Abstract

Mafic microgranular (microgranitoid) enclaves (MMEs) with fine-to medium-grain–size igneous microstructures are abundant in Permian granitoid plutons in the northern North China Block (NCB). They are mainly dioritic in composition with SiO2 contents from 51.0 wt% to 62.7 wt% and are contiguous with their host granitoids (SiO2 = 60.4–68.4 wt%). Their main mineral assemblage of plagioclase (oligoclase and andesine), hornblende, biotite, K-feldspar, and quartz is similar to that of their host granitoids but with different mineral proportions. Zircon laser ablation–inductively coupled plasma–mass spectrometer (LA-ICP-MS) U-Pb dating, hornblende-plagioclase thermobarometry, and Ti-in-zircon thermometry results show that the crystallization ages and pressure-temperature (P-T) conditions of the MMEs in each Permian granitoid pluton are very similar to those of their host granitoids, indicating simultaneous crystallization at the middle to upper crustal levels. Petrographical, geochemical, and Sr-Nd-Hf isotopic data show that the source areas of the Permian granitoid plutons in the northern NCB are temporally and spatially different and magma mixing was likely involved during formation of some granitoid plutons. However, the whole-rock Sr-Nd and zircon Hf isotopic compositions of the MMEs in each pluton are very similar to those of their host granitoids, indicating complete isotopic equilibration between the host granitoids and their enclaves. Combined with similar mineral assemblage and mineral chemistry between the MMEs and their host granitoids and absence of coeval mantle-derived rocks in or nearby the plutons, we confirm that the MMEs were crystallized from a coeval magma that gave rise to the host granitoids, and not by mixing and/or mingling between mantle-derived mafic and crustal-derived felsic magmas or by synplutonic injections of mafic magma into the host granitoid magma. We conclude that magma isotopic equilibration between host granitoids and their enclaves was achieved prior to emplacement and the magma mixing process during formation of some Permian granitoid plutons occurred mainly in the melting, assimilation, storage, and homogenization (MASH) zone in the lowermost crust or mantle-crust transition, not in the magma conduits or chambers at middle to upper crustal levels. A two-stage model that includes rapid cooling within the cogenetic host granitoid magma operating at pluton margins to form solid to sub-solid dioritic rocks and crystal accumulations of the host granitoid magma at the bottom of the pluton to form cumulates, and dioritic rocks and cumulates then fragmentated and interacted with progressive evolved host granitoid magma to change their shapes and orientations is proposed for origin of the MMEs in the Permian granitoid plutons in the northern NCB. This model may be more widely applicable to the generation of many MMEs in granitic rocks, especially where no direct evidence exists for the presence of mafic magmas coeval with granitoids and where there is a lack of isotopic contrast between hosts and enclaves. Our new results on origin of the enclaves in the granitoid plutons in the northern NCB show that some MMEs previously used as evidence for magma mixing and/or mingling are “autoliths (cognate xenoliths)” crystallized from the same magma as their host granitoids. There is an essential need for caution in using the MMEs as evidence for crustal-mantle interaction or mixing and/or mingling between crustal-derived felsic and mantle-derived mafic magmas during formation of granitic rocks, especially where no independent evidence is found for such a process. The role of MMEs in evaluating the mixing and/or mingling processes between crustal-derived felsic and mantle-derived mafic magmas has been much more likely overestimated by some previous studies and should be reconsidered. Instead, the MMEs in this origin can provide valuable information on emplacement mechanism and incremental growth of granitoid plutons in continental crust.

1. INTRODUCTION

Mafic microgranular enclaves (MMEs, Didier, 1973, 1991; Didier et al., 1982; Poli and Tommasini, 1991), also termed microgranitoid enclaves (Vernon, 1983, 1984, 1990; Eberz and Nicholls, 1988; Vernon et al., 1988; Elburg and Nicholls, 1995; Elburg, 1996a; Paterson et al., 2004), mafic magmatic enclaves (Barbarin, 1988, 1990, 2005), microgranular magmatic enclaves (Słaby and Martin, 2008), magmatic enclaves (Perugini and Poli, 2012), or igneous enclaves (Ilbeyli and Pearce, 2005), are very common in calc-alkaline granitoids. They are usually darker colored than their host granitoids and show microstructures characteristic of fine- or medium-grain–size igneous rocks (Vernon, 1984; Barbarin, 2005). Studies on MMEs can potentially give us valuable information on the origin and/or the evolution of the magma in which they are found, the mechanism of production of granitic melt, and interactions between continental crust and mantle (e.g., Didier et al., 1982; Poli and Tommasini, 1991; Elburg, 1996b; Yang et al., 2004, 2007).

Although MMEs have been identified in granitic rocks around the world and have been the focus of study for several decades, their origin is still a matter of hot debate (e.g., Phillips, 1880; Grout, 1937; Didier, 1973, 1991; Didier et al., 1982; Grasset and Albarède, 1994; Cheng et al., 2012; Flood and Shaw, 2014, and references therein). Several models have been proposed for the origin of MMEs in granitic rocks: (1) a hybrid or magma mixing and/or mingling model arguing that MMEs represent extraneous mafic magma blobs likely produced from the mantle that have mingled or partly mixed with felsic magmas derived from the crust or felsic magmas derived by fractional crystallization (e.g., Didier, 1973; Reid et al., 1983; Vernon, 1983, 1984, 1990, 2014; Cantagrel et al., 1984; Eberz and Nicholls, 1988; Vernon et al., 1988; Barbarin, 1990; Castro et al., 1990a, 1990b; Dorais et al., 1990; Eberz et al., 1990; Fourcade and Javoy, 1991; Holden et al., 1991; Orsini et al., 1991; Poli and Tommasini, 1991; Elburg and Nicholls, 1995; Stimac et al., 1995; Elburg, 1996a, 1996b; Maas et al., 1997; Collins et al., 2000; Ratajeski et al., 2001; Perugini et al., 2003; Tepper and Kuehner, 2004; Ilbeyli and Pearce, 2005; Kocak, 2006; Chen et al., 2009a; Qin et al., 2010; Sun et al., 2010; Perugini and Poli, 2012; Xiong et al., 2012; Zi et al., 2012; Clemens and Elburg, 2013; Dan et al., 2015); (2) a restitic origin model interpreted the enclaves as refractory pods of the source rock of the granitoid (e.g., White and Chappell, 1977; Chappell et al., 1987; Chen et al., 1990; Chappell and White, 1991; White et al., 1999; Chappell and Wyborn, 2012; Wyborn, 2013); (3) as having formed as cumulates or chilled margin from the magmas that gave rise to the granitoid pluton (e.g., Pabst, 1928; Fershtater and Borodina, 1977, 1991; Phillips et al., 1981; Dodge and Kistler, 1990; Flood, 1993; John and Blundy, 1993; Flood and Shaw, 1995, 2006, 2014; Dorais et al., 1997; Tobisch et al., 1997; Dahlquist, 2002; Ilbeyli and Pearce, 2005; Chen et al., 2007; Shellnutt et al., 2010; Esna-Ashari et al., 2011; Niu et al., 2013; Huang et al., 2014; Chen et al., 2015; Lee et al., 2015); (4) a crystallization process by rapid cooling within the host granitoid magma operating at the margins of magma conduits (Donaire et al., 2005); and less commonly (5) a xenolithic origin (e.g., Grout, 1937; Elburg, 1996b; Clemens and Elburg, 2013). Among them the most popular issue is the magma mixing and/or mingling model, and in many cases, MMEs are widely used as main evidence for magma mixing and/or mingling between mantle-derived mafic and crustal-derived felsic magmas during formation of calc-alkaline granitoids (e.g., Vernon, 1983, 2014; Poli and Tommasini, 1991; Wiebe et al., 1997; Yang et al., 2004, 2007; Barbarin, 2005; Kocak, 2006; Kumar and Rino, 2006; Kaygusuz and Aydınçakır, 2009; Zhao et al., 2010; Cheng et al., 2012; Dan et al., 2015). In some granitoid plutons, MMEs from different origins coexist and have provided important constraints on formation processes of the granitoid plutons in continental crust (e.g., Grout, 1937; Tindle and Pearce, 1983; Didier, 1987; Didier and Barbarin, 1991; Fornelli, 1994; Stimac et al., 1995; Elburg, 1996b; Schödlbauer et al., 1997; Yang et al., 2004; Barbarin, 2005; Ilbeyli and Pearce, 2005; Esna-Ashari et al., 2011; Clemens and Elburg, 2013). Using MMEs as one of the most important indicators of mixing and/or mingling between mantle-derived mafic and crustal-derived felsic magmas, most researchers considered magma mixing as a common phenomenon in generation of intermediate magmatic rocks such as andesites and diorites (e.g., Cantagrel et al., 1984; Ussler and Glazner, 1989; Castro et al., 1990a; Cole et al., 2001; Janoušek et al., 2004; Alpaslan et al., 2005; Kawabata and Shuto, 2005). However, recent results show that andesites are dominantly produced by crystal-liquid fractionation, and crystal-liquid segregation appears to be the dominant process in generating intermediate magmas, with mixing playing a secondary role (Lee and Bachmann, 2014). Geochemical studies of plutons in the Peninsular Ranges Batholith in southern California, as an example, show that the entire range of magma compositions can be generated by fractionation, but at different levels (Lee and Morton, 2015).

This paper presents new petrographical, zircon U-Pb geochronolocal, geochemical, and Sr-Nd-Hf data as well as field observations, mineral chemistry, and P-T estimates of the Permian granitoids and their MMEs from the northern margin of the North China Block (NCB). The new results provide important constraints on origin of the MMEs in calc-alkaline granitoids and formations of the granitoid batholiths in continental crust, as well as the mechanism for generating the large volumes of intermediate magmatic rocks in continental or island arcs.

2. GEOLOGICAL BACKGROUND

The northern NCB is located south to the Solonker suture zone that marks the final closure of the Paleo-Asian ocean between the NCB and the southern Mongolia composite terranes during the Late Permian to earliest Triassic (e.g., Wang and Liu, 1986; Xiao et al., 2003, 2009; Li, 2006; Windley et al., 2007; Wu et al., 2007; Zhang et al., 2007a, 2009a, 2009b; Li et al., 2009; Chen et al., 2009b; Eizenhöfer et al., 2014). It is composed mainly of two tectonic units including the North China Craton (NCC) and the Bainaimiao arc belt, which are separated by the E-W–trending Bayan Obo-Chifeng-Kaiyuan fault zone (Fig. 1). The basement of the NCC is composed of highly metamorphosed Archean and Paleoproterozoic rocks, which were covered by the Meso-Neoproterozoic and Cambrian–Ordovician marine clastic and carbonate platformal sediments, Middle Carboniferous to Triassic fluvial and deltaic sediments, and Jurassic–Cretaceous and younger volcanic and sedimentary rocks. The early Paleozoic Bainaimiao arc belt was built upon a Precambrian microcontinent that has a tectonic affinity to the Tarim or Yangtze cratons and was accreted to the northern NCC during Late Silurian–earliest Devonian by arc-continent collision (S.H. Zhang et al., 2014). The northern margin of the NCB was strongly involved in evolution of the Central Asian orogenic belt (CAOB) with emplacement of abundant granitoid plutons during the late Paleozoic period.

Carboniferous–Permian calc-alkaline granitoid plutons are very common in the northern NCB and occupy over 20% of the outcrop areas (Fig. 1B). They intruded into the Precambrian rocks in the northern margin of the North China Craton as well as the early Paleozoic rocks in the Bainaimiao arc belt. They exhibit calc-alkaline or high-K calc-alkaline, metaluminous or weak peraluminous geochemical features, and most researchers considered these plutons to reflect arc magmatism along an Andean-type active continental margin (e.g., Wang and Liu, 1986; Xiao et al., 2003, 2009; Li, 2006; Wang et al., 2007; Zhang et al., 2007a, 2009a, 2009b; Bai et al., 2013; Ma et al., 2013a; Zhang and Zhao, 2013), whereas others favored a postcollisional or postsubduction environment for these rocks (e.g., Luo et al., 2009; Zhang et al., 2012). They are composed mainly of diorite, quartz diorite, granodiorite, and granite; other rocks are gabbro and tonalite. Except for a few small intrusions (Zhang et al., 2009b, 2009c; Zhou et al., 2009; Zhao et al., 2011), Carboniferous–Permian mafic-ultramafic rocks are seldom distributed in the northern NCB. MMEs are very common in these plutons, especially the dioritic-granodioritic plutons in the central-western segments of the northern margin of the NCB. Although some research has been conducted on petrogenesis of these plutons (e.g., Zhang et al., 2007a, 2009a, 2009b, 2012; Luo et al., 2009; Wang et al., 2011; Bai et al., 2013; Ma et al., 2013a), the MMEs within these plutons have not be investigated. Therefore, the origin of these MMEs and their roles in formation of the large volumes of granitoid batholiths in the northern margin of the NCB are still not very clear.

3. FIELD OCCURRENCE, PETROLOGY, AND SAMPLES

Samples were collected from five Permian plutons (Rentaihe, Tonghetai, Xuniwusu, Hongge’er, and Yinhao) in the northern margin of the NCB (Fig. 1). Among them the Rentaihe, Tonghetai, Hongge’er, and Yinhao plutons are located in the North China Craton, and the Xuniwusu pluton is located in the Bainaimiao arc belt (Fig. 1B). These plutons consist mainly of quartz diorite and granodiorite with abundant MMEs that are diorite, monzodiorite, or gabbroic diorite in composition. The Yinhao pluton exhibits concentric zoning with a thin dioritic margin and quartz dioritic (tonalitic) to granodioritic interior, and MMEs are especially rich in quartz diorite and tonalite (Fig. 2). Usually, abundance of the MMEs decreases from the margins to the central portion of the pluton. The contact of enclaves with the host granitoids is commonly sharp (Fig. 3). Most enclaves exhibit typical igneous textures with main mineral assemblage of plagioclase, hornblende, biotite, K-feldspar, and quartz. Pyroxene is scarce in both MMEs and host granitoids. Some enclaves were strongly deformed with long axes parallel to the magmatic foliation in their host granitoids and have aspect ratios ranging from 2 to >10 (Figs. 3M–3O).

3.1 Rentaihe Pluton

The Rentaihe pluton is located near Rentaihe village, 20 km north to the Guyang County (Fig. 1). It was previously considered as late Paleozoic (BGMRIM, 1972) or Neoarchean (IMIGS, 2003) in age. The long axis of the pluton trends WNW-ESE, and its area is ∼50 km2. It is composed mainly of medium-grained quartz diorite (Fig. 4A) and exhibits weak magmatic foliation in its southern part (Fig. 3A). The host rocks of the pluton are Archean–Paleoproterozoic metamorphic rocks, and it is unconformably overlain on its northern side by Early Cretaceous strata. Sample 07173-1 is a medium-grained quartz diorite collected from the southern margin of the pluton, which consists of plagioclase (65 vol%), hornblende (5–10 vol%), biotite (10 vol%), K-feldspar (15 vol%), quartz (5–10 vol%), and accessory magnetite, apatite, and zircon. Its zircons are transparent, euhedral, long yellowish prisms that are 70–250 µm long.

The MMEs vary from a few cm to 30 cm across and are ellipsoidal to lenticular in shape (Fig. 3A). They are especially common near the margins of the pluton and are rare in the central portion of the pluton. They are mainly diorite and are finer grained than the host quartz diorite (Fig. 4B). Their long axes are parallel to the magmatic foliation in their host rocks. Sample 07173-2 is a fine-grained dioritic enclave collected from the southern margin of the pluton. It is composed of plagioclase (55 vol%), hornblende (25–30 vol%), biotite (10 vol%), K-feldspar (5 vol%), quartz (4 vol%), and accessory magnetite, titanite, apatite, and zircon. The zircons display typical magmatic features and are yellowish, idiomorphic, with long to short prisms. The grain sizes are from 100 to 300 µm.

3.2 Tonghetai Pluton

The Tonghetai pluton is located near the Tonghetai village 20 km south to the Urad Zhongqi with an exposure area of 5 km2 (Fig. 1B). It intruded Early Permian metasedimentary rocks in the southern side, and the other sides are covered by Cenozoic sediments. It consists mainly of medium- to coarse-grained granodiorite and quartz monzodiorite (Figs. 3B, 3C, and 4C). Sample 07201-1 is a quartz monzodiorite collected from southwestern part of the pluton. Its mineral assemblage is plagioclase (50 vol%), K-feldspar (20 vol%), quartz (10–15 vol%), hornblende (8 vol%), and biotite (8 vol%), with accessory magnetite, apatite, and zircon. The zircons are transparent, pink, and euhedral, long to short prisms with grain sizes from 50 to 150 µm.

The MMEs are very common throughout the pluton and vary from a few cm to 40 cm across. Most of them are ellipsoidal to lenticular in shape, and others are irregular (Figs. 3B and 3C). They are composed of medium- to fine-grained diorite or monzodiorite (Fig. 4D). Sample 07201-2 is a quartz monzodiorite enclave collected from the southwestern part of the pluton. It is composed of plagioclase (45–50 vol%), hornblende (20–25 vol%), K-feldspar (15–20 vol%), biotite (10 vol%), quartz (5 vol%), and accessory magnetite, titanite, apatite, and zircon. The zircons display typical magmatic features and are yellowish, idiomorphic, long to short prisms, and the grain size is from 50 to 120 µm. Sample 07201-3 is a monzodiorite enclave from the southwestern part of the pluton. It consists of plagioclase (35 vol%), hornblende (20 vol%), biotite (30 vol%), K-feldspar (15 vol%), quartz (<1 vol%), and accessory magnetite, apatite, and zircon. The zircons are transparent, pink euhedral prisms that are 50–150 µm long.

3.3 Xuniwusu Pluton

The Xuniwusu pluton is located near the Xuniwusu village, 5 km south to the town of Bainaimiao (Fig. 1B). It was emplaced into the latest Silurian to earliest Devonian Xibiehe Formation, and its exposure area is ∼4 km2. It is composed mainly of coarse-grained porphyric quartz diorite (Figs. 3D, 3E, and 4E). Sample 08435-1 is a porphyric quartz diorite collected from the northern part of the pluton. It is composed of plagioclase (60 vol%, including 10 vol% as phenocrysts), quartz (15 vol%), biotite (10 vol%), K-feldspar (10 vol%), hornblende (5 vol%), and accessory magnetite, titanite, apatite, and zircon. The zircons are yellowish, euhedral prisms, and the grain size is from 100 to 700 µm.

The MMEs vary from a few cm to 50 cm across and are ellipsoidal to lenticular in shapes (Figs. 3D and 3E). Most of the enclaves are diorite and contain phenocrysts of hornblende and feldspar (Fig. 4F). Sample 08435-2 is a porphyric diorite enclave collected from the northern part of the pluton. Its mineral assemblage is plagioclase (40 vol%, including 5 vol% as phenocrysts), biotite (30 vol%), hornblende (15 vol%, including 5 vol% as phenocrysts), quartz (10 vol%), and K-feldspar (5 vol%), with accessory magnetite, apatite, and zircon. The zircons display typical magmatic features and are transparent, yellowish, with euhedral long to short prisms. The grain sizes are from 50 to 400 µm.

3.4 Hongge’er Pluton

The Hongge’er pluton is located near the Hongge’er village, 55 km north to Siziwangqi (Fig. 1B). It is intruded into the late Paleoproterozoic to early Mesoproterozoic sedimentary rocks of the Bayan Group, and its area is ∼300 km2. It is composed of medium- to coarse-grained granodiorite (Figs. 3F–3I, 4H, and 4J). Sample 08458-1 is a medium- to coarse-grained granodiorite collected from the northeastern part of the pluton. It is composed of plagioclase (50 vol%), quartz (20 vol%), biotite (20 vol%), K-feldspar (10 vol%), and accessory magnetite, titanite, apatite, and zircon. The zircons are yellowish, euhedral to subhedral prisms, and have a grain size from 50 to 300 µm.

Mafic micogranular enclaves (MMEs) are very common within the Hongge’er pluton. They vary from a few cm to 80 cm across and are ellipsoidal to lenticular in shapes (Figs. 3F–3I). The MMEs are mainly diorite, and some of them contain phenocrysts of plagioclase, hornblende, and quartz (Figs. 3G and 3I). Most of the enclaves are rich in biotite with minor or no hornblende (Figs. 4G, 4I, and 4K). Sample 08458-2 is a medium- to fine-grained diorite enclave collected from the northeastern part of the pluton. Its mineral assemblage is plagioclase (65 vol%), biotite (15–20 vol%), K-feldspar (10–15 vol%), and quartz (5 vol%), with accessory magnetite, apatite, and zircon. Its zircons are yellowish, euhedral to subhedral prisms that are 50–300 µm long.

3.5 Yinhao Pluton

The Yinhao pluton is ∼25 km east to the Guyang County with an area of 220 km2 (Figs. 1B and 2). The host rocks of the pluton are Archean–Paleoproterozoic metamorphic rocks and late Paleoproterozoic to early Mesoproterozoic sedimentary rocks of the Bayan Group. Its northeastern and southeastern sides were intruded by Triassic granites. It is composed mainly of medium- to coarse-grained quartz diorite, tonalite, and granodiorite with minor diorite in its margins (Figs. 2, 3J–3O, 4L, 4O, and 4Q). Field relations indicate an emplacement sequence from diorite, quartz diorite (tonalite) to granodiorite. Magmatic foliation is very common within the pluton, especially in its southern and western sides. Sample 08603-1 is a quartz diorite collected from northwestern part of the pluton. It is composed of plagioclase (60 vol%), hornblende (10 vol%), biotite (10–15 vol%), K-feldspar (10 vol%), quartz (5–10 vol%), and accessory magnetite, titanite, apatite, and zircon. The zircons display typical magmatic features and are yellowish, idiomorphic, with long to short prisms, and the grain size is from 60 to 500 µm. Samples 08602-1, 08604-1, 08605-1, and 08606-1, which were used for hornblende-plagioclase thermobarometry, are quartz diorites and granodiorite collected from the northern and northeastern margins of the Yinhao pluton. Samples 10224-1 and 12123-1, also used for hornblende-plagioclase thermobarometry, are from another two quartz diorites collected from the western and southwestern margins of the pluton.

The MMEs are very common in quartz diorite (tonalite) and granodiorite within the Yinhao pluton and are rare in its dioritic margins. Abundance of the MMEs decreases from quartz diorite (tonalite) to granodiorite. They vary from a few cm to 40 cm across and are ellipsoidal to lenticular in shape (Figs. 3J–3O). The MMEs near the western margin of the pluton are deformed with ratios of long/short axes ranging from 2 to >10. Their long axes are parallel to the magmatic and high-temperature subsolidus foliations within the host rocks (Figs. 3M–3O). Sample 08603-2 is a dioritic enclave collected from the northwestern part of the pluton (Fig. 4M). Its mineral assemblage is plagioclase (65 vol%), hornblende (10–15 vol%), biotite (10–15 vol%), K-feldspar (5 vol%), and quartz (5 vol%), with accessory magnetite, apatite, and zircon. The zircons are transparent, yellowish, with euhedral long to short prisms and have grain sizes from 50 to 500 µm. Sample 08603-3 is a gabbroic diorite enclave collected from northwestern part of the pluton (Fig. 4N). It consists of plagioclase (55 vol%), hornblende (30 vol%), biotite (10 vol%), K-feldspar (<5 vol%), quartz (<1 vol%), and accessory magnetite, apatite, and zircon. The zircons are transparent, yellowish, subhedral long or short prisms that are 80–400 µm long. Samples 08602-2, 08604-2, 08605-2, and 08606-2 used for hornblende-plagioclase thermobarometry are dioritic enclaves collected from the northern and northeastern margins of the Yinhao pluton. Samples 10224-2 and 12123-2 were used for hornblende-plagioclase thermobarometry and are another two dioritic enclaves collected from the western and southwestern margins of the pluton (Figs. 4P and 4R).

4. ANALYTICAL METHODS

4.1 Sample Preparation and Imaging

Zircons were separated using conventional crushing and separation techniques and were then handpicked under a binocular microscope. They were mounted in epoxy resin and polished to expose the cores of the grains in readiness for photomicrograph, cathodoluminescence (CL), LA-ICP-MS U-Pb, and Lu-Hf isotopic analyses. Zircons were imaged using the Hitachi S3000-N electron microscope with Gatan Chroma CL detector at the Beijing sensitive high-resolution ion microprobe (SHRIMP) Center. Cathodoluminescence images of representative zircon grains are shown in Figure 5.

4.2 LA-ICP-MS U-Pb and Trace-Element Analyses

Laser ablation–inductively coupled plasma–mass spectrometer (LA-ICP-MS) U-Pb and trace-element analyses were performed on an excimer (193 nm–wavelength) LA-ICP-MS at the State Key Laboratory of Continental Dynamics, Northwest University, Xi’an, and State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences, Wuhan, following the method described by Yuan et al. (2004) and Liu et al. (2010), using an Agilent 7500a ICP-MS. The GeoLas 200M laser ablation system was used for the laser ablation experiments. Helium was applied as a carrier gas. Argon was used as the make-up gas and was mixed with the carrier gas via a T-connector before entering the ICP. Nitrogen was added into the central gas flow (Ar + He) of the Ar plasma to decrease the detection limit and improve precision (Hu et al., 2008). Each analysis incorporated a background acquisition of ∼20–30 s (gas blank) followed by 50 s data acquisition from the sample. Sites for dating were selected on the basis of CL and photomicrograph images. The spots used were 30–40 μm in diameter. U, Th, and Pb concentrations were calibrated using 29Si as an internal standard and National Institute of Standards and Technology (NIST) Standard Reference Materials (SRM) 610 as the reference standard. Isotopic ratios were calculated using GLITTER 4.0 (Macquarie University) and ICPMSDataCal 5.0 (Liu et al., 2010), which were then corrected for both instrumental mass bias and depth-dependent elemental and isotopic fractionation using Harvard zircon 91500 as an external standard. GEMOC GJ-1 zircon standard (thermal ionization mass spectrometer [TIMS] U-Pb age = 608.5 ± 0.4 Ma, Jackson et al., 2004) was used as a monitor of data quality during analyses. Concordia diagrams and weighted mean ages were produced using the program Isoplot/Ex 3.23 (Ludwig, 2003). For statistical interpretations of the U-Pb data, strategies recommended by Spencer et al. (2016) were used.

4.3 Major-and Trace-Element Geochemistry

Major elements were analyzed on fused glass discs by X-ray fluorescence spectrometry at Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing. Trace elements (including rare-earth elements [REEs]) were determined by ICP-MS (VG Plasma Quad PQ2 Turbo ICP-MS) at the Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing. Samples of ∼100 mg whole-rock powders were dissolved in distilled HF-HNO3 in Teflon screw-cap beakers and high-pressure Teflon bombs at 200 °C for four days, dried, and then digested with HNO3 at 150 °C for one day. Dissolved samples were diluted to 50 ml with 1% HNO3 before analyses. A blank solution was prepared, and the total procedural blank was <50 ng for all trace elements. Indium was used as an internal standard to correct for matrix effects and instrument drift. The Chinese national standards GSR-1 (granite) and GSR-3 (basalt) were used to monitor analyses. Errors for major-element analysis are within 1%, except for P2O5 (5%), and analyses for most trace elements (including REE) are within 10%.

4.4 Rb-Sr and Sm-Nd Isotopic Analyses

Samples for Rb-Sr and Sm-Nd isotopic analyses were dissolved in Teflon bombs after being spiked with 84Sr, 87Rb, 150Nd, and 149Sm tracers prior to HF + HClO4 dissolution. Rb, Sr, Sm, and Nd were separated using conventional ion-exchange procedures and measured using a GV IsoProbe-T multicollector mass spectrometer at the Analytical Laboratory of the Beijing Research Institute of Uranium Geology. Procedural blanks were <100 pg for Sm and Nd and <500 pg for Rb and Sr. 143Nd/144Nd was corrected for mass fractionation by normalization to 146Nd/144Nd = 0.7219, and 87Sr/86Sr ratios were normalized to 86Sr/88Sr = 0.1194. Typical within-run precision (2σ) for Sr and Nd was estimated to be ±0.000015.

4.5 In Situ Lu-Hf Isotope Analyses

In situ Lu-Hf isotope analyses were performed using a Newwave UP213 laser ablation microprobe attached to a Finnigan Neptune multicollector ICP-MS at the Institute of Mineral Resources, Chinese Academy of Geological Sciences, Beijing, using techniques and analytical procedures described by Wu et al. (2006a) and Hou et al. (2007). Analyses were carried out with laser beam diameter of 55 µm, 6 Hz repetition rates, and 30 s ablation times. During analysis, raw count rates for 172Yb, 173Yb, 175Lu, 176(Hf + Yb + Lu), 177Hf, 178Hf, 179Hf, 180Hf, and 182W were collected. Isobaric interference corrections for 176Lu and 176Yb on 176Hf must be determined precisely. 175Lu was calibrated using 176Yb on 176Hf. The 176Yb/172Yb value of 0.5887 and mean βYb value obtained during Hf analysis on the same spot were applied for the interference correction of 176Yb on 176Hf (Iizuka and Hirata, 2005). The measured 176Hf/177Hf ratio on the standard zircon (GJ-1) was 0.282009 ± 0.000004 (N = 96), similar to the commonly accepted 176Hf/177Hf ratio of 0.282000 ± 0.000005 measured using the solution method (Morel et al., 2008). For the calculation of initial Hf isotope ratio, the decay constant for 176Lu proposed by Söderlund et al. (2004) was used. For the calculation of εHf (t) values, we have adopted the chondritic values of Blichert-Toft and Albarede (1997). Hafnium model ages (TDM and TDMC) were calculated on the basis of the depleted mantle model described by Griffin et al. (2002) and Yang et al. (2006).

4.6 Mineral Analyses

Electron microprobe analyses on plagioclase and hornblende were carried out with a JEOL JXA-8100 outfitted with a five wavelength-dispersive X-ray spectrometer at the Microprobe Analytical Laboratory, Beijing Research Institute of Uranium Geology and the Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing. Quantitative analyses of rock-forming minerals were performed with a 20 kV accelerating voltage, a 20 nA beam current, and a 2–5 µm beam size. The counting time at each peak was 20–30 s.

5. RESULTS

5.1 Zircon U-Pb Geochronology

5.1.1 Rentaihe Pluton

Fifteen spots on 15 zircon grains from the Rentaihe quartz diorite sample 07173-1 were measured, and most of them are concordant (Supplemental Table 11 and Fig. 6A). Except for one discordant spot (01) and three anomalous spots (03, 04, and 05), the remaining spots give a weighted mean 206Pb/238U age of 280 ± 3 Ma (95% confidence, mean square of weighted deviates [MSWD] = 1.6, N = 11). Long euhedral prismatic morphology, well-developed oscillatory zoning (Fig. 5A), and high Th/U ratios (0.64–0.94) indicate that the zircons are magmatic in origin. Therefore, we infer the mean age of 280 ± 3 Ma to date the time of crystallization of the Rentaihe quartz diorite in the Early Permian period.

For the dioritic enclave sample 07173-2 from the Rentaihe pluton, 15 spots on 14 zircon grains were measured, and all of them were concordant (Supplemental Table 1 [see footnote 1] and Fig. 6B). Except for five anomalous analyses (spots 01, 04, 07, 10, and 12), the remaining analyses yield a weighted mean 206Pb/238U age of 278 ± 3 Ma (95% confidence, MSWD = 1.9, N = 10). Their zircons are characterized by long euhedral prismatic morphology, well-developed oscillatory zoning (Fig. 5B), and high Th/U ratios (0.43–1.19), indicating a magmatic origin. Therefore, we conclude that 278 ± 3 Ma reflects the crystallization age of the dioritic enclaves within the Rentaihe pluton.

5.1.2 Tonghetai Pluton

Fifteen spots on 15 zircon grains from the Tonghetai quartz monzodiorite sample 07201-1 were analyzed, and most of them give concordant ages (Supplemental Table 1 [see footnote 1] and Fig. 6C). Except for four discordant analyses (spots 03, 05, 06, and 11), the remaining analyses yield a weighted mean 206Pb/238U age of 283 ± 4 Ma (95% confidence, MSWD = 1.8, N = 11). Their high Th/U ratios (0.49–1.11) and well-developed oscillatory zoning (Fig. 5C) indicate that they are magmatic in origin. Therefore, we conclude that a mean age of 283 ± 4 Ma can be interpreted as the emplacement age of the Tonghetai quartz monzodiorite.

For the quartz monzodiorite enclave sample 07201-2, 15 spots on 15 zircon grains were measured, and most of them are concordant (Supplemental Table 1 [see footnote 1] and Fig. 6D). Except for three discordant analyses (spots 03, 04, and 10) and two slightly older analyses (spots 13 and 15), the remaining analyses yield a weighted mean 206Pb/238U age of 286 ± 4 Ma (95% confidence, MSWD = 1.8, N = 10). Their zircons are characterized by long to short euhedral prismatic morphology and well-developed oscillatory zoning (Fig. 5D) with high Th/U ratios (0.49–1.30), thus indicating a magmatic origin. Therefore, we conclude that 286 ± 4 Ma reflects the crystallization age of the quartz monzodiorite enclave within the Rentaihe pluton.

For the monzodiorite enclave sample 07201-3, 15 spots on 15 zircon grains were measured (Supplemental Table 1 [see footnote 1]). Except for eight discordant spots (01, 04, 05, 06, 07, 10, 11, and 14), the remaining spots give a weighted mean 206Pb/238U age of 288 ± 3 Ma (Fig. 6E; 95% confidence, MSWD = 0.48, N = 7). Their high Th/U ratios (0.43–2.66) and well-developed oscillatory zoning (Fig. 5E) indicate that they are magmatic in origin. Therefore, their mean age of 288 ± 3 Ma is interpreted as the crystallization age of the monzodiorite enclave within the Rentaihe pluton.

5.1.3 Xuniwusu Pluton

Fifteen spots on 15 zircon grains from the Xuniwusu porphyric quartz diorite sample 08435-1 were analyzed, and most of them were concordant (Supplemental Table 1 [see footnote 1] and Fig. 6F). Except for two discordant analyses (spots 09 and 15) and three slightly older analyses (spots 05, 06, and 11), the remaining analyses yielded a weighted mean 206Pb/238U age of 269 ± 1 Ma (95% confidence, MSWD = 1.2, N = 10). Their high Th/U ratios (0.24–0.64) and well-developed oscillatory zoning (Fig. 5F) indicate that they are magmatic in origin. Therefore, we conclude that the mean age of 269 ± 1 Ma dates the time of crystallization of the Xuniwusu porphyric quartz diorite.

For the porphyric diorite enclave sample 08435-2, 15 spots on 15 zircon grains were measured, and most of them give concordant ages (Supplemental Table 1 [see footnote 1] and Fig. 6G). Except for one discordant spot (04) and two slightly older spots (01 and 02), the remaining spots give a weighted mean 206Pb/238U age of 271 ± 1 Ma (95% confidence, MSWD = 1.2, N = 12). Their well-developed oscillatory zoning (Fig. 5G) and high Th/U ratios (0.28–0.55) indicate that they are magmatic in origin. Therefore, we also conclude that 271 ± 1 Ma reflects the crystallization age of the porphyric diorite enclave within the Xuniwusu pluton.

5.1.4 Hongge’er Pluton

For the granodiorite sample 08458-1 from the Hongge’er pluton, 15 spots on 15 zircon grains were measured, and most of them are concordant (Supplemental Table 1 [see footnote 1] and Fig. 6H). Three analyses (spots 01, 07, and 14) yield old inherited ages ranging from 1729 ± 24 Ma (207Pb/206Pb age) to 284 ± 2 Ma (206Pb/238U age). Except for the three inherited zircons (spots 01, 07, and 14) and three discordant analyses (spots 03, 05, and 12), the remaining analyses yield a weighted mean 206Pb/238U age of 262 ± 2 Ma (95% confidence, MSWD = 1.9, N = 9). Although these zircons exhibit variable Th/U ratios from 0.04 to 1.59, their well-developed oscillatory zoning (Fig. 5H) indicates that they are magmatic in origin. Therefore, we determine that 262 ± 2 Ma reflects the emplacement age of the Hongge’er granodiorite.

Fifteen spots on 15 zircon grains from the diorite enclave sample 08458-2 were measured, and most of them give concordant ages (Supplemental Table 1 [see footnote 1] and Fig. 6I). Except for six discordant spots (04, 08, and 10–13), the remaining spots give a weighted mean 206Pb/238U age of 264 ± 2 Ma (95% confidence, MSWD = 1.7, N = 9). Although their Th/U ratios vary from 0.05 to 0.82, they exhibit well-developed oscillatory zoning (Fig. 5I), indicating a magmatic origin. Hence, we infer the mean age of 264 ± 2 Ma to be interpreted as the crystallization age of the diorite enclave within the Hongge’er pluton.

5.1.5 Yinhao Pluton

For the quartz diorite sample 08603-1 from the Yinhao pluton, 15 spots on 15 zircon grains were measured, and most of them are concordant (Supplemental Table 1 [see footnote 1] and Fig. 6J). Except for two discordant analyses (spots 09 and 10), the remaining analyses yield a weighted mean 206Pb/238U age of 283 ± 2 Ma (95% confidence, MSWD = 1.4, N = 13). Euhedral prismatic morphology, well-developed oscillatory zoning (Fig. 5J), and high Th/U ratios (0.55–1.01) indicate that the zircons are magmatic in origin. Therefore, we find the mean age of 283 ± 2 Ma as dating the time of crystallization of the Yinhao quartz diorite.

Fifteen spots on 15 zircon grains from the dioritic enclave sample 08603-2 within the Yinhao pluton were analyzed (Supplemental Table 1 [see footnote 1]). All of them are concordant and give a weighted mean 206Pb/238U age of 283 ± 1 Ma (Fig. 6K; 95% confidence, MSWD = 0.52, N = 15). Their zircons are characterized by euhedral prismatic morphology, well-developed oscillatory zoning (Fig. 5K), and high Th/U ratios (0.62–1.00), indicating a magmatic origin. Therefore, we conclude that 283 ± 1 Ma reflects the crystallization age of the dioritic enclave within the Yinhao pluton.

For the gabbroic diorite enclave sample 08603-3 within the Yinhao pluton, 15 spots on 15 zircon grains were measured, and most of them are concordant (Supplemental Table 1 [see footnote 1] and Fig. 6L). Except for two discordant spots (06 and 10) and two anomalous spots (03 and 04), the remaining spots give a weighted mean 206Pb/238U age of 289 ± 2 Ma (95% confidence, MSWD = 0.86, N = 11). Their well-developed oscillatory zoning (Fig. 5L) and high Th/U ratios (0.60–1.60) indicate a magmatic origin. Therefore, we conclude that mean age of 289 ± 2 Ma should be interpreted as the crystallization age of the gabbroic diorite enclave within the Yinhao pluton.

5.2 Major-and Trace-Element Compositions

Major- and trace-element compositions of the Permian granitoids and their microgranular enclaves are listed in Table 1 and plotted in Figs. 79. In the total alkali (K2O + Na2O) versus silica (SiO2) classification diagram (Fig. 7A), the host granitoid samples plot in the fields of monzonite, quartz monzonite, and granodiorite, and most of them are in the subalkaline field. According to the classification of Peccerillo and Taylor (1976), all the host granitoid samples belong to the calc-alkaline and high-K calc-alkaline series (Fig. 7B). The MMEs are classified as monzogabbro, monzodiorite, gabbroric diorite, diorite, monzonite, and quartz monzonite in the same total alkali (K2O + Na2O) versus silica (SiO2) classification diagram (Fig. 7A), and most of them plot in the fields of the calc-alkaline and high-K calc-alkaline series in the K2O versus SiO2 diagram (Fig. 7B). A molecular Al2O3 (Na2 + K2O) (A/NK) versus A/CNK diagram (not presented) indicates that all the MMEs and their host granitoids are metaluminous or weakly peraluminous (A/CNK <1.1 and A/NK >1). Compared with their host granitoids, the MMEs exhibit relatively low contents of SiO2 (51.0–62.7 wt%), high contents of Fe2O3T (5.5–10.2 wt%), MgO (2.3–5.6 wt%), CaO (3.6–7.5 wt%), Cr (13.9–258.5 ppm), V (76.6–179.6 ppm), Co (11.1–28.2 wt%), Ni (14.0–82.3 wt%), and high Mg# values of 41–61. On the Harker diagrams (Fig. 8), MgO, TiO2, CaO, and P2O5 abundances of the MMEs and their host granitoids tend to decrease with increasing SiO2, consistent with fractionation trends expected of plagioclase, ferromagnesian minerals, Fe-Ti oxides, and apatite.

The MMEs from each Permian pluton exhibit similar REE and trace-element compositions to their host granitoids (Fig. 9). On chondrite-normalized REE diagrams (Figs. 9A, 9C, 9E, 9G, and 9I), most of them display significant light REE (LREE) enrichment (LaN/YbN = 3.22–17.69) and absent to slightly negative Eu anomalies (EuN/EuN* = 0.34–1.06). On primitive mantle-normalized spidergrams (Figs. 9B, 9D, 9F, 9H, and 9J), they exhibit enrichment of Rb, Ba, and K and depletion of Nb, Ta, P, and Ti. Most of the host granitoids exhibit high contents of Sr (341.1–654.4 ppm), low contents of Y (14.2–25.3 ppm), Yb (1.4–2.5 ppm), and high Sr/Y ratios (14.3–46.3).

5.3 Sr and Nd Isotopic Data

Whole-rock Rb-Sr and Sr-Nd isotopic data of the Permian granitoids and their microgranular enclaves are listed in Table 2 and plotted in Figure 10. The results show that Sr-Nd isotopic compositions of the MMEs and their host granitoids from different plutons are very different. However, for the MMEs and their host granitoids from the same pluton, Sr and Nd isotopic compositions are very similar.

The quartz diorite and dioritic enclave samples (07173-1 and 07173-2) from the Rentaihe pluton exhibit low initial 87Sr/86Sr ratios of 0.70496–0.70501, low negative εNd(t) values from –18.3 to –17.7, and old Nd isotopic TDM model ages of 2.91–2.73 Ga. The quartz monzodiorite and monzodiorite enclaves (07201-2 and 07201-3) and their host quartz monzodiorite (07201-1) from the Tonghetai pluton are characterized by low initial 87Sr/86Sr ratios of 0.70564–0.70658, higher but still negative εNd(t) values from –12.4 to –11.3, and Nd isotopic TDM model ages of 2.06–1.82 Ga. The porphyric quartz diorite and porphyric diorite enclave samples (08435-1 and 08435-2) from the Xuniwusu pluton exhibit moderate initial 87Sr/86Sr ratios of 0.70712–0.70789, high negative εNd(t) values from –4.0 to –3.2, and Nd isotopic TDM model ages of 2.17–1.16 Ga. The granodiorite and diorite enclave samples (08458-1 and 08458-2) from the Hongge’er pluton are characterized by high initial 87Sr/86Sr ratios of 0.71138–0.71264, low negative εNd(t) values from –10.7 to –10.2, and Nd isotopic TDM model ages of 1.70–1.65 Ga. The dioritic and gabbroic diorite enclaves (08603-2 and 08603-3) and their host quartz diorite (08603-1) from the Yinhao pluton exhibit low initial 87Sr/86Sr ratios of 0.70517–0.70596, low negative εNd(t) values from –11.4 to –10.5, and Nd isotopic TDM model ages of 1.89–1.67 Ga.

5.4 Lu-Hf Isotopes of Zircons

Zircon Lu-Hf isotopic data of the Permian granitoids and their microgranular enclaves are listed in Supplemental Table 22 and plotted in Figure 11. Although the Permian granitoids and their MMEs collected from different plutons exhibit different Hf isotopic compositions, Hf isotopic compositions of the MMEs and their host granitoid in each pluton are very similar.

5.4.1 Rentaihe Pluton

Zircons from the Rentaihe quartz diorite (sample 07173-1) are characterized by low initial 176Hf/177Hf ratios from 0.281952 to 0.282006, low negative εHf(t) values from –22.8 to –20.9, old Hf isotopic TDM model ages from 1.84 Ga to 1.73 Ga, and old TDMC (crustal model ages) from 2.73 Ga to 2.61 Ga. Zircons from the dioritic enclave within the Rentaihe pluton (sample 07173-2) exhibit very similar Hf isotopic compositions with low initial 176Hf/177Hf ratios from 0.281879 to 0.282006, low negative εHf(t) values from –25.5 to –21.1, old Hf isotopic TDM model ages from 1.94 Ga to 1.73 Ga, and old TDMC from 2.90 Ga to 2.62 Ga.

5.4.2 Tonghetai Pluton

Zircons from the Tonghetai quartz monzodiorite (sample 07201-1) exhibit initial 176Hf/177Hf ratios from 0.282330 to 0.282434, negative εHf(t) values from –9.4 to –5.9, Hf isotopic TDM model ages from 1.30 Ga to 1.15 Ga, and TDMC from 1.90 Ga to 1.67 Ga. Zircons from the quartz monzodiorite enclave within the Tonghetai pluton (sample 07201-2) are characterized by initial 176Hf/177Hf ratios from 0.282263 to 0.282416, negative εHf(t) values from –11.5 to –6.6, Hf isotopic TDM model ages from 1.37 Ga to 1.18 Ga, and TDMC from 2.04 Ga to 1.71 Ga. Zircons from the monzodiorite enclave within the Tonghetai pluton (sample 07201-3) exhibit initial 176Hf/177Hf ratios from 0.282302 to 0.282433, negative εHf(t) values from –10.3 to –5.7, Hf isotopic TDM model ages from 1.46 Ga to 1.14 Ga, and TDMC from 1.96 Ga to 1.67 Ga.

5.4.3 Xuniwusu Pluton

Zircons from the Xuniwusu porphyric quartz diorite (sample 08435-1) exhibit high initial 176Hf/177Hf ratios from 0.282607 to 0.282765, near zero to positive εHf(t) values from 0.0 to 5.7, young Hf isotopic TDM model ages from 0.92 Ga to 0.69 Ga, and young TDMC from 1.29 Ga to 0.93 Ga. Zircons from the porphyric diorite enclave within the Xuniwusu pluton (sample 08435-2) are characterized by high initial 176Hf/177Hf ratios from 0.282593 to 0.282401783, near zero to positive εHf(t) values of –0.4–6.3, young Hf isotopic TDM model ages from 0.93 Ga to 0.66 Ga, and young TDMC from 1.32 Ga to 0.89 Ga.

5.4.4 Hongge’er Pluton

Synmagmatic zircons from the Hongge’er granodiorite (sample 08458-1) exhibit a wide range of initial 176Hf/177Hf ratios, from 0.282343 to 0.282845, negative to positive εHf(t) values of –9.3–8.4, Hf isotopic TDM model ages from 1.35 Ga to 0.62 Ga, and TDMC from 1.88 Ga to 0.76, respectively. Zircons from the diorite enclave within the Hongge’er pluton (sample 08458-2) show a wide range of initial 176Hf/177Hf ratios, from 0.282330 to 0.282667, negative to positive εHf(t) values from –9.8 to 2.2, Hf isotopic TDM model ages from 1.33 Ga to 0.87 Ga, and TDMC from 1.91 Ga to 1.16, respectively.

5.4.5 Yinhao Pluton

Zircons from the Yinhao quartz diorite (sample 08603-1) exhibit initial 176Hf/177Hf ratios from 0.282197 to 0.282401, negative εHf(t) values from –14.2 to –7.0, Hf isotopic TDM model ages from 1.48 Ga to 1.19 Ga, and TDMC from 2.19 Ga to 1.74. Zircons from the dioritic enclave within the Yinhao pluton (sample 08603-2) are characterized by initial 176Hf/177Hf ratios from 0.282281 to 0.282378, negative εHf(t) values from –11.2 to –7.7, Hf isotopic TDM model ages from 1.36 Ga to 1.23 Ga, and TDMC from 2.01 Ga to 1.79. Zircons from the gabbroic diorite enclave within the Yinhao pluton (sample 08603-3) exhibit initial 176Hf/177Hf ratios from 0.282205 to 0.282523, negative εHf(t) values from –13.7 to –2.4, Hf isotopic TDM model ages from 1.46 Ga to 1.05 Ga, and TDMC from 2.17 Ga to 1.46.

5.5 Mineral Compositions of Plagioclase and Hornblende and P-T Conditions of Crystallization

Electron microprobe analytical results of plagioclase and hornblende from the MMEs and their host granitoids are listed in Supplemental Tables 33 and 44, respectively. The compositional ranges of plagioclases from host granitoids and their MMEs completely overlap. They range from oligoclase and andesine with An numbers (the atomic ratio [Ca/(Na+Ca + K)]) ranging from 0.159 to 0.438. Hornblendes from host granitoids and their MMEs are very similar in composition and are calcic amphiboles. According to the nomenclature of Leake et al. (1997), the majority of the hornblende is classified as magnesiohornblende, ferrohornblende, ferro-edenite, and edenite; only a few analyzed crystals are magnesiohastingsite.

Most host granitoids and their MMEs from the Permian plutons in the northern NCB contain a mineral assemblage of plagioclase, hornblende, K-feldspar, quartz, biotite, titanite, and magnetite, which is suitable for using hornblende-plagioclase thermobarometry to estimate the P-T conditions of crystallization (e.g., Johnson and Rutherford, 1989; Schmidt, 1992; Anderson and Smith, 1995). The Fe/(Fe + Mg) ratios of most hornblende range from 0.40 to 0.57, which is within the recommended range (0.40–0.65) for hornblende barometry as presently calibrated (Anderson and Smith, 1995).

Estimated crystallization P-T conditions of host granitoids and their MMEs by hornblende-plagioclase thermobarometry (Holland and Blundy, 1994; Anderson and Smith, 1995) are listed in Table 3 and plotted in Figure 12. Crystallization pressures were also calculated by the methods of Johnson and Rutherford (1989) and Schmidt (1992) and were compared with results obtained from the Anderson and Smith (1995) calibration. In many cases, crystallization pressures calculated from Anderson and Smith (1995) calibration tend to be higher than those of Johnson and Rutherford (1989) but lower than those of Schmidt (1992).

The estimated temperatures (edenite-richterite thermometers) of the host granitoids and their MMEs are 634–749 °C and 654–787 °C, respectively. Average crystallization pressures of the Permian host granitoids estimated from the Anderson and Smith (1995) calibration vary from 2.04 kbar to 5.45 kbar, which are in good agreement with the ages and metamorphic conditions of their country rocks. Average crystallization pressures of the MMEs within the Permian plutons estimated from the Anderson and Smith (1995) calibration range from 2.08 kbar to 5.25 kbar. As shown in Figure 12, almost all the Permian granitoid plutons and their MMEs yielded P-T solutions at or above the granite to tonalite solidus. Except for the Rentaihe pluton, there are no significant differences between P-T conditions of the MMEs and their host granitoids in each pluton (Fig. 12).

6. DISCUSSION

6.1 Almost Coeval Crystallization of MMEs and Their Host Granitoids

Although compositions of the MMEs from the Permian granitoid plutons in northern NCB range from gabbroic diorite, diorite, monzodiorite, to quartz monzodiorite, their zircon U-Pb ages are very similar to those of their host granitoids, indicating simultaneous crystallization of the MMEs and their hosts in each granitoid pluton.

As shown in Table 3 and Figure 12, estimated crystallization P-T conditions by hornblende-plagioclase thermobarometry of the MMEs in each pluton are also very similar to their host granitoids. Similar zircon U-Pb ages and crystallization P-T conditions of the MMEs and their host granitoids from each granitoid pluton clearly indicate that their crystallization was almost coeval and occurred simultaneously at similar emplacement depth. However, emplacement depths of different Permian granitoid plutons in the northern NCB range from middle to upper crustal levels, as indicated by their crystallization pressures of 2.04–5.45 kbar.

6.2 Source Region Characteristics and Petrogenesis of the Permian Granitoid Plutons

Rocks from the Permian granitoid plutons in the northern NCB exhibit SiO2 contents of 60.35–68.40 wt%, calc-alkaline or high-K calc-alkaline, metaluminous or weakly peraluminous geochemical features, and are mainly classified as I-type granite. Most of them are characterized by LREE enrichment, minimal Eu anomaly, high contents of Al2O3, Sr, and Ba, low contents of Y and Yb, and high Sr/Y ratios. These features are suggestive of a source with residual garnet, amphibole and/or pyroxene, but little or no olivine and plagioclase (Stern and Kilian, 1996). These adakitic geochemical conditions can be achieved by partial melting of the mafic lower crust (e.g., Atherton and Petford, 1993; Stern and Kilian, 1996; Castillo, 2006), subducted slab melting (e.g., Defant and Drummond, 1990; Rapp et al., 1999; Prouteau et al., 2001; Martin et al., 2005), partial melting of ancient lower crust (Jiang et al., 2007), low-degree partial melting of a metasomatized lithosphere (Jiang et al., 2006), or magma mixing (e.g., Qin et al., 2010; Chen et al., 2013). As shown in many research results in western South America, western North America, and NCC, materials from the subducted slab, mantle wedge, and continental crust could be involved in the generation of arc magmatism in the Andean-type convergent continental margins, and their source areas could be temporally and spatially different (e.g., Hildreth and Moorbath, 1988; McMillan et al., 1989; Stern and Kilian, 1996; Ramos, 1999; Gómez-Tuena et al., 2003, 2007, 2014; Mamani et al., 2010; Zhang et al., 2016).

Moderate initial 87Sr/86Sr ratios (0.70789), high negative to positive εNd(t) and εHf(t) values (–4.0 and 0–5.7, respectively), and young Nd and Hf isotopic model ages (1.16 Ga and 1.29–0.93 Ga, respectively) of samples from the Xuniwusu pluton in the Bainaimiao arc belt indicate that it was mainly produced by partial melting of the subducted slab of the Paleo-Asian oceanic plate or lower arc continental crust of the Bainaimiao arc belt. Relatively high initial 87Sr/86Sr ratios are likely related to modest upper crust assimilation due to their shallower emplacement depth at upper crustal levels.

Samples from the Rentaihe pluton in the northern NCC are characterized by low initial 87Sr/86Sr ratios (0.70501), low negative εNd(t) and εHf(t) values (–17.7 and –22.8 to –20.9, respectively), and old Nd and Hf isotopic model ages (2.73 Ga and 2.73–2.61 Ga, respectively), In εNd(t) and versus εHf(t) crystallization age diagrams (Figs. 10 and 11), they fall into the evolutionary trend lines of the Neoarchean–Paleoproterozoic metamorphic basement rocks in the northern NCC, indicating that the Rentaihe pluton was produced by partial melting of ancient lower crust of the NCC.

For the Tonghetai and Yinhao plutons in the northern NCC (Fig. 1B), they are also characterized by low initial 87Sr/86Sr ratios (0.70596–0.70618), low negative εNd(t) and εHf(t) values (–14.4 and –14.2 to –5.9, respectively), and old Nd and Hf isotopic model ages (1.89–1.82 Ga and 2.19–1.67 Ga, respectively). These geochemical characteristics indicate that they were mainly produced by partial melting of ancient lower crust. However, compared with the Rentaihe pluton and other Permian–Triassic granitoids produced by partial melting of ancient lower crust (e.g., Jiang et al., 2007), their εNd(t) and εHf(t) values are slightly younger, suggesting involvement of some mantle component in the petrogenesis of the Tonghetai and Yinhao plutons.

The Hongge’er pluton located near the northern margin of the NCC (Fig. 1B) exhibits high initial 87Sr/86Sr ratios of 0.71264, low negative εNd(t) values of –10.7, and old Nd isotopic model ages of 1.70. Different from other Permian granitoid plutons, the Hongge’er pluton is characterized by a wide range of zircon Hf isotopic compositions with variable εHf(t) values of –9.3 to 8.4 and Hf isotopic model ages (TDMC) of 1.88–0.76 Ga. These geochemical characteristics indicate that they were likely produced by interactions and mixing of melts from subducted slab of the Paleo-Asian oceanic plate and continental crust of the NCC. High initial 87Sr/86Sr ratios are likely a result of upper crust assimilation as indicated by their shallower emplacement depth at upper crustal levels.

6.3 Isotopic Equilibration between Host Granitoids and Their Enclaves

As stated above, although the five Permian granitoid plutons are all located along the active continental margin of the northern NCB, their source regions and crystallization pressures were very different. For the MMEs and host granitoids from the same pluton, their whole-rock Sr-Nd and zircon Hf isotopic compositions are very similar (Table 2 and Supplemental Table 2 [see footnote 2]; Figs. 10 and 11), indicating complete isotopic equilibration between host granitoids and their enclaves in each pluton. These similar Sr-Nd-Hf isotopic compositions and isotopic equilibration between MMEs and their host granitoids more likely require either a common source or complete equilibration of dissimilar materials.

Although some researchers believe that similar chemical affinities and isotopic equilibration between MMEs and their host granitoids can be produced by magma mingling and mixing of mantle-derived mafic magma and crust-derived granitic magma at the emplacement level through diffusion and percolation processes (e.g., Barbarin and Didier, 1992; Barbarin, 2005; Kaygusuz and Aydınçakır, 2009; Turnbull et al., 2010), many results show that this isotopic equilibration can only be partially achieved, in many cases, especially for Nd (e.g., Campbell and Turner, 1986; Sparks and Marshall, 1986; Turner and Campbell, 1986; Holden et al., 1987, 1991; Eberz et al., 1990; Allen, 1991; Metcalf et al., 1995; Elburg, 1996a; Perugini and Poli, 2000; Waight et al., 2000a, 2001a; Pankhurst et al., 2011) and zircon Hf isotopes (e.g., Yang et al., 2006, 2007; Shaw and Flood, 2009). Moreover, it is physically unlikely for isotopic ratios to be homogenized, whereas major and trace elements are not (Niu et al., 2013). Therefore, we conclude that Sr-Nd-Hf isotope equilibration between the host granitoids and their enclaves from the Permian granitoid plutons indicate a cognate origin for the host granitoids and their enclaves and that the enclaves were crystallized from a coeval magma that gave rise to the host granitoid plutons.

6.4 Implications for Origin of the MMEs in Calc-Alkaline Granitoids

Although enclaves in some S-type granites could be of restite or xenolithic origin (White and Chappell, 1977; Chappell et al., 1987; Chen et al., 1990; Chappell and White, 1991; Schödlbauer et al., 1997; White et al., 1999; Chappell and Wyborn, 2012; Wyborn, 2013), most enclaves in granitic rocks show many features of an igneous origin (e.g., Vernon, 1983, 1984, 1990, 2014; Barbarin, 1988, 2005; Pin et al., 1990; Didier, 1991; Paterson et al., 2004; Ilbeyli and Pearce, 2005; Clemens and Elburg, 2013). MMEs resulting from different origins could coexist in a single granitoid pluton (e.g., Grout, 1937; Didier, 1987; Fornelli, 1994; Stimac et al., 1995; Elburg, 1996b; Schödlbauer et al., 1997; Kadioğlu and Güleç, 1999; Waight et al., 2001b; Yang et al., 2004; Barbarin, 2005; Ilbeyli and Pearce, 2005; Esna-Ashari et al., 2011; Clemens and Elburg, 2013). Among the two popular models proposed for the origin of the MMEs in granitic rocks, the magma mixing and mingling model of mantle- and crust-derived magmas accounts for the igneous textural features, finer grain size, and chilled margins of the enclaves and isotopic differences between enclaves and their host granitoids (e.g., Holden et al., 1987, 1991; Eberz et al., 1990; Elburg and Nicholls, 1995; Metcalf et al., 1995; Maas et al., 1997; Altherr et al., 1999; Yang et al., 2004, 2007; Chen et al., 2009a; Shaw and Flood, 2009; Shin et al., 2009; Rajaieh et al., 2010; Qin et al., 2010; Zhao et al., 2010; Cheng et al., 2012; Jiang et al., 2013; Liu et al., 2013). The cognate origin model for crystallization of the MMEs from a coeval magma that gave rise to the host granitoids accounts for similar mineral assemblage and similarities in chemical and isotopic compositions between enclaves and their hosts (e.g., Fershtater and Borodina, 1977, 1991; Dodge and Kistler, 1990; Pin et al., 1990; Dorais et al., 1997; Dahlquist, 2002; Donaire et al., 2005; Ilbeyli and Pearce, 2005; Chen et al., 2007; Shellnutt et al., 2010; Esna-Ashari et al., 2011; Flood and Shaw, 2014). In hybridism or magma mixing and/or mingling model for origin of the MMEs within the granitic pluton, the mineral, chemical, and isotopic similarities between enclaves and their hosts are commonly interpreted in terms of complete thermal and chemical equilibration between coeval, compositionally contrasted magmas in a slowly cooling plutonic body (e.g., Holden et al., 1991; Barbarin and Didier, 1992; van der Laan and Wyllie, 1993; Barbarin, 2005; Kaygusuz and Aydınçakır, 2009; Xiong et al., 2012; Zi et al., 2012; Dan et al., 2015; Ghaffari and Rashidnejad-Omran, 2015). However, others interpreted these similarities as evidence for cognate origin of the MMEs from a coeval magma that gave rise to the host granitoids (e.g., Fershtater and Borodina, 1977, 1991; Dodge and Kistler, 1990; Pin et al., 1990; Dorais et al., 1997; Dahlquist, 2002; Donaire et al., 2005; Ilbeyli and Pearce, 2005; Chen et al., 2007; Shellnutt et al., 2010; Esna-Ashari et al., 2011; Niu et al., 2013; Flood and Shaw, 2014; Huang et al., 2014; Chen et al., 2015), especially when coeval mantle-derived rocks (such as basalts, gabbros, etc.) are absent in or nearby the granitoid plutons.

Although reversely zoned crystals were usually considered as important evidence of the mixing of magmas of contrasting composition, temperature and origin (e.g., Anderson, 1976; Hibbard, 1981; Barton et al., 1982; Andersson and Eklund, 1994; Pearce, 1994; Davidson and Tepley, 1997; Tepley et al., 1999; Waight et al., 2000b; Stewart and Fowler, 2001; Wallace and Bergantz, 2002; Janoušek et al., 2004; Landi et al., 2004; Xie et al., 2004; Browne et al., 2006; Kocak et al., 2011; Shcherbakov et al., 2011; Xiong et al., 2012; Ginibre and Davidson, 2014; Temizel, 2014; J.Y. Zhang et al., 2014), they can develop in response to a number of different magmatic processes (e.g., Perugini et al., 2005; Ginibre et al., 2007; Ruprecht and Wörner, 2007; Pietranik and Waight, 2008; Pietranik and Koepke, 2014). Many results show that such features are not dependent on magma mixing (e.g., Singer et al., 1993; Brophy et al., 1996; Barbey et al., 2005; Pietranik and Waight, 2008) and can be caused by convection within a magma body with a single composition that is heated from below and cooled from above (e.g., Couch et al., 2001), interaction between end-member magmas that had similar compositions (self-mixing events, e.g., Alves et al., 2009), or decompression-driven crystallization (e.g., Nelson and Montana, 1992; Blundy et al., 2006; Chu et al., 2006; Humphreys et al., 2006; Pietranik and Waight, 2008; Ustunisik et al., 2014). The relatively fine-grained microstructures and chilled margins of the enclaves were usually interpreted as quenching or rapid crystallization of hot mafic magma injected into granitic magma and as evidence for magma mixing (e.g., Didier, 1973; Vernon, 1983, 1984, 1990; Furman and Spera, 1985; Wiebe et al., 1997, 2002; Kim et al., 2014). However, recent results show that they can be formed by chemical reaction between the solidified enclaves and a hydrous K-rich residual melt or fluid formed after progressive crystallization and cooling of the host magma body (e.g., Farner et al., 2014) or a combination of in situ crystal fractionation of isolated magma globules, mass transfer by diffusion, and metasomatic exchange (e.g., Eberz and Nicholls, 1990).

As stated above, the Permian granitoid plutons in the northern NCB were produced by different sources during southward subduction of the Paleo-Asian oceanic plate and were emplaced at different depths from middle to upper crustal levels, as indicated by their different Sr-Nd-Hf isotopes and crystallization pressures (Tables 2, 3, and Supplemental Table 2 [see footnote 2]). However, Sr-Nd-Hf isotopes of the MMEs and their host granitoids from the same pluton are very similar, indicating isotopic equilibration between enclaves and their hosts. Although enclaves from the Permian granitoid plutons are rich in dark-colored minerals such as hornblende and biotite, they are mainly dioritic rocks in compositions with SiO2 contents of 51.03–62.73 wt%. These silica contents are contiguous with their host granitoids (SiO2 = 60.35–68.40 wt%). Their mineral assemblage of plagioclase, hornblende, biotite, K-feldspar, and quartz is also similar to that of the host granitoids but with different mineral proportions. The compositional ranges of plagioclases from host granitoids and their MMEs overlap completely and are oligoclase and andesine with An numbers ranging from 0.159 to 0.438. These plagioclase compositions are very different from those in mafic-ultramafic rocks that are usually characterized by An numbers over 0.50 (e.g., Brophy et al., 1996; Best, 2003). All of these geological and geochemical characteristics indicate that the MMEs were crystallized from a coeval magma that gave rise to the host granitoids, not by mixing and/or mingling between mantle-derived mafic and crustal-derived felsic magmas or synplutonic injections of mafic magma into the host granitoid magma. Crystallization of the MMEs from a coeval magma that gave rise to the host granitoids was also confirmed by similar zircon REE and trace-element patterns between the MMEs and their hosts in each pluton (Fig. 13). Fractionation of plagioclase, ferromagnesian minerals, Fe-Ti oxides, and apatite as shown by the Harker diagrams (Fig. 8) may play important roles in diversity of mineral compositions between host granitoids and their MMEs.

Crystallization temperatures calculated by Ti-in-zircon thermometer calibration (Fig. 14 and Supplemental Table 55) provide further evidence for crystallization of the MMEs from a coeval magma that gave rise to the host granitoids. On average, zircons from mafic igneous rocks have higher Ti concentrations and Ti-in-zircon temperatures than those from felsic rocks (Fu et al., 2008). As shown in Figure 14, although zircons from different plutons are characterized by different Ti content and crystallization temperatures calculated by Ti-in-zircon thermometer calibration (Watson et al., 2006), there is no significant difference between zircon Ti content and crystallization temperatures of the MMEs and their hosts from the same pluton. Similar Ti-in-zircon temperatures of the MMEs and their hosts from each pluton further support our inference on cognate origin of the the MMEs and their host granitoids.

As indicated by geochemical and isotopic data, source areas of the Permian granitoid plutons in the northern NCB are temporally and spatially different, and magma mixing was likely involved during formation of some plutons (Hongge’er, Tonghetai, and Yinhao); and other plutons such as Rentaihe and Xuniwusu were produced by partial melting of ancient lower crust of the NCC and the subducted slab of the Paleo-Asian oceanic plate (or lower arc continental crust of the Bainaimiao arc belt), respectively. However, the above magma mixing process occurred mainly in the melting, assimilation, storage, and homogenization (MASH) zone in the lowermost crust or mantle-crust transition, where basaltic magmas become neutrally buoyant, induce local melting, assimilate, and mix extensively (Hildreth and Moorbath, 1988). In the MASH zone, conditions were favorable for high-volume dehydration partial melting of preexisting ancient lower-crustal rocks (Lackey et al., 2005). Magmas generated in the MASH zone could have then moved up and emplaced in the middle-upper crust. As magmas passed through the continental crust, fractional crystallization and possibly crust assimilation processes likely occurred. The higher initial 87Sr/86Sr ratios of 0.70709–0.71263 from the Xuniwusu and Hongge’er plutons are likely a result of upper crust assimilation because their emplacement depths were much shallower than those of the other Permian plutons.

As for the MMEs crystallized from a coeval magma as the host granitoids, they can be produced by several processes including crystal accumulations (e.g., Fershtater and Borodina, 1977, 1991; Dorais et al., 1997; Phillips et al., 1981; Dodge and Kistler, 1990; Flood, 1993; Flood and Shaw, 1995, 2006, 2014; Dahlquist, 2002; Ilbeyli and Pearce, 2005; Shellnutt et al., 2010), fragmentated chilled margins (e.g., Visonà, 1986; Esna-Ashari et al., 2011), or a crystallization process by rapid cooling within the host granitoid magma operating at the margins of magma conduits (e.g., Donaire et al., 2005). Lithological zoning of some plutons with a dioritic margin and quartz dioritic (tonalitic)–granodioritic interior (Fig. 2) and relatively fine-grained igneous textures of the enclaves (Figs. 3A–3C, 3H, and 3J–3O) indicates that the MMEs in these plutons were mainly produced by fragmentated chilled margins. For the MMEs characterized by porphyric structure (fine-grained groundmass and euhedral hornblende and plagioclase phenocrysts, Figs. 3D–3G, and 3I), they are likely a product of crystal accumulations. Therefore, we propose a two-stage process model including rapid cooling within the cogenetic host granitoid magma operating at pluton margins to form solid to sub-solid dioritic rocks and crystal accumulations of the host granitoid magma at the bottom of the pluton to form cumulates (Fig. 15A), and then fragmentated and interacted with progressive evolved host granitoid magma to change their shapes and orientations (Figs. 15B–15D) for the origin of the MMEs in the Permian granitoid plutons in the northern NCB. The processes proposed in this model may be more widely applicable to the generation of many microgranular enclaves in granitic rocks, especially where no direct evidence exists for the presence of basic magmas coeval with granitoids and where there is a lack of isotopic contrast between hosts and enclaves.

The above model for origin of the MMEs explains the geochemical and textural characteristics of the microgranular enclaves and their host granitoids, including shape, grain size, mineralogy (e.g., dioritic in composition, lack of pyroxene, and low plagioclase An numbers <0.50 in the MMEs), texture, major and trace chemistry, and Sr-Nd-Hf isotopic composition. It also explains the decreasing trend of MMEs abundance from pluton margins to its central portion. Since solidifical and near solidifical chilled margins can drop during their interactions with host granitoid magma due to their relatively high density (Fig. 14D), their crystallization pressures could be less than their host granitoids (Table 3). Ellipsoidal or rounded shapes and chilled margins of the MMEs were most likely generated by chemical reactions between the solidified enclave and a hydrous K-rich residual melt or fluid formed after progressive crystallization and cooling of the host magma body as suggested by recent results from Farner et al. (2014). Therefore, some MMEs in calc-alkaline granitoids previously used as evidence for magma mixing and/or mingling between mantle-derived mafic and crustal-derived felsic magmas are “autoliths (cognate xenoliths)” (Pabst, 1928; Fershtater and Borodina, 1977; Tindle and Pearce, 1983; Dodge and Kistler, 1990; Schönenberger et al., 2006) that are crystallized from the same magma as their host granitoids. There is an essential need for caution in using the MMEs as evidence for crustal-mantle interaction or mixing and/or mingling between crustal-derived felsic and mantle-derived mafic magmas during formation of granitic rocks, especially where no independent evidence is found for such a process. In some previously studied granitoid plutons with enclaves that are in the intermediate composition range and exhibit similar mineral assemblage and isotopic compositions (e.g., Holden et al., 1991; Barbarin and Didier, 1992; Sergi, 1997; Asrat et al., 2004; Barbarin, 2005; Kaygusuz and Aydınçakır, 2009; Turnbull et al., 2010; Wolska, 2012; Xiong et al., 2012), the role of MMEs in evaluating the mixing and/or mingling processes between crustal-derived felsic and mantle-derived mafic magmas may have been overestimated. Instead, the MMEs in this setting can provide valuable information on emplacement mechanisms and incremental growth of granitoid plutons in continental crust (e.g., Paterson et al., 2003, 2004; Donaire et al., 2005; Žák and Paterson, 2005, 2010; Žák et al., 2005, 2007, 2013; Barbey et al., 2008; Economos et al., 2009; Caricchi et al., 2012; Zhang and Zhao, 2013).

7. CONCLUSIONS

  • (1) Mafic microgranular enclaves are very common in the Permian granitoid plutons in the northern NCB. They are mainly dioritic rocks in compositions with SiO2 contents from 51.0 wt% to 62.7 wt% and are contiguous with their host granitoids (SiO2 = 60.4–68.4 wt%). Their mineral assemblage is similar to that of the host granitoids but with different mineral proportions.

  • (2) Zircon LA-ICP-MS U-Pb dating, hornblende-plagioclase thermobarometry, and Ti-in-zircon thermometry results show that the crystallization ages and P-T conditions of the MMEs in the Permian granitoid plutons in northern NCB are very similar to those of their host granitoids, indicating simultaneous crystallization of the MMEs and their hosts at similar emplacement depth in the middle to upper crustal levels.

  • (3) Geochemical and isotopic data show that the source areas of the Permian granitoid plutons in the northern NCB are temporally and spatially different and magma mixing was involved during formation of some plutons. However, the Sr-Nd-Hf isotopic compositions of the MMEs from each pluton are very similar to those of their host granitoids, indicating complete isotopic equilibration between the host granitoids and their enclaves. Combined with similar mineral assemblage and mineral chemistry between the MMEs and their host granitoids and absence of coeval mantle-derived rocks in or nearby the plutons, we suggest that the MMEs were crystallized from a coeval magma that gave rise to the host granitoids, not by mixing and/or mingling between mantle-derived mafic and crustal-derived felsic magmas or synplutonic injections of mafic magma into the host granitoid magma. We conclude that magma isotopic equilibration between host granitoids and their enclaves was achieved prior to emplacement and the magma mixing process occurred mainly in the MASH zone in the lowermost crust or mantle-crust transition, not in the magma conduits or chambers at middle to upper crustal levels.

  • (4) A two-stage model including rapid cooling within the cogenetic host granitoid magma operating at pluton margins to form dioritic rocks and crystal accumulations of the host granitoid magma at the bottom of the pluton to form cumulates, and then differentiated and interacted with progressive evolved host granitoid magma to change their shapes and orientations is proposed for origin of the MMEs in the Permian granitoid plutons in the northern NCB. This model may be more widely applicable to the generation of many microgranular enclaves in granitic rocks, especially where no direct evidence exists for the presence of mafic magmas coeval with granitoids and where there is a lack of isotopic contrast between hosts and enclaves.

  • (5) Our new results show that some MMEs previously used as evidence for magma mixing and/or mingling are “autoliths (cognate xenolith)” crystallized from the same magma as their host granitoids. Therefore, there is an essential need for caution in using the MMEs as evidence for crustal-mantle interaction or mixing and/or mingling between crustal-derived felsic and mantle-derived mafic magmas during formation of granitic rocks, especially where no independent evidence is found for such a process. The role of MMEs in evaluating the mixing and/or mingling processes between crustal-derived felsic and mantle-derived mafic magmas has been much more likely overestimated by some previously studies and should be reconsidered.

This research was financially supported by the National Natural Science Foundation of China (grants 41372230 and 41572204) and the International Science and Technology Cooperation Program of China (grant 2014DFR21270). We thank Y.S. Liu, Z.C. Hu, X.M. Liu, C.Y. Diwu, K.J. Hou, M. Liu, X.D. Jin, and H. Li for their analytical assistance. We are grateful to J. Lawford Anderson, Jiří Žák, and Calvin G. Barnes for their thorough, critical, and constructive reviews and comments that significantly improved the quality of the manuscript and to Cin-Ty Lee for comments on the early version of the paper.

1Supplemental Table 1. LA-ICP-MS U-Pb dating results of the Permian granitoids and their microgranular enclaves. Please visit http://dx.doi.org/10.1130/GES01407.S1 or the full-text article on www.gsapubs.org to view Supplemental Table 1.
2Supplemental Table 2. Hf isotopic results of zircons from the Permian granitoids and their microgranular enclaves. Please visit http://dx.doi.org/10.1130/GES01407.S2 or the full-text article on www.gsapubs.org to view Supplemental Table 2.
3Supplemental Table 3. Hornblende compositions of the Permian granitoids and their mafic microgranular enclaves. Please visit http://dx.doi.org/10.1130/GES01407.S3 or the full-text article on www.gsapubs.org to view Supplemental Table 3.
4Supplemental Table 4. Plagioclase compositions of the Permian granitoids and their mafic microgranular enclaves. Please visit http://dx.doi.org/10.1130/GES01407.S4 or the full-text article on www.gsapubs.org to view Supplemental Table 4.
5Supplemental Table 5. Trace-element compositions of synmagmatic zircons from the Permian granitoids and their mafic microgranular enclaves. Please visit http://dx.doi.org/10.1130/GES01407.S5 or the full-text article on www.gsapubs.org to view Supplemental Table 5.