Search

Amphibole exerts a fundamental control on arc magma petrogenesis, differentiation, and the long-term evolution of the arc crust. This study identifies two texturally distinct amphibole populations within andesitic lavas and entrained hornblendite cumulates at the Quillacas monogenetic volcanic center in the Eastern Altiplano, Bolivia. Within the hornblendites, all amphiboles are tschermakitic, large (≤800 μm) with thick, granular reaction rims (avg. 27 μm thickness). In the host andesites, tschermakites are also the dominant amphibole species but are smaller (250–400 μm) with thin, symplectic reaction rims (avg. 7–9 μm thickness). An intergrowth of symplectic and granular reaction rims is also observed in this population. The amphibole populations within the Quillacas magmatic system also record irregular volumetric decomposition where amphibole is replaced by mineral aggregates of plagioclase, pyroxene, and oxide within the crystal. This suggests the occurrence of a relatively slow reaction between the amphibole and melt trapped in fractures and cleavages during decompression-induced degassing. Geothermobarometry indicates that the hornblendite cumulate tschermakites crystallized at P-T conditions ranging from 467 to 598 ± 12 % MPa and 945–991 ± 22 °C. The host andesite tschermakites crystallized at P-T conditions ranging from 448 to 570 ± 12 % MPa and 928–1004 ± 22 °C. These geothermobarometric constraints correspond to depths of 16–24 km, which, within this region of the Central Andean crust, coincides with a regionally extensive low-seismic velocity zone. The texturally distinct amphibole populations imply that a multi-stage trans-crustal magmatic system is likely present beneath the Quillacas volcanic center. In this scenario, a crystal mush zone exists at upper crustal depths where the hornblendite cumulate tschermakites initially crystallized. Magma recharge into this mush zone initiated a reaction between hornblendite cumulates and the melt which formed the amphibole granular rims. This recharge event also transported the host andesite tschermakites that subsequently developed symplectic rims due to heating and ascent-driven decompression. This study supports the presence of amphibole-dominated mush filters in the upper crust of the Central Andean arc and advances our understanding of amphibole's role in the evolution of arc magmatic systems. • Distinct amphibole populations identified in arc andesites and hornblendites. • Amphibole P-T conditions consistent with low seismic velocity zone in Central Andes. • Amphibole breakdown textures indicate complex multi-stage magmatic processes. • Amphibole-dominated mush filters likely exist in Central Andean arc crust. [ABSTRACT FROM AUTHOR]

Amphibole is thought to be a key control on the differentiation of arc magmas but the fractionation of this mineral remains enigmatic because it is rarely present in erupted lavas. This study presents detailed petrological, mineralogical, and geochronological data for newly identified Late Triassic (217–215 Ma) mafic intrusive rocks from the Gangdese arc of southern Tibet. These gabbros host an amphibole-rich cumulate enclave and clinopyroxene crystal clots, including two mineral assemblages in chemical disequilibrium: (1) amphibole and Type-1 clinopyroxene within the enclave and clots; and (2) small crystals of Type-2 clinopyroxene, plagioclase, and biotite within the host. Textural and geochemical evidence indicate that the enclave and clots represent captured fragments of amphibole- and clinopyroxene-dominated lower crustal crystal mushes, respectively, whereas small host crystals formed by the shallow crystallization of evolved melts extracted from these crystal mushes. The compositional variations between the two types of clinopyroxene also reveal this melt segregation process. Amphibole with similar rare earth element patterns to associated Type-1 clinopyroxene inclusions suggest that the former formed through peritectic reactions consuming clinopyroxene. Enhanced amphibole fractionation may also have occurred as a result of the presence of high-density clinopyroxene inclusions. This indicates that the fractionation of amphibole together with precursory clinopyroxene inclusions within the deep crust controls the early chemical evolution of mafic arc magmas, a process that is consistent with the geochemical variations in lavas from three other active arc volcanoes. The associated cumulates can efficiently delaminate back into the mantle, ultimately leading to the generation of SiO 2 -rich continental crustal material. • Late Triassic mafic rocks represent early products of Indus–Yarlung Zangbo Tethyan subduction. • The amphibole-rich enclave and clinopyroxene crystal clots were derived by crystal fractionation. • The fractionation of amphibole and clinopyroxene precursors occurs in the deep arc crust. • This fractionation controls the early chemical evolution of arc magmas. [ABSTRACT FROM AUTHOR]

We report here for the first time aluminous Ca-amphibole exsolution lamellae in an anhydrous chromite host of podiform chromitite from the Nakhl massif of the Semail Ophiolite. Transmission electron microscopy observations show that the Ca-amphibole lamellae have a crystallographic topotaxy and d -spacing relationship with the host chromite, i.e., [103] am //[11 8 _ ] chr , (020) am //(2 2 _ 0) chr and d 060am ≈ d 220chr , that indicate an exsolution origin. The presence of Ca-amphibole exsolution implies the incorporation of Si4+ and Ca2+ cations and hydroxyl in the host precursor chromite. The reconstructed precursor chromite had SiO 2 concentrations of 0.33–0.58 wt%, suggesting upwelling from a minimum pressure of 8 GPa (depth of ~250 km). Hydroxyl groups occupy octahedral vacancies, which can be introduced in high-pressure chromite by substituting 4 Cr3+ by 3 Si4+ and a vacancy. During upwelling, vacancies decrease when Si4+ cations exsolve stepwise from chromite. It leads to a coupled decrease of OH occupied vacancies and exsolution of Ca-amphibole lamellae at pressures below 3 GPa. Our results provide new evidence for a deep origin of the chromitite-bearing mantle section of the Semail ophiolite. • Aluminous Ca-amphibole exsolution lamellae are firstly found in chromite from Semail ophiolite. • The reconstructed precursor chromite indicates Oman podiform chromitite upwelling from a depth > 250 km. • Exsolved Ca-amphibole indicates OH can incorporate in chromite by Si4+ 3 / 1 Cr3+ −4 substitution in M site at HP condition. • Oman podiform chromitite from Semail ophiolite has a deep origin. [ABSTRACT FROM AUTHOR]

The Edolo diabase dykes intruded the Variscan basement in the central Southalpine domain. Biotite 40Ar / 39Ar dating shows that the emplacement occurred during the final stage of Permian-Triassic lithospheric thinning. In this phase on the Adria passive margin, middle Triassic magmatism with disputed arc affinity is followed by a late Triassic alkaline event associated with continental rifting. The Edolo diabase dykes are Ti-rich and alkaline, with high Nb and Ta contents and ratios in igneous amphibole, biotite, and ilmenite. These features are typical of the late Triassic magmatism. In contrast, the early-crystallized clinopyroxene shows low Nb and Ta contents and ratios and an apparent arc affinity may be thus envisaged in the clinopyroxene parental melt. However, given the coexistence of both features in the same dykes, their composition cannot be interpreted in terms of transition between tectonic environments. Instead, it is likely attributed to the heterogeneity of the Adriatic lithosphere inherited from the Variscan to Permian evolution, which underwent asthenospheric upwelling due to Permian-Triassic extensional tectonics. • Alkaline mafic dyke intruded the Adria passive margin in the late Triassic. • Clinopyroxene and amphibole are not in textural and compositional equilibrium. • The clinoyroxene parental melt reveals an arc-signature. • The amphibole parental melt is alkaline and Nb Ta rich. • The dykes were derived from chemically heterogeneous source rocks. [ABSTRACT FROM AUTHOR]

Abstract Natural occurrences of Cl-bearing amphibole indicate crystallization in the presence of a highly saline fluid. Amphibole with varying Cl contents (0.1 to 5 wt% Cl) found in lower oceanic crustal rocks demonstrates the activity of saline hydrothermal fluids at depth. Here we present an experimental study done in cold seal pressure vessels (CSPV) and internally heated pressure vessels (IHPV) to illustrate the process by which gabbro-hosted amphibole-rich parageneses evolve in the presence of a hydrothermal saline fluid. The starting material used was a natural gabbroic rock containing high-Ti magnesio-hastingsite with the addition of a moderately to highly saline fluid (NaCl-H 2 O) with X NaCl = 0.02, 0.07 and 0.24 (6, 20 and 50 wt% NaCl). We evaluated a range of conditions from hydrothermal (500–750 °C) to magmatic (900 °C) at pressures of 200 MPa and f O 2 close to NNO. Fluid/rock mass ratios used were 0.2 and 1 in subsolidus (hydrothermal) experiments, and 0.07 in partial melting (magmatic) experiments. New amphibole was formed on rims of the starting high-Ti magnesio-hastingsite, with product compositions corresponding to magnesio-hastingsite, high-Si magnesio-hornblende, tschermakite, edenite, hastingsite and ferro-pargasite with varying Cl contents up to 0.47 wt% Cl. Our results from subsolidus experiments demonstrate a decrease in olivine, plagioclase and clinopyroxene in the starting rock, and formation of new amphibole, with decreased IVAl and Ti with respect to starting amphibole. Product amphibole does not display any correlation between Cl and Fe2+, IVAl and K, in contrast to natural highly Cl-rich amphibole, suggesting that these correlations hold true only at Cl contents higher than those attained in our experiments. In addition, we found that increasing NaCl in the fluid correlates with increased maximum Cl contents in amphibole. Compositional variations found in product amphibole highlight the heterogeneities in fluid infiltration and Cl activity that account for the complexity of hydrothermal fluid/rock interactions in deep oceanic geological systems. Highlights • Seawater-derived fluids interact with gabbro in the lower oceanic crust. • Gabbro/hydrothermal fluid interaction leads to a wide range of amphibole compositions. • Hydrothermal fluid composition and water/rock ratio affect Cl content of amphibole. • No structural control on Cl incorporation in amphibole when Cl <0.5 wt%. • Experimental amphiboles formed from 500 to 900 °C overlap widely with those from nature. [ABSTRACT FROM AUTHOR]

Abstract Hydrothermal veins and dykelets that cross-cut layered olivine gabbros deep in the plutonic section of the Samail Ophiolite, Sultanate of Oman, point towards the occurrence of hydrothermal circulation in the deep oceanic crust, and these features record interactions between rock and high temperature seawater-derived fluids or brines. Deep penetration of seawater-derived fluids down to 100 m above the Moho transition zone and the consequent interactions with the host rock lead to hydrothermal alteration from granulite facies grading down to greenschist facies conditions. Here we present a study of veins and dykelets formed by hydrothermal interaction cutting layered gabbro in the Wadi Wariyah, using petrographic, microanalytical, isotopic, and structural methods. We focus on amphiboles, which show a conspicuous compositional variation from high-Ti magnesiohastingsite and pargasite via magnesiohornblende and edenite, to Cl-rich ferropargasite and hastingsite (up to 1.5 a.p.f.u. Cl) and actinolite. These minerals record a wide range of formation conditions from magmatic to hydrothermal, and reveal a complex history of interactions between rock and hydrothermal fluid or brine in a lower oceanic crustal setting. Large variations in Cl content and cation configurations in amphibole suggest formation in equilibrium with fluids of different salinities at variable fluid/rock ratios. The presence of subsolidus amphibole extremely enriched in chlorine implies phase separation and brine/rock interactions. 87Sr/86Sr values of 0.7031 to 0.7039 and stable δ18O isotopic compositions of 4.1 to 5.6‰ of the different amphibole types suggest a rock-dominated environment, i.e. with low fluid/rock ratios. However, the slight departure from mean Oman isotope values may indicate there was some influence of seawater in the aforementioned fluid-rock interactions. Our study provides new petrological data for the subsolidus evolution of gabbro-hosted amphibole-rich veins in the presence of a seawater-derived fluid. Highlights • Cl-rich amphibole (<1.5 a.p.f.u. Cl) indicates equilibrium with Cl-rich fluid. • Interaction of amphibole with Cl-bearing fluid occurs in rock-dominated environment. • Fluid evolution and changing temperatures recorded in variations in amphibole composition. • Veins and dykelets in lower oceanic crust as signs of hydrothermal alteration. [ABSTRACT FROM AUTHOR]

Abstract Eclogite lenses and boudins are volumetrically minor, but petrologically important, features of peridotite massifs worldwide. In the Western Gneiss Region of the Scandinavian Caledonides, eclogites in the Almklovdalen and Raubergvik peridotites originated as basaltic to picrobasaltic dikes, comprising both olivine–normative and nepheline–normative types, with a wide variation in Mg–number from 34 to 65. Positive anomalies for Pb and Sr and negative anomalies for Zr and Hf reflect a subduction signature in the basic melts, and rare–earth element modelling requires 20% to 70% fractional crystallization, combined with 20% to 70% assimilation of peridotite. Clinopyroxenes in eclogites have a wide variation in ε Nd (0) from +68 to −26, which is comparable to that for associated garnet peridotites and pyroxenites, +55 to −38, and a range in 87Sr/86Sr from 0.7021 to 0.7099, which is much larger than that in peridotites and pyroxenites, 0.7014 to 0.7033. Plagioclase and amphibole inclusions in eclogite garnet provide evidence for prograde metamorphism, which attained a maximum temperature of ~775 °C and pressure of ~25 kb. Such conditions are allofacial with those of associated garnet peridotites and pyroxenites, which equilibrated at ~825 °C and ~37 kb. Eclogites yield mixed Sm-Nd isochron ages, as do the peridotites and pyroxenites, but ages in eclogites are <1000 Ma, and those in peridotites and pyroxenites are >1000 Ma. Three eclogites yield Ordovician U-Pb ages for rutile at 440 ± 12, 445 ± 51, and 480 ± 29 Ma, which are coeval with the Taconic Orogeny and are consistent with a Laurentian provenance for the host peridotites. Eclogites in both Norwegian and Czech peridotites originated from melts passing through a mantle wedge above a subduction zone, and both suites exhibit subduction geochemical signatures, although they differ dramatically in petrogenesis. Eclogites in Norwegian peridotites initially crystallized as relatively low–pressure, plagioclase–bearing basaltic or gabbroic dikes and subsequently recrystallized to high–pressure eclogite, whereas most eclogites in Variscan Moldanubian peridotites crystallized directly from magmas at high pressure to produce eclogite facies assemblages. Highlights • Eclogite layers in Norwegian garnet peridotites originated as basalt or gabbro dikes • AFC modelling yields 20–70% crystallization and 20–70% assimilation of peridotite • The dikes recrystallized to eclogite at 775 °C and 25 kb during the Taconic Orogeny • Eclogite in Czech peridotites originated by HP crystallization of garnet and pyroxene [ABSTRACT FROM AUTHOR]

Early Cretaceous northwest (NW)-trending dolerite and amphibole lamprophyre dykes exposed in NW Zhejiang Province provide a number of new insights into the nature of the subcontinental mantle on the northeast (NE) South China Block (SCB). These dykes have a high Al 2 O 3 (14.04–17.89 wt%) and K 2 O (0.66–2.69 wt%) contents but relatively low Na 2 O (2.48–4.61 wt%) and TiO 2 (1.33–2.79 wt%) makeup alongside moderate K 2 O/Na 2 O ratios between 0.26 and 1.04. These amphibole lamprophyre dykes also have higher MgO, Cr, and Ni contents than those of comparable dolerites that have SiO 2 content ranging from 46.32 to 49.87 wt%. The most striking feature of these intrusions is that they contain higher contents of Rb, Th, U, Nb, Ta, and LREE compared to their dolerite counterparts, although both amphibole lamprophyres and dolerites do exhibit similar geochemical patterns that are indicative of subduction-related origins. These features imply that an ambient peridotitic mantle that acted as the source for the amphibole lamprophyre magma source may have reacted with silicate-rich melts leading to olivine consumption while maintaining orthopyroxene. The geochemical composition of these dolerites are likely influenced to a variable extent by the fractionation of olivine, orthopyroxene, clinopyroxene, Fe–Ti oxides, and apatite, while their amphibole lamprophyre counterparts have been modified to a minor degree by amphibole fractionation. Measured Sr–Nd isotopic compositions suggest relatively constant Nd isotopic compositions (−0.36 to +1.52) with more variable Sr isotopic compositions (0.7071 to 0.7306). We hence propose that both the dolerite and amphibole lamprophyre dykes in this region are the products of mantle source metasomatism by the subducted Paleo-Pacific slab. The dolerite dykes are mainly associated with slab-derived fluids, while the lamprophyre dykes are related to both slab-derived fluids and sediment melts. Evidence in support of metasomatism comprising distinct two-stage processes including a fluid-dominated phase followed by a sediment melts-dominated metasomatism stage, further suggests that these mafic dykes most likely formed in a back-arc setting. [ABSTRACT FROM AUTHOR]

The Charco Rico prospective comprises a segment of the Sierra de Majé massif in Central Panama, an intrusive complex constructed during a lull in volcanic activity between 40 and 16 Ma at the 75–0 Ma Central American Volcanic arc system (CAVAS). Charco Rico represents an intermediate suite (55–65 wt% SiO 2) but with inclusion of related Majé intrusives, range from gabbro to granodiorite. Zircon U Pb dating of Charco Rico granitoids yield LA-ICP-MS ages of 20.2 ± 0.9 to 19.1 ± 1.5 Ma. Their formation is consistent with maximum partial melting of ∼10% of a Sr-enriched (i.e., amphibolite) N-MORB like source followed by fractional crystallization of hornblende at about 0.8 GPa (∼29 km depth). Zirconium saturation thermometry shows a temperature of magma crystallization of 700–800 °C which essentially is a record of amphibole crystallization temperature. Whole rock chemistry of the Charco Rico and Majé intrusives and other 32–20 Ma arc-related suites in Panama record chemical parameters similar to adakite, e.g., Sr/Y that ranges to 80, high LREE/HREE ratios, and depletion in HREE (Yb, Lu). A fundamental chemical attribute of these suites, however, is depletion of MREE, and particularly Ho, relative to LREE and HREE, and the absence of Eu-anomalies. These features, together with positive and negative correlations of La/Yb and Dy/Yb with SiO 2 , respectively, infer a dominant role of amphibole fractionation from basaltic magma and/or residue during partial melting of older calc-alkaline bodies in the lower crust. 176Hf/177Hf ratios in zircon grains range from 0.282925 to 0.283192 and record εHf t between +5.8 and + 15.3, which are indicative of a depleted mantle (juvenile) source and/or recycling (partial melting) of older juvenile crust. T DM incubation ages range to higher than 300 Ma, which may reflect old basement recycling. Partial melting of the depleted mantle ± lower crust may have been facilitated by a combination of hot asthenospheric upwelling during slab rollback, slab tear or lithospheric delamination, and simultaneous thinning of the continental crust under extension. Whole rock chemistry and reported cumulate hornblende in host granites with adakitic-like signatures, infer that hornblende differentiation controlled the magma chemistry and migration to high-SiO 2 rocks with 'adakite' signatures. [Display omitted] • Charco Rico was built during a lull in volcanic activity at 40–16 Ma at the CAVAS. • Intermediate intrusive bodies yield zircon ages of 20.2 ± 0.9 to 19.1 ± 1.5 Ma. • Charco Rico and similar suites infer a dominant role of amphibole fractionation. • Zircon grains record εHft between +5.8 and + 15.3. • Old TDM incubation age ranges may reflect ancient basement recycling. [ABSTRACT FROM AUTHOR]

In this paper, U-Pb zircon geochronology, Lu-Hf and Sm-Nd isotope systematics, geochemistry and geothermobarometry of metaigneous basement rocks exposed in the southeastern Chiapas Massif Complex are presented. Geologic mapping of the newly defined “El Triunfo Complex” located at the southeastern edge of the Chiapas Massif reveals (1) partial melting of a metamorphic basement mainly constituted by mafic metaigneous rocks (Candelaria unit), (2) an Ediacaran metasedimentary sequence (Jocote unit), and (3) occurrence of massif-type anorthosite. All these units are intruded by undeformed Ordovician plutonic rocks of the Motozintla suite. Pressure and temperature estimates using Ca-amphiboles, plagioclase and phengite revealed prograde metamorphism that reached peak conditions at ~ 650 °C and ~ 6 kbar, sufficient for partial melting under water saturated conditions. Relict rutile in titanite and clinopyroxene in amphibolite further indicate a previous metamorphic event at higher P-T conditions. U-Pb zircon ages from felsic orthogneiss boudins hosted in deformed amphibolite and migmatite yield crystallization ages of ~ 1.0 Ga, indicating that dry granitic protoliths represent remnants of Rodinia-type basement. Additionally, a mid-Tonian (~ 920 Ma) metamorphic overprint is suggested by recrystallized zircon from a banded gneiss. Zircon from folded amphibolite samples yield mainly Ordovician ages ranging from ~ 457 to ~ 444 Ma that are indistinguishable from the age of the undeformed Motozintla plutonic suite. Similar ages between igneous- and metamorphic- zircon suggest a coeval formation during a high-grade metamorphic event, in which textural discrepancies are explained in terms of differing zircon formation mechanisms such as sub-solidus recrystallization and precipitation from anatectic melts. In addition, some amphibolite samples contain inherited zircon yielding Stenian-Tonian ages around 1.0 Ga. Lu-Hf and Sm-Nd isotopes and geochemical data indicate that the protoliths of the amphibolite have E-MORB characteristics and were derived from a depleted mantle source younger than the Rodinia-type basement. Inasmuch as similar amphibolites also occur in the Ediacaran metasedimentary rocks as dykes or lenses, Late Neoproterozoic magmatism in a rift setting is suggested. Hence, the geologic record of the El Triunfo Complex includes evidences for Rodinia assemblage, Tonian circum-Rodinia subduction, and breakup during the Late Neoproterozoic. Metamorphism, and partial melting are interpreted in terms of a convergent margin setting during the Ordovician. The results place the southern Chiapas Massif along with Oaxaquia and similar Northern Andes terranes on the NW margin of Gondwana interpreted as the extension of the Famatinian orogen that evolved during the closure of the Iapetus Ocean. [ABSTRACT FROM AUTHOR]

A careful review of petrological and geochemical data on the Paranà-Etendeka igneous province is reported, with particular attention being devoted to the relationships between high-MgO CFB (tholeiitic basalts-picrites) and nearly coeval alkaline-carbonatite complexes linked to the same extensional tectonics on a regional scale. At 135–130 Ma, the tectonomagmatic activity was focused in Etendeka, the centre of the restored province, and characterised by an exclusive occurrence of the hottest and deepest high-MgO CFB (potential temperature T p up to 1590 °C and pressure up to 5 GPa) possessing the same Sr-Nd-Pb isotopic composition of the “Gough” geochemical component, a marker of the initial Tristan plume activity. Etendeka high-MgO CFB thus represent the most genuine proxies of sublithospheric melts generated at the plume axis and are relatively unaffected by lithospheric contamination. Nearly coeval (133–128 Ma) alkaline‑carbonatite complexes cluster around the extensional structures of the Ponta Grossa Arch (e.g., Jacupiranga and Juquia in Brazil) and the Damara Belt (e.g., Erongo, Okurusu, Okenyenya and Paresis in Namibia), both of which intersect the early track of the south Atlantic opening. Compared to high-MgO CFB, alkaline magmas display distinctive isotopic signatures and an incompatible element distribution consistent with their generation from lithospheric mantle sources, which were variably metasomatised (veined?) by amphibole and phlogopite. Metasomes of alkaline mantle sources have a HIMU affinity and are dominated by amphibole in Namibia, whereas they display EM1 tendency and a more relevant role of phlogopite in Brazil, which implies important lithospheric differences at a regional scale. The tectonomagmatic features of Paranà-Etendeka –also shared by other Gondwana LIPs, such as Deccan and Karoo– can be reconciled by a generalized model where a hot plume impinging on a relatively thick lithosphere caused, in the axial zone, the contemporaneous generation and rise of high-MgO CFB and alkaline magmas from distinct asthenospheric and lithospheric mantle sources, respectively. In the asthenosphere, the volatile-poor solidus was crossed under an adiabatic thermal regime, mostly in the range of 4 to 5 GPa and T p 1500 to 1600 °C with the development of high-MgO CFB. In the overlying metasomatised lithosphere, the plume effects caused a perturbation of the conductive thermal regime and a crossing of volatile-rich solidus (mostly P 2–3 GPa, T p 1300–1400 °C) with the generation of alkaline melts from the most fusible (hydrated and carbonated) mantle domains. [ABSTRACT FROM AUTHOR]

Major and trace element compositions of amphibole in igneous environments commonly reflect evolving magma compositions. In this study, we use the amphibole-group minerals from the Strange Lake, REE-enriched peralkaline granitic pluton to gain insights into the evolution of the magma. This 1240 Ma old pluton consists of two main intrusive facies, an early hypersolvus granite, which occurs as separate northern and southern intrusions, and a more evolved transsolvus granite. In the hypersolvus granite the amphibole is a late interstitial phase, whereas in the transsolvus granite, it is present as phenocrysts. The amphibole compositions vary from calcic–sodic (ferro–ferri-katophorite) in the southern hypersolvus granite to sodic (arfvedsonite, ferro–ferri-leakeite) in the other, more evolved granitic units. High Na, Si, Li, and low Al and Ca concentrations in the amphibole phenocrysts of the transsolvus granite indicate formation from a more evolved magma compared to the hypersolvus granite, despite the fact that these crystals formed early. We interpret the increasing Fe 3+ /Fe 2+ ratios in the amphibole of the hypersolvus granite to reflect crystal chemical effects (Na/Ca-ratio) and increasingly oxidizing conditions in the magma, whereas in the phenocrysts of the transsolvus granite, the increasing ratio was the product of increasing proportions of F − and OH − in the melt. The amphiboles of all the granite units have elevated Nb, Zr, Hf and REE concentrations compared to the bulk rock, suggesting that these elements are compatible in amphibole. By contrast the much lower Ti concentration was due to saturation of the magma in sodium-titanosilicates. The amphibole REE concentrations vary greatly among the granite units. Amphibole of the southern and northern hypersolvus granite contains 0.16 and 0.07 wt.% ∑ REE + Y, on average, respectively, and in the transsolvus granite, the average ∑ REE content is only 0.01 wt.%, despite the more evolved nature of its host transsolvus granite. We intrepret this compositional difference to be due to the fact that the latter represents phenocrysts, which crystallized early, whereas the hypersolvus arfvedsonite is a late interstitial phase. Chondrite-normalized REE profiles emphasise the wide range in LREE-, and the narrow range in HREE-concentrations of the amphiboles. The variations in the LREE-profiles reflect the variable crystallization of primary LREE-bearing phases, including monazite-(Ce), pyrochlore group minerals and gagarinite-(Ce), prior to or contemporaneous with the amphibole, as well as the exsolution of a LREE-rich fluoride melt. The LREE are incompatible in the amphibole structure (apparent D < 0.01) and are preferably accommodated by the octahedral C-site, whereas the HREE occupy the B-site. The chondrite-normalized HREE profiles are steep and display an increasing relative enrichment that culminates in compatible behavior for Yb and Lu (apparent D > 1). Owing to their small ionic radius and their compatibility with the amphibole structure, HREE concentrations were more controlled by partitioning (crystal chemical effects) than by the concentrations in the corresponding magma. Large proportions of the bulk HREE content (up to 70%) reside in the amphibole, and their later release through hydrothermal replacement helps to explain the extreme and unusual HREE enrichment of the Strange Lake pluton. [ABSTRACT FROM AUTHOR]

Mantle xenoliths in the 20 Ma Wilcza Góra basanite (Lower Silesia, NE Bohemian Massif) are mostly harzburgites, some with amphibole which is exceptional in the region. Forsterite content in olivine defines two Groups of peridotites: Group A (Fo 89.1–91.5 ) and Group B (Fo 84.2–89.2 ). Hornblende-clinopyroxenite, websterite and one composite xenolith consisting of dunite, olivine-hornblendite and pyroxene-hornblende-peridotite contain olivine with Fo 77.3–82.5 and are classified as Group C. Group A xenoliths contain Al-poor orthopyroxene and some contain LREE-enriched clinopyroxene with negative Ti, Zr–Hf and Nb–Ta anomalies. Spinel (Cr# 0.57–0.68) is scarce in Group A, and Cr-rich pargasite occurs in only two xenoliths. Group B xenoliths contain less magnesian orthopyroxene and clinopyroxene. The REE patterns of Group B clinopyroxene are convex downward, less enriched in LREE and have smaller negative Ti, Zr–Hf and Nb–Ta anomalies than those in Group A. The Cr# in Group B spinel is 0.26–0.56, while pargasite is Ti-rich and Cr-poor. Clinopyroxene from Group C is low magnesian, slightly enriched in LREE and has no negative Ti, Zr–Hf and Nb–Ta anomalies. Group C pargasite is rich in Ti and poor in Cr. Equilibration temperatures recorded in all groups vary within the range of 905–970 °C. Xenoliths from Wilcza Góra record a polyphase lithospheric mantle evolution, starting with melting which extracted ca. 30% melt from the protolith and left a harzburgite residuum depleted in Al, lacking clinopyroxene and containing rare Cr-rich spinel. This residuum was later overprinted by chromatographic metasomatism by carbonated hydrous silicate melt related to Cenozoic volcanism. The metasomatic agent was locally hydrous enough to enable amphibole to crystallize. The Group C pyroxenites formed directly from the metasomatic melt or during peridotite–melt reactions at high melt–rock ratio. The melt is inferred to have percolated through the wall-rock peridotite, decreasing its amount and changing composition (and that of the crystallizing clinopyroxene and amphibole) as it differentiated chromatographically from Groups B to A. Enrichment in carbonatite component occurred in the further parts of a chromatographic column. Group B peridotites closest to the source of the metasomatic agent were percolated by an iron-rich melt. The latter, aside from crystallizing clinopyroxene and amphibole, caused Fe-enrichment in the host harzburgite. Such a metasomatic history is typical for the lithospheric mantle located beneath the northern margin of the Bohemian Massif, but Wilcza Góra is the only locality, where activity of a hydrous metasomatic agent is recorded. Thus, the lithospheric mantle in this area was affected by mafic silicate metasomatic agents of variable compositions. [ABSTRACT FROM AUTHOR]

We present new trace element compositions of amphiboles, Sr–Nd–Hf isotope compositions of clinopyroxenes and mineral modes for spinel peridotite xenoliths that were entrained in a Miocene alkali basalt (Gundeweyn, northwestern Ethiopian plateau), in order to understand the geochemical evolution and variation occurring within the continental lithospheric mantle (CLM) in close proximity to the East African Rift system, and its dynamic implications. With the exception of a single amphibole-bearing sample that is depleted in LREE (La/Yb N = 0.45 × Cl), amphiboles in lherzolites and in one harzburgite show variable degrees of LREE enrichment (La/Yb N = 2.5–12.1 × Cl) with flat HREE (Dy/Yb N = 1.5–2.1 × Cl). Lherzolitic clinoyroxenes have 87 Sr/ 86 Sr (0.70227 to 0.70357), 143 Nd/ 144 Nd (0.51285 to 0.51346), and 176 Hf/ 177 Hf (0.28297 to 0.28360) ranging between depleted lithosphere and enriched mantle. LREE-enriched clinopyroxenes generally have more enriched isotope compositions than depleted ones. While lherzolites with isotope compositions similar to those of the Afar plume result from the most recent metasomatic overprint, isotope compositions more depleted than present-day MORB can be explained by an older melt extraction and/or isotopic rehomogenisation event, possibly related to the Pan-African orogeny. Several generations of amphibole are recognized in accord with this multi-stage evolution. Texturally unequilibrated amphibole occurring within the peridotite matrix and in melt pockets attest to continued hydration and refertilization of the lithospheric mantle subsequent to Oligocene flood basalt magmatism, during which an earlier-emplaced inventory of amphibole was likely largely consumed. However, a single harzburgite contains amphibole with the highest Mg# and lowest TiO 2 content, which is interpreted as sampling a volumetrically subordinate mantle region beneath the Ethiopian plateau that was not tapped during flood basalt magmatism. Strikingly, both trace-element enriched and depleted lherzolites have high clinopyroxene and orthopyroxene and low olivine contents (median 15, 24 and 56 vol.%), combined with primitive olivine Mg# (median 89.5), indicating the presence of refertilized mantle beneath Gundeweyn. Despite its fertility and FeO-rich character (hence high inferred density), and impingement by the Afar plume, the CLM beneath the Ethiopian plateau, though apparently thinned through thermochemical erosion, has so far resisted whole-sale delamination or dripping. This is tentatively ascribed to insufficient stress and density contrasts at the periphery of the Afar plume, which reached its greatest thermochemical buoyancy in the Afar region, northeast of Gundeweyn. [ABSTRACT FROM AUTHOR]

Calcic amphibole phenocrysts in the Tsurumidake summit (TS) lava, which was produced during the most recent eruption at Tsurumi Volcano (SW Japan) at around 7.5–10.5 ka, have been analyzed to determine the pre-eruptive conditions, such as temperature ( T ), pressure ( P ), oxygen fugacity ( f O 2 ), SiO 2 content (SiO 2 melt ), and FeO*/MgO ratio (FeO*/MgO melt ), of coexisting silicate melts in the magma reservoir beneath the volcano. Although most of the amphibole phenocrysts have been completely decomposed to a fine-grained opaque symplectite, ~ 6% of the grains remain intact. The degree of amphibole breakdown (DAB), defined as the ratio of the area of symplectite to the area of symplectite plus relict amphibole in each phenocryst, varies from ~ 20% to 100%. Compositional zoning was not observed in the amphibole grains, however, two distinct groups of amphibole phenocrysts have been identified based on their chemical compositions: group-I amphiboles, which are relatively Si-poor, and [6] Al-rich, and have a relatively high Mg# [100Mg/(Mg + Fe 2 + )]; and group-II amphiboles, which are Si-rich, and [6] Al-poor, and have a relatively low Mg#. Empirical equations for geothermobarometry, oxygen barometry and chemometry that exclusively rely on the amphibole composition were applied to estimate the T – P – f O 2 –SiO 2 melt –FeO*/MgO melt conditions of the silicate melts with which the amphibole crystals equilibrated. The results show that group-I and group-II amphiboles equilibrated with andesitic melts (group-I melts) and dacitic–rhyolitic melts (group-II melts), respectively. The T – P – f O 2 conditions of group-I melts were estimated as ~ 374–483 MPa (~ 13.9–17.9 km depth), ~ 950 °C, and NNO + 1.3, respectively, and those of group-II melts were ~ 93–242 MPa (~ 3.4–9.0 km depth), 824–913 °C, and NNO + 0.6–1.7, respectively. The estimated T – P – f O 2 –SiO 2 melt –FeO*/MgO melt conditions were almost constant for group-I melts, whereas the T , Δ NNO , and FeO*/MgO melt values of group-II melts decreased with increasing SiO 2 melt . The DAB decreased as the temperature increased; low-DAB grains are absent in group-II amphiboles. Our results suggest that (1) a deeper reservoir (at a depth of ~ 16 km) stored an andesitic melt coexisting with group-I amphiboles at ~ 950 °C; (2) a chemically zoned reservoir at shallower depth (~ 6.4 km) was filled with dacitic–rhyolitic melts that coexisted with group-II amphiboles at ~ 824–913 °C; (3) the TS lava was the product of magma mixing of at least three components, including the andesitic melt derived from the deeper reservoir, the dacitic and rhyolitic melts from the shallower reservoir and a mafic magma derived from the lower crust and/or upper mantle; and (4) the TS lava eruption occurred within a month or less of the magma mixing. [ABSTRACT FROM AUTHOR]

Mid-ocean ridge magma reservoirs are mush-dominated systems, where reactive porous flow is a dominant post-cumulus process. However, the length scale of this process is poorly constrained, especially in ultraslow-spreading ridges. Here, we seek to address this question by presenting a comprehensive brown amphibole dataset of gabbros in IODP Hole U1473A drilled on the Atlantis Bank ocean core complex, Southwest Indian Ridge (SWIR). The results suggest a magmatic origin of the brown amphiboles. Intra-grain chemical variations and compositional zoning of these amphibole grains suggest they were crystallized in a closed system. Nevertheless, geochemical modelings indicate the remarkable Zr/Nd ratio variations in brown amphibole can only be explained by assimilation-fractional crystallization processes triggered by the reactive porous flow. Thus, the brown amphibole in olivine gabbros was crystallized from trapped melts residual after earlier melt-mush interaction. In this light, the reactive porous flow should have ceased prior to the brown amphibole precipitation. This, in turn, suggests the stratigraphic variations in amphibole trace elements reflect the final-stage reactive melt migration. The stratigraphic variations in amphibole trace elements reflect a major episode of reactive melt migration through an ∼800-m section upwardly. This implies that the porosity of the entire gabbroic section remained connected, providing a minimum estimate of the size of the melt-present zones in the lower oceanic crust during the spreading of the 57-60°E magmatic segment of SWIR. Our study reveals that large-scale reactive porous flow can occur in the magmatic segments of ultraslow-spreading ridges, and has important implications for the thermal and rheological evolution of the lower oceanic crust. • Brown amphibole records the final-stage reactive porous flow. • Reactive porous flow occurred in a length scale of >800 m. • >800-m melt-present zones once occurred in ultraslow-spreading ridges. [ABSTRACT FROM AUTHOR]

At the Eastern part of the Southwest Indian Ridge (SWIR), plate divergence is accommodated by large offset normal faults, also called detachment faults, that exhume mantle-derived rocks on the seafloor. A third of the ultramafic samples dredged on- and off-axis in this nearly amagmatic ridge setting present amphibole-bearing secondary mineralogical assemblages indicative of hydration, and for the most part predating the growth of serpentine minerals. The deepest evidence of hydration is the occurrence of small amounts of syn -kinematic amphibole in microshear zones with strongly reduced grain size, which record deformation at high stress and high temperatures (>800 °C) at the root zone of the detachment. The composition of these amphiboles is consistent with a hydrothermal origin, suggesting that seawater derived fluids percolated down to the root of detachment faults, at the Brittle-Ductile Transition (BDT). We propose that the constant exhumation of new mantle material to the seafloor, and the limited lifetime of each detachment (1–3 Myrs) prevent a more pervasive deep hydration of mid-ocean ridge detachment root regions, as proposed at transform fault plate boundaries. • Synkinematic amphiboles crystallized at the root of Eastern SWIR axial detachments. • Seawater-derived fluids percolate along the detachment down to the BDT. • Variations in amphibole composition are controlled by the protolith. • Deep fluid percolation is controlled by detachment fault activity. [ABSTRACT FROM AUTHOR]

How the low niobium/tantalum (Nb/Ta) ratios of continental crust formed is a hotly debated issue. Here, we report the Nb-Ta features of a well-characterized magmatic rock series, mainly composed of amphibole and biotite cumulates, from the Northern Dabie orogenic belt in China. The Nb/Ta ratios of whole-rock (19.5–26.2), amphibole (16.8–41.1), and biotite (21.1–49.6) are negatively correlated with their Mg# values, indicating that the separation of amphibole and biotite could not only cause Nb-Ta fractionation but also reduce the Nb/Ta ratios of residual magmas during magmatic evolution. Furthermore, we compile the Nb/Ta ratios of global arc magmas (n = 15,696) to examine the role of amphibole and biotite in the differentiation of arc magmas. The Nb/Ta ratios of global arc magmas display a negative correlation with SiO 2 contents and positive correlations with Dy/Yb and Ba/Rb ratios. Additionally, the compositions of the upper continental crust (UCC) exactly fall on the evolution line of Nb/Ta vs. SiO 2 and Nb/Ta vs. Ba/Rb, yet significantly falls below the evolution line of Nb/Ta vs. Dy/Yb. These observations show that amphibole and biotite are two important magmatic phases necessary for processes of arc magma differentiation and that they play a role together in the formation of low Nb/Ta ratios of the continental crust. As a whole, we suggest that amphibole- and biotite-rich rocks stored in the deep crust are one of the potential hidden reservoirs with higher Nb/Ta ratios, which may contribute to the depleted Nb during the formation process of continental crust. • Nb-Ta fractionation of arc magmas could be induced by amphibole and biotite. • Amphibole- and biotite-rich rocks in deep crust are potential high Nb/Ta reservoirs. [ABSTRACT FROM AUTHOR]

Astrophyllite is a common "agpaitic" accessory mineral in evolved peralkaline alkali feldspar granites from the Mandira and Papanduva plutons in the Graciosa Province (S-SE Brazil). Textural and in-situ chemical (major and trace element) data provide new insights on the mechanism of formation and evolution of astrophyllite in magmatic and post-magmatic environments. Magmatic astrophyllite forms isolated or aggregates of micaceous crystals that, in part, co-precipitated with sodic amphiboles (arfvedsonite and riebeckite) and aegirine, and with Ti Zr and/or REE-rich accessories. It presents relatively Nb- and Zr-rich and Ti-poor compositions and the Mn contents are significantly higher in the Papanduva crystals. These crystals have significant REEs contents (up to 3949 ppm), and efforts were made to obtain rough estimates for the REEs partition coefficients (ast/melt D REE) between astrophyllite and modeled evolved melts to examine their controls. Except for Nd to Tb, the calculated ast/melt D REE are higher for the Na-richer crystals from the Mandira samples, suggesting a crystal chemical control on their partitioning, especially the preference of trivalent REEs to charge-balance Na+ incorporation by the coupled substitution Na+ + REE3+ ↔ 2Ca2+. Lattice strain modeling supports the entrance of the REEs mainly in the [10]B cationic site of the astrophyllite structure. Post-magmatic astrophyllite occurs mainly as acicular and radiating crystals by replacing magmatic arfvedsonite, riebeckite, and aegirine. They are Ti-rich, close to the ideal end member, with lower Zr, Hf, Th, U, Nb, Ta, Ga, Li, Sc, Y, and REE contents than the primary crystals. Their formation occurred mainly via disequilibrium processes and substitution reactions of early sodic amphiboles and clinopyroxenes triggered by the circulation of relatively K- and Ti-rich hydrothermal fluids. Compositional variations between both the astrophyllite groups reveal that the parameters Zr/(Zr + Ti) and Mn/(Mn + Fe) can be used as chemical indicators to discriminate magmatic and post-magmatic (hydrothermal) crystallization environments, as well as to evaluate their redox conditions. • First insights into astrophyllite REE partition coefficients. • Lattice strain modeling supports REE entrance into astrophyllite [10]B site. • Zr/(Zr + Ti) ratio distinguish magmatic and post-magmatic astrophyllite. [ABSTRACT FROM AUTHOR]

Amphiboles are ubiquitous minerals in the altered oceanic crust. During subduction, their breakdown is governed by continuous reactions up to eclogitic facies conditions. Amphiboles thus contribute to slab-derived fluid throughout prograde metamorphism and continuously record information about volatile exchanges occurring between the slab and the mantle wedge. However, the fate of volatile elements and especially halogens, such as F and Cl, in amphibole during subduction is poorly constrained. We studied metagabbros from three different localities in the Western Alps: the Chenaillet ophiolite, the Queyras Schistes Lustrés and the Monviso meta-ophiolitic complexes. These samples record different metamorphic conditions, from greenschist to eclogite facies, and have interacted with different lithologies (e.g. sedimentary rocks, serpentinites) from their formation at mid-oceanic ridge, up to their devolatilization during subduction. In the oceanic crust, the initial halogen budget is mostly stored in magmatic amphibole (F = 300–7000 ppm; Cl = 20–1200 ppm) or in amphibole corona (F = 100–7000 ppm; Cl = 80–2000 ppm) and titanite (F = 200–1500 ppm; Cl < 200 ppm) formed during hydrothermal seafloor alteration. It is thus the fate of these phases that govern the halogen fluxes between the crust and the overlying mantle and/or the plate interface in subduction zones. Li and B are poorly stored in the oceanic crust (< 5 ppm). In subduction zones, prograde metamorphism of metagabbros is first marked by the crystallization of glaucophane at the expense of magmatic and amphibole coronas. This episode is accompanied with a decrease of halogen concentrations in amphiboles (< 200 ppm of F and Cl) suggesting that these elements can be transferred to the mantle wedge by fluids. In the Queyras Schistes Lustrés complex, the intense deformation and the abundant devolatilization of metasedimentary rocks produce large fluid flows that promote rock chemical hybridization (metasomatic mixing with hybrid composition between metasedimentary rock and metagabbro) at the metasedimentary rock/metagabbro contacts. Such fluid/rock interactions result in a strong addition of Li in glaucophane (up to 600 ppm) whereas halogen concentrations are unaffected. At eclogite facies conditions, metagabbros display low halogens concentrations (< 20 ppm of F and < 100 ppm of Cl) relative to altered oceanic crust (F = 40–650 ppm; Cl = 40–1400 ppm) suggesting that these elements are continuously released by fluids during the first 30–80 km of subduction whatever the tectonic environment (e.g. slab, plate interface) and the considered fluid/rock interactions. [ABSTRACT FROM AUTHOR]

Mafic–ultramafic cumulates enclosed in gabbroic-dioritic rocks form part of the Mesozoic Ditrău Alkaline Massif in the Eastern Carpathians, Romania. The poikilitic olivine- and pyroxene-rich and nearly mono mineralic hornblendite rocks display typical cumulate textures with early crystallised olivine (Fo 75–73 ), diopside and augite. In the early stages of their genesis the amphibole was intercumulus whilst in later stages it acquired cumulus status as the fractionating magma evolved. Using major and trace element compositions of minerals and whole-rock samples the origin of these cumulates is determined and the parental magma composition and depth of emplacement are calculated. Cumulus clinopyroxene has more primitive composition than intercumulus amphibole suggesting closed system fractionation for the evolution of poikilitic olivine- and pyroxene-rich cumulates. The evolution of the amphibole-rich mesocumulates is more clearly the result of closed system crystallisation dominated by the precipitation of clinopyroxene and amphibole cumulus crystals. Lamprophyre dykes of the Ditrău Alkaline Massif are proposed to reflect multiple basanitic parental magma batches from which the cumulus olivine and clinopyroxene crystallised. Relative to these dykes the calculated equilibrium melts for intercumulus amphibole in the cumulates was more primitive whilst that for the cumulus amphibole was more evolved. The calculated crystallisation temperature and pressure of ~ 1000–1050 °C and ~ 0.7 GPa, based on the composition of the amphiboles, indicate crystallisation at lower crustal depths. Rare earth element compositions are consistent with an intra-plate tectonic setting. [ABSTRACT FROM AUTHOR]

Dependencies of plagioclase/amphibole Al-Si partitioning, D Al / Si plg / amp , and amphibole/liquid Mg partitioning, D Mg amp / liq , on temperature, pressure and phase compositions are investigated employing robust regression methods based on MM-estimators. A database with 92 amphibole-plagioclase pairs – temperature range: 650–1050 °C; amphibole compositional limits: > 0.02 apfu (23O) Ti and > 0.05 apfu Al – and 148 amphibole-glass pairs – temperature range: 800–1100 °C; amphibole compositional limit: CaM 4/( CaM 4 + NaM 4) > 0.75 – compiled from experiments in the literature was used for the calculations (amphibole normalization scheme: 13-CNK method). Statistical analysis reveals a significant dependence of D Al / Si plg / amp on pressure, temperature, Al fraction in amphibole T1-site, X Al T 1 , and albite fraction in plagioclase, X Ab , leading to the barometric expression: P kbar = 8.3144 T K ln D A l / S i p l g / a m p − 8.7 T K + 23377 X A l T 1 + 7579 X Ab − 11302 / − 274 (precision: ± 1.5 to ± 2.3 kbar; expressed at 1 s). The D Mg amp / liq was found to have a significant dependence on temperature and the logarithm of X Ca liq /( X Ca liq + X Al liq ) ratio (glass composition expressed as anhydrous mole fraction of cation components) that led to the thermometer: T ℃ = 71975 − 11896 ln X C a l i q / X C a l i q + X A l l i q 8.3144 ln D Mg a m p / l i q + 58 − 273 (precision: ± 35 to ± 45 °C). Amphibole-liquid compositional relationships revealed that the temperature dependence of D Mg amp / liq was mostly controlled by a positive correlation of lnX Mg liq with temperature that led to the thermometer: T (℃) = 107 lnX Mg liq − 108 ln [ X Ca liq /( X Ca liq + X Al liq )] + 1184 (precision: ± 37 to ± 42 °C). The plagioclase-amphibole barometer is suitable for a large diversity of amphibole-plagioclase-bearing assemblages from metamorphic (amphibolites and mafic granulites) and igneous (metaluminous granitoids to gabbros) rocks, whereas the liquid-only and the amphibole-liquid thermometers are applicable to alkaline and subalkaline igneous rocks. The calibrated expressions yield P-T estimates consistent with those obtained with widely used barometers and thermometers. Plagioclase-amphibole barometry extends pressure determinations to quartz-free and/or garnet-free assemblages. [ABSTRACT FROM AUTHOR]

Late Cretaceous (~ 106–76 Ma) adakite-like intrusive rocks in the middle-eastern Gangdese belt occur in an E–W trending belt paralleling the Indus–Yarlung suture, south Tibet. Their petrogenesis and geodynamic processes have been a subject of debate. We report here U–Pb zircon ages, geochemical and Sr–Nd–Hf isotopic data for adakite-like intrusive rocks as well as the normal arc rocks (gabbros and gabbroic diorites) in the middle Gangdese belt. LA-ICPMS U–Pb zircon analyses yielded an identical age of ~ 88 Ma for two adakite-like rocks, which are slightly younger than the gabbro and gabbroic diorite (ca. 94–90 Ma). Both the adakite-like rocks and the normal arc rocks have similar whole-rock Sr–Nd and zircon Hf isotope compositions, indicating that they have been derived from a common source. Similarly, the adakite-like and normal arc intrusive rocks in the eastern Gangdese belt also show similar Sr–Nd–Hf isotope compositions. In the middle-eastern Gangdese belt, the > 85 Ma Late Cretaceous intrusive rocks consist of a magma series from gabbro to granodiorite, including both normal arc rocks and adakite-like rocks. These rocks overlap in space and time that conform to a normal arc differentiation trend. In terms of major and trace elements, they also show a clear evolution from the normal arc magmatic into adakitic field. Thus, we suggest that these > 85 Ma Late Cretaceous intrusive rocks were ultimately derived from melting of the hydrated mantle wedge and the adakite-like rocks can be generated in normal arc magmas by amphibole-dominated fractionation. Taking into accounting for the spatial and temporal distribution of the Cretaceous magmatic rocks in the Lhasa terrane, we prefer a model of early Late Cretaceous rollback following Early Cretaceous low-angle oceanic slab subduction. At intermediate pressure and H 2 O-rich conditions, fractionation of amphibole changes the major and trace element compositions of arc magmas, and will efficiently drives basaltic composition to andesitic composition in arc magmas. [ABSTRACT FROM AUTHOR]

Adrar Izzilatène in the Egéré-Aleksod terrane of the LATEA metacraton (Hoggar, Algeria) exposes one of the best preserved examples of eclogite facies metamorphism in Hoggar. Three distinct stages of metamorphic development are recognised, namely, the pre-peak stage (M1), characterised by garnet, amphibole, epidote, quartz and rutile, the peak eclogite facies stage (M2), consisting of omphacite, garnet, edenite, epidote, quartz and rutile, and the retrograde stage (M3), where initial decompression resulted in the appearance of plagioclase, the development of pargasite + plagioclase kelyphites and finally the formation of anhydrous plagioclase + diopside coronas. Porphyroblastic omphacite has a jadeite content of up to X Jd = 0.36, which is the highest yet observed for eclogite facies rocks from the Tuareg Shield. Garnet growth zoning patterns are characterised by flat profiles in the cores ( X Alm = 0.55–0.60; X Prp = 0.12–0.16; X Grs = 0.26–0.30) before showing a decrease in almandine to X Alm = 0.45, coupled to an increase in pyrope to X Prp = 0.29 and decrease in grossular to X Grs = 0.26 at the rims. Calculated P–T–M H2O pseudosections show that the prograde M1 assemblage equilibrated at 13–14 kbar and 580 °C, before pressure and temperature increased to 19 kbar and 650–700 °C at fluid-saturated conditions during peak metamorphism. Retrogression involved near-isothermal decompression to 8–9 kbar and 700–750 °C at fluid-undersaturated conditions. Prograde-to-peak metamorphism of the Izzilatène eclogite could have involved either oceanic or continental subduction, followed by exhumation as the area was obducted towards the LATEA metacraton during the Pan-African orogeny and the assembly of Western Gondwana. [ABSTRACT FROM AUTHOR]

The lower layered nepheline syenite sequence (kakortokites) of the Mesoproterozoic alkaline to peralkaline Ilímaussaq complex, South Greenland shows spectacular rhythmic meter-scale igneous layering. The 29 exposed units have sharp contacts against each other and each of these units consists of three modally graded layers dominated by arfvedsonitic amphibole, eudialyte-group minerals, and alkali feldspar, respectively. This study uses field observations on changes in mineral orientation, recurrent mineral textures, compositional data from eudialyte-group minerals and amphibole, and settling rate calculations based on a modified Stokes' equation to explain the igneous layering of the kakortokites. We propose that the three major cumulus minerals (amphibole, eudialyte s.l., and alkali feldspar) were separated from each other by density contrasts, resulting in modally graded layers within each unit. The densest of these three minerals (amphibole) formed crystal mats within the cooling magma body. These crystal mats acted as barriers that inhibited large-scale vertical migration of melts and crystals with increasing effectiveness over time. The sub-volumes of magma captured in between the crystal mats evolved largely as geochemically independent sub-systems, as indicated by the observed trends in mineral composition. [ABSTRACT FROM AUTHOR]

The concentration of halogens in apatite, biotite and amphibole is investigated for a large variety of mantle-derived plutonic rocks (gabbros, diorites, monzonites, olivine- and pyroxene-bearing monzonitic to granitic rocks, syenites, carbonatites and a phoscorite). In all rocks studied, apatite occurs as an early magmatic phase, whereas biotite and amphibole may occur either as a late magmatic phase or as late-stage, potentially hydrothermal product replacing precursor olivine, pyroxene and Fe–Ti oxides (ilmenite and magnetite). Based on electron microprobe analyses for F and Cl and detailed textural observations, we test existing models of halogen partitioning between apatite and biotite. Bromine concentration data for apatite, biotite and amphibole are used to further refine our understanding of the geochemical similarities and differences between Cl and Br during magmatic and hydrothermal processes. Our data suggests that F and Cl contents in apatite, biotite and amphibole can indeed be useful monitors of the halogen systematics in magmas, but they may also be subject to post-magmatic changes to variable extents. The relatively small radius and compatible F cation seems to be less prone to post-magmatic alteration and is likely to best reflect the original magmatic halogen abundances — especially in apatite. However, the larger and probably more incompatible Cl anion, is more easily re-mobilized as reflected by strong redistribution of Cl in biotite and amphibole which have been clearly overprinted by hydrothermal fluids. In certain cases, the ability of halogens to re-distribute themselves after magmatic equilibrium partitioning (as emphasized by our data) suggests that observed partitioning (especially between apatite and biotite) may also be used as a very sensitive indicator for post-magmatic hydrothermal processes. [ABSTRACT FROM AUTHOR]

Large silicic explosive eruptions are one of Nature's most hazardous phenomena. Very few have been witnessed and unravelling the complex conditions that lead to these eruptions remains a difficult task because the characteristics of the feeding magma reservoirs are still insufficiently constrained from geophysical imaging. Here we show that a barometer based on the composition of amphibole in equilibrium with plagioclase and biotite, common minerals in the eruptive products of large silicic eruptions, records the thickness and depth of magma reservoirs with uncertainties of 0.8 and 2.7 km, respectively. Pressures are given by the equation: P (MPa) = 892 · VIAl + 101, where VIAl is the octahedral aluminum content of the amphibole. With the example of a Miocene Turkish ignimbrite, we show that reservoirs feeding large silicic eruptions can be pancake-shaped. Our new barometer, valid between 650 and 950 °C, can be used in combination with volcanological and geophysical data to infer the size, shape and depth of magma reservoirs and may serve as a tool for monitoring future activity. This temperature-independent barometer is also applicable to any volcanic or plutonic rock containing amphibole + plagioclase + biotite and is in excellent agreement with previously published temperature-dependent barometers within their calibration range. • Developpement of a new temperature-independent Al-in amphibole geobarometer • Depth of emplacement of plutons and magma reservoirs with an uncertainty of 2.7 km • Shape of erupted magma reservoirs accessible from petrological and volcanological observations [ABSTRACT FROM AUTHOR]

Lamprophyres of different age showing distinctive mineralogy, geochemistry and isotopic ratios are exposed in northwestern Iran. They can be divided into Late Cretaceous sannaite, Late Oligocene–Early Miocene camptonite (amphibole-bearing) and Late Miocene minette (mica-bearing) and spessartite (amphibole-bearing) lamprophyres. Sannaites have high-Ti amphibole along with high-Ti and Al clinopyroxene, and they are characterised by homogeneous enrichment in incompatible trace elements with troughs at Pb. Spessartites have hornblende and low-Al and Ti clinopyroxene, and they are characterised by enriched incompatible trace element pattern with depletions of Nb, Ta, Pb, and Ti with respect to large ion lithophile elements. Minettes have high-Ti and Al brown mica and low-Al and Ti clinopyroxene, and similar to spessartite, are characterised by fractionation of high field strength elements with respect to large ion lithophile elements, with troughs at Nb, Ta, and Ti and a peak at Pb. Minettes show high initial 87 Sr/ 86 Sr values up to 0.70760 and low initial 143 Nd/ 144 Nd down to 0.512463 with a negative correlation, consistent with the trace element distribution related with an enriched mantle source modified after sediment recycling during subduction and continental collision. Cretaceous sannaites and Early Miocene spessartites show low initial 87 Sr/ 86 Sr approaching 0.70447 and high 143 Nd/ 144 Nd values up to 0.512667, which are consistent with a depleted within-plate mantle source. Minette and spessartite lamprophyres show high initial 206 Pb/ 204 Pb, 207 Pb/ 204 Pb, and 208 Pb/ 204 Pb values, whereas sannaites have lower, but variable, initial 206 Pb/ 204 Pb, 207 Pb/ 204 Pb, and 208 Pb/ 204 Pb values with respect to those of calc-alkaline lamprophyres. Minettes originated by partial melting of a metasomatised lithospheric mantle following siliciclastic sediment recycling by subduction. In contrast, sannaites were generated from the partial melting of a similar lithospheric mantle that was metasomatised by within-plate agents. [ABSTRACT FROM AUTHOR]

The presence of variable amounts of amphibole ± phlogopite in a garnet websterite and a garnet clinopyroxenite from the Beni Bousera peridotite massif provides evidence for post-formation metasomatism. Textural observations associated with major- and trace-element mineral compositions allowed us to distinguish two metasomatic episodes, which occurred at different stages of the Beni Bousera massif evolution. The garnet websterite has recorded interaction with LREE-rich silicate melts before the uplift of the massif. Amphibole/clinopyroxene and amphibole/garnet trace-element ratios closely approach partition coefficient values, indicating that chemical equilibrium was attained between amphibole and pyroxenite matrix minerals. The geochemical signatures of the putative alkaline interacting melts are similar to those of recent basaltic magmas erupted in Morocco, suggesting a common peridotite mantle source. In contrast, amphibole from the garnet clinopyroxenite is in chemical disequilibrium with the pyroxenite matrix minerals. In this clinopyroxenite the crystallization of amphibole and plagioclase occurred at lower T (and P) conditions, most probably during the ascent of the Beni Bousera massif and its emplacement into the crust. The melt responsible for this later metasomatic episode was LREE-depleted and HREE-enriched, suggesting that it resulted from decompression melting of a garnet-bearing source (with garnet as a melting phase), similar to the garnet-bearing pyroxenites outcropping in the Beni Bousera massif. [ABSTRACT FROM AUTHOR]

The continental crust (CC) and the depleted mantle (DM) are generally assumed to be complementary reservoirs within the Earth. However, the mixture between CC and upper mantle does not generate the Nb/Ta and Nb/La ratios of chondrites. A reservoir with superchondritic ratios for Nb/Ta and Nb/La is thus required in the Earth's system. The occurrence of a hidden amphibole reservoir in the lower arc crust has been recently proposed. This, coupled with the capability of calcic amphibole to give rise to a superchondritic Nb/Ta and Nb/La reservoir, led us to determine to what extent amphibole-rich ultramafic rocks can account for the Nb (and Nb/Ta, Nb/La as well) imbalance on Earth. We have considered lower crust mafic and ultramafic amphibole-rich intrusive rocks from collisional settings worldwide. Because CC is considered to have primarily formed in collisional setting these rocks are important for its genetic model. We modeled Nb, Ta and La contents of the hidden Nb reservoir by mass balance calculations between continental crust, depleted mantle and primitive mantle. Modeling shows that amphibole-rich mafic lower crust can solve the so-called Nb paradox if large volumes of materials are supposed to be returned into the mantle during the Earth's history. A possible mechanism is recycling, particularly in Precambrian times, of eclogites that underwent pre-eclogitic melting in the amphibolite facies field and then recrystallized under eclogite-facies conditions. [ABSTRACT FROM AUTHOR]

Cretaceous lamprophyres in the Catalonian Coastal Ranges are isolated, cm- to m-thick intrusions of camptonite, exceptionally rich in crystal populations. Most lamprophyres are strongly porphyritic, with large (up to cm) crystals of clinopyroxene, amphibole, olivine pseudomorphs and opaque minerals embedded in a microcrystalline groundmass primarily composed of amphibole and feldspars. Detailed petrography and chemical analysis of crystal populations reveals complex compositional zoning patterns in large crystals, most of which are antecrysts recycled from earlier stages of the magmatic system at depth. The growth stratigraphy of clinopyroxene and amphibole antecrysts requires an intricate succession of magmatic processes in the plumbing system, including magma recharge, magma mixing and polybaric fractionation during magma ascent and emplacement. In particular, significant leaps in MgO with crystal growth indicate events of recharge of more primitive magma into the system. The groundmass composition is similar in the whole lamprophyre suite, but the accumulation of antecrysts triggers whole rock enrichments in MgO (up to 14 wt.%) and compatible elements that could be erroneously interpreted as parental magma compositions. This way, the variable entrainment of antecrysts in ascending lamprophyre magmas modulates whole rock compositional variability in the lamprophyre suite. The present study widens our previous results from one of the lamprophyres and shows that i) much information on magmatic processes may be derived from detailed mineral chemistry in addition to whole rock investigations, and ii) interpretation of whole rock evolutionary trends should be preceded by a close textural understanding of the rocks and their crystal cargo. [ABSTRACT FROM AUTHOR]

HP mafic and felsic adakitic granulite bodies in the Dulan area of the North Qaidam ultrahigh-pressure (UHP) terrane record high-pressure (HP) granulite-facies metamorphism and anatexis and provide temporal and tectonic constraints on deep subduction of continental crust and its subsequent exhumation. Mafic HP granulite components dominate the main outcrop and preserve features diagnostic of anatexis and locally may be described as migmatite. The HP mafic granulites comprise garnet, clinopyroxene, plagioclase and quartz. The felsic granulite (leucosome) is mainly composed of K-feldspar+plagioclase+quartz+kyanite+garnet. Detailed zircon U–Pb and amphibole Ar–Ar geochronology, combined with trace element geochemistry, indicate peak metamorphism for the mafic HP granulite at 434±3 to 435±3Ma and peak metamorphism and partial melting for the felsic HP granulite at 433±5 to 438±4Ma, which overlaps the ages of UHP metamorphism for adjacent eclogite (430–446Ma). 40Ar/39Ar amphibole ages of 423 to 432Ma represent amphibolite-facies retrograde metamorphism and indicate rapid cooling during exhumation of the HP granulite bodies. Our geochronological data, combined with field relationships, petrology and geochemistry suggest that HP granulite-facies metamorphism and the partial melting that produced adakitic melts represent the same tectonic event. In this case, the felsic HP granulites (leucosome) formed from an adakitic melt derived from partial melting of mafic HP granulite in the overriding plate in a relatively higher geothermal gradient (15–18°C/km), leaving garnet-cumulate and/or meta-ultramafic (mainly garnet pyroxenite) as the residual component. In contrast, the nearby UHP eclogite is thought to have formed in the subducted plate in a relatively lower geothermal gradient (6–10°C/km). Penecontemporaneous metamorphic ages but different geothermal gradients between HP granulites and related UHP eclogite define a possible paired metamorphic belt generated in a subduction–collision setting associated with the North Qaidam continental collisional orogeny during the Late Ordovician–Early Silurian. [ABSTRACT FROM AUTHOR]

Abstract: A Cretaceous sill of alkaline lamprophyre (camptonite) cropping out in the northern Catalonian Coastal Ranges includes macrocrysts and microcrysts of clinopyroxene and amphibole with complex zoning patterns. The in-depth petrographic, compositional and barometric study of the different crystal populations, including statistical evaluation of the compositional dataset by principal component analysis, provides insights into the magmatic plumbing system. Macrocryst cores are inherited antecrysts crystallised in a deep (500–800MPa) magma chamber or conduit. Their reverse zoning patterns reveal repeated injection and mixing of batches of a more primitive and water-rich magma with the resident magma in the magma chamber. Macrocryst rims and microcrysts, in contrast, crystallised during the ascent and shallow emplacement of the magma below 50MPa pressure. They define normal zoning patterns that can be related to progressive fractionation of the magma. This study shows that an apparently simple porphyritic rock can actually contain a significant amount of recycled crystals, reflecting an open-system magmatic behaviour. Only the groundmass of the sill represents a closed-system, where microphenocryst cores are in chemical equilibrium with the bulk composition of the groundmass. Using trace element compositions obtained by LA–ICP–MS and geochemical modelling, a complete and consistent partition coefficient dataset is obtained for the first time for clinopyroxene and amphibole in camptonite melts. Both minerals show a similar partitioning behaviour for most trace elements. Clinopyroxene-melt and amphibole-melt partition coefficients are overall consistent with data from basic rocks, although clinopyroxene partition coefficients are generally higher than data from basalt and basanite melts. The results from this research support the use of naturally occurring mineral-melt equilibrium pairs for partitioning studies but emphasise the need for a detailed study of the phases used to define the partition relationships. [Copyright &y& Elsevier]

Abstract: At Okataina volcano, a caldera-collapse event at 46ka followed by frequent intra-caldera eruptions (every 1–5ka) provides an opportunity to examine the temperature and pressure conditions from amphiboles (and other phases) through the life cycle of a rhyolite magma system. In each of 18 eruptions examined, the amphibole population is heterogeneous (e.g., Al2O3 =4.45–10.88wt.%; and MgO=10.81–16.35wt.%), and lacks trends in space or time or with magma composition and volume. The crystals are a ‘cargo’ gathered rapidly from a range from T-P-X conditions prior to eruption, perhaps due to buoyancy instabilities in the system causing overturn and blending. Nearly all of the compositional variation can be accounted for by temperature-sensitive edenite and Ti-Tschermak exchange mechanisms, and ‘Plagioclase exchange’ that is influenced by co-crystallizing plagioclase composition. The pressure-sensitive Al-Tschermak exchange is insignificant. This highlights that the application of Al-in-amphibole barometers would produce erroneous conclusions if atomic exchange mechanisms were not assessed. Variations in temperature implied by atomic substitutions can be explained by petrographic evidence for frequent mafic intrusions. Amphibole temperature estimates based on empirical models are supported by equilibrium calculations involving plagioclase and melt, and indicate some amphiboles are relicts from a hotter thermal regime than that of the dominant plagioclase population and melt. Fast equilibrating melts and Fe–Ti oxides show a dramatic change to high temperature rhyodacitic compositions immediately following caldera-collapse (42–36ka), before returning to typical low temperature high-SiO2 rhyolite production over the last ~30ka. Throughout this cycle, the amphiboles record a contiguous thermally zoned system (700–900°C), buffered from eruption events, perhaps reflecting its large size or a lack of connectivity in the melt generation zones. Hence, the crystallization state of the system has little bearing on its potential for eruption. [Copyright &y& Elsevier]

Abstract: Akarem Intrusion in the southern part of the Eastern Desert of Egypt is one of Alaskan-type ultramafic-mafic complexes of the Neoproterozoic age. The olivine and pyroxene-rich ultramafic core is surrounded by hornblende gabbro and hornblendite margin with gradational contacts. The pyroxene chemistry indicates that the intrusion was crystallized from tholeiitic arc magmas. The calculated liquids in equilibrium with clinopyroxene show enrichment in Th, U and LREE over the HREE, similar to arc magma formed from the mantle metasomatized by fluids from subducted sediments. Both olivine and pyroxene compositions indicate fractional crystallization from the ultramafic core to the mafic margin. However, amphiboles show systematic increase in concentrations of compatible elements, such as V, and shift in REE patterns from a bowed-up shape to bowed-down shape from the mafic margin to the ultramafic core. Such variation in trace elements of amphiboles indicates progressive crystallization of amphiboles from the mafic margin to the ultramafic core. The variation in amphibole chemistry has implication on both development of Alaskan-type intrusion and subduction zone geochemical signature of the evolved magma. The formation of the Akarem Intrusion started with an early fractional crystallization stage which led to the formation of a stratified magma chamber. The circulation in the magma chamber resulted in the solidification of the mafic magma along the margin where the early-stage amphiboles crystallized along with plagioclase. Subsequently, an evolved liquid from the crystallizing mafic margin percolated into the liquid-rich ultramafic core to form the later generation of interstitial amphiboles. The calculated trace element composition of the liquids in equilibrium with amphibole shows a progressive development of Nb and Ti negative anomalies and increase in La/Yb ratios from the margin to the core of the intrusion. Such variations manifest the role of amphibole fractionation in the evolution of the geochemical signature of arc magma. [Copyright &y& Elsevier]

Abstract: S-type granites always contain more Al than the amounts of Na, Ca and K in the rock required to form feldspars, primarily owing to their derivation from source components that had previously been weathered. Those rocks are therefore always saturated in Al, or peraluminous. Many I-type granites are also peraluminous, despite I-type source rocks typically not being saturated in Al. It has previously been suggested that this may result from the fractional crystallisation of amphibole. However, data from compositionally zoned high-temperature plutons in the Lachlan Fold Belt show that it is difficult to generate large quantities of peraluminous melt by removal of amphibole. Most of the I-type granites (~95%) in the Lachlan Fold Belt formed at lower temperatures and almost half of those rocks for which bulk chemical compositions are available are peraluminous. Among these granites there are 98 separate suites for which there are chemical data for two or more samples, with 47 suites that include both metaluminous more mafic and peraluminous more felsic compositions. The origin of those peraluminous compositions is fundamental to any understanding of I-type granite petrogenesis in this region. Compositional variations caused by the assimilation or by partial melting of supracrustal rocks are very small as the isotopic variations within these rocks are dominantly between different suites, not within suites. The partial melting of more mafic source rocks, rather than the fractional crystallisation of more mafic magmas, is favoured for the origin of these rocks. Partial melting is the most likely process involved in the petrogenesis of felsic granites where broadly granodioritic–monzogranitic batholiths are associated with lesser amounts of tonalite and very minor amounts of mafic rock. Experimental studies have shown that the melts generated by the partial melting of basaltic to andesitic rocks under crustal conditions are mostly peraluminous. During the dehydrational melting of I-type granite source rocks at pressures below the garnet stability field, biotite and amphibole melt incongruently to yield pyroxenes. The excess Al is incorporated into the felsic liquid, resulting in the generation of peraluminous melts. In this instance, the excess Al in felsic I-type granites is a function of the melting process, and unrelated to the bulk composition of the source. The observed gradation from peraluminous felsic granites to metaluminous compositions in less felsic rocks in largely isotopically closed systems could happen in two ways. At higher temperatures of partial melting, Ca and other components of clinopyroxene could dissolve in the melt, with the melt eventually becoming metaluminous. Alternatively, minerals residual from partial melting, dominantly pyroxenes and plagioclase, could be incorporated in suspension in the melt, so that the resulting bulk magma is metaluminous. Examples of these two extremes, and of intermediate cases, are well developed among the granites of southeastern Australia. [Copyright &y& Elsevier]

Abstract: Stable isotope and trace element compositions of igneous amphiboles from different tectonic settings (ocean island basalts, intraplate alkaline basalts, subduction-related andesitic complexes) were compiled to help understand the role of fluids and melts in subduction-related mantle metasomatism and to evaluate the use of selected trace element ratios (Pb/Pb*(N)=Pb/(√(Ce·Pr)) and Ba/Nb(N), normalized to primitive mantle) to help detect possible metasomatism. Comparisons of stable H and O isotope compositions and trace element ratios of amphiboles from ocean island basalts (Canary Islands), intraplate basalts, and subduction-related calc-alkaline andesitic series (Carpathian–Pannonian Region, CPR) indicate systematic distributions in δD–δ 18O–Pb/Pb*(N)–Ba/Nb(N) diagrams that are related to metasomatic processes in the mantle and the migration of fluids and melts derived from subducted crustal slabs. In order to interpret these data for the amphiboles from the CPR, ophiolites of the Penninic and the Meliata–Vardar complexes as potential sources of subducted crustal melts and fluids in the mantle of the Carpathian–Pannonian Region were also analyzed. On the basis of published fluid/rock partition coefficients the compositions of fluids emanating from subducted ophiolites were calculated. The calculated fluid compositions—especially for blueschists of the Meliata complex—fit the amphibole trends, indicating that such fluids could have been responsible for the mantle metasomatism beneath the CPR. [Copyright &y& Elsevier]

Abstract: The Mt Gibraltar intrusion near Mittagong and Bowral in New South Wales, Australia (lat. 34°27′54″S, long. 150°25′44″) is a small intrusive body of hypersolvus microsyenite emplaced into the Triassic Hawkesbury Sandstone of the Sydney Basin in Jurassic time, possibly related to extensional faulting. The rock itself consists of intermediate alkali feldspar with minor titanomagnetite and interstitial pyroxene ranging from nearly pure hedenbergite to ≈Hd34Aeg65 in composition. It is crosscut by an irregular system of late-magmatic veins consisting of homogeneous alkali feldspar (≈Ab50Or50), clinopyroxene evolving from sodic hedenbergite to zirconium-rich aegirine, arfvedsonite and siderite. During postmagmatic evolution of the veins, microcrystalline or amorphous silica precipitated together with calcite filling miarolitic cavities. The late-magmatic mineral assemblage of the veins indicate crystallisation (at assumed 700bar pressure) at T=650–670°C, log fO2 =−22. This corresponds to conditions very close to the magnetite–wustite curve. Zirconium-bearing pyroxene has a restricted stability field in the system SiO2–ZrO2–FeO–FeO1.5–NaO0.5–HO0.5, at moderately elevated peralkalinity and intermediate silica activity. Under such conditions, pyroxene and amphibole will act as effective sinks for Zr, preventing crystallisation of magmatic zircon or more exotic Zr silicates. The Mt Gibraltar microsyenite is therefore a rare example of an igneous rock in which zirconium is camouflaged in pyroxene and amphibole rather than forming its own minerals. [Copyright &y& Elsevier]

Abstract: The Late Neoproterozoic Greendale Complex, located within the Avalon terrane of Nova Scotia, is a suite of appinitic rocks ranging from ultramafic to felsic in composition that were intruded during regional ensialic arc magmatism and crystallized at shallow crustal levels under conditions of high pH2O. Amphibole is the dominant mafic mineral in ultramafic to mafic rocks and displays the extraordinary variability in texture and modal abundance that is characteristic of appinite suites. These features allow sensitivity of amphibole composition (major, trace and REE) to the evolution of water-rich magma to be investigated. All amphiboles in mafic and ultramafic rocks are calcic, with (Ca+Na)B ≥1.34 and SiIV between 6.1 and 7.3. They predominantly range in composition from tschermakite to magnesiohornblende and display a dominance of edenite (Na,KA +AlIV =SiIV) substitution. Although each sample exhibits remarkably uniform Mg# over a wide range in Si of up to one formula unit, the mafic rock amphiboles are characterized by lower (0.5 to 0.7) Mg#, compared to the ultramafic rocks (0.7 and 0.9). REE profiles are bow-shaped, and are characterized by depletion in LREE (La/Sm≈0.61), a slight depletion in HREE (Gd/Yb≈1.55) as well as a negative Eu anomaly, which is attributed to co-precipitation of plagioclase. REE and trace element profiles of ultramafic amphiboles are divided into two groupings. Group A amphiboles occur in all specimens analyzed and their REE profiles are very similar to the whole-rock analyses of the mafic rocks and to those predicted from amphibole/melt partition coefficients. In contrast, Group B amphiboles display relative enrichment in light REEs (La/Sm≈2.05), have lower ΣREE, and lack a negative Eu anomaly relative to Sm and Gd. Group B amphiboles are more enriched in Th and U and show a more pronounced depletion in Nb, Ti, Y and HREE. Group B amphiboles probably grew in a reaction relationship with olivine and pyroxene, and their contrasting profiles suggest that partition coefficients traditionally used for petrogenetic modeling may not be appropriate for such amphiboles. Groups A and B are virtually indistinguishable with respect to the major elements used for amphibole classification, suggesting that REE and selected trace element data, when combined with textural observations, may provide important additional insights into phase equilibria and conditions of crystallization. [Copyright &y& Elsevier]

Abstract: The present work is a first comprehensive study of the trace-element composition and zoning in clinopyroxene- and amphibole-group minerals from carbonatites, incorporating samples from 14 localities worldwide (Afrikanda, Aley, Alnö, Blue River, Eden Lake, Huayangchuan, Murun, Oka, Ozernaya Varaka, Ozernyi, Paint Lake, Pinghe, Prairie Lake, Turiy Mys). The new electron-microprobe data presented here significantly extend the known compositional range of clinopyroxenes and amphiboles from carbonatites. These data confirm that calcic and sodic clinopyroxenes from carbonatites are not separated by a compositional gap, instead forming an arcuate trend from nearly pure diopside through intermediate aegirine–augite compositions confined to a limited range of CaFeSi2O6 contents (15–45mol%) to aegirine with <25mol% of CaMgSi2O6 and a negligible proportion of CaFeSi2O6. A large set of LA-ICPMS data shows that the clinopyroxenes of different composition are characterized by relatively low levels of Cr, Co and Ni (≤40ppm) and manifold variations in the concentration of trivalent lithophile and some incompatible elements (1–150ppm Sc, 26–6870ppmV, 5–550ppm Sr, 90–2360ppm Zr, and nil to 150ppm REE), recorded in some cases within a single crystal. The relative contribution of clinopyroxenes to the whole-rock Rb, Nb, Ta, Th and U budget is negligible. The major-element compositional range of amphiboles spans from alkali- and Al-poor members (tremolite) to Na–Al-rich Mg- or, less commonly, Fe-dominant members (magnesiohastingsite, hastingsite and pargasite), to calcic–sodic, sodic and potassic–sodic compositions intermediate between magnesio-ferrikatophorite, richterite, magnesioriebeckite, ferri-nyböite and (potassic-)magnesio-arfvedsonite. In comparison with the clinopyroxenes, the amphiboles contain similar levels of tetravalent high-field-strength elements (Ti, Zr and Hf) and compatible transition elements (Cr, Co and Ni), but are capable of incorporating much higher concentrations of Sc and incompatible elements (up to 500ppm Sc, 43ppm Rb, 1470ppm Sr, 1230ppm Ba, 80ppm Pb, 1070ppm REE, 140ppm Y, and 180ppm Nb). In some carbonatites, amphiboles contribute as much as 25% of the Zr+Hf, 15% of the Sr and 35% of the Rb+Ba whole-rock budget. Both clinopyroxenes and amphiboles may also host a significant share (~10%) of the bulk heavy-REE content. Our trace-element data show that the partitioning of REE between clinopyroxene (and, in some samples, amphibole) and the melt is clearly bimodal and requires a revision of the existing models assuming single-site REE partitioning. Clinopyroxenes and amphiboles from carbonatites exhibit a diversity of zoning patterns that cannot be explained exclusively on the basis of crystal chemistry and relative compatibility of different trace-element in these minerals. Paragenetic analysis indicates that in most cases, the observed zoning patterns develop in response to removal of selected trace elements by phases co-precipitating with clinopyroxene and amphibole (especially magnetite, fluorapatite, phlogopite and pyrochlore). With the exception of magnesiohastingsite–richterite sample from Afrikanda, the invariability of trace-element ratios in the majority of zoned clinopyroxene and amphibole crystals implies that fluids are not involved in the development of zoning in these minerals. The implications of the new trace-element data for mineral exploration targeting REE, Nb and other types of carbonatite-hosted rare-metal mineralization are discussed. [Copyright &y& Elsevier]

Abstract: The mechanism of formation of blueschist–eclogite belts and their space-time distribution are important in understanding the tectonics associated with convergent plate boundaries. Here we investigate the garnet-bearing blueschists from Rongma area of central Qiangtang in northern Tibet. The mineral assemblage in these rocks is characterized by porphyroblastic garnet set within a matrix of fine-grained amphibole, white mica, epidote, chlorite, albite and quartz with accessory rutile, titanite and apatite. The garnet porphyroblasts exhibit core and rim portions, and the cores carry abundant inclusions of Na amphibole, quartz and rutile, as well as rhomb-shaped inclusions of paragonite and epidote which are interpreted as pseudomorphs after lawsonite. The rims are characterized by coarse-grained inclusions of epidote as well as the absence of paragonite and epidote aggregates, clearly suggesting that the transition from garnet core to rim marks a metamorphic transformation from lawsonite- to epidote-stability field. The Mn content of the garnet porphyroblasts decreases from core to rim, whereas the Fe and Mg contents show an increasing trend. In the matrix, we identify two stages of Na amphibole rimmed by Na-Ca amphibole and albite. Retrograde chlorite is rimmed by fine-grained biotite. Based on microstructural observations and pseudosection modelling, we trace the P–T path for the Rongma garnet blueschist from 1.92GPa and 490°C (lawsonite eclogite field) to about 1.68GPa and 535°C (epidote eclogite field), marking an initial increase in temperature and decrease in pressure. This stage is followed by a decrease of pressure through the blueschist facies down to P–T conditions of about 0.6GPa and 530°C. In combination with previous work including the available isotopic age data, the P–T path obtained in the present study suggests the deep subduction of palaeo-Tethyan oceanic crust between southern and northern Qiangtang blocks, supporting the model that the blueschist belt defines the location of the palaeo-Tethyan suture zone within northern Tibet. T-X(O) pseudosection modelling further reveals that the transformation of lawsonite to epidote occurred under high O contents. Our results provide important clues to better understand the subduction-related late Permian to early Triassic magmas in northern Tibet. [Copyright &y& Elsevier]

Abstract: Granitoids varying in composition from biotite–amphibole granodiorite to monzogranite with rapakivi textures have recently been recognized in the Qinling orogen (central China). The granitoids are magnesian, calc-alkalic and metaluminous (A/CNK=0.71–0.95). They contain many mafic to intermediate microgranular enclaves. The enclaves are quenched mafic melts in granitic host magma as evidenced by fine-grained sinuous margins against or transitional contacts with the host granitoids as well as disequilibrium textures and mineral assemblages. U–Pb zircon dating of the granitoids yields: 209±2Ma and 212±2Ma and the enclaves 210±2Ma and 211±2Ma, respectively. The host granitoids and enclaves show obvious silica gap but similar REE and trace element patterns. The whole-rock 87Sr/86Sr(t), ε Nd(t) and zircon ε Hf(t) values for the granitoids vary from 0.70516 to 0.70634, −2.3 to −4.4 and 0.7 to −7.6, and for the enclaves from 0.70467 to 0.70600, −1.1 to −4.0, and 1.2 to −8.8, respectively, showing overlapped ε Nd(t) and zircon ε Hf(t) values. All these suggest that the two magmatic systems represented by the host rock and enclave are probably derived from distinct sources with sufficient interaction, or common origin but underwent different degrees of crustal contamination. Mineralogical and geochemical data point to enclave dissolution in silicic magma as a major process accounting for the origin of the rapakivi texture. The possible bimodal magmatism suggested by the coeval granitoids and lamprophyre dykes, combined with the structural pattern, geochemical features and regional tectonics, suggest a post-collision setting for the Qinling rapakivi-textured granitoid plutons. [Copyright &y& Elsevier]

Abstract: A petrological study of the Punta Orvili metabasite in NE Sardinia has been carried out, integrating quantitative pseudosection modelling with reaction balancing, with the aim of reconstructing the metamorphic evolution and P–T path. The Punta Orvili metabasite preserves microstructural evidence of: (i) a pre-symplectite polyphase and prograde stage (M1) indicated by the occurrence of amphibole inclusions in garnet, by Na-rich diopside and by compositional zoning in garnet, clinopyroxene and amphibole; (ii) a symplectite stage (M2) represented by the occurrence of Cpx2 +Pl1 symplectite; (iii) a corona stage (M3) documented by the formation of micrometre-thick Pl2 ±Am3 ±Ilm coronas around garnet; and (iv) a late stage (M4), documented by the growth of epidote and albite and by the replacement of biotite and clinopyroxene by chlorite. The M1 pre-symplectite stage has been modelled by P–T pseudosections calculated in the NCKFMASTH model system at a(H2O)=1 (for a(H2O)=0.5 temperature values are ~50°C lower). Mg and Ca zoning in garnet and Na zoning in clinopyroxene testify to a progressive increase in temperature and pressure during garnet and clinopyroxene growth from 610

Abstract: Spinel lherzolite and spinel harzburgite mantle xenoliths occur in the 4.56±0.2Ma Lutynia basanite in SW Poland. Only one studied xenolith contains minor pargasitic amphibole. Minerals forming the xenoliths are chemically unzoned. Olivine (Fo [(Mg/(Mg+Fetot))*100] 89.8–92.4) contains 0.34–0.45wt.% NiO and <780ppm Ca; orthopyroxene is Al enstatite (#mg [(Mg/(Mg+Fetot)*100] 0.90–0.92, Al 0.06–0.17a.pfu); clinopyroxene is Al–Cr diopside (#mg 0.91–0.93, Al 0.104–0.197a.pfu). Exsolved orthopyroxene occurs in the clinopyroxene and vice versa. Al–Mg spinel occurs in symplectites with clinopyroxene. Second generation crystals of olivine (Fo88.0–91.7, Ca up to 1800ppm), clinopyroxene and spinel occur in small interstitial patches containing feldspar. Clinopyroxene II is Al-poor (0.018–0.070a.pfu, #mg 0.921–0.932) when coexisting with alkali feldspar, but Al-rich (0.046–0.261a.pfu; #mg 0.907–0.925) when found with plagioclase. Four types of REE (rare earth elements) patterns are found in the clinopyroxene: (group A) flat HREE (heavy REE) with LREE (light REE) content increasing smoothly; (group B) flat HREE with an abrupt increase of MREE (medium REE) and LREE; (group C) LREE-enriched, flat HREE with negative inflection at MREE; (group D) smoothly LREE-enriched with no flat HREE pattern. The exception is clinopyroxene from xenolith MM30 which is extremely depleted in LREE. Clinopyroxene I from all the xenoliths (excluding MM30) contains high amounts of Th and U. Clinopyroxene trace element compositions record 8–15% of partial melting. Major and trace element compositions of minerals record later cryptic metasomatism induced by a CO2-bearing alkaline melt. Variable REE patterns for clinopyroxene I from groups A, B and C are due to chromatographic enrichment, with group A peridotites located close to the metasomatic source and group C being the furthest. The pargasite-bearing MM04 was probably the closest to the metasomatic source and thus recorded incipient modal metasomatism. Clinopyroxene–spinel symplectites and ortho-/clinopyroxene exsolutions suggest that the peridotites were transported from the garnet-lherzolite facies into the spinel one where they cooled and equilibrated at 960–1000°C. The patches formed by the second generation of minerals are the effect of pre-eruption infiltration by the basanitic melt. [Copyright &y& Elsevier]

Abstract: Ultramafic rocks in contact to gneisses in the Mt. Hochwart HP mélange (Eastern Italian Alps) preserve a series of metasomatic mineral zones. A phlogopitite with minor tremolite and accessory zircon and apatite forms close to the gneiss (Zone 1). Zone 2 consists of tremolite, phlogopite and anthophyllite followed by Zone 3 with anthophyllite plus minor chlorite and talc. Zone 3 grades into an amphibole–garnet peridotite lens. This reaction zone has been generated by infiltration of hydrous fluids at T of 660–700°C and P <1.2GPa, which occurred during exhumation of coupled continental crust and mantle peridotites. The reaction zone between a trace element-rich (gneiss) and a trace element depleted reservoir (peridotite) allows assessment of local trace element mobility in aqueous fluids. We present the results of in situ LA-ICP-MS trace element analysis of minerals from the reaction zone. Phlogopite is the main host for Large Ion Lithophile Elements (LILE) and contributes significantly to the Li, Ti, Nb, Ta, Pb and Sc budget. Anthophyllite is the main host for Li whereas all other trace elements including Rare Earth Elements (REE) are preferentially incorporated into tremolite. Combined with the abundance of these minerals over the contact zone, the mineral trace element data suggests that the LILE and REE were mobile on a small scale of a few centimetres only. Limited mobility of Ta, which is generally regarded as barely mobile in fluids, is documented in elevated contents of Ta in anthophyllite coupled with low Nb/Ta. The high Li content in minerals throughout the reaction zone suggests that Li was the most mobile element. The studied metasomatic zones mirror geochemical processes occurring in subduction zones at the slab–mantle interface. Phlogopite crystallization at the slab–mantle interface is an efficient mechanism to filter LILE from the aqueous fluid. Thus, such reaction zones, forming at temperatures <660–700°C, likely prevents that the typical slab signature with enriched LILE is transported by aqueous fluids over long distances in the mantle wedge. However, if coupled to the downgoing slab, phlogopite- and tremolite-rich rocks from such reaction zones might be able to act as carriers of trace elements and water into deeper parts of the subduction zone. [Copyright &y& Elsevier]

Abstract: Plagioclase crystals with an anorthite content up to An99 have been discovered in mantle peridotites from the Yungbwa ophiolite, which crops out along the Indus–Yarlung Zangbo Suture, southwestern Tibetan Plateau. Amphiboles of pargasitic composition also occur in these plagioclase peridotites. Microtextures support the view that both plagioclase and amphibole were formed metasomatically by hydrous melts entering the Yungbwa peridotites, during which pyroxene was dissolved and olivine was precipitated. The extreme Ca-rich nature of the plagioclase indicates that the metasomatic melts had high Ca/Na ratios, whereas low Ti contents in accompanying spinel and amphibole indicate severe depletion of Ti in the hydrous melts. This suggests that the metasomatic melts were derived from a refractory mantle source. Clinopyroxene trace element data indicate that the metasomatic melts were also enriched in light rare earth elements relative to both middle and heavy rare earth elements. Therefore, the melts show geochemical characteristics similar to arc magmas. This implies that the mantle peridotites in the Yungbwa ophiolite experienced a two-stage evolution; they were formed at a mid-ocean ridge and subsequently entered a subduction zone setting. Their history thus records the opening and closing of the Neo-Tethys Ocean. [Copyright &y& Elsevier]

Abstract: Amphibole in metasomatised mantle peridotites is commonly associated with inclusions and pools of silicate glass that contain microlites of spinel, olivine and clinopyroxene. There are two contrasting hypotheses for the origin of this association: 1) glass is the remnant of a metasomatic fluid or melt that is responsible for the crystallisation of amphibole. 2) Glass and the associated minerals are the result of amphibole breakdown during heating and decompression associated with the transport of the xenolith to the surface. Experiments over the range 1 atm to 1.5 GPa at temperatures from 950–1150 °C reproduce the textures, mineral assemblages and mineral and glass compositions observed in natural samples. Thus, it is most likely that the amphibole+glass+microlite assemblage in mantle xenoliths is an artefact of heating during transport rather than a clue to the agent of mantle metasomatism. [Copyright &y& Elsevier]

Abstract: In the Western Alps, the ocean-derived Lago di Cignana Unit of the Piemonte Zone has experienced ultrahigh-pressure metamorphism. We have studied the minerals of two eclogite samples from this unit using transmission electron microscopy (TEM) in order to characterise their microstructures. These microstructures are the result of deformation, phase transformation and reactions and allow conclusions on the processes that had generated them. The goal of this process-oriented paper is to contribute to the knowledge of the formation and exhumation of ultrahigh pressure eclogites. In our TEM-study we found omphacite, amphibole (barroisite, glaucophane, hornblende), clinozoisite, garnet, albite, and thin layers of chlorite and mica in amphibole. The omphacite is well-ordered and has the space-group P2/n. We observed antiphase domains (APDs), dislocations that are organised in small-angle grain boundaries, and – very rare – crystallographic shear planes parallel to (010) and (110). Deformation twin lamellae on (100) have not been observed. Most interesting is the first observation of faults parallel to (110) in a natural omphacite. They are due to deformation. Chain multiplicity faults are common in the amphibole grains. In one case, the orientation of crystallographic shear planes was not only parallel to (010), but also parallel to (130) and (−110). Clinozoisite showed deformation twin lamellae on (100) with widths of a few nm up to about 50 nm. Dislocations organised into a small-angle grain boundary have been found, which have not been described before. The garnet is almost free of dislocations. While there are differences regarding the deformation microstructures from the UHP Lago di Cignana eclogite and other HP occurrences, it is unlikely that these are due to higher pressures. It is much more likely that the whole formation and exhumation history of the different geological units and the individual positions of the collected rocks within them are reflected in their microstructures. A tentative correlation of the omphacite microstructures with the P–T path of the Lago di Cignana Unit is presented. An important point is the early crystallisation of C2/c-omphacite at a temperature below 500 °C, perhaps around 400 °C. The low temperature prevents rapid diffusion-controlled ordering to P2/nomphacite and thus enables deformation of the metastable C2/c-omphacite. Since C2/c-omphacite is more easily to deform than P2/n-omphacite, this may influence the interplate coupling. The onset of the transition to P2/n-omphacite, however, will increase the shear strength of eclogite and the coupling with the overriding plate, which may slow-down the subduction and finally stop it. [Copyright &y& Elsevier]

Abstract: The composition of mineral phases and their modal proportions have been determined for three representative Eoalpine eclogites from the Saualpe type locality/Eastern Austria (sample SKP31) and the Pohorje Massif/Slovenia (CM31/03 and CM15/01) using electron microprobe, laser ICP-MS, IR spectroscopy and modal analysis to evaluate possible mechanisms for the formation of composite oriented calcic amphibole+quartz inclusions (COIs) in omphacite and to assess the relative importance of hydrous and nominally anhydrous phases as H2O carriers in these eclogites. For omphacites in CM31/03 with a zonal distribution of COIs, a comparison of water and trace element concentrations of areas containing COIs and those free of COIs and a comparison with the trace element concentration of calcic amphibole indicate that COIs have formed through an open-system alteration of clinopyroxene and not through a closed system exsolution process. In sample SKP31, both textural and mineral chemical evidence suggests that COIs did not form by exsolution involving a Ca-Eskola component in clinopyroxene but formed by progressive growth under eclogite-facies P-T conditions and prior to the onset of retrogressive symplectite formation analogous to the formation of poikiloblastic quartz–calcic amphibole grains in the matrix. Bulk H2O contents of the eclogites are between ca. 750 and 2150 ppm with 6–25% of the total water contributed by nominally anhydrous minerals (NAMs). Because of high modal amounts of 37–65%, omphacite is the major nominally anhydrous water carrier, containing 145–580 ppm H2O with significant concentration variations on a thin section scale. Due to their very low H2O concentrations of <5–10 ppm (garnet, kyanite) or insignificant modal amounts ≤3% (rutile) the remaining NAMs contribute less than 1.5% to the bulk eclogite H2O content. Calcic amphibole forming part of COIs may be a major carrier of H2O as evidenced by CM31/03 containing both COIs and texturally primary calcic amphibole. In this sample calcic amphibole of the COIs contributes 63% whereas primary calcic amphibole only accounts for 13% of the bulk water. The relative order of H2O concentrations in NAMs is H2Orutile >H2Oomphacite ≫H2Ogarnet for CM31/03 and CM15/01 and H2Orutile ≈H2Oomphacite ≫H2Ogarnet in SKP31. [Copyright &y& Elsevier]

Abstract: The Mid-Proterozoic composite Ilímaussaq complex, South Greenland, is a classic locality to study magmatic layering in evolved peralkaline magmas. Most of the rock units show magmatic layering to differing extents, but ‘kakortokites’ – generally medium-grained agpaitic nepheline syenites – show a spectacular recurrence of black, red and white layers, which is due to regular changes in the modal contents of arfvedsonitic amphibole, eudialyte (sensu lato), alkali feldspar and nepheline, respectively. These three-layer units are found in the lower part of the intrusion and recur 29 times before grading into the overlying lujavrites (melanocratic agpaitic nepheline syenites), which are generally fine-grained and fissile with less-developed layering. The compositional trends observed in amphibole and eudialyte throughout the stratigraphic sequence reflect various processes including the chemical evolution of the melt by crystal fractionation, changes in the crystallising mineral assemblage and sub-solidus alteration. Eudialyte is the first mineral to crystallise in the investigated sequence and is therefore appropriate for recording evolution trends within the melt. Amphibole, on the other hand, always crystallises later and is therefore affected by other crystallising minerals. A detailed microprobe study of both minerals through the whole kakortokite stratigraphy displays surprisingly little change in mineral compositions within the kakortokites, but strong fractionation trends in the overlying lujavrites. Although various models have been proposed to explain the recurrence of the 29 rhythmic units, the origin of this prominent magmatic layering in the kakortokites and the lack of mineralogical and strong mineral chemical changes has not been quantitatively explained. We propose, that in the kakortokites, minerals were probably separated from each other as a result of their different densities. The interior cooled, resulting in crystallization but only a very small proportion of crystals (0.1–0.3% for each of the four minerals) could remain suspended in the melt before gravity forced them to settle down in a stagnant layer of reduced convection. A combination of volatile pressure variations caused by eruptive activity and repeated replenishment can explain the oscillating liquidus temperature, the small changes in mineral compositions and such a process would produce enough crystals to form the 29 layers. [Copyright &y& Elsevier]