An extensive dataset of major- and trace-element and B-isotope composition related to the compositional variation of the tourmalines from the S-type granite and pegmatite from the Upper Bonai and Gangpur Group granitoids, eastern India, is provided. The Gangpur Schist Belt, consisting of the Upper Bonai and Gangpur Group, lies at the eastern end of the Proterozoic Central Indian Tectonic Zone. An attempt to constrain the source of the boron isotope and describe the magmatic–hydrothermal evolution of the granite–pegmatite system is made. The tourmalines record generally low trace-element concentrations, with the exception of some elements such as Zn, Li, Ga, Mn and Ti. The Zn and Li concentration increases from the tourmalines in the granites to the tourmalines in the pegmatite. Trace-element compositions also vary slightly from the core tourmaline to the rim tourmaline. Tourmalines in granites/pegmatites are mostly Fe-rich schorl, and have S-type B-isotopic signature (δ11B = −10‰ to −12‰). Their chemistries reflect changing salinities of the granitic melts as well as vapour exsolution during progressive crystallization. This study indicates that the tourmalines in the pegmatites crystallized from evolved flux-element-rich (F, B, Li) melt, probably after the crystallization of granite-forming melt.

Detrital zircon U–Pb ages for sediments in and around the Palaeo-Tethyan convergence zone in northern Thailand provide constraints for tectonic interpretations of the Indochina Block, the Sibumasu Block, the Inthanon Zone accretionary complex and the Nan Back-arc Basin during the Triassic. In sedimentary rocks of the Indochina Block, almost all of the Palaeozoic and Triassic zircons were sourced from the collision zone between the Indochina and South China blocks, and an active continental margin in the western Indochina Block. Sediments of the Sibumasu Block were supplied by erosion of Archaean basement and from the Grenville and the Pan African orogenies, but show no record of Permian to Triassic igneous activity. Accretionary complex sediments have provenances of both the Sukhothai Arc and the Indochina and South China blocks, with detrital zircons of various ages being supplied from crustal uplift and erosion related to the Indosinian I orogeny. Sedimentary rocks of the Nan Back-arc Basin are widely distributed not only in the Nan–Uttaradit but also in northern Sukhothai areas. The origin of the Pha Som Metamorphic Complex and associated formations can be traced to basin-filling sediments in the Nan Back-arc Basin. These detrital zircon U–Pb ages have also allowed identification of the changing tectonic setting in the Palaeo-Tethys convergence zone from the ‘erosion of Proterozoic continental basement’ to ‘Palaeozoic active continental margin in the western Indochina Block’ and ‘Palaeozoic, Permian to Triassic collision zone between the South China and Indochina blocks’ through to ‘Triassic active Sukhothai Arc’.

Our field-based geochemical studies of the Triassic, Jurassic and Cretaceous granitoids on Hainan Island indicate that their magmas had different geochemical affinities, changing from alkaline in the Triassic through ocean island basalt (OIB) in the Jurassic, to calc-alkaline in the Cretaceous. We show that these changes in the geochemical affinities of the Mesozoic granitoids on Hainan and in SE China reflect different melt sources and melt evolution patterns through time. Our new geodynamic model suggests that: (1) Triassic geology was controlled by flat-slab subduction of the palaeo-Pacific plate beneath SE China. This slab dynamics resulted in strong coupling between the lower and upper plates, causing push-over tectonics and contractional deformation in SE China. Flat subduction-induced edge flow and aesthenospheric uprising led to the production of high-K granites, syenites and mafic rocks. (2) Slab foundering, accelerated subduction rates and subduction hinge retreat in the Early Jurassic caused rapid rollback of the downgoing slab. Strong decoupling of the upper and lower plates resulted in pull-away tectonics, producing extensional deformation in SE China. Decompression melting of the upwelling aesthenosphere produced OIB-type melts, which interacted with the subcontinental lithospheric mantle (SCLM) to form A- and I-type granitoids. (3) Segmentation of the palaeo-Pacific plate in the Early Cretaceous resulted in steeply dipping slabs and their faster rollback, facilitating lithospheric-scale extension and oceanward migration of calc-alkaline magmatism. This extensional deformation played a significant role in the formation of metamorphic core complexes, widespread crustal melting and development of a Basin and Range-type tectonics and landscape evolution in SE China.

Two ammonites have been found in porcelaneous limestones of the White Lias Formation traditionally considered to be of Rhaetian (Late Triassic) age at Lavernock Point, south Wales (ST 181 681). These ammonites, named here Neophyllites lavernockensis. sp. nov, are the earliest recorded from the UK. This horizon is located directly above a major negative δ13C isotope anomaly (CIE1) in the Upper Cotham Member that marks the top of the Triassic System and below another negative δ13C isotope anomaly (CIE3). Both correlate with negative δ13C isotope anomalies in the Triassic–Jurassic global boundary stratotype section and point (GSSP) at Kuhjoch, Austria. This establishes the base of the tilmanni Chronozone, Hettangian Stage and Jurassic System in SW Britain within the White Lias Formation at Lavernock Point.

This paper provides insight into the Late Ordovician to earliest Silurian evolution of sedimentary environments in the southern Holy Cross Mountains (SE Poland), which at that time were a part of the SW periphery of Baltica. The facies layout in this area was influenced by the basement block faulting, which differentiated the basin bathymetry into submarine horst and grabens, controlling facies distribution. However, the local tectonism was insufficient to fully mask the global eustatic events. Therefore, it is possible to correlate some facies changes in the Upper Ordovician and lower Llandovery sedimentary record of the southern Holy Cross Mountains with eustatic and palaeoceanographic changes reported worldwide. The most noticeable influence of eustasy on the sedimentary record in the studied area occurs at the Ordovician/Silurian boundary. It is manifested by Hirnantian regressive coarse-grained clastic sediments overlain by a post-glacial anoxic/dysoxic interval represented by the Rhuddanian transgressive black cherts and shales. It is noteworthy that the pre- and post-Hirnantian sedimentary environments in the southern Holy Cross Mountains were affected by upwelling induced by the SE trade winds.

In the Transdanubian Range, Pannonian Basin, Hungary, karstic sinkholes on a planation surface of Triassic carbonates are filled by grey clayey–silty kaolin deposits. The provenance and accumulation age of these strongly altered terrestrial karst-filling sediments are constrained by X-ray powder diffraction, heavy mineral analysis and zircon U–Pb dating. The heavy minerals of the Southern Bakony Mountains samples are dominated by the ultra-stable zircon–rutile–tourmaline association. Zircon U–Pb data indicate accumulation between 20 and 16 Ma. Furthermore, Archaean to Palaeogene grains were also determined, reflecting the principally fluvial recycling of Eocene bauxites and their cover sequences. In contrast, the sample from the Keszthely Hills consists almost exclusively of airborne material including zircons of 18–14 Ma, reflecting a dominant contribution from the Carpathian–Pannonian Neogene volcanism. The shift in the Miocene age components is inferred to have been caused by the landscape evolution and burial history of the planation surface remnants controlled by local block tectonics.

The newly discovered Bibole banded iron formations are located within the Nyong Group at the northwest of the Congo Craton in Cameroon. The Bibole banded iron formations comprise oxide (quartz-magnetite) and mixed oxide-silicate (chlorite-magnetite) facies banded iron formations, which are interbedded with felsic gneiss, phyllite and quartz-chlorite schist. Geochemical studies of the quartz-magnetite banded iron formations and chlorite-magnetite banded iron formations reveal that they are composed of >95 wt % Fe2O3 plus SiO2 and have low concentrations of Al2O3, TiO2 and high field strength elements. This indicates that the Bibole banded iron formations were not significantly contaminated by detrital materials. Post-Archaean Australian Shale–normalized rare earth element and yttrium patterns are characterized by positive La and Y anomalies, a relative depletion of light rare earth elements compared to heavy rare earth elements and positive Eu anomalies (average of 1.86 and 1.15 for the quartz-magnetite banded iron formations and chlorite-magnetite banded iron formations, respectively), suggesting the influence of low-temperature hydrothermal fluids and seawater. The quartz-magnetite banded iron formations display true negative Ce anomalies, while the chlorite-magnetite banded iron formations lack Ce anomalies. Combined with their distinct Eu anomalies consistent with Algoma- and Superior-type banded iron formations, we suggest that the Bibole banded iron formations were deposited under oxic to suboxic conditions in an extensional basin. SIMS U–Pb data indicate that the Bibole banded iron formations were deposited at 2466 Ma and experienced metamorphism and metasomatism at 2078 Ma during the Eburnean/Trans-Amazonian orogeny. Overall, these findings suggest that the studied banded iron formations probably marked the onset of the rise of atmospheric oxygen, also known as the Great Oxidation Event in the Congo Craton.

The Haimur area represents the central part of the Wadi Allaqi region in the southern block of the Egyptian Eastern Desert near Nasser Lake and the Nile valley. It is made up of ophiolitic assemblage comprising serpentinite and talc carbonate, listwenite, metagabbro/amphibolite and metabasalt and island arc assemblages. The orogenic gold deposits in the Haimur area occur in the form of smoky/white sulphide-bearing quartz / quartz-carbonate veins or lenses cutting through the listwenite zone-related rocks. The NE-trending auriferous veins were formed due to an extensional to transtensional shearing related to NW Najd shear tectonics. Gold was observed in association with late-pyrite and chalcopyrite paragenetically formed in the gold phase (second phase), while the first phase is the sulphide phase including early-pyrite and arsenopyrite. Arsenopyrite was formed during early metamorphic recrystallization at a high-temperature range between 405 and 512 °C. However, gold was formed at lower temperature through retrograde metamorphism. Three types of fluid inclusions have been recognized: aqueous (type-I), mixed aqueous–carbonic (type-II) and hydrocarbonic (type-III). The P–T conditions of trapping were obtained from the isochore lines of intersection between T = 300–320 °C and pressure range of 60–180 MPa. The Haimur gold deposit was supposed to be derived from metamorphic fluids created by dehydration and decarbonation of ophiolitic mélange assemblages and volcano-sedimentary rocks, in which the gold was transported as Au(HS)2– complexes and precipitated in reduced environment. The strong linkage between the deformation and metamorphism triggered gold deposition in structurally favourable sites related to the Najd fault system.

An integrated approach involving Sr–Nd isotope, trace and rare earth element analyses tracks multiple sources of the Mesozoic sediments of the Kutch Basin at the western continental margin of India. High (87Sr/86Sr)t (ratio at time of deposition), negative εNd and high concentrations of large-ion lithophile elements (LILEs) indicate the upper continental source. Ratios of Nb/Ta and Zr/Hf suggest sedimentary and felsic igneous sources of sediments. The moderate to high concentration of La, Th and Sc, light rare earth elements (LREE-) enrichment, weak negative Eu anomalies and the relationship between εNd(0) and Th/Sc indicate the dominantly felsic composition of source rocks. However, low contents of Th, low values of (87Sr/86Sr)t and depleted mantle model age TDM < 1600 Ma indicate input from a younger mafic source. Increasing concentrations of Zr, Hf and Nd isotopes and a gradual increase in mean TDM from the older to the younger formations indicate erosional unroofing at the source terrain. The increasing (87Sr/86Sr)t through time relates to increased weathering of the source rock. The overwhelmingly southwesterly palaeocurrent direction of current-generated sedimentary structures, and the mean TDM ages trace suggest source areas of the Kutch Basin to Precambrian rocks in the north and NE of this basin. The TDM ages highlight the dominance of late Palaeoproterozoic source rocks. Nd isotope composition indicates that Proterozoic rocks of Marwar Supergroup and Erinpura Granite, in particular, served as main sediment contributors for the Mesozoic sediments in Kutch. We therefore conclude that the Mesozoic sediments in the Kutch Basin are predominantly of late Palaeoproterozoic age with lesser inputs from rocks of early Mesoproterozoic and early Palaeoproterozoic age.

A detailed geochronological study was conducted on zircons from a diorite sample of the Posets pluton (Axial Zone, Pyrenees). The extracted igneous zircons constrain the emplacement of the pluton to 302 ± 2 Ma and 301 ± 3 Ma, by means of U–Pb sensitive high-resolution ion microprobe (SHRIMP) and laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) analyses, respectively. Considering the syn- to late-tectonic emplacement of the Posets pluton during the main Variscan deformation event (D2), the obtained ages constrain the long-lasting D2, associated with the dextral transpression registered through the Axial Zone of the Pyrenees.

The South Qinling block, a segment of the Yangtze craton involved in the Qinling–Dabie orogen, is critical for understanding the tectonic evolution of eastern China. However, the tectonic setting of the South Qinling block and the northern margin of the Yangtze block during middle Neoproterozoic time has long been the subject of debate, with two distinctly different models (continental rift or volcanic arc) proposed. Here, a comprehensive study of zircon U–Pb geochronology and geochemistry has been carried out on the Chengwan granitic pluton from the Suizao terrane in the South Qinling block. The granites are monzogranite and syenogranite in lithology, and are mainly composed of potash feldspar, quartz, plagioclase and biotite. This suite has long been regarded as a Palaeozoic magmatic pluton, but zircon U–Pb ages of 809 ± 9 Ma and 816 ± 4 Ma are obtained in this study. The granites are metaluminous to strongly peraluminous with high alkali contents, and exhibit highly fractionated features, including high SiO2, low Zr/Hf ratios, rare earth element tetrad effects and enrichment of K and Rb. They show Hf–Nd isotopic decoupling, which may be genetically related to their petrogenetic process. Based on the geochemical features and the positive εHf(t) values of the zircons, it is indicated that the granites may have been derived from partial melting of juvenile tonalitic rocks by biotite breakdown under fluid-absent conditions. The Chengwan granite geochemically belongs to the A2-subtype granites, suggesting that it might have formed in a post-orogenic tectonic setting. The highly fractionated A-type granite in this study may represent extensional collapse shortly after the collisional events in the South Qinling block, and thus indicate a tectonic regime switch, from compression to extension, as early as middle Neoproterozoic time. Integrating our new data with documented magmatic, metamorphic and sedimentary events during middle Neoproterozoic time in the region may support a continental rift model, and argues against arc models.

Myanmar is occupied by the N-wards continuation of the Sunda arc and by the Shan Plateau and its continuation through Yunnan into Tibet. Our new tectonic interpretation of the ophiolite–flysch belts, world-famous jadeite and tin deposits in Myanmar west of the Salween adopts previous proposals that, before 450-km post-early Oligocene dextral displacement along the Sagaing Fault, the ophiolite belt in NE Myanmar continued through the topography that is now located west of the fault in the Indo-Burman Ranges. Differences in cross-section through Mogok and the Shan Scarps are reconciled by the recently proposed emplacement, in our view during Permian time, of the Mogok Metamorphic Group onto the Slate Belt to form Sibumasu. We argue that during Early Jurassic time a Neo-Tethys ophiolite nappe was obducted over turbidites on Sibumasu’s passive western margin. Following reversal in tectonic polarity, the remaining Neo-Tethys subducted E-wards generating the 113–128 Ma Mondaung Arc. During ocean closure the Victoria–Katha Block and its Triassic flysch subducted beneath Sibumasu, resulting in jadeite veins in overlying serpentinite that ascended in the subduction zone and were exhumed at Hpakant and Nat Hmaw, bordering the Jade Mines Uplift. Subduction of the Indian Ocean since Albian time generated the Popa–Loimye arc, while extensional faulting led to uplift of the Indo-Burman Ranges and to the formation of the Western Tin Belt granites. Tectonic effects in Myanmar of the India–Asia collision may be confined to the Disang thrust belt in the Naga Hills.

Rare earth elements (REE) in marine minerals have been widely used as proxies for the redox status of depositional and/or diagenetic environments. Phosphate nodules, which are thought to grow within decimetres below the sediment–water interface and to be able to scavenge REE from the ambient pore water, are potential archives of subtle changes in REE compositions. Whether their REE signals represent specific redox conditions or they can be used to track the overlying water chemistry is worth exploring. Through in situ laser ablation – inductively coupled plasma – mass spectrometry (LA-ICP-MS), we investigate the REE compositions of a drill-core-preserved phosphate nodule from the lower Cambrian Niutitang Formation in the Daotuo area, northeastern Guizhou Province, South China. REE distributions of the nodule show concentric layers with systematic decreases in Ce anomalies (Ce/Ce*) from the core to the rim. The lowest Ce/Ce* appears in the outer rim where REE concentrations are relatively high. These results are interpreted to reflect REE exchange with pore water at a very early stage or bathymetric variation during apatite precipitation. The origin of the shale-normalized middle REE (MREE) enrichment in our sample is less constrained. Possible driving factors include preferential MREE substitution for Ca in the apatite lattice, degradation of organic matter and deposition beneath a ferruginous zone. Although speculative, the last possibility is consistent with the chemically stratified model for early Cambrian oceans, in which dynamic fluctuations of the chemocline provided an ideal depositional context for phosphogenesis.

The Jameson Land Basin lies within the Greenland–Norway rift, to the south of Triassic basins on the Halten Bank and East Greenland shelf, and can therefore offer insights into the development of these less-well-documented offshore basins and the wider North Atlantic. This study focuses on the stratigraphic development of the Jameson Land Basin. Broad-scale stratigraphic logging is augmented with heavy mineral analysis that constrains the dominant source catchments during this period. A thickness of over 1.5 km of Middle–Late Triassic strata fill the Jameson Land Basin, predominantly comprising continental deposits laid down during arid to semi-arid climatic conditions. The basal Pingo Dal Group thickens westwards into the Stauning Alper Fault, forming a typical syn-rift succession. Thickness variations have also highlighted the presence of an intrabasinal high and an associated unconformity between the Pingo Dal Group and the overlying Gipsdalen Group, which provides the first direct evidence for Middle–Late Triassic rifting in East Greenland. Thickness variations in the overlying Gipsdalen Group are more subdued, and the Fleming Fjord Group, and its formations, display remarkably tabular geometries, with little evidence for significant lateral facies variance; this is suggestive of reduced topography and a more regional thermal subsidence signature. Provenance data indicate sediment sourcing from both the western and eastern basin margins during the deposition of the Pingo Dal Group, above which the western source area becomes dominant, alongside some axial input from the south.

Late Cretaceous tracks attributable to deinonychosaurs in North America are rare, with only one occurrence of Menglongipus from Alaska and two possible, but indeterminate, occurrences reported from Mexico. Here we describe the first probable deinonychosaur tracks from Canada: a possible trackway and one isolated track on a single horizon from the Upper Cretaceous Wapiti Formation (upper Campanian) near Grande Prairie in Alberta. The presence of a relatively short digit IV differentiates these from argued dromaeosaurid tracks, suggesting the trackmaker was more likely a troodontid. Other noted characteristics of the Wapiti specimens include a rounded heel margin, the absence of a digit II proximal pad impression, and a broad, elliptical digit III. Monodactyl tracks occur in association with the didactyl tracks, mirroring similar discoveries from the Early Cretaceous Epoch of China, providing additional support for their interpretation as deinonychosaurian traces. Although we refrain from assigning the new Wapiti specimens to any ichnotaxon because of their relatively poor undertrack preservation, this discovery is an important addition to the deinonychosaur track record; it helps to fill a poorly represented geographic and temporal window in their known distribution, and demonstrates the presence of a greater North American deinonychosaur ichnodiversity than has previously been recognized.

The formation of magmatic plumbing systems in the crust involves mass and heat transfer from deep to shallow levels. This process modifies the local geotherm and increases the thermal maturation of the crust, affecting the rheological state of the host rock and the composition of magma. Here, we report a petrological, geochemical, isotopic and geochronological integrated study of the Huaco (354 Ma) and Sanagasta (353 Ma, from a new U–Pb zircon age) units from the Carboniferous (Lower Mississippian) Huaco Intrusive Complex, NW Argentina. Similar values of ϵNdt and δ18O, of −3.2 ± 0.7 and +11.2‰ ± 0.3‰ (V-SMOW), respectively, for both units indicate that they shared the same source, as a result of mixing and later homogenization of a crustal component at the Late Devonian (378 to 366 Ma), with metasomatized mantle-derived melts. Slightly higher contents of TiO2, FeO, MgO, CaO and rare earth elements for the Sanagasta unit in comparison with the Huaco unit suggest an increase in the degree of partial melting, which may have been caused by a higher temperature at the lower crust. In addition, the previous structural model of the Huaco Intrusive Complex points to an increase in thermal maturation in the upper crust, which drives a change in the emplacement style from tabular subhorizontal (Huaco) to vertically elongated (Sanagasta) bodies. Therefore, the evolution of the intrusive complex may reflect a generalized thermal maturation of the complete magmatic column, at both upper and lower crustal levels.

Zircon U–Pb and Lu–Hf isotope microanalysis was conducted in (meta)-igneous units of Ammouliani island to characterize crust-forming and reworking events in the Serbo-Macedonian massif. Zircon grains from an orthogneiss of the Vertiskos unit yielded a U–Pb crystallization age for the igneous precursor at 458.8 ± 11 Ma with dominantly subchondritic ϵHf values indicating reworking of Neoproterozoic basement. A weighted mean ϵHf value of 0.7 ± 2.4 from oscillatory zoned zircon grains of the Ouranoupoli granodiorite indicates juvenile crustal input at 52.1 ± 0.6 Ma. The U–Pb–Hf zircon archive records discrete stages in the crustal evolution of the Serbo-Macedonian massif.

The Sargur Group has been considered to be the oldest group (>3.0 Ga) in the Archaean sequence of the Dharwar Craton in south India, whereas the rocks of the Dharwar Supergroup are younger (between 3.0 and 2.55 Ga). The supracrustal units of the Sargur Group were deposited during the Archaean period. The Banavara quartzite forms part of the supracrustal Sargur Group and contains significant amounts of chromian spinel (Cr-spinel). Here, U–Pb and Hf isotopes of detrital zircons are integrated with compositional data and X-ray refinement parameters for Cr-spinels to decipher the provenance of the metasediments. Zircons show an age spectrum from 3.15 to 2.50 Ga, and juvenile Hf isotopic compositions (ϵHf = +0.8 to +6.4) with model ages between 3.3 and 3.0 Ga. Major- and trace-element contents of the Cr-spinels do not resemble those in the Sargur ultramafic rocks, but resemble well-characterized Archaean anorthosite-hosted chromites. Cr-spinel trace-element signatures indicate that they have undergone secondary alteration or metamorphism. X-ray refinement parameters for the Cr-spinels also resemble the anorthosite-hosted chromites. We conclude that the detrital minerals were probably derived from gneissic and anorthositic rocks of the Western Dharwar Craton, and that the Sargur Group sequences have experienced a younger (2.5 Ga) metamorphic overprint.

The Golshekanan granitoid body is situated in the central part of the Urumieh–Dokhtar magmatic arc (UDMA) in central Iran, and includes granite and granodiorite with minor monzonite and diorite. Zircon U–Pb dating yields a late Eocene (Priabonian) crystallization age of 37.6 ± 0.2 Ma. The body is calc-alkaline and metaluminous to weakly peraluminous (A/CNK ≤ 1.10) with SiO2 ranging from 61.1 to 71.5 wt% and MgO from 0.8 to 3.3 wt%, with Na2O + K2O of 4.0–8.5 wt%. Primitive mantle-normalized trace-element patterns display enrichments in the large-ion lithophile elements (LILE), such as Rb, Cs, Ba and K, and depletion from the high-field-strength elements (HFSEs), such as Nb, Ti, Ta and P. The rocks are enriched in LREEs relative to HREEs (average (La/Yb)CN = 4.3) and exhibit weak negative Eu anomalies (average Eu/Eu* = 0.75), revealing typical active continental margin arc affinity. The low initial 87Sr/86Sr ratios (0.70440–0.70504) and notable positive ϵNd(t) values (+4.0 to +5.2) indicate an origin by partial melting of juvenile rocks in the lower crust, possibly with some involvement of sub-continental lithospheric mantle beneath Central Iran. These processes probably occurred due to the Neo-Tethys oceanic slab retreat and (or) rollback during late Eocene time.

The Zhangguangcai Range in the Xing’an Mongolian Orogenic Belt, NE China, contains Early Jurassic (c. 188 Ma) Dabaizigou (DBZG) porphyritic dolerite. Compared with other island-arc mafic rocks, the DBZG dolerite is characterized by high trace-element contents, relatively weak Nb and Ta enrichments, and no Zr, Hf or Ti depletions, similar to OIB-type rocks. Analysed rocks have (87Sr/86Sr)i ratios of 0.7033–0.7044, relatively uniform positive ɛNd(t) values of 2.3–3.2 and positive ɛHf(t) values of 8.5–17.1. Trace-element and isotopic modelling indicates that the DBZG mafic rocks were generated by partial melting of asthenospheric mantle under garnet- to spinel-facies conditions. The occurrence of OIB-like mafic intrusion suggests significant upwelling of the asthenosphere in response to lithospheric attenuation caused by continental rifting. These processes occurred in an incipient continental back-arc environment in the upper plate of a palaeo-Pacific slab subducting W–NW beneath East Asia.