. This coincides with the Early This agrees well with the two episodes of gabbros from Cretaceous giant igneous event in eastern China (132~120 Laiyuang-Linxian and Yunmengshan-Hanxing, which Ma, peaked at 125 Ma; Wu et al., 2005). While the west- were dated at 125–145 Ma and 150–160 Ma, respectively ward subduction of the Pacific plate beneath the Eura- (Davis et al., 1998; Wang, Y. J. et al., 2005; Zhang et al., sian continent may provide a geodynamic setting for de- 2004). The typical igneous oscillatory zonation of the zir- velopment of Mesozoic magmatism along the eastern edge cons (Fig. 6B) in this work indicates that the mean 206Pb/ of the China continent (Wu et al., 2005), the intensive

456 X.-C. Wang et al. occurrence of Early Cretaceous magmatism in this wide 2004; Chen and Zhai, 2003; Cai et al., 2003; Davis, 2003; area points to a thermal event that may be associated with Peng et al., 2004; Li and Li, 2004). This implies that the the Pacific superplume activity as advocated by Zhao et lithosphere of the Trans-North China Orogen might have al. (2004, 2005) for the post-collisional magmatism in not thinned as the east part of the East China suggested the Dabie orogen. Furthermore, the estimated tempera- by Gao et al. (2002). We speculate that basaltic tures of pyroxenite and granulite xenoliths from the underplating, and possibly the resultant chemical erosion, Hannuoba basalts could be high up to 900–1100°C (Chen could dominate over much of the Trans-North China et al., 2001; Liu et al., 2003). This implies that the basal- Orogen. SiO2-rich melt-mantle peridotite interaction (Liu tic underplating can provide the necessary heat to induce et al., 2005) and the resultant chemical erosion (Xu, 2001) partial melting of the preexisting mafic lower crust within the gradually upward moving lithosphere- (Guffanti et al., 1996; Petford and Gallagher, 2001; asthenosphere interface played an important role in gen- Rushmer, 1991). erating a thickened lower continental crust due to basal- If the above model is correct, the Baiqi adakitic rocks tic underplating. should match the partial melts of ~150–160 Ma gabbros or mafic granulites. Batch melting models were calcu- CONCLUSION lated assuming ~150–160 Ma gabbros (Zhang, H. F, 2005, unpublished data) and mafic granulite xenoliths from the The Baiqi volcanics possess the same geochemical Hannuoba basalts (Liu et al., 2001) as starting materials, characteristics as those of typical adakites in modern arcs. respectively (Fig. 7). The results indicate that the Baiqi However, they have relatively lower Cr and Ni contents adakitic rocks match well the partial melts formed by 20– and Mg#, suggesting that their parental magma did not 40% partial melting of gabbros (Table 4) leaving the interact with the lithospheric mantle. Combined with the restite of 50% Grt + 50% Cpx (Figs. 4, 7C and 7D). How- geological setting of the studying area, these geochemical ever, partial melts of mafic granulites present lower Ba, features suggest that the Baiqi adakitic rocks formed by Th, U, Nb, Ta and LREE (Figs. 7A and B). This diver- partial melting of a thickened lower continental crust at a gence indicates that the mafic granulites could suffer from depth of >40 km in central North China. More impor- granulite-facies metamorphism/partial melting at 80–140 tantly, the formation of the Baiqi adakitic volcanics im- Ma resulted in decrease in these elements. As a result, we plies that partial melting of the underplated mafic lower suggest that the source rock of the Baiqi adakitic rocks crust of ~150–160 Ma, rather than the Precambrian base- could be represented by the ~150–160 Ma gabbro. ment rocks, was caused by enormous conductive heating The above observations and discussions outline such from a subsequent basaltic underplating event at 80–140 a process that the thickened lower crust generated by ba- Ma. saltic underplating at ~150–160 Ma experienced subsequent granulite-facies metamorphism and thus partial Acknowledgments—We thank Dr. S. Gao for the assistance in melting, which could have been caused by heating of an- field work, and J. Q. Wang for XRF whole rock analyses. Con- other phase of congenetic basaltic underplating at ~80– structive comments by Dr. X. H. Li and Dr. S. Gao have sub- 140 Ma. However, where was the basaltic melt derived stantially improved the manuscript. We also thank Dr. H. F. from remains enigmatic due to its evolved Sr-Nd isotopic Zhang for provide their unpublished data. Two anonymous re- viewers are thanked for their helpful comments that improved compositions compared with the mantle peridotites the manuscript significantly. This work was financially sup- (Rudnick et al., 2004) (Fig. 5). ported by National Nature Science Foundation of China (Nos. 40521001, 40473013) and the Program for Changjiang Schol- Implications for the Mesozoic deep geodynamic process ars and Innovative Research Team in University (IRT0441). in the Trans-North China Orogen differing from the east part of the East China Gao et al. (2004) argued that partial melting of REFERENCES delaminated lower crust in the mantle from Jurassic vol- Andersen, T. (2002) Correction of common lead in U-Pb analy- canic rocks at Liaoxi indicates that lithospheric foundering ses that do not report 204Pb. Chem. Geol. 192, 59–79. in the east part of the North China Craton could have Atherton, M. P. and Petford, N. (1993) Generation of sodium- served as the heat engine. During the delamination proc- rich magmas from newly Underplated basaltic crust. Na- ess, the upwelling of asthenospheric mantle heated the ture 362, 144–146. overlying crust and resulted in the formation of juvenile Black, L. P., Kamo, S. L., Allen, C. M., Aleinikoff, J. N., Davis, crust by underplating of mantle-derived basaltic magma. D. W., Korsch, R. J. and Foudoulis, C. (2003) TEMORA 1: a new zircon standard for Phanerozoic U-Pb geochronology. However, neither Mesozoic high-Mg andesite nor high- Chem. Geol. 200, 155–170. Mg adakitic rock has been identified in the Trans-North Cai, J. H., Yan, G. H., Chang, Z. S., Wang, X. F., Shao, H. X. China Orogen (Li et al., 2001; Chen et al., 2002, 2003, and Chu, Z. Y. (2003) Petrological and geochemical char-

Mesozoic adakites: Partial melting of underplated basalts 457 acteristics of the Wang’an complex and discussion on its facies garnet-pyroxenite xenoliths in Hannuoba area: new genesis. Acta Petrologica Sin. (in Chinese with English evidence of magma underplating. Acta Petrologica Sin. 17, abstract) 19, 81–92. 1–6 (in Chinese with English abstract). Chen, B. and Zhai, M. G. (2003) Geochemistry of late Mesozoic Feeley, T. C. and Hacker, M. D. (1995) Intracrustal derivation lamprophyre dykes from the Taihang Mountains, north of Na-rich andesite and dacite magmas: an example from China, and implications for the sub-continental lithospheric Volcan Ollagüe Andean Central Volcanic Zone. J. Geol. 103, mantle. Geol. Mag. 140, 87–93. 213–225. Chen, B., Zhai, M. G. and Shao, J. (2002) Petrogenesis and Gao, S., Rudnick, R. L., Carlson, R. W., McDonough, W. F. significance of the Mesozoic North Taihang complex: ma- and Liu, Y. S. (2002) Re-Os evidence for replacement of jor and trace element evidence. Sci China (D) 46, 941–953. ancient mantle lithosphere beneath the North China craton. Chen, B., Jahn, B. M. and Zhai, M. G. (2003) Sr-Nd isotopic Earth Planet. Sci. Lett. 198, 307–322. characteristics of the Mesozoic magmatism in the Taihang- Gao, S., Rudnick, R. L., Yuan, H. L., Liu, X. M., Liu, Y. S., Xu, Yanshan orogen, north China craton, and implications for W. L., Ling, W. L., Ayers, J., Wang, X. C. and Wang, Q. H. Archean lithosphere thinning. J. Geol. Soc. Lond. 160, 963– (2004) Recycling lower continental crust in the North China 970. craton. Nature 432, 892–897. Chen, B., Jahn, B. M., Arakawa, Y. and Zhai, M. G. (2004) Green, T. G. and Ringwood, A. E. (1968) Genesis of the calc- Petrogenesis of the Mesozoic intrusive complexes from the alkaline igneous rocks suit. Contrib. Mineral. Petrol. 18, southern Taihang Orogen, North China Craton: elemental 105–162. and Sr-Nd-Pb isotopic constraints. Contrib. Mineral. Pet- Griffin, W. L., Zhang, A., O’Reilly, S. Y. and Ryan, C. G. (1998) rol. 148, 489–501. Phanerozoic evolution of the lithosphere beneath the Sino- Chen, S. H., O’Reilly, S. Y., Zhou, X., Griffin, W. L., Zhang, Korean Craton. Mantle Dynamics and Plate Interactions in G., Sun, M., Feng, J. and Zhang, M. (2001) Thermal and East Asia (Flower, M., Chung, S. L., Lo, C. H. and Lee, T. petrological structure of the lithosphere beneath Hannuoba, Y., eds.), 107–126, American Geophysical Union, Washing- Sino-Korean Craton, China: Evidence from xenoliths. Lithos ton, D.C. 56(4), 267–301. Guffanti, M., Clynne, M. A. and Muffler, L. J. P. (1996) Ther- Chen, Y. J., Pei, S., Zhao, D., Yin, A., Ning, J. and Chen, X. F. mal and mass implications of magmatic evolution in Lassen (2005) Tomographic Structure of East Asia: 2. Stagnant slab volcanics region, California, and constraints on basalt in- above 660-km discontinuity and its geodynamic implica- flux to the lower crust. J. Geophys. Res. 101, 3003–3013. tions. J. Geophys. Res. (revised in February 2005). Hirose, K. (1997) Melting experiments on lherzolite KLB-1 Condie, K. C. (2005) TTGs and adakites: are they both slab under hydrous conditions and generation of high-magnesian melts? Lithos 80, 33–44. andesitic melts. Geology 25, 42–44. Cong, B.-L. (1996) Ultrahigh-Pressure Metamorphic Rocks in Hirose, K. and Kushiro, I. (1993) Partial melting of dry the Dabieshan-Sulu Region of China. Science Press, Beijing, peridotites at high pressures: determination of compositions 224 pp. of melts segregated from peridotite using aggregates of dia- Davis, G. A. (2003) The Yanshan belt of North China: Tecton- mond. Earth Planet. Sci. Lett. 114, 477–489. ics, adakitic magmatism, and crustal evolution. Earth Sci- Huang, X. L., Xu, Y. G. and Liu, D. Y. (2004) Geochronology, ence Frontiers 10(4), 373–384. petrology and geochemistry of the granulite xenoliths from Davis, G. A., Zheng, Y. D., Wang, C., Darby, B. J., Zhang, C. Nushan, east China: Implication for a heterogeneous lower and Gehrels, G. (1998) Geometry and Geochronology of crust beneath the Sino-Korean Craton. Geochim. Yanshan Belt Tectonics, Collected Works of International Cosmochim. Acta 68(1), 127–149. Symposium on Geological Science. Seismic Press, Beijing, Jahn, B.-m. and Zhang, Z. Q. (1984) Archean granulite gneisses 275–292 pp. from eastern Hebei Province, China: rare earth geochemistry Defant, M. J. and Drummond, M. S. (1990) Derivation of some and tectonic implications. Contrib. Mineral. Petrol. 85, 224– modern arc magmas by melting of young subducted 243. lithosphere. Nature 347, 662–665. Jahn, B.-m., Auvray, B., Shen, Q. H., Liu, D. Y., Zhang, Z., Falloon, T. J. and Green, D. H. (1987) Anhydrous partial melt- Dong, Y. J., Ye, X. J., Zhang, Q. Z., Cornichet, J. and Mace, ing of MORB pyrolite and other peridotite compositions at J. (1988) Archean crustal evolution in China: the Taishan 10 kbar and implications for the origin of primitive MORB complex, and evidence for juvenile crustal addition from glasses. Contrib. Mineral. Petrol. 37, 181–219. long-term depleted mantle. Precamb. Res. 38, 381–403. Falloon, T. J., Green, D. T., Hatton, C. J. and Harris, K. L. (1988) Jahn, B.-m., Rumble, D. and Liou, J. G. (2003) Geochemistry Anhydrous partial melting of a fertile and depleted peridotite and isotope study of UHP metamorphic rocks. Ultrahigh from 2 to 30kbar and application to basalt petrogenesis. J. Pressure Metamorphism (Carswell, D. A. and Compagnoni, Petrol. 29, 1257–1282. R., eds.), EMU Notes in Mineralogy 5, 365–414. Fan, Q. C., Liu, R. X., Li, H. M., Li, N., Sui, J. L. and Lin, Z. Kay, R. W. and Kay, S. M. (1991) Creation and destruction of R. (1998) Zircon geochronology and rare earth element lower continental crust. Geol. Rundsch. 80, 259–278. geochemistry of granulite xenoliths from Hannuoba. Chi- Kay, R. W. and Kay, S. M. (1993) Delamination and nese Sci. Bull. 43, 133–137 (in Chinese with English ab- delamination magmatism. Tectonophysics 219, 177–189. stract). Klemme, S., Blundy, J. D. and Wood, B. D. (2002) Experimen- Fan, Q. C., Sui, J. L., Liu, R. X. and Zhou, X. M. (2001) Eclogite tal constraints on major and trace element partitioning dur-

458 X.-C. Wang et al. ing partial melting of eclogite. Geochim. Cosmochim. Acta Peacock, S. M., Rushmer, T. R. and Thompson, A. B. (1994) 66, 3109–3123. Partial melting of subducting oceanic crust. Earth Planet. Kushiro, I. (2001) Partial melting experiments on peridotites Sci. Lett. 121, 227–244. and origin of mid-ocean ridge basalt. An. Rev. Earth Planet. Pei, S., Chen, Y. J., Zhao, D., Yin, A. Ning, J. and Chen, X. F. Sci. 29, 71–107. (2005) Tomographic Structure of East Asia: 1. No fast (slab) Li, W. P. and Li, X. H. (2004).Petrogenesis and its implication anomalies beneath 660-km discontinuity. J. Geophys. Res. for the middle-late Jurassic intermediate-acidic volcanic (revised in February 2005). rocks in the middle section of Yanshan orogen. Acta Peng, T. P., Wang, Y. J., Fan, W. M., Peng, B. X. and Guo, F. Petrologica Sin. 20, 501–510 (in Chinese with English ab- (2004) SHRIMP zircon U-Pb geochronology of the diorites stract). for the southern Taihang Mountains in Central North Inte- Li, W. P., Lu, F. X. and Li, X. H. (2001) Genesis of Late Jurassic rior and its Petrogenesis. Acta Petrologica Sin. 20, 1253– Trachyandesite in Western Hills of Beijing and Its Geologi- 1262 (in Chinese with English abstract). cal Implications. Acta Petrol. Mineral 20, 247–254 (in Chi- Petford, N. and Atherton, M. (1996) Na-rich partial melts from nese with English abstract). newly underplated basaltic crust: the cordillera blanca Ling, W.-L., Gao, S., Zhang, B. R., Li, H. M., Liu, Y. and Cheng, batholith, Peru. J. Petrol. 37, 1491–1521. J. P. (2003) Neoproterozoic tectonic evolution of the north- Petford, N. and Gallagher, K. (2001) Partial melting of mafic western Yangtze craton, South China: implications for amal- (amphibolitic) lower crust by periodic influx of basaltic gamation and break-up of the Rodinia Supercontinent. magma. Earth Planet. Sci. Lett. 193, 483–499. Precamb. Res. 122, 111–140. Rapp, R. P. (1995) Amphibole-out phase boundary in partially Liu, Y. S., Gao, S., Jin, S. Y., Hu, S. H., Sun, M., Zhao, Z. B. melted metabasalt, its control over liquid fraction and com- and Feng, J. L. (2001) Geochemistry and petrogenesis of position, and source permeability. J. Geophys. Res. 100, lower crustal xenoliths from Hannuoba, North China: im- 15601–15610. plications for the continental lower crustal composition and Rapp, R. P. and Watson, E. B. (1995) Dehydration melting of evolution at convergent margin. Geochim. Cosmochim. Acta metabasalt at 8–32 kbar: Implications for continental growth 65(15), 2589–2604. and Crust-mantle recycling. J. Petrol. 36, 891–931. Liu, Y. S., Gao, S., Liu, X. M., Chen, X. M., Zhang, W. L. and Rapp, R. P., Watson, E. B. and Miller, C. F. (1991) Partial melt- Wand, X. C. (2003) Thermodynamic evolution of ing of amphibolite/eclogite and the origin of Archean lithosphere of the North China Craton: Records from lower trondhjemites and tonalites. Precamb. Res. 51, 1–25. crust and upper mantle xenoliths from Hannuoba. Chinese Rapp, R. P., Shimizu, N., Norman, M. D. and Applegate, G. S. Sci. Bull. 48(21), 2371–2377. (1999) Reaction between slab-derived melts and peridotite Liu, Y. S., Gao, S., Yuan, H. L., Zhou, L., Liu, X. M., Wang, X. in the mantle wedge: experimental constraints at 3.8 GPa. C., Hu, Z. C. and Wang, L. S. (2004) U-Pb zircon dates and Chem. Geol. 160, 335–356. Nd, Sr and Pb isotopes of lower crustal xenoliths from North Rudnick, R. L., Gao, S., Ling, W. L., Liu, Y. S. and McDonough, China Craton: Insights on evolution of lower continental W. F. (2004) Petrology and geochemistry of spinel peridotite crust. Chem. Geol. 211(1–2), 87–109. xenoliths from Hannuoba and Qixia, North China craton. Liu, Y. S., Gao, S., Lee, C. T., Hu, S. H., Liu, X. M. and Yuan, Lithos 77, 609–637. H. L. (2005) Melt-peridotite interactions: Links between Rushmer, T. (1991) Partial melting of two amphibolites: con- garnet pyroxenite and high-Mg# signature of continental trasting experimental results under fluid-absent conditions. crust. Earth Planet. Sci. Lett. 234(1–2), 39–57. Contrib. Mineral. Petrol. 107, 41–59. Ludwig, K. R. (2003) ISOPLOT 3.00: A Geochronological Sen, C. and Dunn, T. (1994) Dehydration melting of a basaltic Toolkit for Microsoft Excel. Berkeley Geochronology composition amphibolite at 1.5 and 2.0 Gpa: implications Center, Berkeley, California, 39 pp. for the origin of adakites. Contrib. Mineral. Petrol. 117, Martin, H. (1999) Adakitic magmas: modern analogues of 394–409. Archaean granitoids. Lithos 46, 411–429. Shaw, D. M. (1970) Trace element fraction during anatexis. Martin, H., Smithies, R. H. and Rapp, R. (2005) An overview Geochim. Cosmochim. Acta 34, 237–243. of adakite, tonalite-trondhjemite-granodiorite (TTG), and Spulber, S. D. and Rutherford, M. J. (1983) The origin of sanukitoid: relationships and some implications for crustal rhyolite and plagiogranite in oceanic crust: An experimen- Evolution. Lithos 79, 1–24. tal study. J. Petrol. 24, 1–25. McKenzie, D. and O’Nions, R. K. (1991) Partial melt distribu- Stern, C. R. and Kilian, R. (1996) Role of the subducted slab, tions from inversion of rare Earth element concentrations. mantle wedge and continental crust in the generation of J. Petrol. 32, 1021–1091. adakites from the Andean Austral Volcanic Zone. Contrib. Meijer, A. (1983) The origin of low-K rhyolites from the Mineral. Petrol. 123, 263–281. Mariana frontal arc. Contrib. Mineral. Petrol. 83, 45–51. Sun, M., Armstrong, R. L. and Lambert, R. J. (1992) Petro- Menzies, M. A., Fan, W. M. and Zhang, M. (1993) Paleozoic chemistry and Sr, Pb, and Nd isotopic geochemistry of Early and Cenozoic lithoprobes and the loss of >120 km of Precambrian rocks, Wutaishan and Taihangshan areas, Archean lithosphere, Sino-Korean craton, China. Magmatic China. Precamb. Res. 56, 1–31. Processes and Plate Tectonics (Prichard, H. M., Alabaster, Sun, S.-s. and McDonough, W. F. (1989) Chemical and iso- T., Harris, N. B. W. and Neary, C. R., eds.), 71–81, Geol. topic systematics of oceanic basalts: implications for man- Soc. Spec. Pub., London. tle composition and processes. Magmatism in the Ocean

Mesozoic adakites: Partial melting of underplated basalts 459 Basins (Saunders, A. D. and Norry M. J., eds.), 313–345, nal of Geostandards and Geoanalysis 28, 353–370. Geol. Soc. London, Spec. Publ. Zhang, G. H., Zhou, X. H., Sun, M., Chen, S. H. and Feng, J. L. Walter, M. J. (1998) Melting of garnet peridotite and the origin (1998) Sr, Nd and Pb isotopic characteristics of granulite of komatiite and depleted lithosphere. J. Petrol. 39, 29–60. and pyroxenite xenoliths in the Hannuoba basalts, Hebei Wang, Q., McDermott, F., Xu, J. F., Bellon, H. and Zhu, Y. T. province: implications for geological processes. Acta (2005) Cenozoic K-rich adakitic volcanic rocks in the Petrologica Sin. 14, 190–197 (in Chinese with English ab- Hohxil area, northern Tibet: Lower-crustal melting in an stract). intracontinental setting. Geology 33, 465–468. Zhang, H.-F., Sun, M., Zhou, X. H., Fan, W. M. and Yin, J. F. Wang, Y. J., Fan, W. M., Zhang, H. F. and Peng, T. P. (2005) (2002) Mesozoic lithosphere destruction beneath the North Early Cretaceous gabbroic rocks from the Taihang Moun- China Craton: evidence from major-, trace-element and Sr- tains: Implications for a paleosubduction-related Nd-Pb isotope studies of Fangcheng basalts. Contrib. Min- lithospheric mantle beneath the central North China Craton. eral. Petrol. 144, 241–253. Lithos (in press). Zhang, H. F., Sun, M., Zhou, M. F., Fan, W. M., Zhou, X. H. Wilde, S. A., Zhou, X. H., Nemchin, A. A. and Sun, M. (2003) and Zhai, M. G. (2004) Highly heterogeneous Late Mesozoic Mesozoic crust-mantle interaction beneath the North China lithospheric mantle beneath the North China Craton: Evi- craton: A consequence of the dispersal of Gondwanaland dence from Sr-Nd-Pb isotopic systematics of mafic igne- and accretion of Asia. Geology 31(9), 817–820. ous rocks. Geol. Mag. 141, 55–62. Winther, K. T. (1996) An experimentally based model for the Zhang, Q., Wang, Y., Qian, Q., Yang, J. H., Wang, Y. L., Zhao, origin of tonalitic and trondhjemitic melts. Chem. Geol. 127, T. P. and Guo, G. J. (2001a) The characteristics and tectono- 43–59. metallogenic significance of the adakites in Yanshan pe- Wolf, M. B. and Wyllie, P. J. (1994) Dehydration-melting of riod from eastern China. Acta Petrologica Sin. 17, 236–244 amphibolite at 10 kbar: the effects of temperature and time. (in Chinese with English abstract). Contrib. Mineral. Petrol. 115, 369–383. Zhang, Q., Wang, Y., Wang, E. C., Wang, Y., Zhao, T. P., Hao, Wu, F. Y., Walker, R. J., Ren, X. W., Sun, D. Y. and Zhou, X. H. J. and Guo, G. J. (2001b) Existence of east China Plateau (2003) Osmium isotopic constraints on the age of in mid-late Yanshan period: implication from adakite. lithospheric mantle beneath northeastern China. Chem. Scientia Geologica Sin. 36, 248–255 (in Chinese with Eng- Geol. 196, 107–129. lish abstract). Wu, F. Y., Lin, J. Q., Wilde, S. A., Zhang, X. O. and Yang, J. H. Zhao, G.-C., Cawood, P. A., Wilde, S. A., Sun, M. and Lu, L. (2005) Nature and significance of the Early Cretaceous gi- Z. (2000) Metamorphism of basement rocks in the Central ant igneous event in eastern China. Earth Planet. Sci. Lett. Zone of the North China Craton: implications for 233, 103–119. Paleoproterozoic tectonic evolution. Precamb. Res. 103, 55– Xiao, L., Rapp, P. R. and Xu, J. F. (2004) The role of deep 88. processes controls on variation of compositions of adakitic Zhao, G.-C., Wilde, S. A., Cawood, P. A. and Sun, M. (2001) rocks. Acta Petrologica Sin. 20, 219–228 (in Chinese with Archean blocks and their boundaries in the North China English abstract). Craton: lithological, geochemical, structural and P-T path Xiong, X. L., Li, X. H., Xu, J. F., Li, W. X., Zhao, Z. H. and constraints and tectonic evolution. Precamb. Res. 107, 45– Wang, Q. (2003) Extremely high-Na adakite-like magmas 73. derived from alkali-rich basaltic underplate: the Zhantang Zhao, Z.-F., Zheng, Y.-F., Wei, C.-S. and Wu, Y.-B. (2004) Zir- andesites in Huichang basin, SE China. Geochem. J. 37, con isotope evidence for recycling of subducted continen- 233–252. tal crust in post-collisional granitoids from the Dabie terrane Xu, J. F., Shinjo, R., Defant, M. J., Wang, Q. and Rapp, R. P. in China. Geophys. Res. Lett. 31(L22602), doi:10.1029/ (2002) Origin of Mesozoic adakitic intrusive rocks in the 2004GL021061. Ningzhen area of east China: Partial melting of delaminated Zhao, Z.-F., Zheng, Y.-F., Wei, C.-S., Wu, Y.-B., Chen, F. and lower continental crust? Geology 30(12), 1111–1114. Jahn, B.-m. (2005) Zircon U-Pb age, element and C-O iso- Xu, Y. G. (2001) Thermo-tectonic destruction of the Archaean tope geochemistry of post-collisional mafic-ultramafic rocks lithospheric keel beneath the Sino-Korean craton in China: from the Dabie orogen in east-central China. Lithos 83, 1– Evidence, timing and mechanism. Phys. Chem. Earth (A) 28. 26(9–10), 747–757. Zheng, Y.-F., Fu, B., Gong, B. and Li, L. (2003) Stable isotope Yu, J. H., Xu, X. S., O’Reilly, S. Y., Griffin, W. L. and Zhang, geochemistry of ultrahigh pressure metamorphic rocks from M. (2003) Granulite xenoliths from Cenozoic Basalts in SE the Dabie-Sulu orogen in China: Implications for China provide geochemical fingerprints to distinguish lower geodynamics and fluid regime. Earth Sci. Rev. 62, 105–161. crust terranes from the North and South China tectonic Zhou, X. H., Sun, M., Zhang, G. H. and Chen, S. H. (2002) blocks. Lithos 67, 77–102. Continental crust and lithospheric mantle interaction be- Yuan, H. L., Gao, S. and Liu, X. M. (2004) Accurate U-Pb age neath North China: isotopic evidence from granulite and trace element determinations of zircon by laser abla- xenoliths in Hannuoba, Sino-Korean craton. Lithos 62, 111– tion-inductively coupled plasma mass spectrometry. Jour- 124.

460 X.-C. Wang et al. Appendix 1. Major element analyses of the international rock standards(a)

BCR-2(n = 4) GSR-3(n = 3) GSR-1(n = 4)

Meas. 1σ Rec. Meas. 1σ Rec. Meas. 1σ Rec.

SiO 2 54.01 0.24 54.10 44.60 0.02 44.64 72.81 0.02 72.83 TiO2 2.26 0.02 2.26 2.40 0.03 2.37 0.30 0.03 0.29 Al2O3 13.37 0.08 13.50 13.86 0.10 13.83 13.50 0.20 13.40 Fe 2O3* 13.85 0.04 13.80 13.29 0.08 13.4 2.15 0.07 2.14 MnO 0.18 0.01 0.19 0.16 0.01 0.17 0.06 0.01 0.06 MgO 3.69 0.01 3.59 7.79 0.04 7.77 0.44 0.04 0.42 CaO 7.17 0.02 7.12 8.81 0.03 8.81 1.54 0.03 1.55

Na2O 3.10 0.10 3.16 3.47 0.09 3.38 3.07 0.08 3.13 K2O 1.80 0.01 1.79 2.31 0.02 2.32 5.06 0.06 5.01 P2O5 0.35 0.00 0.35 0.94 0.01 0.95 0.09 0.00 0.09 a)n = number of analysis; Meas. = measured value; Rec. = recommended value (http://minerals.cr.usgs.gov/geo_chem_stand); Units are wt% for all elements.

Appendix 2. Analyses of international rock standards BHVO-1, AGV-1 and GSR-1 by ICP-MS(a)

BHVO-1 (n = 5) AGV-1 (n = 5) GSR-1 (n = 3)

Meas. 1σ Rec. Meas. 1σ Rec. Meas. 1σ Rec.

Li 5.08 0.17 4.60 10.5 0.34 12.0 147.4 5.00 131 Be 0.99 0.01 1.10 2.21 0.06 2.10 13.5 1.80 12.4 Sc 31.8 0.07 31.8 12.0 0.06 12.2 6.00 0.80 6.10 V 314 0.17 317 122 0.38 121 25.3 1.20 24.0 Cr 285 5.00 289 12.0 0.13 10.1 9.80 2.30 5.00 Co 45.0 0.29 45.0 15.0 0.08 15.3 3.20 0.80 3.40 Ni 121 3.35 121 15.0 1.68 16.0 5.60 3.20 2.30 Cu 138 0.65 136 56.0 0.48 60.0 3.54 0.40 3.20 Zn 110 0.68 105 80.0 0.25 88.0 25.5 1.60 28.0 Ga 21.0 0.06 21.0 20.3 0.15 20.0 19.9 1.20 19.0 Ge 1.65 0.03 1.64 1.28 0.04 1.25 2.10 0.08 2.00 Rb 9.60 0.13 11.0 66.0 0.18 67.3 454 4.60 466 Sr 399 2.06 403 662 1.33 662 114 4.50 106 Y 27.3 0.05 27.6 21.0 0.09 20.0 65.8 1.00 62.0 Zr 173 0.67 179 233 0.87 227 170 1.35 167 Nb 19.3 0.07 19.0 15.0 0.10 15.0 43.6 0.96 40.0 Cs 0.11 0.01 0.13 1.34 0.01 1.28 36.6 0.10 38.4 Ba 138 0.32 139 1234 8.00 1226 353 16.0 343 La 15.6 0.05 15.8 38.4 0.11 38.0 52.6 1.90 54.0 Ce 38.3 0.01 39.0 68.4 0.10 67.0 100 1.90 108 Pr 5.44 0.02 5.70 8.40 0.03 7.60 11.9 0.41 12.7 Nd 25.6 0.10 25.2 32.8 0.07 33.0 44.5 1.10 47.0 Sm 6.24 0.03 6.20 5.90 0.04 5.90 9.10 0.80 9.70 Eu 2.01 0.01 2.06 1.69 0.02 1.64 0.80 0.06 0.85 Gd 6.17 0.06 6.40 5.40 0.05 5.00 9.10 1.60 9.30 Tb 0.96 0.003 0.96 0.70 0.00 0.70 1.56 0.03 1.65 Dy 5.20 0.01 5.20 3.63 0.01 3.60 9.79 0.12 10.2 Ho 0.98 0.00 0.99 0.68 0.01 0.67 2.11 0.15 2.05 Er 2.36 0.01 2.40 1.75 0.03 1.70 6.18 0.20 6.50 Tm 0.32 0.00 0.33 0.25 0.00 0.34 1.10 0.02 1.10 Yb 2.03 0.02 2.02 1.70 0.02 1.72 7.40 0.16 7.40 Lu 0.30 0.00 0.29 0.26 0.01 0.27 1.20 0.01 1.20 Hf 4.41 0.03 4.38 5.10 0.03 5.10 5.70 0.12 6.30 Ta 1.23 0.005 1.23 0.90 0.01 0.90 6.90 0.04 7.20 Pb 2.33 0.11 2.60 36.3 0.70 36.0 32.2 0.02 31.0 Th 1.25 0.02 1.08 6.39 0.05 6.50 51.6 0.02 54.0 U 0.42 0.01 0.42 1.88 0.02 1.92 18.2 0.04 18.8

(a)n = number of analysis; Meas. = measured value; Rec. = recommended value (http://minerals.cr.usgs.gov/geo_chem_stand); Units are ppm for all elements.

Mesozoic adakites: Partial melting of underplated basalts 461