In this study petrographic characteristics Fig a d suggest
In this Quinacrine Dihydrochloride cost study, petrographic characteristics (Fig. 2a–d) suggest that the tourmalines from the Hudongliang and Cuonadong Neogene magmatic rocks directly crystallized from felsic magmas, and their δ11B values should be similar to those of the magmas. The tourmalines of the Hudongliang Pliocene two mica rhyolite sample (1P2JD7-1) have a mean δ11B value of −10.47 ± 0.54‰, which is consistent with those (−9.71 ± 0.04) of whole rock sample (1P2JD7-1) within the range of errors (Wang et al., unpublished data). The tourmalines from the Cuonadong Miocene two mica granite sample (11SN36) in the Himalayan Block have a mean δ11B value of −12.48 ± 1.04‰, which is in the range of tourmaline δ11B values (−16.3 to −10.3‰) of the Himalayan leucogranites (Chaussidon and Albarède, 1992; Yang et al., 2015) (Fig. 4). The δ11B values of granitic tourmalines were commonly considered to be similar to those of the melt (Jiang and Palmer, 1998; Jiang et al., 2008; Talikka and Vuori, 2010; Zhao et al., 2011; Yang and Jiang, 2012; Yang et al., 2015). The tourmalines of both the Hudongliang rhyolites and Cuonandong granites have δ11B values ranging from −12.48 to −10.47‰, which are clearly lower than those (>0 in general) of marine sedimentary rocks or magmatic rocks undergoing seawater alteration or tourmalines from above rocks or fluids from them (Fig. 4) (Marschall and Jiang, 2011). E.g., the metasomatic tourmaline owing to influx of hydrous fluids released from the subducted oceanic slab in the Syros Island has exceptionally high δ11B (+18 to +28‰) (Marschall et al., 2006). However, their δ11B values are similar to those (−10.3 to −15.4‰) of magmatic tourmalines from continental crust-derived granites and volcanic rocks (Fig. 4) (Jiang, 2001; Jiang et al., 2008; Talikka and Vuori, 2010; Zhao et al., 2011; Yang and Jiang, 2012; Guo et al., 2014; Yang et al., 2015), and very close to those (−10 ± 3‰) of average continental crust (Fig. 4) (Marschall and Jiang, 2011). Alternatively, some other studies have suggested that the δ11B values of tourmaline may be higher than those of melt (e.g., Macgregor et al., 2013), which would imply that the melts of both the Hudongliang rhyolites and Cuonandong granites had δ11B values lower than −12.48 to −10.47‰. In other words, both the Hudongliang rhyolites and Cuonandong granites may have had crustal sources with lower δ11B values than those of the tourmalines or the (−10 ± 3‰) average continental crust (Fig. 4) (Marschall and Jiang, 2011). Miocene–Pliocene peraluminous lavas including the Hudongliang rhyolites in the Songpan-Ganzi block have higher εNd (−5.8 to −8.6) and lower 87Sr/86Sr (0.7125–0.7178) (McKenna and Walker, 1990; Wang et al., 2012; Zhang et al., 2012) than the respective values (−13.0 to −20.0 and 0.7256 to 0.8547) for Late Oligocene–Miocene leucogranites, including the Cuonandong granites in the Himalayan block (e.g., Inger and Harris, 1993; Harris and Massey, 1994; Guillot and Le Fort, 1995; Harris et al., 1995; Knesel and Davidson, 2002; Zhang et al., 2004; Liao et al., 2007; Guo and Wilson, 2012; Gao and Zeng, 2014; Liu et al., 2014; Wu et al., 2015). This indicates that Neogene peraluminous rocks in the northern and southern margins of Tibet were generated by partial melting of metasedimentary rocks with different components. Petrological and Sr–Nd isotopic data also suggest, however, that secondary cell wall were both most likely derived from Precambrian–Mesozoic metasedimentary rocks (e.g., metapelites or metagreywackes) on the northern and southern margins of Tibet, respectively (e.g., Le Fort et al., 1987; Burchfiel et al., 1989; McKenna and Walker, 1990; Inger and Harris, 1993; Harris and Massey, 1994; Guillot and Le Fort, 1995; Harris et al., 1995; Patiño Douce and Harris, 1998; Knesel and Davidson, 2002; Zhang et al., 2004, 2012; Liao et al., 2007; Guo and Wilson, 2012; Wang et al., 2012; Gao and Zeng, 2014; Liu et al., 2014; Wu et al., 2015). The question of whether their source rocks were sedimentary rocks formed in marine or continental settings has been more difficult to resolve. Our new results demonstrate that the Hudongliang rhyolites and Cuonandong granites exhibit tourmaline δ11B values close to those of average continental crust or continental crust-derived granites (Fig. 4), indicating that they were mainly generated by partial melting of the metasedimentary rocks with continental affinities. Moreover, these source rocks were most probably subducted or overridden in the northern and southern margins of Tibet during contraction associated with the collision of the Indian plate with the Asian plate (Yin and Harrison, 2000; Tapponnier et al., 2001; Ding et al., 2003; Kapp et al., 2003, 2005).