Pfaffenbergite, KNa3(Al4Si12)O32, is a new mineral found in nanogranitoids included in garnets from HP and UHP eclogites in the Saxo-Thuringian Zone of the Bohemian Massif (Saxony, Germany). The occurrence of a new mineral phase was initially inferred from its unique micro-Raman spectrum, characterized by a very strong vibrational mode at 412 cm-1 and minor peaks at 105, 832, 130, and 470 cm-1. Subsequently, three-dimensional electron diffraction revealed that pfaffenbergite is a hexagonal mineral, crystallizing in space group P6/mcc, isostructural with kokchetavite (KAlSi3O8) and wodegongjieite [KCa3(Al7Si9)O32]. These two sheet silicates have feldspar stoichiometry and chemical composition, and pfaffenbergite corresponds chemically to an unmixed binary K-Na feldspar. Microstructural and experimental constraints suggest that it formed as a result of melt crystallization during cooling, after entrapment within metamorphic garnets. The precise pressure-temperature conditions of formation of pfaffenbergite are currently unknown. However, they must have been below the P-T conditions of entrapment of the inclusions, i.e., 1000-1050 degrees C and 2.2-4.5 GPa, as determined in the present work, based on the values estimated for each individual case study.We interpret pfaffenbergite as a metastable phase crystallizing rapidly in a silicate melt enclosed in a small pore under non-equilibrium conditions, also based upon previous studies reporting the occurrence of other metastable phases in the same or neighboring nanogranitoids (kokchetavite, kumdykolite, dmisteinbergite, etc.). The increasing number of findings of metastable phases in the last years suggests that these minerals are more common than expected. We propose that it is even possible they may represent rock-forming minerals in natural rocks that experienced rapid cooling/rapid crystallization, for instance, lavas and ignimbrites, along with experimental products involving silicate melts.

Garnets in the eclogites of Pfaffenberg, Granulitgebirge (Bohemian Massif, Germany) contain primary granitic melt inclusions with a continental crust signature. The inclusions are up to 30 µm in diameter and polycrystalline with a main mineral assemblage dominated by phlogopite/biotite, kumdykolite, quartz/cristobalite, two unknown phases with main Raman peaks at 412 and 430 cm-1 respectively, osumilite and plagioclase. In minor amounts, the inclusions contain also white mica, K-feldspar, amphibole and kokchetavite with the local presence of a fluid phase composed of CO2, CH4 and N2. The inclusions were successfully re-homogenized at 975ºC and 2.7 - 3 GPa and the melt is from trondhjemitc to granitic, peraluminous and hydrous (average H2O = 4.82 wt%). The melt trace elements patterns revealed similarities with melts produced by partial melting of metasediments part of the continental crust. The melt is in fact enriched in Cs, Pb, Rb, Th, U, Li and B and most likely it originated from the continental crust itself. Interestingly, in situ analyses of Cl and calculation of F partitioning between apatite and melt show that the melt is exceptionally halogens-rich with an average Cl content of 0.41 wt% and a calculated F content of 0.23 wt%. Pfaffenberg eclogites occur as lenses in garnet peridotite and they are surrounded by continental rocks. They can be regarded as the product of crust-mantle interaction taking place during subduction at mantle depth with the agent of the interaction, i.e., the melt, now preserved as inclusions in the eclogite garnets. The melt is responsible for crustal material mobilization and transfer in the mantle and can be used to constrain and quantify the elements, especially volatiles, transported from the crust to the mantle. This research is part of the project No. 2021/43/P/ST10/03202 co-funded by the National Science Centre of Poland and the European Union Framework Programme for Research and Innovation Horizon 2020 under the Marie Skłodowska-Curie grant agreement No. 945339.

The Kerala Khondalite Belt – part of the Southern Granulite Belt of SW India – is well known for its excellent outcrops of metapelites which underwent partial melting during the Panafrican orogeny at lower crustal conditions with temperatures up to 1000°C. Several groups have worked on the metamorphism and petrology of this area which is now complemented by our study of nanogranitoids, i.e., unmodified melt inclusions (MI) of anatectic origin, in zircons from Khondalite. The investigated samples originate from the locations Koliakkode and Pottangodu quarries. Whereas nanogranitoids in peritectic garnets from Koliakkode quarry have already been investigated (Cesare et al., 2009; Ferrero et al., 2012), in Pottangodu we report the first finding of such inclusions. Preliminary results from optical microscopy investigation reveal that the inclusions are always primary, with an azonal distribution in zircon and a cluster distribution in the garnets from both localities. MicroRaman spectroscopy shows the occurrence of a constant phase assemblage in MI in zircon consisting mainly of kokchetavite, white mica, cristobalite/quartz ± kumdykolite. Phase 430 and phase 412 (both new phases and currently under investigation), carbonate and graphite may be present, in some cases also CO2 and N2. The MI hosted by garnet are quite similar but here we also find biotite in the main assemblage. Additionally, the MI in garnet may contain osumilite, andalusite, apatite and rutile. In numerous studies by our group we have demonstrated that the identified metastable polymorphs, like kokchetavite (hexagonal k-feldspar), kumdykolite (orthorhombic albite) and cristobalite, are an excellent indicator that melt inclusions are pristine i.e. unadulterated due to decrepitation or by post entrapment infiltration of material along cracks. In order to obtain new data on the pristine melt compositions these nanogranitoids will be then re-homogenized via piston cylinder experiments. The detailed investigation of these inclusions, coupled with geochronological studies on the host zircon from both localities will unravel the microchemistry of the original unmodified anatectic melt preserved in inclusions, as well as its evolution over time at the host rock scale. In particular, these data will clarify 1) which zircon domains originated before, during and after melt production, as well as 2) the age of the melting event(s?) at HT-UHT conditions. Moreover, to date this represents one of the few examples of anatectic MI in zircons from metamorphic rocks to be investigated in detail via in situ MicroRaman spectroscopy. References: Cesare, B., Ferrero, S., Salvioli-Mariani, E., Pedron, D. & Cavallo, A. 2009. Nanogranite and glassy inclusions: the anatectic melt in migmatites and granulites. Geology, 37, 627–630, https://doi.org/10.1130/G25759A.1 Ferrero, S., Bartoli, O., Cesare, B., Salvioli-Mariani, E., Acosta-Vigil, A., Cavallo, A., Groppo, C., Battiston, S., 2012. Microstructures of melt inclusions in anatectic metasedimentary rocks. Journal of Metamorphic Geology, 30, 303–322, https://doi.org/10.1111/j.1525-1314.2011.00968.x

Primary granitic melt inclusions are trapped in garnets of eclogites in the garnet peridotite body of Pfaffenberg, Granulitgebirge (Bohemian Massif, Germany). These polycrystalline inclusions, based on their nature and composition, can be called nanogranitoids and contain mainly phlogopite/biotite, kumdykolite, quartz/rare cristobalite, a phase with the main Raman peak at 412 cm−1, a phase with the main Raman peak at 430 cm−1, osumilite and plagioclase. The melt is hydrous, peraluminous and granitic and significantly enriched in large ion lithophile elements (LILE), Th, U, Li, B and Pb. The melt major element composition resembles that of melts produced by the partial melting of metasediments, as also supported by its trace element signature characterized by elements (LILE, Pb, Li and B) typical of the continental crust. These microstructural and geochemical features suggest that the investigated melt originated in the subducted continental crust and interacted with the mantle to produce the Pfaffenberg eclogite. Moreover, in situ analyses and calculations based on partition coefficients between apatite and melt show that the melt was also enriched in Cl and F, pointing toward the presence of a brine during melting. The melt preserved in inclusions can thus be regarded as an example of a metasomatizing agent present at depth and responsible for the interaction between the crust and the mantle. Chemical similarities between this melt and other metasomatizing melts measured in other eclogites from the Granulitgebirge and Erzgebirge, in addition to the overall similar enrichment in trace elements observed in other metasomatized mantle rocks from central Europe, suggest an extended crustal contamination of the mantle beneath the Bohemian Massif during the Variscan orogeny.

Subduction is the main process that recycles surface material into the mantle. Fluids and melts derived by dehydration and partial melting reactions of subducted continental crust, a major reservoir of volatiles (i.e., H 2 O and CO 2 ) and incompatible elements, can substantially metasomatize and refertilize the mantle. Here, we investigate glassy inclusions of silicate melt of continental origin found in Variscan ultrahigh-pressure eclogites to assess the continental crust contribution to mantle metasomatism and the journey of volatiles, carbon in particular, to the deep roots of mountain belts. We argue that the melt preserved in these inclusions is the agent responsible for mantle metasomatism and subsequent ultrapotassic magmatism in the Variscides. We propose that continental subduction can redistribute a substantial volume of carbon in the continental lithosphere, which is subsequently transferred to the continental crust during postcollisional magmatism and stored for a time length longer than that of the modern carbon cycle.

Abstract We identified dmisteinbergite, the rare trigonal polymorph of CaAl2Si2O8, for the first time in high-grade metamorphic rocks. Dmisteinbergite occurs as a crystallization product of silicate melt inclusions (nanogranitoids) in garnet from three host rocks with different protoliths and re-equilibration conditions, i.e., from 1.0 to 4.5 GPa. Raman spectra and compositions of the dmisteinbergite here investigated are overall identical to those of previously characterized artificial and natural dmisteinbergite. In nanogranitoids, this phase coexists with other metastable polymorphs of feldspar (kumdykolite, kokchetavite) and SiO2 (quartz, cristobalite), recently interpreted as the result of undercooling, supersaturation and rapid crystallization of a silicate melt confined in a micrometric pore. Dmisteinbergite formation likely results from a similar process, and thus it should be regarded as a kinetically controlled phase. Moreover, the investigation of dmisteinbergite as well as of other metastable feldspar polymorphs offers new insights into the behavior of natural materials under non-equilibrium conditions.

Melt inclusions of very unusual nature occur in garnets of eclogites of the Granulitgebirge, Bohemian Massif. This is one of the first direct characterization of a preserved metasomatic melt responsible for the formation of eclogites enclosed in garnet peridotites. The inclusions are micrometric, from glassy to fully crystalized as nanogranitoids and randomly distributed in the garnet core. Nanogranitoids contain kumdykolite/albite, phlogopite, osumilite and kokchetavite with a variable amount of quartz, pyroxene, carbonate and rare white mica. The melt has a granitic composition rather than basaltic or tonalitic/trondhjemitic as would be expected from the partial melting of ultramafic or mafic rocks and it is as well hydrous and peraluminous. The trace elements composition is also unusual for melts in mantle rocks with elements typical of continental crust (Cs, Li, B, Pb and Rb) and subduction zone (Th and U). Similar signatures, i.e. continental crust and subduction, are visible also in the whole rock trace elements in the form of high amounts of LILE and U. The eclogite major elements composition is similar to a Ca- and Fe - rich mafic rock akin more to the crust than to the mantle.The peculiar melt composition and the lack of a clear residue of a melting reaction in the eclogites suggest that this melt is external, i.e. metasomatic. It infiltered the peridotites during subduction of the continental crust at mantle depth and aided the transformation of basic layers, already in the peridotite, to eclogite. In addition, similar trace elements patterns to the melt reported here can be found in the so-called durbachite -ultrapotassic melanosyenite present in the high-grade Variscan basement- and in the garnet peridotites and garnet pyroxenites of the T-7 borehole. In both case metasomatism was suggested but the agent was just inferred based on the geochemical signature. All these occurrences suggest that mantle contaminated by melts from deeply subducted continental crust is widespread beneath the Bohemian Massif.

The garnet in the ultra-high pressure (UHP) eclogites of the Erzgebirge (Bohemian Massif, Germany) trapped primary inclusions of metasomatic melt originated by the partial melting of the continental crust. The study of these inclusions alow us to estimate the contribution of the subducted continental crust to mantle metasomatism and deep carbon fluxes. The inclusions are randomly distributed in the inner part of the garnet, they are micrometric and occur as both polycrystalline, i.e. nanogranitoids, and glassy, often with a shrinkage bubble. Nanogranitoids consist of kumdykolite, quartz, kokchetavite, biotite, white mica, calcite and rare graphite. The inclusions share their microstructural position in the garnet with inclusions of polycrystalline quartz interpreted as quartz pseudomorph after coesite that indicate the entrapment at UHP conditions. The melt composition, measured on glassy inclusions and rehomogenized nanogranitoids, is granitic. The melt is also hydrous, slightly peraluminous and the trace element enrichments observed are consistent with an origin from the continental crust, testified by the high amount of incompatible elements such as Cs, Pb, Th, U, Li and B. Similar signatures were also reported elsewhere in the Bohemian Massif, e.g. in other metasomatic melts hosted in HP mantle eclogites, in metasomatized mantle rocks and in post-collisional ultrapotassic magmatic rocks, suggesting that mantle metasomatism from melts originated in the continental crust is widespread in the orogen. The melt H2O and CO2 contents were measured with the NanoSIMS. The CO2 values in particular were corrected reintegrating the vapor contained in the shrinkage bubble and are in average 19552 ± 772 ppm, the highest content of CO2 measured so far in crustal melt inclusions. The modelled endogenic carbon flux associated with the subduction of the continental crust of the Variscan Orogenic Cycle is 22 ± 8 Mt C yr-1. This flux within error is similar to the endogenic carbon fluxes in the serpentinized mantle (~ 14 Mt C yr-1) and to the exogenic fluxes in mid-oceanic ridges (~ 16 Mt C yr-1) and arc volcanoes (~ 24 Mt C yr-1). Hence, in collisional settings, deeply subducted continental crust carried a large amount of volatiles to the mantle and the lower crust. Due to the absence of post collisional arc volcanism, most of these volatiles remained trapped in the root of mountain belts. This long-term storage of the carbon in the orogen roots prevents ultimately the closure of the carbon cycle.

Inclusions of relic high pressure melts provide information on the fate of crustal rocks in the deep roots of orogens during collision and crustal thickening, including at extreme temperature conditions exceeding 1000°C. However, discoveries of high pressure melt inclusions are still a relative rarity among case studies of inclusions in metamorphic minerals. Here we present the results of experimental and microchemical investigations of nanogranitoids in garnets from the felsic granulites of the Central Maine Terrane (Connecticut, US). Their successful experimental re-homogenization at ~2 GPa confirms that they originally were trapped portions of deep melts and makes them the first direct evidence of high pressure during peak metamorphism and melting for these felsic granulites. The trapped melt has a hydrous, granitic, and peraluminous character typical of crustal melts from metapelites. This melt is higher in mafic components (FeO and MgO) than most of the nanogranitoids investigated previously, likely the result of the extreme melting temperatures – well above 1000°C. This is the first natural evidence of the positive correlation between temperature and mafic character of the melt, a trend previously supported only by experimental evidence. Moreover, it poses a severe caveat against the common assumption that partial melts from metasediments at depth are always leucogranitic in composition. NanoSIMS measurement on re-homogenized inclusions show significant amounts of CO2, Cl and F. Halogen abundance in the melt is considered to be a proxy for the presence of brines (strongly saline fluids) at depth. Brines are known to shift the melting temperatures of the system toward higher values, and may have been responsible for delaying melt production via biotite dehydration melting until these rocks reached extreme temperatures of more than 1000°C, rather than 800-850°C as commonly observed for these reactions.

Three half-meter-wide pelitic granulite interlayers are found within mafic, intermediate, and felsic granulites in Chicheng of the Trans-North China Orogen, they help to understand the assembly of the Eastern and Western blocks of the the North China Craton (NCC). Detailed petrological study shows that all have identical highpressure mineral assemblages of garnet + kyanite + K-feldspar + plagioclase + quartz + rutile. Inclusions of mesoperthite in garnet indicate the former presence of ternary feldspar at peak conditions, which, in the matrix, have recrystallized. Kyanite is commonly surrounded by a narrow corona or chain of small garnets, enveloped by a zone of plagioclase and an outermost rim of potassium feldspar. Phase equilibria modeling using THERMO-CALC yiels a peak metamorphic pressure of 1.6 GPa at 880 degrees C (M2a) and a peak temperature of 950 degrees C at 1.5 GPa (M2b). The ternary feldspar thermometer also indicates a peak temperature of 900-950 degrees C. The pelitic and mafic granulites exhibit the same peak metamorphic pressure, while the pelitic granulites show a 150 degrees C higher peak metamorphic temperature than the interlayered mafic granulites of 750-790 degrees C. Absence of sillimanite in the Chicheng pelitic granulite and orthopyroxene in the mafic granulite indicates a fast exhumation process of cooling, which is different from other high pressure rocks from the Trans-North China Orogen. Analysis of metamorphic monazite and zircon yields concordant SIMS U-Pb ages of ca. 1.91 Ga. This study suggests that the Chicheng HP granulite terrane was subjected to fast tectonic exhumation after the collision that occurred at ca. 1.91 Ga. The results provide a reliable metamorphic record for the final collisiion of the Paleoproterozoic TransNorth China Orogen in the NCC.

The garnet megacrysts of Gore Mountain (Adirondacks, US) are world-renown crystals due to their size, up to 1 m in historical record, which makes them the largest known garnets on the planet. We show here that they are also host to the first primary inclusions of trondhjemitic melt found in natural mafic rocks. The petrological and experimental investigation of the inclusions, coupled with phase equilibrium modelling, shows that this melt is the result of H2O-fluxed partial melting at T > 900 degrees C of a lower crustal gabbro. The compositional similarity between the trondhjemitic melt inclusions and tonalitic-trondhjemitic-granodioritic (TTGs) melts makes these inclusions a direct natural evidence that melting of mafic rocks generates TTG-like melts, and provides us with the possibility to clarify processes responsible for the formation of the early continental crust. These TTG embryos represent the trondhjemitic end-member of the melts whose emplacement at upper crustal levels, after being modified by mixing and crystallization-related processes, leads to the formation of the TTG terranes. Moreover, our study shows how the melt from H2O-fluxed melting of mafic lower crust has mismatched major and trace element signatures, previously interpreted as evidence of melting at very different pressures. This poses serious limitations to the established use of some chemical features to identify the geodynamic settings (e.g. subduction versus thickened crust) responsible for TTGs generation and the growth of early crust. (C) 2021 Elsevier B.V. All rights reserved.

The garnet megacrysts (Fig. 1a) of Gore Mountain (Adirondacks, US) are world-renown crystals due to their size, up to 1 m in historical record, which makes them the largest known garnets on the planet. We show here that they are also host to the first primary inclusions of trondhjemitic melt found in natural mafic rocks (Fig. 1b). The petrological and experimental investigation of the inclusions, coupled with phase equilibrium modelling, shows that this melt is the result of H 2 O-fluxed partial melting at T > 900 °C of a lower crustal gabbro. The compositional similarity between the trondhjemitic melt inclusions and tonalitic–trondhjemitic–granodioritic (TTGs) melts makes these inclusions the first direct natural evidence that melting of mafic rocks generates TTG-like melts, and provide us with the possibility to clarify processes responsible for the formation of the early continental crust (Fig. 1c). These TTG embryos represent the trondhjemitic end-member of the melts whose emplacement at upper crustal levels, after being modified by mixing and crystallization-related processes, leads to the formation of the TTG terranes. Our study also shows how the melt from H 2 O-fluxed melting of mafic lower crust has mismatched

Small portions of pristine melt with diameters of 2 to 50µm are increasingly recognized as a rather common occurrence in high grade metamorphic terranes which experienced melting. Their study delivers crucial chemical information on partial melts at depth. But they are also unique "natural experimental charges" where the behaviour of the silicate melt can be investigated, directly in the natural rocks, under P-T-t conditions which cannot be completely reproduced in the laboratory. Each nanogranitoid case study has consistently shown H2O-bearing, silica and alkali-rich melt. However, rather than a classic granitoid assemblage consisting mainly of quartz and feldspar(s), on cooling these isolated melt droplets produce a plethora of mineral phases identified via microRaman spectroscopy that are rarely –or never- observed as rock-forming minerals. Cristobalite (tetragonal) and tridymite (orthorhombic) are often present as SiO2 polymorphs, and hexagonal kokchetavite as a polymorph of KAlSi3O8. NaAlSi3O8 occurs as orthorhombic kumdykolite, whereas CaAl2Si2O8 may occur either as monoclinic svyatoslavite or trigonal dmisteinbergite. Two presently unidentified phases have been also recognized via Raman and analysed via electron microprobe. One has the main peak at 426-430 cm-1 and has the composition of a granitic glass, whereas the second has a main peak at 412 cm-1 and a variable composition depending on the inclusion in which it occurs. As their main peaks occur in the same region of most tectosilicates, it is likely that they are two new polymorphs of feldspar, to the best of our knowledge never reported before. These polymorphs have been so far identified in inclusions mainly hosted in garnet, zircon and, in one case, sapphirine and trapped under an extremely variable range of metamorphic conditions (from low P migmatites to UHP eclogites) in very different rock types (metagranitoids, metasediments, mafic and ultramafic rocks). Microstructures confirm that all of these phases crystallize directly from the trapped melt on cooling, independently of the internal P of the inclusions or the original conditions of melt entrapment. They appear to be the result of metastability in the inclusions, possibly during rapid crystallization of a melt, not caused by rapid cooling but by the peculiar undercooled and supersaturated conditions achieved on cooling by a melt confined in a small cavity (Ferrero & Angel, 2018). According to this possibility, these polymorphs can be regarded as kinetically stabilized, yet possibly thermodynamically metastable, phases as recently proposed by Zolotarev et al. (2019) for dmisteinbergite. A preliminary crystallization experiment on a haplogranitic melt at undercooled conditions however failed to reproduce such phases. Another possibility is that under natural cooling the confined inclusions experience underpressurization, and the system (i.e. the trapped melt) reacts crystallizing phases, i.e. the polymorphs, less dense than their common counterparts. This would result in the decreasing of the P gradient between inclusions and surrounding rock, equivalent to reducing the free energy of the system. References Ferrero, S. & Angel, R. 2018. JPet 59, 1671–1700. Zolotarev, A.A. et al. 2019. Minerals 9, 570.

Melt inclusions have been for almost 150 years an exclusive feature of magmatic rocks. However, intensive research activity in the last decade has shown that melt inclusions, or nanogranitoids, are also a widespread feature of high grade metamorphic rocks. Such inclusions rapidly became fundamental tools to unravel partial melting and melt-related processes taking place during orogenesis. One of the latest discoveries in this field has been the identification of nanogranitoids and glass inside the mega almandine-pyrope garnets of Barton Mine (Gore Mountain, NY State, US). These crystals are arguably the world’s largest garnets and occur within garnet hornblendite. Their size is ca. 35 cm in average, while garnet diameters up to 1 m were reported in historical record. Fluid is often invoked in the formation of large crystals, but so far no study has identified clear witnesses for the presence of fluid during garnet formation, e.g. primary fluid inclusions. Polycrystalline inclusions of primary nature were instead reported by Darling et al. (1997) to occur inside the garnet: such inclusions are the main target of our study. Their shape ranges from tubular (2-100 µm in length) to negative crystal shape (2-50 µm). They mainly contain cristobalite/quartz, kumdykolite and amphibole. Minor phases such as biotite/phlogopite, enstatite, rutile, ilmenite and a second, Ca-richer plagioclase (or its rare polymorphs dmisteinbergite and svyatoslavite) may be also present. The inclusions were re-homogenized to a silicate-rich glass via piston cylinder experiments at 1.0-1.5 GPa and 925-940°C. Experimental results prove that such inclusions are former droplets of melt, in agreement with the finding of preserved residual glass in one single inclusion before the experimental runs. The melt composition measured in situ via electron microprobe is tonalitic-trondhjemitic with 5-6 wt% H2O. The identification of melt inclusions points toward a melt rather than a fluid as the medium which favored extreme garnet growth under low nucleation rate conditions. The elements necessary to grow garnets – mainly Fe, Al, Si, Mg- are indeed far more effectively transported by a silicate melt rather than simple aqueous fluid, at least at the limited depth envisioned for this process. In conclusion, the finding of melt inclusions in metamorphic rocks brought us forward along the path toward the solution of the enigma represented by the formation of these giant garnets. References Darling, R.S., Chou, I.M., Bodnar, R.J., 1997. An Occurrence of Metastable Cristobalite in High-Pressure Garnet Granulite. Science 276, 91.

Abstract Himalayan high-pressure metamorphic rocks are restricted to three environments: the suture zone; close to the suture zone; and (mostly) far (>100 km) from the suture zone. In the NW Himalaya and South Tibet, Cretaceous-age blueschists (glaucophane-, lawsonite- or carpholite-bearing schists) formed in the accretionary wedge of the subducting Neo-Tethys. Microdiamond and associated phases from suture-zone ophiolites (Luobusa and Nidar) are, however, unrelated to Himalayan subduction–collision processes. Deeply subducted and rapidly exhumed Indian Plate basement and cover rocks directly adjacent to the suture zone enclose eclogites of Eocene age, some coesite-bearing (Kaghan/Neelum and Tso Morari), formed from Permian Panjal Trap, continental-type, basaltic magmatic rocks. Eclogites with a granulite-facies overprint, yielding Oligocene–Miocene ages, occur in the anatectic cordierite ± sillimanite-grade Indian Plate mostly significantly south of the suture zone (Kharta/Ama Drime/Arun, north Sikkim and NW Bhutan) but also directly at the suture zone at Namche Barwa. The sequence carpholite-, coesite-, kyanite- and cordierite-bearing rocks of these different units demonstrates the transition from oceanic subduction to continental collision via continental subduction. The granulitized eclogites in anatectic gneisses preserve evidence of former thick crust as in other wide hot orogens, such as the European Variscides.

Garnet of eclogite (formerly termed garnet clinopyroxenite) hosted in lenses of orogenic garnet peridotite from the Granulitgebirge, NW Bohemian Massif, contains unique inclusions of granitic melt, now either glassy or crystallized. Analysed glasses and re‐homogenized inclusions are hydrous, peraluminous, and enriched in highly incompatible elements characteristic of the continental crust such as Cs, Li, B, Pb, Rb, Th, and U. The original melt thus represents a pristine, chemically evolved metasomatic agent, which infiltrated the mantle via deep continental subduction during the Variscan orogeny. The bulk chemical composition of the studied eclogites is similar to that of Fe‐rich basalt and the enrichment in LILE and U suggest a subduction‐related component. All these geochemical features confirm metasomatism. In comparison with many other garnet+clinopyroxene‐bearing lenses in peridotites of the Bohemian Massif, the studied samples from Rubinberg and Klatschmühle are more akin to eclogite than pyroxenites, as reflected in high jadeite content in clinopyroxene, relatively low Mg, Cr, and Ni but relatively high Ti. However, trace elements of both bulk rock and individual mineral phases show also important differences making these samples rather unique. Metasomatism involving a melt requiring a trace element pattern very similar to the composition reported here has been suggested for the source region of rocks of the so‐called durbachite suite, that is, ultrapotassic melanosyenites, which are found throughout the high‐grade Variscan basement. Moreover, the Th, U, Pb, Nb, Ta, and Ti patterns of these newly studied melt inclusions (MI) strongly resemble those observed for peridotite and its enclosed pyroxenite from the T‐7 borehole (Staré, České Středhoři Mountains) in N Bohemia. This suggests that a similar kind of crustal‐derived melt also occurred here. This study of granitic MI in eclogites from peridotites has provided the first direct characterization of a preserved metasomatic melt, possibly responsible for the metasomatism of several parts of the mantle in the Variscides.

Abstract The central European Bohemian Massif has undergone over two centuries of scientific investigation which has made it a pivotal area for the development and testing of modern geological theories. The discovery of melt inclusions in high-grade rocks, either crystallized as nanogranitoids or as glassy inclusions, prompted the re-evaluation of the area with an ‘inclusionist’ eye. Melt inclusions have been identified in a wide range of rocks, including felsic/perpotassic granulites, migmatites, eclogites and garnet clinopyroxenites, all the result of melting events albeit over a wide range of pressure/temperature conditions (800–1000°C/0.5–5 GPa). This contribution provides an overview of such inclusions and discusses the qualitative and quantitative constraints they provide for melting processes, and the nature of melts and fluids involved in these processes. In particular, data on trace-element signatures of melt inclusions trapped at mantle depths are presented and discussed. Moreover, experimental re-homogenization of nanogranitoids provided microstructural criteria allowing assessment of the conditions at which melt and host are mutually stable during melting. Overall this work aims to provide guidelines and suggestions for petrologists wishing to explore the fascinating field of melt inclusions in metamorphic terranes worldwide, based on the newest discoveries from the still-enigmatic Bohemian Massif.