Jadeitite, known as 'hisui' in Japan, has been esteemed as sacred stone in both ancient and modern Japanese cultures. Although it was thought that the source material of Japanese jadeitite was brought from China, the identification of jadeite in 1939 changed this interpretation. Japanese jadeitites and jadeite-rich metasomatic rocks are found in Paleozoic and also Mesozoic geotectonic units. All localities are situated in serpentinite melange with high-pressure metamorphic rocks and/or serpentinite lenses within a high-pressure metamorphosed complex. Outcrop exposures of contact between jadeitite and host serpentinite are extremely rare. Normally the jadeitites show lithological heterogeneity in the same locality due to multiple deformation, recrystallization, and metasomatic fluid infiltration. Studies over the last two decades have interpreted jadeitite in worldwide either as the direct aqueous fluid precipitate (P-type) from subduction channel into the overlying mantle wedge, as the metasomatic replacement (R-type) by such fluids of oceanic plagiogranite, graywacke, or metabasite along the channel margin, or a combination of these two processes. Japanese jadeitites are classified into one or the other type. Multiple stable isotope characterization analyses for jadeitite and related metasomatic rocks and serpentinite become increasingly important to decode fluid behaviors in past subduction zone. However, available geochemical data on Japanese jadeitite are very limited in comparison with other studied localities. More systematic research will unlock new insights about fluid flow and its impacts at the bottom of forearc mantle where jadeitites form. Chemical differentiation and transportation of the fluids involved in jadeitite-formation are crucial topics requiring further research. Nevertheless, the designation of jadeite (and jadeitite) as the national stone of Japan by the Japan Association of Mineralogical Sciences in 2016 should bolster education of the public about this revered stone and its role in subduction zone processes.

Paleozoic jadeitite-bearing serpentinite-matrix melange represents the oldest mantle wedge record of a Pacific-type subduction zone of proto-Japan. Most jadeitites are fluid precipitates (P-type), but some jadeitites are metasomatic replacement (R-type) which preserve relict minerals and protolith textures. The beauty and preciousness of some gem-quality, semi-translucent varieties of jadeitites in the Itoigawa-Omi area led to the designation of jadeitite as the national stone of Japan by the Japan Association of Mineralogical Sciences. Zircon geochronology indicates jadeitite formed prior to Late Paleozoic Renge metamorphism that formed blueschist and rare eclogite. For example, in the Itoigawa-Omi and Osayama localities, older jadeitites and younger high-pressure/ low-temperature metamorphic rocks in a single melange complex imply different histories for the subduction channel and jadeite-bearing serpentinite-matrix melange. This suggests that the jadeitite-hosted melange (or serpentinized peridotite) can stay within the mantle wedge for a considerable time; thus recrystallization, resorption, and re-precipitation of jadeitite can continue in the mantle wedge environment. Therefore, studies of Paleozoic jadeitites in Japan have great potential to elucidate the earliest stages of orogenic growth (oceanward-accretion and landward-erosion) associated with the subduction of the paleo-Pacific oceanic plates, and to test geophysical observations of modern analogues from a mixture of fossilized mantle wedges and subduction channels.

Jadeitite from the New Idria serpentinite body of California is a fluid precipitation-to-metasomatic product. Optical cathodoluminescence (CL) microscopy of the jadeitite revealed that vein-filling 'pure' jadeites (mostly 97-99.9 mol% jadeite) exhibit bright luminescence, whereas 'impure' jadeites (mostly 75-95 mol% jadeite) in pale-greenish matrix show dark luminescence. The 'pure' jadeites in the veins are composed of mixtures of red, blue and dull blue CL-colored domains, showing growth textures (oscillatory bands). The 'impure' jadeites in the pale-greenish matrix with dark luminescence have a higher augite component (up to 5.37 wt% FeO), implying that the CL property is due to significant amount of Fe2+ to act as a quencher. CL spectra of the blue CL-colored domains of the vein-filling 'pure' jadeite have a doublet broad emission peak centered at similar to 320 and similar to 360 nm in the ultraviolet (UV) to blue region. In the red CL-colored domains, a broad asymmetric emission peak at similar to 700 nm is also recognized together with the doublet UV-blue emission peak. Comparing monochromatic CL images in the UV-blue (300-400 nm) and red (650-750 nm) emission regions with X-ray elemental maps, luminescence centers contributing the UV-blue and red CL emission peaks were assigned. The red emission peak of the 'pure' jadeite with subtle augite component would be attributed to lattice defects related to Ca2+, Fe2+ (or Fe3+) and Mg2+ deficiency and/or excess centers in M1 or M2 sites. Alternatively, transition metal ions (Mn2+ and Fe3+) or rare earth elements in the M1 and M2 sites as impurity centers, might contribute to the red emission peak. As the UV-blue emissions correlate with Al3+ content, i.e. purity of jadeite component, they might be related to Na+ and/or Al3+ defect centers.

Field and petrologic characteristics of two new eclogite localities within the Guatemala Suture Complex (GSC) north of the Motagua Fault are presented. The Tuncaj Hill locality exposes a coherent body of retrogressed eclogite hundreds of metres long that is associated with serpentinite of the North Motagua Unit. The Tanilar River locality exposes numerous bands and lenses of eclogite hosted in sialic gneisses of the Chuacus Complex. The Tuncaj eclogite has a two-stage prograde evolution containing the peak assemblage Grt + Omp + Ttn + Czo + Zo +/- Am, formed at temperatures <720 degrees C. In contrast, eclogites of the Tanilar unit are characterized by the paragenesis Omp + Grt + Rt +/- Phg +/- Qtz +/- Ep giving higher peak conditions of T=720-830 degrees C and P=2.1-2.7 GPa, near the stability field of coesite. Previously obtained data and our thermobaric calculations suggest distinct petrotectonic evolutions for the various types of eclogites within the suture. The lawsonite eclogites south of the Motagua Fault were probably produced in a mature Farallon subduction zone during the Early Cretaceous. The northern high-pressure (HP) blocks in serpentinite melange and coherent amphibolite bodies with eclogite relics were generated by the Early Cretaceous subduction of the proto-Caribbean lithosphere under the Great Caribbean Arc. A continental block, the North American passive margin, reached the arc's trench in the Campanian and was subducted to ca. 80km depth, producing the eclogites of the Chuacus Complex. As the slab was delaminated and partially exhumed, the continental Chuacus became tectonically juxtaposed with HP blocks of the proto-Caribbean that had been accreted to the Caribbean plate forming the North Motagua Unit. The juxtaposed group migrated to mid-crustal level and was contemporaneously retrogressed under epidote-amphibolite facies conditions.

Serpentinites (massive and schistose) and listvenite occur as tectonic sheets and lenses within a calcareous metasedimentary melange of the Tulu Dimtu, western Ethiopia. The massive serpentinite contains high-magnesian metamorphic olivine (forsterite [fo] similar to 96mol%) and rare relict primary mantle olivine (Fo(90-93)). Both massive and schistose serpentinites contain zoned chromian spinel; the cores with the ferritchromite rims preserve a pristine Cr/(Cr+Al) atomic ratio (Cr#=0.79-0.87), suggesting a highly depleted residual mantle peridotite, likely formed in a suprasubduction zone setting. Listvenite associated with serpentinites of smaller ultramafic lenses also contain relict chromian spinel having identical Cr# to those observed in serpentinites. However, the relict chromian spinel in listvenite has significantly higher Mg/(Mg+Fe2+) atomic ratios. This suggests that a nearly complete metasomatic replacement of ultramafic rocks by magnesite, talc, and quartz to prevent Mg-Fe2+ redistribution between relict chromian spinel and the host, that is, listvenite formation, took place prior to re-equilibration between chromian spinel and the surrounding mafic minerals in serpentinites. Considering together with the regional geological context, low-temperature CO2-rich hydrothermal fluids would have infiltrated into ultramafic rocks from host calcareous sedimentary rocks at a shallow level of accretionary prism before a continental collision to form the East African Orogen (EAO).

Intense devolatilization and chemical-density differentiation attended accretion of planetesimals on the primordial Earth. These processes gradually abated after cooling and solidification of an early magma ocean. By 4.3 or 4.2 Ga, water oceans were present, so surface temperatures had fallen far below low-pressure solidi of dry peridotite, basalt, and granite, similar to 1300, similar to 1120, and similar to 950 degrees C, respectively. At less than half their T solidi, rocky materials existed as thin lithospheric slabs in the near-surface Hadean Earth. Stagnant- lid convection may have occurred initially but was at least episodically overwhelmed by subduction because effective, massive heat transfer necessitated vigorous mantle overturn in the early, hot planet. Bottom-up mantle convection, including voluminous plume ascent, efficiently rid the Earth of deep-seated heat. It declined over time as cooling and top-down lithospheric sinking increased. Thickening and both lateral extensional + contractional deformation typified the post-Hadean lithosphere. Stages of geologic evolution included (i) 4.5-4.4 Ga, magma ocean overturn involved ephemeral, surficial rocky platelets; (ii) 4.4-2.7 Ga, formation of oceanic and small continental plates were obliterated by return mantle flow prior to similar to 4.0 Ga; continental material gradually accumulated as largely sub-sea, sialic crust-capped lithospheric collages; (iii) 2.7-1.0 Ga, progressive suturing of old shields + younger orogenic belts led to cratonal plates typified by emerging continental freeboard, increasing sedimentary differentiation, and episodic glaciation during transpolar drift; onset of temporally limited stagnant-lid mantle convection occurred beneath enlarging supercontinents; (iv) 1.0 Ga-present, laminar-flowing asthenospheric cells are now capped by giant, stately moving plates. Near-restriction of komatiitic lavas to the Archean, and appearance of multicycle sediments, ophiolite complexes +/- alkaline igneous rocks, and high-pressure-ultrahigh-pressure (HP-UHP) metamorphic belts in progressively younger Proterozoic and Phanerozoic orogens reflect increasing negative buoyancy of cool oceanic lithosphere, but decreasing subductability of enlarging, more buoyant continental plates. Attending supercontinental assembly, density instabilities of thickening oceanic plates began to control overturn of suboceanic mantle as cold, top-down convection. Over time, the scales and dynamics of hot asthenospheric upwelling versus lithospheric foundering + mantle return flow (bottom- up plume-driven ascent versus top-down plate subduction) evolved gradually, reflecting planetary cooling. These evolving plate-tectonic processes have accompanied the Earth's thermal history since similar to 4.4 Ga.

We want to know when plate tectonics began and will consider any important Earth feature that shows significant temporal evolution. Kimberlites, the primary source of diamonds, are rare igneous features. We analyze their distribution throughout Earth history; most are young (similar to 95% are younger than 0.75 Ga), but rare examples are found as far back as the Archean (older than 2.5 Ga). Although there are differing explanations for this age asymmetry (lack of preservation, lack of exposure, fewer mantle plumes, or lack of old thick lithosphere in the Archean and Proterozoic), we suggest that kimberlite eruptions are a consequence of modernstyle plate tectonics, in particular subduction of hydrated oceanic crust and sediments deep into the mantle. This recycling since the onset of modern-style plate tectonics ca. 1 Ga has massively increased mantle CO2 and H2O contents, leading to the rapid and explosive ascent of diamond-bearing kimberlite magmas. The age distribution of kimberlites, combined with other large-scale tectonic indicators that are prevalent only in the past similar to 1 Ga (blueschists, glaucophane-bearing eclogites; coesite-or diamond-bearing ultrahigh-pressure metamorphic rocks; lawsonite-bearing metamorphic rocks; and jadeitites), indicates that plate tectonics, as observed today, has only operated for <25% of Earth history.

The Late Jurassic English Peak plutonic complex was emplaced in an upper crustal retro-arc setting in the central Klamath Mountains province, northern California. Emplacement of the main, central pluton was preceded by intrusion of two satellite bodies: the Uncles Creek pluton crystallized from H2O-rich quartz dioritic magma with hornblende as the liquidus mafic phase; in contrast, the Heiney Bar pluton is a c. 2.5km diameter body zoned from gabbro to granodiorite. Al-in-hornblende barometry from these two plutons indicates a stage of magma storage at c. 600-500 MPa. The central English Peak pluton is a c. 15km diameter body composed of early and late stages. Early stage rocks range from gabbro to tonalite, with variable proportions of augite, orthopyroxene, hornblende and biotite. The early stage lacks discernible zoning and rock types vary at the outcrop scale. This diversity is reflected in bulk-rock compositions, which do not form a compositional array. The late-stage intrusion consists of three concentric units that are zoned from outer, more mafic rocks (quartz diorite, tonalite, quartz monzodiorite) to inner, compositionally evolved rocks (granodiorite and granite). Late-stage samples plot in smooth, typically linear arrays for most major and trace elements. Al-in-hornblende pressures indicate that late-stage hornblende cores grew in a reservoir at c. 400MPa and that rims grew at the level of final emplacement (c. 250 MPa). The mid-crustal reservoir was the site of late-stage magma evolution, including episodic magma mixing. Oxygen and Sr isotopes indicate initial evolution of English Peak pluton magmas in a deep crustal region of mixing, assimilation, storage, and homogenization (MASH zone), where they were contaminated by metasedimentary rocks. Thus, the English Peak pluton represents a crustalscale system, with mantle-derived magmas that differentiated near the Moho, storage and crystallization of satellite-pluton magmas in the middle crust (c. 600-500 MPa), development of a large, episodically recharged, magma chamber in the upper middle crust (c. 400 MPa) and final emplacement in the upper crust.

We report the origin of isotope fractionation induced by the choice of laser parameters and a method for accurate in situ determination of lithium (delta Li-7) and boron (delta B-11) isotope ratios in glasses and minerals using laser ablation multiple Faraday collector inductively coupled plasma mass spectrometry (LA-MFC-ICPMS). Laser ablation parameters were examined using 266 nm femtosecond (266FsLA) and 193 nm nanosecond excimer (193ExLA) laser ablation systems for crater diameters of 30-200 mu m. We found that higher laser repetition rates and larger crater diameters have led to enhanced fractionation of lighter isotopes, as much as -8 parts per thousand for both delta Li-7 and delta B-11. Fractionation was primarily affected by the ICP aerosol loading and secondly by the thermal fractionation at the LA site. The former was accounted for by mass loading effects, which lowered the plasma temperature and led to insufficient aerosol vaporisation. The latter was related to the molten layer on the crater walls, which resulted in coarser and heavier delta Li-7 and delta B-11 aerosols that did not reach the ICP. Both processes can result in Rayleigh fractionation during aerosol formation and vaporisation. Controlled ablation using a constant crater size, repetition rate, and high laser fluence of 193ExLA enabled reproducible ablation for the standard NIST SRM 61X glasses and unknown basalt glasses. Based on the principles of isotopic fractionation deduced from our experiments, we propose a novel ablation volume correction (AVC) protocol for accurate isotopic analyses of various samples with different matrices. Both the repeatability and the laboratory bias of the delta Li-7 and delta B-11 measurements using the new AVC protocol were better than 1 parts per thousand for samples containing a few tens to a few tens of thousands ppm Li and B. We also report significant local heterogeneity of up to several parts per thousand found in some basalt glasses, but not in NIST SRM 612 and 610.

White mica (phengite and paragonite) K-Ar ages of eclogite-facies Sanbagawa metamorphic rocks (15 eclogitic rocks and eight associated pelitic schists) from four different localities yielded ages of 84-89Ma (Seba, central Shikoku), 78-80Ma (Nishi-Iratsu, central Shikoku), 123 and 136Ma (Gongen, central Shikoku), and 82-88Ma (Kotsu/Bizan, eastern Shikoku). With the exception of a quartz-rich kyanite-bearing eclogite from Gongen, white mica ages overlap with the previously known range of phengite K-Ar ages of pelitic schists of the Sanbagawa metamorphic belt and can be distinguished from those of the Shimanto metamorphic belt. The similarity of K-Ar ages between the eclogites and surrounding pelitic schists supports a geological setting wherein the eclogites experienced intense ductile deformation with pelitic schists during exhumation. In contrast, phengite extracted from the Gongen eclogite, which is less overprinted by a ductile shear deformation during exhumation, yielded significantly older ages. Given that the Gongen eclogite is enclosed by the Higashi-Akaishi meta-peridotite body, these K-Ar ages are attributed to excess Ar-40 gained during an interaction between the eclogite and host meta-peridotite with mantle-derived noble gas (very high Ar-40/Ar-36 ratio) at eclogite-facies depth. Fluid exchange between deep-subducted sediments and mantle material might have enhanced the gain of mantle-derived extreme Ar-40 in the meta-sediment. Although dynamic recrystallization of white mica can reset the Ar isotope system, limited-argon-depletion due to lesser degrees of ductile shear deformation of the Gongen eclogite might have prevented complete release of the trapped excess argon from phengites. This observation supports a model of deformation-controlled K-Ar closure temperature.

Jadeitite is a relatively rare, very tough rock composed predominantly of jadeite and typically found associated with tectonic blocks of high-pressure/low-temperature metabasaltic rocks (e.g., eclogite, blueschist) in exhumed serpentinite-matrix melanges. Studies over the past similar to 20 years have interpreted jadeitite either as the direct hydrous fluid precipitate from subduction channel dewatering into the overlying mantle wedge or as the metasomatic replacement by such fluids of oceanic plagiogranite, graywacke, or metabasite along the channel margin. Thus, jadeitites directly sample and record fluid transport in the subduction factory and provide a window into this geochemical process that is critical to a major process in the Earth system. They record the remarkable transport of large ion lithophile elements, such as Li, Ba, Sr, and Pb, as well as elements generally considered more refractory, such as U, Th, Zr, and Hf. Jadeitite is also the precious form of jade, utilized since antiquity in the form of tools, adornments, and symbols of prestige.

Newly recognized occurrences of ultrahigh-pressure (UHP) minerals including diamonds in ultrahigh-temperature (UHT) felsic granulites of orogenic belts, in chromitites associated with ophiolitic complexes, and in mantle xenoliths suggest the recycling of crustal materials through deep subduction, mantle upwelling, and return to the Earth's surface. This circulation process is supported by crust-derived mineral inclusions in deep-seated zircons, chromites, and diamonds from collision-type orogens, from eclogitic xenoliths in kimberlites, and from chromitities of several Alpine-Himalayan and Polar Ural ophiolites; some of these minerals contain low-atomic number elements typified by crustal isotopic signatures. Ophiolite-type diamonds in placer deposits and as inclusions in chromitites together with numerous highly reduced minerals and alloys appear to have formed near the mantle transition zone. In addition to ringwoodite and inferred stishovite, a number of nanometric minerals have been identified as inclusions employing state-of-the-art analytical tools. Reconstitution of now-exsolved precursor UHP phases and recognition of subtle decompression microstructures produced during exhumation reflect earlier UHP conditions. For example, Tibetan chromites containing exsolution lamellae of coesite + diopside suggest that the original chromitites formed at P > 9-10 GPa at depths of >250-300 km. The precursor phase most likely had a Ca-ferrite or a Ca-titanite structure; both are polymorphs of chromite and (at 2000 degrees C) would have formed at minimum pressures of P > 12.5 or 20 GPa respectively. Some podiform chromitites and host peridotites contain rare minerals of undoubted crustal origin, including zircon, feldspars, garnet, kyanite, andalusite, quartz, and rutile; the zircons possess much older U-Pb ages than the time of ophiolite formation. These UHP mineral-bearing chromitite hosts evidently had a deep-seated evolution prior to extensional mantle upwelling and partial melting at shallow depths to form the overlying ophiolite complexes. These new findings together with stable isotopic and inclusion characteristics of diamonds provide compelling evidence for profound underflow of both oceanic and continental lithosphere, recycling of surface 'organic' carbon into the lower mantle, and ascent to the Earth's surface through mantle upwelling. Intensified study of UHP granulite-facies lower crustal basement and ophiolitic chromitites should allow a better understanding of the geodynamics of subduction and crustal cycling. (C) 2014 Elsevier Ltd. All rights reserved.

Oxygen isotope compositions are reported for the first time for the Himalayan metabasites of the Kaghan Valley, Pakistan in this study. The highest metamorphic grades are recorded in the north of the valley, near the India-Asia collision boundary, in the form of high-pressure (HP: Group I) and ultrahigh-pressure (UHP: Group II) eclogites. The rocks show a step-wise decrease in grade from the UHP to HP eclogites and amphibolites. The protoliths of these metabasites were the Permian Panjal Trap basalts (ca. 267 +/- 2.4 Ma), which were emplaced along the northern margin of India when it was part of Gondwana. After the break-up of Gondwana, India drifted northward, subducted beneath Asia and underwent UHP metamorphism during the Eocene (ca. 45 +/- 1.2 Ma). At the regional scale, amphibolites, Group I and II eclogites yielded delta O-18 values of +5.84 and +5.91 parts per thousand, +1.66 to +424 parts per thousand, and -2.25 to +0.76 parts per thousand, respectively, relative to VSMOW. On a more local scale, within a single eclogite body, the delta O-18 values were the lowest (-2.25 to-1.44%.) in the central, the best preserved (least retrograded) parts, and show a systematic increase outward into more retrograded rocks, reaching up to +0.12 parts per thousand. These values are significantly lower than the typical mantle values for basalts of + 5.7 +/- 0.3 parts per thousand. The unusually low or negative delta O-18 values in Group II eclogites potentially resulted from hydrothermal alteration of the protoliths by interactions with meteoric water when the Indian plate was at southern high latitudes (similar to 60 degrees S). The stepwise increase in delta O-18 values, among different eclogite bodies in general and at single outcrop-scales in particular, reflects differing degrees of resetting of the oxygen isotope compositions during exhumation-related retrogression. (C) 2014 Elsevier B.V. All rights reserved.

Lawsonite-bearing metamorphic/metasomatic rocks form at high-pressure-ultrahigh pressure (HP-UHP) and low-temperature (LT) conditions, commonly in Pacific-type subduction zones. The P-T stability fields of lawsonite blueschists and lawsonite eclogites represent subfacies of the blueschist and eclogite facies, respectively. Although the lawsonite-epidote transition boundary has a positive Clapeyron (dP/dT) slope, the blueschist-to-eclogite transformation within the lawsonite stability field in metabasaltic rocks is gradual and cannot be defined by a specific discontinuous reaction in P-T space. The oldest occurrences of lawsonite-bearing blueschists are latest Neoproterozoic, suggesting that subduction-zone thermal structures evolved towards the necessary LT conditions for lawsonite formation only by the late Neoproterozoic. A clear difference in frequency between Phanerozoic lawsonite and epidote blueschists does not exist, but our new compilation found a global lawsonite hiatus in the Permian that is a robust indication of relatively warm subduction-zone thermal regimes. Lawsonite eclogites have been confirmed from at least 19 localities; they are classified as L- (lawsonite only), E- (lawsonite+epidote), and U-type (lawsonite+coesite). Complete preservation of L-type lawsonite eclogites attending their return to the surface is uncommon. Rare evidence of progressive eclogitization within the lawsonite stability field is preserved in some zoned garnets, as growth isolates a significant volume of precursor phases and textures during incipient eclogitization. Brittle fracturing and fluid infiltration are common during prograde eclogite facies metamorphism. Certain lawsonite-bearing metasomatic rocks record multiple fluid-infiltration events. Significant cooling and continuous H2O supply from the dehydrating oceanic plate to exhuming HP serpentinite melange may cause lawsonite blueschist facies overprinting and prevent breakdown of lawsonite during decompression. The subduction records of lawsonite blueschists and eclogites agree with numerical modelling of subduction zones.

The Shanderman eclogites and related metamorphosed oceanic rocks mark the site of closure of the Palaeotethys ocean in northern Iran. The protolith of the eclogites was an oceanic tholeiitic basalt with MORB composition. Eclogite occurs within a serpentinite matrix, accompanied by mafic rocks resembling a dismembered ophiolite. The eclogitic mafic rocks record different stages of metamorphism during subduction and exhumation. Minerals formed during the prograde stages are preserved as inclusions in peak metamorphic garnet and omphacite. The rocks experienced blueschist facies metamorphism on their prograde path and were metamorphosed in eclogite facies at the peak of metamorphism. The peak metamorphic mineral paragenesis of the rocks is omphacite, garnet (pyrope-rich), glaucophane, paragonite, zoisite and rutile. Based on textural relations, post-peak stages can be divided into amphibolite and greenschist facies. Pressure and temperature estimates for eclogite facies minerals (peak of metamorphism) indicate 15-20kbar at similar to 600 degrees C. The pre-peak blueschist facies assemblage yields <11kbar and 400-460 degrees C. The average pressure and temperature of the post-peak amphibolite stage was 5-6kbar, similar to 470 degrees C. The Shanderman eclogites were formed by subduction of Palaeotethys oceanic crust to a depth of no more than 75km. Subduction was followed by collision between the Central Iran and Turan blocks, and then exhumation of the high pressure rocks in northern Iran.

Investigations of microstructures are crucial if we are to understand the seismic anisotropy of subducting oceanic crust, and here we report on our systematic fabric analyses of glaucophane, lawsonite, and epidote in naturally deformed blueschists from the Diablo Range and Franciscan Complex in California, and the Hida Mountains in Japan. Glaucophanes in the analyzed samples consist of very fine grains that are well aligned along the foliation and have high aspect ratios and strong crystal preferred orientations (CPOs) characterized by a (1 00)[001] pattern. These characteristics, together with a bimodal distribution of grain sizes from some samples, possibly indicate the occurrence of dynamic recrystallization for glaucophane. Although lawsonite and epidote display high aspect ratios and a strong CPO of (001)[010], the occurrence of straight grain boundaries and euhedral crystals indicates that rigid body rotation was the dominant deformation mechanism. The P-wave (AV(P)) and S-wave (AV(S)) seismic anisotropies of glaucophane (AV(P) = 20.4%, AV(S) = 11.5%) and epidote (AV(P) = 9.0%, AV(S) = 8.0%) are typical of the crust; consequently, the fastest propagation of P-waves is parallel to the [001] maxima, and the polarization of S-waves parallel to the foliation can form a trench-parallel seismic anisotropy owing to the slowest V-S polarization being normal to the subducting slab. The seismic anisotropy of lawsonite (AV(P) = 9.6%, AV(S) = 19.9%) is characterized by the fast propagation of P-waves subnormal to the lawsonite [001] maxima and polarization of S-waves perpendicular to the foliation and lineation, which can generate a trench-normal anisotropy. The AV(S) of lawsonite blueschist (5.6-9.2%) is weak compared with that of epidote blueschist (8.4-11.1%). Calculations of the thickness of the anisotropic layer indicate that glaucophane and lawsonite contribute to the trench-parallel and trench-normal seismic anisotropy beneath NE Japan, but not to that beneath the Ryukyu arc. Our results demonstrate, therefore, that lawsonite has a strong influence on seismic velocities in the oceanic crust, and that lawsonite might be the cause of complex anisotropic patterns in subduction zones. (c) 2013 Elsevier B.V. All rights reserved.

Newly recognized occurrences of ultrahigh-pressure minerals in ultrahigh-temperature felsic granulites and in chromitite associated with ophiolite complexes lead to speculation about the recycling of supracrustal materials through deep subduction, mantle upwelling, and return to the Earth's surface. This idea is supported by possible "organic" light carbon isotopes observed in diamond and moissanite in kimberlite xenoliths and chromitite. These findings, presage another renaissance in the study of UHP granulite fades lower-crustal basement and ophiolites and in our understanding of the geodynamics of continental subduction and mantle cycling.

The gemstones jadeite and ruby generally form as a result of the plate tectonic processes subduction and collision. Jade made of jadeite (jadeitite) forms when supercritical fl uids released from subducting oceanic crust condense in the overlying mantle wedge, 20-120 km deep in the Earth. Jadeitite deposits thus mark the location of exhumed fossil subduction zones. Ruby, the red gem variety of corundum, forms during amphibolite- and granulite-facies metamorphism or melting of mixed Al-rich and Si-poor protoliths, 10-40 km deep in the crust. Suitable conditions generally exist where passive-margin carbonates and shales are involved in continental collision. Most ruby deposits formed during Ediacaran-Cambrian (ca. 550 Ma) collisions that produced the East African-Antarctic orogen and the supercontinent Gondwana, or during Cenozoic collisions in south Asia. Ruby is thus a robust indicator of continental collision. As a result of these diagnostic properties, we propose the term "plate tectonic gemstones" (PTGs) for jadeitite and ruby. The PTGs are a new type of petrotectonic indicator that are mostly found in Neoproterozoic and younger rocks. The PTGs as petrotectonic indicators that form deep in the Earth have the added advantage that their record is unlikely to be obliterated by erosion, although the possibility of destruction via retrogression needs to be further assessed. Recognition of the PTGs links modern concepts of plate tectonics to economic gemstone deposits and ancient concepts of beauty, and may aid in exploration for new deposits. © 2013 Geological Society of America.

Deformation microstructures for a lawsonite blueschist from the New Idria serpentinite body, Diablo Range, are investigated to clarify rheological behaviors of glaucophane and lawsonite, which are main mineral assemblages of subducting oceanic crust at relatively cold geotherm. Developments of crystal-preferred orientations (CPOs) with small grain size, irregular grain boundary and high aspect ratio of glaucophane indicate deformation mechanism as recovery and dynamic recrystallization possibly accommodated by dislocation creep, while lawsonite deforms by rigid body rotation based on euhedral grains with angular or straight grain boundaries. Higher aspect ratios, lower angle to foliation, and stronger CPOs of both minerals in the glaucophane-rich layer rather than those in the lawsonite-rich layer suggest the strain localization into the glaucophane-rich layer. Additionally fabric strength (the degree of crystal alignment) and seismic anisotropy are higher in the glaucophane-rich layer than that of the lawsonite-rich layer, which is consistent with the microstructural analyses. All our results imply, therefore, the dominant role of glaucophane rather than lawsonite for rheological behavior and seismic anisotropy of blueschist.

We report ion microprobe U-Th-Pb geochronology of in situ zircon from the Himalayan high- and ultrahigh-pressure eclogites, Kaghan Valley of Pakistan. Combined with the textural features, mineral inclusions, cathodoluminescence image information and the U-Th-Pb isotope geochronology, two types of zircons were recognized in Group I and II eclogites. Zircons in Group I eclogites are of considerably large size (>100 mu m up to 500 mu m). A few grains are euhederal and prismatic, show oscillatory zoning with distinct core-rim luminescence pattern. Several other grains show irregular morphology, mitamictization, embayment and boundary truncations. They contain micro-inclusions such as muscovite, biotite, quartz and albite. Core or middle portions of zircons from Group I eclogites yielded concordant U-Th-Pb age of 267.6 +/- 2.4 (MSWD = 8.5), have higher U and Th contents with a Th/U ratio > 1, indicating typical magmatic core domains. Middle and rim or outer portions of these zircons contain inclusions of garnet, omphacite, phengite and these portions show no clear zonation. They yielded discordant values ranging between 210 and 71 Ma, indicating several thermal or Pb-loss events during their growth and recrystalization prior to or during the Himalayan eclogite-facies metamorphism. Zircons in Group II eclogites are smaller in size, prismatic to oval, display patchy or sector zoning and contain abundant inclusions of garnet, omphacite, phengite, quartz, rutile and carbonates. They yielded concordant U-Th-Pb age of 44.9 +/- 1.2 Ma (MSWD = 4.9). The lower U and Th contents and a lower Th/U ratio (<0.05) in these zircons suggest their formation from the recrystallization of the older zircons during the Himalayan high and ultrahigh-pressure eclogite-facies metamorphism. (C) 2012 Elsevier Ltd. All rights reserved.

The Tia Complex in the southern New England Fold Belt is a poly-metamorphosed Late Paleozoic accretionary complex. It consists mainly of high-P/low-T type pumpellyite-actinolite facies (rare blueschist facies) schists, phyllite and serpentinite (T = 300 degrees C and P = 5 kbar), and low-P/high-T type amphibolite facies schist and gneiss (T = 600 degrees C and P < 5 kbar) associated with granodioritic plutons (Tia granodiorite). White mica and biotite K-Ar ages distinguish Carboniferous subduction zone metamorphism and Permian granitic intrusions, respectively. The systematic K-Ar age mapping along a N-S traverse of the Tia Complex exhibits a gradual change. The white mica ages become younger from the lowest-grade zone (339 Ma) to the highest-grade zone (259 Ma). In contrast, Si content of muscovite changes drastically only in the highest-grade zone. The regional changes of white mica K-Ar ages and chemical compositions of micas indicate argon depletion from precursor high-P/low-T type phengitic white mica during the thermal overprinting and recrystallization by granitoids intrusions. Our new K-Ar ages and available geological data postulate a model of the eastward rollback of a subduction zone in Early Permian. The eastward shift of a subduction zone system and subsequent magmatic activities of high-Mg andesite and adakite might explain formation of S-type granitoids (Hillgrove suite) and coeval low-P/high-T type metamorphism in the Tia Complex. (C) 2012 Elsevier Ltd. All rights reserved.

Records of micrometeorite collisions at down to submicron scales were discovered on dust grains recovered from near-Earth asteroid 25143 (Itokawa). Because the grains were sampled from very near the surface of the asteroid, by the Hayabusa spacecraft, their surfaces reflect the low-gravity space environment influencing the physical nature of the asteroid exterior. The space environment was examined by description of grain surfaces and asteroidal scenes were reconstructed. Chemical and O isotope compositions of five lithic grains, with diameters near 50 mu m, indicate that the uppermost layer of the rubble-pile-textured Itokawa is largely composed of equilibrated LL-ordinary-chondrite-like material with superimposed effects of collisions. The surfaces of the grains are dominated by fractures, and the fracture planes contain not only sub-mu m-sized craters but also a large number of sub-mu m-to several-mu m-sized adhered particles, some of the latter composed of glass. The size distribution and chemical compositions of the adhered particles, together with the occurrences of the sub-mu m-sized craters, suggest formation by hypervelocity collisions of micrometeorites at down to nm scales, a process expected in the physically hostile environment at an asteroid's surface. We describe impact-related phenomena, ranging in scale from 10(-9) to 10(4) meters, demonstrating the central role played by impact processes in the long-term evolution of planetary bodies. Impact appears to be an important process shaping the exteriors of not only large planetary bodies, such as the moon, but also low-gravity bodies such as asteroids.

Jadeitite-bearing serpentinite-matrix melange is distributed in the Caribbean (Guatemala, Cuba, and Dominican Republic), circum-Pacific (Japan, Western USA, and Papua New Guinea), Alpine-Himalayan (Italy, Iran, Greece, and Myanmar), and Caledonian (Russia and Kazakhstan) orogenic belts, and always contains high-pressure, low-temperature (HP-LT) metamorphic rocks. There are also jadeitite xenoliths in kimberlitic pipes in the Colorado Plateau (USA). The oldest occurrences of jadeitite are Early Paleozoic in Japan, Russia, and Kazakhstan, suggesting subduction-zone thermal structures evolved the necessary high pressure/temperature conditions for jadeitite formation since Early Paleozoic; the youngest occurrence is a xenolith from the Colorado Plateau. Major occurrences consist principally of fluid precipitates (P-type) that infiltrated the mantle wedge; fewer occurrences document metasomatic replacement (R-type) of plagiogranite, metagabbro and eclogite, and both types may be possible in the same occurrence or system. The P-T conditions for jadeitite formation can be extended beyond the previously argued limits of blueschist-facies conditions. Some jadeitite formed at epidote amphibolite and others at eclogite facies conditions. Available geochronological data of both jadeitite and associated HP-LT rock show temporal discrepancies between jadeitite formation and HP-LT metamorphism at some localities. The close association between older jadeitite and younger HP-LT rock in a single melange complex implies different histories for the subduction channel and jadeitite-bearing melange. Jadeitite-bearing serpentinite melange can stay at the mantle wedge for a considerable time and, as a result, experience multiple fluid-infiltration events. The subduction channel can occasionally incorporate overlying serpentinized mantle wedge material due to tectonic erosion. With time, the disrupted mantle wedge containing jadeitite veins is mixed with younger blueschists, exhumed eclogites and various fragments of suprasubduction-zone lithologies. Consequently, recrystallization and re-precipitation of jadeitite are reactivated along a slab-mantle wedge interface. All these possible scenarios and their combinations yield a complicated petrological record in jadeitite. With further investigation, the rock association of jadeitite-HP-LT metamorphic rocks-serpentinite has the potential to yield a greater understanding of subduction channels and overlying mantle wedge.

Maruyama <i>et al.</i> (2011) noted; "Some researchers insist that the Dabie-Su Lu metamorphic belt, which was formed by the North China-South China collision, and the Sangun metamorphic belt of Japan were generated along the same trench, based on their similar radiometric ages (Ishiwatari and Tsujimori, 2003; Ernst <i>et al.</i>, 2007). This idea is completely wrong." Omori and Isozaki (2011) also criticized these papers as "a synthesis that misleads to cause confusion" and "a strange paleogeographic reconstruction" using the same logic. These criticisms are based on the idea that the continental collision-type (A-type) and oceanic subduction-type (B-type; also called Cordilleran-type or Pacific-type) orogenic belts are different and exclusive of each other. However, it is evident that both A- and B-type processes took place simultaneously along each of the Appalachian-Caledonian and Hercynian belts (Maruyama <i>et al.</i>, 1996), and the Himalayan continental collision and Indonesian oceanic subduction are currently occurring along the same plate boundary. We provide an objective discussion to counter those criticisms, and point out some discrepancies in the large-scale superficial nappe model of Omori and Isozaki (2011) for the geology of Korea. We believe that our "Yaeyama promontory" hypothesis has inspired constructive international discussions about the configuration of the Earliest Mesozoic collision belt in East Asia, and we think our sinuous configuration model still persists.

Jadeitites form from hydrothermal fluids during high pressure metamorphism in subduction environments; however, the origin of zircons in jadeitite is uncertain. We report ion microprobe analyses of delta(18)O and Ti in zircons, and bulk delta(18)O data for the jadeitite whole-rock from four terranes: Osayama serpentinite m,lange, Japan; Syros m,lange, Greece; the Motagua Fault zone, Guatemala; and the Franciscan Complex, California. In the Osayama jadeitite, two texturally contrasting groups of zircons are identified by cathodoluminescence and are distinct in delta(18)O: featureless or weakly zoned zircons with delta(18)O = 3.8 +/- A 0.6aEuro degrees (2 SD, VSMOW), and zircons with oscillatory or patchy zoning with higher delta(18)O = 5.0 +/- A 0.4aEuro degrees. Zircons in phengite jadeitite from Guatemala and a jadeitite block from Syros have similar delta(18)O values to the latter from Osayama: Guatemala zircons are 4.8 +/- A 0.7aEuro degrees, and the Syros zircons are 5.2 +/- A 0.5aEuro degrees in jadeitite and 5.2 +/- A 0.4aEuro degrees in associated omphacitite, glaucophanite and chlorite-actinolite rinds. The delta(18)O values for most zircons above fall within the range measured by ion microprobe in igneous zircons from oxide gabbros and plagiogranites in modern ocean crust (5.3 +/- A 0.8aEuro degrees) and measured in bulk by laser fluorination of zircons in equilibrium with primitive magma compositions or the mantle (5.3 +/- A 0.6aEuro degrees). Titanium concentrations in these zircons vary between 1 and 19 ppm, within the range for igneous zircons worldwide. Values of delta(18)O (whole-rock) a parts per thousand... delta(18)O (jadeite) and vary from 6.3 to 10.1aEuro degrees in jadeitites in all four areas. These values of delta(18)O and Ti are higher than predicted for hydrothermal zircons, and the delta(18)O values of most zircons are not equilibrated with the coexisting jadeite at reasonable metamorphic temperatures. We conclude that while some zircons may be hydrothermal in origin, a majority of the zircons studied are best explained as relic igneous crystals inherited from precursor rocks; they were not precipitated directly from hot aqueous fluids as previously assumed. Therefore, U-Pb ages from these zircons may date magmatic crystallization and do not establish the timing of high pressure metamorphism or hydrothermal activity.

The Japanese islands were formed during a long-sustained active Pacific-type orogen resulting from subduction, oceanward-accretion, and landward-erosion, which began in the early-Paleozoic. The earlier stage of the accretion of oceanic and continental materials took place between 500-300 Ma, reflecting the convergence of two or more paleo-Pacific plates and the stable, non-subducted Middle and Late Proterozoic continental lithosphere. In Japanese Paleozoic geo-tectonic units, two different high-pressure metamorphic rocks with the Pacific-type protoliths are associated with serpentinite bodies of Early Paleozoic Oeyama ophiolite: blueschist-facies pelitic and mafic schists with phengite K-Ar ages of 350-280 Ma (Renge) and high-pressure epidote-amphibolite-facies amphibolite and pelitic schist with 480-400 Ma hornblende and phengite K-Ar age (Kitomyo-Fuko Pass). The Early Paleozoic Kitomyo-Fuko Pass high-pressure metamorphic rocks provide a petrotectonic constraint on the earliest subduction event in the Japanese orogen. The presence of Middle to Late Paleozoic Renge lawsonite-blueschist and glaucophane-eclogite provides evidence of a cold geotherm in the paleo-subduction zone. Metamorphic and geochronologic data provide important constraints on tectonic development during the earliest stages of orogenic growth associated with the subduction of the paleo-Pacific oceanic plates. The lack of Paleozoic batholith belt and fore-arc sediments coeval with either Renge or Kitomyo-Fuko Pass metamorphism and the presence of blueschist-facies metamorphosed fore-arc ophiolititic materials (fragments of the Oeyama ophiolite) in the Renge metamorphic rocks suggest that a significant landward subduction erosion has occurred since early-Paleozoic time; the eroded material must have been recycled back into the mantle during subduction of the paleo-pacific plate. Thus, the early history of subduction-related orogenesis—after the dramatic tectonic conversion from a passive to active convergent margin—in the Japanese islands is comparable to the modern island arc system occurring worldwide. To further our understanding of the continuous paleo-subduction record in the Japanese Paleozoic geotectonic units, a more detailed and comprehensive approach to geology, petrology, and geochronology of high-pressure metamorphic rocks and associated rocks is required than that documented in previous studies.

Core rocks recovered from the main hole (5158 m deep) of the Chinese Continental Scientific Drilling (CCSD-MH) project, southern Sulu UHP terrane, east-central China, consist of eclogites, various gneisses and minor metaperidotite cumulates; this lithological section underwent subduction-zone UHP metamorphism. Coesite-bearing eclogites are mainly present between the depths of 100-2000 m, but below 2000 m, mafic eclogites are rare. Selected elements (Zr, Nb, Cr, Fe, Si, Mg, Al & Ti) in rutile from 39 eclogite cores from 100 to 2774 m, and major elements of minerals from representative eclogites were analysed by electron microprobe. Zirconium and Nb concentrations of rutile cluster similar to 100-400 and 200-700 ppm respectively. However, Zr and Nb contents in rutile from strongly retrograded eclogites show larger variations than those of fresh or less retrograded eclogites, implying that somehow fluid infiltration affected rutile chemistry during retrograde metamorphism. Zr contents in rutile inclusions in garnet and omphacite are slightly lower than those of the matrix rutile, suggesting that the rutile inclusions formed before or close to the peak temperature. The P-T conditions of the CCSD-MH eclogites were estimated by both Fe-Mg exchange and Zr-in-rutile thermometers, as well as by the Grt-Cpx-Phn-Ky geothermobarometer. The maximum temperature range of 700-811 degrees C calculated at 40 kbar using the Zr-in-rutile thermometer is comparable with temperature estimates by the Fe-Mg exchange thermometer. The temperature estimates of eclogites in a similar to 3000 m thick section define a continuous gradient, and do not show a distinct temperature gap, suggesting that the rocks from 100 to 3000 m depth might belong to a single, large-scale UHP slab. These data combined with P-T calculations for CCSD-MH peridotites yield a low geotherm (similar to 5 degrees C km-1) for the Triassic subduction zone between the Sino-Korean and Yangtze cratons; it lies similar to 30-35 mW m-2 conductive model geotherm.

Ultrahigh-pressure (UHP) metamorphism refers to mineralogical modifications of continental and oceanic crustal protoliths +/- associated mafic-ultramafic rocks initially formed or emplaced in shallow levels of the lithosphere, but which subsequently have experienced P-T conditions within or above the coesite stability field (&gt;similar to 2.7 CPa, similar to 700 degrees C). Typical products include eclogite, garnet peridotite, and UHP varieties of metapelite, quartzite, marble, paragneiss, and orthogneiss. UHP metamorphic assemblages require relatively cold lithospheric subduction to mantle depths; some recrystallization even occurs under "forbidden" P-T conditions, characterized by a geotherm of &lt;5 degrees C/km. In appropriate bulk compositions, UHP metamorphism produces coesite, microdiamond and other indicator phases such as majoritic garnet, TiO(2) with alpha-PbO(2) structure, supersilicic clinopyroxene, high-P clinoenstatite, K-cymrite and stishovite. Globally, at least 20 coesite-bearing eclogitic, eight diamond-bearing, and five majoritic garnet-bearing UHP regions have been documented thus far; they are mostly of Phanerozoic ages. The presence of majoritic garnet, and even apparent stishovite pseudomorph in supracrustal rocks suggests continental subduction to mantle depths exceeding 300 km; such UHP metamorphic terranes should be distinguished from deep-seated mantle xenoliths that contain UHP minerals. Cold subduction zones may be sites of major recycling of H(2)O back into the mantle; high-P experiments on mafic-ultramafic bulk compositions reveal that many important hydrous and formally anhydrous phases are stable under such UHP conditions. The current explosion of research on continental UHP terranes reflects their significance for mantle dynamics and the tectonics of continental subduction, collision, exhumation, mantle-slab interactions, and geochemical recycling. A further characterization of UHP phases and positive identification of UHP minerals requires new experimental studies coupled with state-of-the-art analyses. For example, the very rare occurrence of microdiamond inclusions in zircons from Dabie-Sulu UHP rocks may reflect high f(O2) attending recrystallization inasmuch as epidote is rather common. Rutile needles within garnets from Sulu UHP eclogitic rocks may not be the result of exsolution, so in such cases the apparent UHP pressure may have been over estimated. Hadean igneous microdiamond inclusions in Jack Hills detrital zircons could have originated from mantle xenoliths whereas abundant detrital Phanerozoic diamonds containing inclusions of coesite and other eclogitic minerals from New South Wales might have been derived from unexposed UHP metamorphic terranes. Micro-mineral intergrowth and nano-size minerals may hold important key to deciphering the actual P-T paths of subduction and mantle return flow. Although most exhumed terranes have returned surfaceward relatively rapidly after short time of UHP condition, the long duration of storage at great depth and slow exhumation for the largest UHP terranes remain as major problems. (C) 2008 Elsevier Ltd. All rights reserved.

The Guatemala suture zone is a major east-west left-lateral strike slip boundary that separates the North American and Caribbean plates in Guatemala. The Motagua fault, the central active strand of the suture zone, underwent two major collisional events within a system otherwise dominated by strike-slip motion. The first event is recorded by high-pressure/low temperature (HP/LT) eclogites and related rocks that occur within serpentinites both north and south of the Motagua fault. Lawsonite eclogites south of the fault are not significantly retrograded and give Ar-40/Ar-39 ages of 125-116 Ma and Sm-Nd mineral isochrons of 144-132 Ma. Eclogites north of the fault give similar Sm-Nd isochron ages (131-126 Ma) but otherwise differ in that they are strongly overprinted by a lower pressure assemblage and, along with associated HP/LT rocks, give much younger Ar-40/Ar-39 ages of 88-55 Ma indicating a later amphibolite facies metamorphic event. We propose therefore that all serpentinite hosted eclogites along the Motagua fault formed at essentially the same time in different parts of a laterally extensive Lower Cretaceous forearc subduction system, but subsequently underwent different histories. The southern assemblages were thrust southwards (present coordinates) immediately after HP metamorphism whereas the northern association was retrograded during a later collision that thrust it northward at ca. 70 Ma. They were subsequently juxtaposed opposite each other by major strike slip motion. This model implies that the HP rocks on opposing sides of the Motagua fault evolved along a plate boundary that underwent both dip slip and strike slip motion throughout the Late Cretaceous as a result of oblique convergence. The juxtaposition of a convergent and strike slip system means that HP/LT rocks within serpentinites can be found at depth along much of the modern Guatemala suture zone and its eastward extension into the northern Caribbean. Both sets of assemblages were exhumed relatively recently by the uplift of mountain ranges on both sides of the fault caused by movement along a restraining bend. Recent exhumation explains the apparently lack of offset of surface outcrops along a major strike slip fault. (C) 2009 Elsevier B.V. All rights reserved.

Laser step heating Ar-40/Ar-39 analysis of biotite and muscovite single crystals from a Barrovian type metamorphic belt in the eastern Tibetan plateau yielded consistent cooling ages of ca. 40 Ma in the sillimanite zone with peak metamorphic temperatures higher than 600 degrees C and discordant ages from 46 to 197 Ma in the zones with lower peak temperatures. Chemical Th-U-Total Pb Isochron Method (CHIME) monazite (65 Ma) and sensitive high mass-resolution ion microprobe (SHRIMP) apatite (67 Ma) dating give the age of peak metamorphism in the sillimanite zone. Moderate amounts of excess Ar shown by biotite grains with ages of 46 to 94 Ma at metamorphic grades up to the high-grade part of the kyanite zone probably represent incomplete degassing during metamorphism. In contrast, the high-grade part of the kyanite zone yields biotite ages of 130 to 197 Ma. The spatial distribution of these older ages in the kyanite zone along the sillimanite zone boundary suggests they reflect trapped excess argon that migrated from higher-grade regions. The most likely source is muscovite that decomposed to form sillimanite. The zone with extreme amounts of excess argon preserves trapped remnants of an 'excess argon wave'. We suggest this corresponds to the area where biotite cooled below its closure temperature in the presence of an elevated Ar wave. Extreme excess Ar is not recognized in muscovite suggesting that the entrapment of the argon wave by biotite took place when the rocks had cooled down to temperatures lower than the closure temperature of muscovite. The breakdown of phengite during ultrahigh-pressure (UHP) metamorphism may be a key factor in accounting for the very old apparent ages seen in many UHP metamorphic regions. This is the first documentation of a regional Ar-wave spatially associated with regional metamorphism. This study also implies that resetting of the Ar isotopic systems in micas can require temperatures up to 600 degrees C; much higher than generally thought.

Garnet clinopyroxenite at Rizhao occurs as a lens 1.00 x 225m(2) within a serpentinized peridotite body, in fault contact with felsic gneiss. Minor dunite also occurs within the same peridotite. The clinopyroxenite consists of porphyroblastic Grt-bearing (type 7), megacrystic Grt-bearing (type 2), porphyroblastic Cpx-bearing (type 3), and equigranular (type 4) garnet clinopyroxenite. Lamellae-rich, coarse-grained Cpx occurs as inclusions in megacrystic Grt from type 2 clinopyroxenite, and as porphyroblasts in a matrix of Cpx + Grt + Ilm from type 3 clinopyroxenite; these clinopyroxenes contain abundant exsolution lamellae of Grt (Prp(16-19)Grs(62)Alm (19-21)Sps(1)) + Ilm +/- Mag +/- Amp (Mg#: 0.88-0.98; Na2O: 0.5-3.3 wt%; K2O: 0.7-1.0 wt%; TiO2: 0.1-0.2 wt%). Megacrystic and porphyroblastic garnet (Prp(35-42)Grs(34-45),Alm (19-24) Sps(1)) also occurs in a matrix of fine-grained Cpx + Grt + Ilm; spinel (Mg#: 0.60-0.62) is present as inclusions in both porphyroblastic Grt and lamellae-rich Cpx. Petrochemical data support a mantle origin involving crystallization of majoritic garnet. We propose that the protolith of the Rizhao clinopyroxenite initially was a spinel clinopyroxenite, and was convected to great depths (&gt; 450 km) to form a majoritic garnet that contains high CaO and considerable amounts of FeO and TiO2. During later decompression, the majoritic garnet broke down to form intergrowths of Cpx + Grt + Ilm Mag Amp at T = 700-800 degrees C, P &lt;= 3 GPa. It was then emplaced into a subduction zone and experienced UHP metamorphism at 620-880 degrees C, 3 GPa (luring Triassic continental collision. Megacrystic/porphyroblastic garnet formed during this stage, and lamellaerich Cpx recrystallized to the matrix assemblage of Cpx + Grt + Ilm subsequently. The clinopyroxenite also experienced later retrogression during exhumation. Zircon U-Pb geochronology yields 215 +/- 2 Ma for the Rizhao garnet clinopyroxerute; occurrence of garnet and clinopyroxene inclusions in zircon suggests that this age reflects the UHP metamorphism.

Paleozoic lawsonite-bearing low-grade metabasalts with rare flattened pillow structures occur in the Chugoku Mountains, Southwest Japan. The nieta-mafic rocks are divided into nieta-pillow basalt core (PBC), meta-pillow basalt rim (PBR), and irietabasaltic breeeia (1313). The PBC have MORBlike major and trace element concentrations, contain 5 wt% water, and a Na-amphibole-free mineral assemblage: Ca-Na pyroxene (maximum 29% jadeite component) + lawsonite + chlorite + quartz. In contrast, the PBR and BB contain a lawsonite-blueschist-facies mineral assemblage: Na-amphibole +/- lawsonite +pumpellyite +/- chlorite + albite + quartz. Parageneses and compositions of minerals yield equilibrium P-T conditions of T&lt; 280 degrees C and P = 0.6-0.7 GPa for all three rock types. Phase relations in the 2Al-Ca-2Fe(3+)-(Fe+Mg) tetrahedron suggest that the Na-amphibole-free mineral assemblage in PBC occurs only under relatively oxidized conditions. This mineral assemblage has been described from low-grade nietabasalts in several high-P-low-T terranes, and might be a low-T equivalent of the lawsonite eclogite mineral assemblage. We propose that interaction of fluids with subducting old occanic crust first produces the Ca-Na pyroxene + lawsonite + chlorite assemblage by hydration reactions at shallow depths in a cold subduction zone. With subduction to greater depths, such intensely hydrated inetabasalts with the Na-amphibole-free mineral assemblage transforms to lawsonite eclogite below 300 degrees C.

Convergent plate motion at similar to 320-210 Ma generated the Tongbai-Dabie-Sulu (east-central China)-Imjingang-Gyeonggi (central Korea)-Renge-Suo (Southwestern Japan) -Sikhote-Alin orogen along the paleo-Pacific edge of cratonal Asia. This amalgamated belt reflects collision between the Sino-Korean and Yangtze cratons on the SW portion, and accretion of outboard oceanic arcs sialic fragments against the NTE margin. Subducted Proterozoic-Palcozoic continental and oceanic crustal complexes underwent high- and ultrahigh-pressure metamorphism at low to moderate temperatures. Tectonic slices of sialic crust episodically disengaged from the down-going plate and, driven by buoyancy, ascended rapidly to midcrustal levels from depths exceeding 90-200 km after continental collision in east-central China plus or minus Korea, and from similar to 30-50 km after arrival of far-traveled oceanic terranes in SW Japan and the Russian Far East. On achieving neutral buoyancy and stalling out at 10-20 km depth, later doming, gravitational collapse, and erosion exposed parts of the high- and ultrahigh-pressure complexes. This curvilinear orogen has been segmented and offset by major and minor transverse faults. Also, regional backarc spreading opened marginal basins behind the Permo-Triassic convergent suture zone, further disturbing portions occanward.

The Timor-Tanimbar islands of eastern Indonesia form a non-volcanic arc in front of a 7 km deep fore-arc basin that separates it from a volcanic inner arc. The Timor-Tanimbar Islands expose one of the youngest high P/T metamorphic belts in the world, providing us with an excellent opportunity to study the inception of orogenic processes, undisturbed by later tectonic events. Structural and petrological studies of the high P/T metamorphic belt show that both deformation and metamorphic grade increase towards the centre of the 1 km thick crystalline belt. Kinematic indicators exhibit top-to-the-north sense of shear along the subhorizontal upper boundaries and top-to-the-south sense in the bottom boundaries of the high P/T metamorphic belt. Overall configuration suggests that the high P/T metamorphic rocks extruded as a thin sheet into a space between overlying ophiolites and underlying continental shelf sediments. Petrological study further illustrates that the central crystalline unit underwent a Barrovian-type overprint of the original high P/T metamorphic assemblages during wedge extrusion, and the metamorphic grade ranged from pumpellyite-actinolite to upper amphibolite facies. Quaternary uplift, marked by elevation of recent reefs, was estimated to be about 1260 m in Timor in the west and decreases toward Tanimbar in the east. In contrast, radiometric ages for the high P/T metamorphic rocks suggest that the exhumation of the high P/T metamorphic belt started in western Timor in Late Miocene time and migrated toward the east. Thus, the tectonic evolution of this region is diachronous and youngs to the east. We conclude that the deep-seated high P/T metamorphic belt extrudes into shallow crustal levels as a first step, followed by doming at a later stage. The so-called 'mountain building' process is restricted to the second stage. We attribute this Quaternary rapid uplift to rebound of the subducting Australian continental crust beneath Timor after it achieved positive buoyancy, due to break-off of the oceanic slab fringing the continental crust. In contrast, Tanimbar in the east has not yet been affected by later doming. A wide spectrum of processes, starting from extrusion of the high P/T metamorphic rocks and ending with the later doming due to slab break-off, can be observed in the Timor-Tanimbar region. (C) 2006 Published by Elsevier B.V. on behalf of International Association for Gondwana Research.

Three high-grade tectonic blocks, including jadeite-bearing retrograded eclog-ite, pumpellyite-rich retrograded eclogite, and clinopyroxene-bearing garnet-amphibolite, are newly described in the jadeitite-bearing New Idria serpentinite body. Petrologic analyses reveal two contrasting peak metamorphic stages-eclogite-facies metamorphism (M1E) characterized by garnet + omphacite (∼48 mol% jadeite) + rutile ± epidote + quartz, and amphibolite-facies metamorphism (M1A characterized by garnet + hornblende + augite (∼14 mol% jadeite) + rutile + quartz. Both peak metamorphic events are overprinted by very low-T blueschist-facies minerals (M2), which include glaucophane, lawsonite, pumpellyite, jadeitite (up to 94 mol% jadeite), chlorite, and titanite. Garnet-clinopyroxene geothermometry yields T = ∼580-620 °C at P &gt 1.3 GPa for the M 1E stage and T = ∼630-680 °C at P = ∼0.8-1.0 GPa for the MJ1A stage. The jadeite-and lawsonite-bearing phase equilibria constrain metamorphic conditions of P &gt 1.0 GPa at T = ∼250-300 °C for the M2 stage that is probably synchronous with the formation of nearby jadeitite within serpentinite. The presence of eclogite blocks suggests that the New Idria serpentinite diapir was initiated at mantle depths. The wide range of P-T conditions of tectonic blocks supports the idea that the New Idria serpentinite diapir rose from mantle depths and enclosed tectonic blocks at various mantle-crustal levels during diapiric upwelling and extrusion. ©2007 Geological Society of America. All rights reserved.

Lawsonite eclogites preserve a record of very-low-temperature conditions in subduction zones. All occur at active margin settings, typically characterized by accretionary complexes lithologies and as tectonic blocks within serpentinite-matrix melange. Peak lawsonite-eclogite facies mineral assemblages (gamet+omphacite+lawsonite+rutile) typically occur in prograde-zoned garnet porphyroblasts. Their matrix is commonly overprinted by higher-temperature epidote-bearing assemblages; greenschist- or amphibolite-facies conditions erase former lawsonite-eclogite relics. Various pseudomorphs after lawsonite occur, particularly in some blueschist/eclogite transitional facies rocks. Coesite-bearing lawsonite-eclogite xenoliths in kimberlitic pipes and lawsonite pseudomorphs in some relatively low-temperature ultrahigh-pressure eclogites are known. Using inclusion assemblages in garnet, lawsonite eclogites can be classified into two types: L-type, such as those from Guatemala and British Columbia, contain garnet porphyroblasts that grew only within the lawsonite stability field and E-type, such as from the Dominican Republic, record maximum temperature in the epidote-stability field. Formation and preservation of lawsonite eclogites requires cold subduction to mantle depths and rapid exhumation. The earliest occurrences of lawsonite-eclogite facies mineral assemblages are Early Paleozoic in Spitsbergen and the New England fold belt of Australia; this suggests that since the Phanerozoic, secular cooling of Earth and subduction-zone thermal structures evolved the necessary high pressure/temperature conditions. Buoyancy of serpentinite and oblique convergence with a major strike-slip component may facilitate the exhumation of lawsonite eclogites from mantle depths. (c) 2006 Elsevier B.V. All rights reserved.

The Haiyangsuo Complex in the NE Sulu ultrahigh-pressure (UHP) terrane has discontinuous, coastal exposures of Late Archean gneiss with amphibolitized granulite, amphibolite, Paleoproterozoic metagabbroic intrusives, and Cretaceous granitic dikes over an area of about 15 km(2). The U-Pb SHRIMP dating of zircons indicates that the protolith age of a garnet-biotite gneiss is &gt; 2500 Ma, whereas the granulite-facies metamorphism occurred at around 1800 Ma. A gabbroic intrusion was dated at similar to 1730 Ma, and the formation of amphibolite-facies assemblages in both metagabbro and granulite occurred at similar to 340-46OMa. Petrologic and geochronological data indicate that these various rocks show no evidence of Triassic eclogite-facies metamorphism and Neoproterozoic protolith ages that are characteristics of Sulu-Dabie HP-UHP rocks, except Neoproterozoic inherited ages from post-collisional Jurassic granitic dikes. Haiyangsuo retrograde granulites with amphibolite-facies assemblages within the gneiss preserve relict garnet formed during granulite-facies metamorphism at similar to 1.85 Ga. The Paleoproterozoic metamorphic events are almost coeval with gabbroic intrusions. The granulite-bearing gneiss unit and gabbro-dominated unit of the Haiyangsuo Complex were intruded by thin granitic dikes at about 160 Ma, which is coeval with post-collisional granitic intrusions in the Sulu terrane. We suggest that the Haiyangsuo Complex may represent a fragment of the Jiao-Liao-Ji Paleoproterozoic terrane developed at the eastern margin of the Sino-Korean basement, which was juxtaposed with the Sulu terrane prior to Jurassic granitic activity and regional deformation.

High-grade blocks in the Franciscan complex at Tiburon, California, record relatively low temperature eclogite-facies metamorphism and blueschist-facies overprinting. The eclogite-facies mineral assemblage contains prograde-zoned garnet + omphacite + epidote +/- hornblende (katophoritic and barroisitic Ca-Na amphibole) +/- glaucophane + phengite (similar to 3.5 Si p.f.u.) +/- paragonite + rutile + quartz. The blueschist-facies mineral assemblage contains chlorite + titanite + glaucophane + epidote +/- albite +/- phengite (similar to 3.3 Si p.f.u.). Albite is not stable in the eclogite stage. New calculations based on gamet-omphacite-phengite thennobarometry and THERMOCALC average-P-T calculations yield peak eclogite-facies P-T conditions of P = 2.2-2.5 GPa and T=550-620 degrees C; porphyroclastic omphacite with inclusions of garnet and paragonite yields an average-P-T of 1.8 +/- 0.2 GPa at 490 +/- 70 degrees C for the pre-peak stage. The inferred counterclockwise hairpin P-T trajectory suggests prograde eclogitization of a relatively "cold" subducting slab, and subsequent exhumation and blueschist-facies recrystallization by a decreasing geotherm. Although an epidote-garnet amphibolitic assemblage is ubiquitous in some blocks, P-T pseudosection analyses imply that the epidote-garnet amphibolitic assemblage is stable during prograde eclogite-facies metamorphism. Available geochronologic data combined with our new insight for the maximum pressure suggest an average exhumation rate of similar to 5 km/Ma, as rapid as those of some ultrahigh pressure metamorphic terranes.

The Haiyangsuo area of the NE Sulu ultrahigh-pressure terrane, eastern China, consists of gneisses with minor granulite and amphibolite layers, metagabbros, and granitic dikes. The peak-stage assemblages of the granulites (garnet + orthopyroxene + clinopyroxene + plagioclase +/- pargasite +/- biotite +/- quartz) formed at &gt; 750 degrees C and 9-11 kbar and were overprinted by amphibolite-facies phases characterized by well-developed corona layers of vertical bar garnet vertical bar amphibolite + quartz vertical bar at contacts between plagioclase and clinopyroxene or orthopyroxene, as well as by the exsolution of (orthopyroxene + ilmenite + amphibole) from clinopyroxene. These textures indicate a near-isobaric cooling history of the granulite-bearing gneiss terrane. The metagabbro preserves a relict igneous assemblage (orthopyroxene + clinopyroxene + plagioclase + pargasite +/- ilmenite +/- quartz) in its core, but in its margins has a metamorphic corona texture similar to the granulite that formed at similar to 600-700 degrees C and 7-10 kbar. Sensitive high-resolution ion microprobe (SHRIMP) U-Pb dating of zircons indicates that the protolith age of the garnet-biotite gneiss is older than 2500 Ma, whereas the granulite-facies metamorphism (the first regional metamorphic event) occurred at 1846 +/- 26 Ma. Gabbro intrusion took place at 1734 +/- 5 Ma, and the formation of amphibolite assemblages in both metagabbro and granulite occurred at ca. 340-370 Ma. Both gneiss and metagabbro were intruded by granitic dikes, with one dated at 158 +/- 3 Ma. These data, together with a lack of eclogitic assemblages, suggest that this granulite-amphibolite-facies complex is exotic relative to the Triassic Sulu high-pressure-ultrahigh-pressure terrane; juxtaposition took place in Jurassic time.

青海蛇紋岩メランジュ産の藍閃石-縁れん石エクロジャイト(Tsujimori, 2002)について,Na_2O-CaO-FeO-MgO-Al_2O_3-SiO_2-H_2O系(NCFMASH系)の圧力温度シュードセクションをTHERMOCALC2.75プログラムを使って計算した.その結果,既存の圧力温度見積もり(温度=約550-600℃,圧力>約1.8GPa)に近い条件に,実際に観察される鉱物組み合わせの安定領域が存在することを確認した.現在の沈み込み帯の推定地温勾配を仮定する限り,藍閃石-縁れん石エクロジャイトは若く暖かいスラブの沈み込み変成作用を記録しているのであろう.

The Honvang serpentinite body in the Song Ma fault zone consists mainly of massive serpentinite, altered gabbro and rare chromitite. The serpentinite preserves relict chromian spinel with rare olivine inclusions. The compositional relationship between the Fo content of olivine (Fo(90-92)) and Y-Cr [atomic ratio Cr/(Cr+Al)=0.43-0.44] of chromian spinel suggests that the original peridotite was spinel-bearing Iherzolitic harzburgite. Chromitite is typically a high-Al type, consisting of chromian spinel with Y-Cr=0.43-0.44. Saussuritized fine-grained gabbros display nearly flat rare earth element patterns, suggesting MORB-like affinity. Considering this petrotectonic information, we suggest that the serpentinite body of the Song Ma fault zone represents a remnant of paleo-oceanic lithosphere between the Indochina and South China blocks. The Iherzolitic harzburgite may have formed in an environment with low degrees of melt depletion in a slow-spreading setting similar to some Tethyan paleo-oceanic lithospheres. (c) 2006 Published by Elsevier B.V. on behalf of International Association for Gondwana Research.

Paragonite- and gamet-bearing high-grade epidote-amphibolite (PGEA) in the Ise area of the Hida Mountains, Japan is characterized by the high-pressure (HP) epidote-amphibolite facies parageneses (M-1), gamet+hornblende+clinozoisite+paragonite+quartz+rutile. Paragonite and garnet of the peak M-1 stage are locally replaced by retrograde albite (+oligoclase) and chlorite (M-2), respectively. Phase equilibria constrain peak metamorphic conditions of P=1.1-1.4 GPa and T=530-570 degrees C, and a decompressional P-T path for this rock. Mineral parageneses of prograde epidote-amphibolite facies are comparable to some HP rocks from the Hongan region of western Dabie, but differ from other HP mafic schists with cooling ages of c. 330 Ma in the Hida Mountains. New paragonite K-Ar dating for the PGEA yields a Triassic cooling event at 210 Ma that is coeval with regional cooling and exhumation of the Sulu-Dabic-Qinling (SDQ) belt. Both petrological and geochronological data of the Triassic HP epidote-amphibolite in Hida Mountains support our earlier hypothesis that the SDQ belt extends across the Korean Peninsula to SW Japan. (c) 2005 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.

Early Cretaceous lawsonite eclogites and related high-pressure rocks occur as tectonic inclusions within serpentinite melange south of the Motagua fault zone, Guatemala. Petrologic and microtextural analyses of mafic high-pressure rocks reveal three metamorphic stages linked to several deformational textures. The prograde stage represents an incipient eclogitization and is preserved in prograde garnet, along with an older S-1-S-2 foliation. The prograde assemblage is garnet (X-Mg = similar to 0.22) + omphacite (similar to 52 mol% jadeite) or jadeite (similar to 83 mol % jadeite) + lawsonite + chlorite + rutile + quartz +/- phengite (3.6 Si p.f.u.); some rocks also have ilmenite and rare ferroglaucophane. Lawsonite in garnet of some lawsonite eclogites contains rare pumpellyite inclusions. The presence of synmetamorphic brittle deformation, inclusions of pumpellyite, Fe2+-Mg distribution coefficients between omphacite inclusions and adjacent garnet with Ln(K-D) = 2.7-4.5, and garnet-clinopyroxene-phengite thermobarometry suggest that eclogitization initiated at temperature (T) = similar to 300 degrees C and pressure (P) &gt; 1.1 GPa, and continued to T = similar to 480 degrees C and P = similar to 2.6 GPa. In contrast, the retrograde eclogite-facies assemblage is characterized by reversely zoned garnet rims and omphacite +/- glaucophane + lawsonite + rutile + quartz +/- phengite (3.5 Si p.f.u.) along the S-3 foliation. Garnet-phengite-clinopyroxene thermobarometry yields P = similar to 1.8 GPa and T = similar to 400 degrees C. The youngest, blueschist-facies assemblage (glaucophane + lawsonite + chlorite + titanite + quartz +/- phengite) locally replaces earlier mineral assemblages along S-4 crenulations. The inferred prograde P-T trajectory lies near a geotherm of similar to 5 degrees C km(-1), comparable to the calculated thermal and petrologic structure of the NE Japan subduction zone. These petrologic characteristics indicate: (1) the basalt-eclogite transformation may occur at T = similar to 300 degrees C in cold subduction zones, (2) glaucophane-bearing prograde assemblages are rare during incipient eclogitization in cold subduction zones, and (3) the chlorite-consuming reactions that form Fe-Mg-Mn garnet are more effective than the lawsonite-consuming reaction that forms a grossular component. At depths of similar to 100 km in cold subduction zones, dehydration embrittlement may be caused by such chlorite-consuming reactions.

Kosmochloric diopside with high K content up to 0.56 wt% (0.026 K atom per formula unit) was discovered from the Osayama serpentinite melange in the Chugoku Mountains, SW Japan. K-bearing clinopyroxene fills microcracks (5-150 mu m in wide) together with uvarovite within albite vein of a tremolite rock. Compositions of analyzed clinopyroxene consist mainly of kosmochlor + augite (92-98 mol%; Ko(19-38)Aug(56-76)) components and minor amounts of jadeite (0-6 mol%), aegirine (0-5 mol%), Ca-Tschermak (0-3 mol%), and K-kosmochlor (0-2 mol%). Although the K content in clinopyroxene is also variable and heterogeneous even in a single vein, clinopyroxene with higher K content occurs in Ko-rich part. Higher magnification secondary electron images confirmed that exsolution and inclusion are essentially absent in the analyzed clinopyroxenes. The good negative correlation between Cr + Na + K and Ca + Mg + Fe2+ indicates the Cr incorporation into the octahedral site. Furthermore, K correlates with Na and Cr, indicating a simultaneous enrichment of K for Na and Cr during pyroxene growth. Textual relations, and parageneses and compositions of minerals suggest that the K-Cpx precipitated together with uvarovite in brittle microcracks directly from a Ca- and Cr-rich hydrothermal fluid at approximately P &lt; 0.3 GPa and T &lt; 400 degrees C. Although it has been experimentally concluded that only ultrahigh-P (&gt; 4 GPa) environment permits to host relatively large K+ cation into the clinopyroxene structure, our finding indicates that the incorporation of K into the kosmochlor-diopside series solid solution with at least 0.2 Cr cation p.f.u. is possible even at low P conditions.

Crystals of zircon up to 3 mm in length occur in jadeitite veins in the Osayama serpentinite melange, Southwest Japan. The zircon porphyroblasts show pronounced zoning, and are characterized by both low Th/U ratios (0.2-0.8) and low Th and U abundances (Th = 1-81 ppm; U = 6-149 ppm). They contain inclusions of high-pressure minerals, including jadeite and rutile; such an occurrence indicates that the zircon crystallized during subduction-zone metamorphism. Phase equilibria and the existing fluid-inclusion data constrain P-T conditions to P &gt; 1.2 GPa at T &lt; 350 degrees C for formation of the jadeitite. Most U/Pb ages obtained by SHRIMP-RG are concordant, with a weighted mean (206)Pb/(238)U age of 472 +/- 8.5 Ma (MSWD = 2.7, n = 25). Because zircon porphyroblasts contain inclusions of high-pressure minerals, the SHRIMP U-Pb age represents the timing of jadeitite formation, i.e., the timing of interaction between alkaline fluid and ultramafic rocks in a subduction zone. Although this dating does not provide a direct time constraint for serpentinization, U-Pb ages of zircon in jadeitite associated with serpentinite result in new insights into the timing of fluid-rock interaction of ultramafic rocks at a subduction zone and the minimum age for serpentinization.

Coexisting jadeite and omphacite were found as retrograde minerals in a jadeite-bearing lawsonite-eclogite from the Motagua Fault Zone, Guatemala. The lawsonite-eclogite is characterized by the occurrence of garnet porphyroblasts up to 2.5 cm in size, and the eclogite-facies parageneses, almandine-rich garnet + impure jadeite + lawsonite + rutile + quartz; garnet contains inclusions of impure jadeite, phengite, ferroglaucophane, chlorite, lawsonite, rutile, ilmenite, and quartz. Textural relations and parageneses and compositions of minerals indicate that the lawsonite-eclogite experienced two stages of metamorphism: prograde eclogite-facies stage (M-1) and retrograde stage (M-2). The impure jadeite (Jd-I) of the M-1 eclogite-facies occurs in both the matrix and as inclusions in garnet, and contains considerable amounts of augite and aegirine components (Jd(61-75)Aug(16-24)Aeo(0-18)). It is partly recrystallized to retrograde M-2 jadeite (Jd-II) (Jd(74-87)Aug(9-16)Aeo(0-11)) and omphacite (Jd(42-50)Aug(36-46)Ae(7-16)); some of these two sodic pyroxenes may have crystallized from fluids. Both M-2 jadeite and omphacite show textural equilibrium and are believed to have grown concurrently. Based on the observed compositions and the phase relations of sodic pyroxenes from Carpenter (1980), the M, impure jadeite (Jd-I) may have had a disordered C2/c symmetry at T = ca. 450 degrees C and P = ca. 1.8-2.4 GPa, and was subsequently crystallized into jadeite (Jd-II) plus ordered P2/n omphacite during retrogression with infiltration of fluids at &lt; ca. 300 degrees C and P = ca. 0.7 GPa (M-2). The extreme low-T conditions during retrogression may have prevented reaction between eclogitic jadeite and adjacent minerals. Instead, eclogitic impure jadeite (plus fluid) has recrystallized into the retrograde jadeite + omphacite pair with a wide compositional gap.

An eclogite-facies assemblage garnet + omphacite + rutile + glaucophane + quartz occurs as inclusions in clinozoisite porphyroblasts in a high-grade blueschist block in the Paleozoic Osayama serpentinite melange of the Chugoku Mountains, southwest Japan. Textual relations, and parageneses and compositions of minerals, indicate that the high-grade block experienced peak eclogite-facies metamorphism (M-1) and subsequent blueschist-facies overprinting (M-2). Moreover, low-P amphibolite-facies minerals (M-0) that may represent remnants of pre-subduction conditions are locally preserved in the protolith. Omphacite + adjacent garnet inclusion pairs in clinozoisite give a minimum P = 1.1-1.3 GPa at T = similar to 480-550 degrees C. Many petrologic similarities between the Osayama eclogite and the Hida Mountains eclogile provide evidence for a regional eclogile-facies metamorphism in Paleozoic terranes of southwest Japan. Considering available petrotectonic information regarding eastern Asia, we suggest two alternative models: (1) Japanese Paleozoic eclogiles may represent "Paeffic-type" subduction prior to collision between the Sino-Korean and Yangtze blocks. which occurred on both western and eastern sides of the Qinling-Dabie-Sulu belt; or (2) Japanese Paleozoic eclogites may represent part of a Paleozoic "Pacific -type" subduction chain, which is continuous from the China-Russia border through Japan southward to North Korea.

Eclogites interlayered with gneiss and minor quartzite in the Qinglongshan near Donghai are characterized by unusually abundant (15-40 vol%) hydrous phases including talc, phengite, and epidote; many also contain kyanite. Garnet hosts both prograde (paragonite, amphibole, epidote) and peak stage (omphacite, epidote, phengite, kyanite) mineral inclusions. Several eclogites contain talc rimmed by barroisite; optically and compositionally similar coarse-grained amphibole in other samples indicates that the reaction Omp + Tlc = Amp has completely consumed talc. Estimated peak conditions of 30-35 kbar, 600-700 degreesC, are consistent with polycrystalline quartz pseudomorphs after coesite included in garnet, omphacite, epidote, and kyanite, and LIP to 3.6 Si pfu (110 atom basis) in phengite. Garnet-epidote oxygen barometry on the peak metamorphic assemblalge indicates oxygen fugacities above the Hem-Mag buffer, consistent with the epidote + talc assemblage and 5-20 mol% aegerine component in omphacite. The high oxygen fugacity calculated in this study as well as previously documented negative oxygen isotope values recorded by these rocks may both reflect alteration by oxidizing, meteoric water in a hydrothermal system. Oxidized conditions during peak metamorphism may explain the extreme scarcity of microdiamond in this area. The Ep + Tlc assemblage is stabilized by high oxygen fugacity, and demonstrates that talc-bearing eclogites are not restricted only to unusually Mg-rich bulk compositions.

Early Palaeozoic kyanite-staurolite-bearing epidote-amphibolites including foliated epidote-amphibolite (FEA), and nonfoliated leucocratic or melanocratic metagabbros (LMG, MMG), occur in the Fuko Pass metacumulate unit (FPM) of the Oeyama belt, SW Japan. Microtextural relationships and mineral chemistry define three metamorphic stages: relict granulite facies metamorphism (M1), high-P (HP) epidote-amphibolite facies metamorphism (M2), and retrogression (M3). M1 is preserved as relict Al-rich diopside (up to 8.5 wt.% Al2O3) and pseudomorphs after spinel and plagioclase in the MMG, suggesting a medium-P granulite facies condition (0.8-1.3 GPa at &gt; 850 degreesC). An unusually low-variance M2 assemblage, Hbl + Czo + Ky +/- St + Pg + Rt +/- Ab +/- Crn, occurs in the matrix of all rock types. The presence of relict plagioclase inclusions in M2 kyanite associated with clinozoisite indicates a hydration reaction to form the kyanite-bearing M2 assemblage during cooling. The corundum-bearing phase equilibria constrain a qualitative metamorphic P-T condition of 1.1-1.9 GPa at 550-800 degreesC for M2. The M2 minerals were locally replaced by M3 margarite, paragonite, plagioclase and/or chlorite. The breakdown of M2 kyanite to produce the M3 assemblage at &lt; 0.5 GPa and 450-500 degreesC suggests a greenschist facies overprint during decompression. The P-T evolution of the FPM may represent subduction of an oceanic plateau with a granulite facies lower crust and subsequent exhumation in a Pacific-type orogen.

Continental crust (density similar to2.8 g(.)cm(-3)) resists subduction into the earth&apos;s mantle (similar to3.3 g(.)cm(-3)) because of buoyancy. However, more than 20 recognized ultrahigh-pressure (UHP) terranes have been documented; these occurrences demonstrate that not only is continental crust subducted to depths as great as 150 km, but also that some supracrustal rocks were then exhumed to the earths surface. UHP terranes are composed of mainly supracrusial rocks that contain minor amounts of minerals such as coesite or diamond, indicative of P &gt; 2.5 GPa. to general, quartzofeldspathic units are thoroughly back reacted, and only mafic eclogite lenses and boudins retain scattered UHP phases. These index minerals are restricted to micron-scale inclusions in chemically and mechanically resistant zircon, garnet, and a few other strong container minerals, and are difficult to identify by conventional petrologic studies. The continental rocks were subjected to UHP metamorphism at T ranging from similar to700 to 950degreesC and P &gt; 2.8 to 5.0 GPa, corresponding to depths of similar to100 to 150 km. These UHP units were subsequently exhumed to crustal depths and subjected to intense hydration and amphibolite-facies overprint. Widespread Barrovian-type metamorphism in many collisional orogens may mask an earlier, higher-pressure metamorphic history. We suspect that coesite-bearing UHP rocks were once generated in the majority of exhumed collisional orogens. The recent finding of coesite inclusions in rare Himalayan eclogites and country rock gneisses is a typical example. We use the Himalayan model to illustrate UHP metamorphism and subduction of continental crustal rocks to mantle depths and later Barrovian-lype overprint during exhumation. Himalayan UHP eclogites and adjacent gneisses were formed at mantle depths &gt;100 km at 46 to 52 Ma. These rocks were exhumed to crustal depths and subjected to Barrovian amphibolite- to granulite-facies metamorphism; associated magmatism occurred at 30 to 15 Ma. The Himalayan metamorphic belt was domally uplifted and the mountain-building process initiated since 11 Ma, when underthrusting of the Indian tectosphere beneath the Lesser Himalayas occurred.

Chromian clinopyroxenes with exsolution textures were found in high-pressure (HP) tremolite schist from the Osayama serpentinite melange in the Chugoku Mountains, Southwestern Japan. Chromian omphacite [jadeite (Jd)(23.3-41.4) diopside (Di)(46.7-60.7) kosmochlor (Ko)(4.5-19.6) aegirine (Ae)(&lt;7.0)] with up to 6.6 wt% Cr2O3 occurs as neoblastic crystals in a foliated tremolite-rich matrix, or as pseudomorphs after relict chromian spinel, and contains irregular-shaped thin lamellae of chromian diopside (Jd(3.8-18.2)Di(72.5-93.6)Ko(1.1-13.9)Ae(&lt;3.4)). Chromian omphacite contains roughly constant jadeite plus kosmochlor components at 38.2-51.6 mol%; this is equivalent to the jadeite component of ordered P2/n omphacite. Systematic analyses of coexisting chromian pyroxenes yield a clear immiscibility gap between "omphacite" and "diopside." The compositional gap becomes much narrower with increasing Ko component; addition of only 10 mol% Ko component narrows the omphacite-diopside gap by an order of magnitude. Such an effect is similar to, but more effective than, the introduction of Fe3+ on the omphacite-diopside immiscibility gap. Chromian pyroxenes replacing relict chromian spinel are associated with other chromian silicates including phengite, chlorite, and pumpellyite. The wide compositional gap of chromian pyroxenes and the presence of chromian pumpellyite and Si-rich chromian phengite indicate T &lt; 300-400 degreesC and P &gt; 0.8 GPa. This P-T estimate is consistent with parageneses of minerals in the host serpentinite. The variation of Cr content in chromian silicates reflects the extent of Cr &lt;----&gt; Al substitution, and may be related to a chemical heterogeneity of the Cr-bearing fluid.

飛騨山地,青海蛇紋岩メランジュを構成する超苦鉄質岩は著しく蛇紋岩化し,その起源や変成作用を明らかにするための情報に乏しかった.しかし,塊状のアンチゴライト蛇紋岩と,それに伴うクロミタイト脈から,初生的なCr/(Cr+Al)原子比(0.70-0.77)をコアに保持した組成累帯クロムスピネルと,包有物としての初生的なパーガス閃石(〜3.8wt.% Na_2O)を初めて見出した.蛇紋岩の源岩は変成作用を被っており,クロムスピネルのコアは低Mg#(0.20-0.43)で特徴づけられる.さらに,リムにおいて初生鉱物包有物のドロマイト化とTiの付加が部分的に認められる.これらの特徴から,青海蛇紋岩メランジュを構成する蛇紋岩は,沈み込み帯のマントルかんらん岩を起源とした低〜中程度の温度の変成かんらん岩が,より低温で蛇紋岩化したものと推測される.

New zircon U-Pb and mica Ar-40/Ar-39 dating combined with structural studies in the Longmenshan orogen confirm that most of the upper crustal deformation in the eastern margin of Tibet is Mesozoic. However, at lower structural levels, apatite U-Pb and monazite electron microprobe dating reveals a previously unknown domain of Cenozoic (ca. 65 Ma) Barrovian-type metamorphism and deformation. This discovery shows that the crust in the eastern margin of Tibet was already a substantial thickness around the time of the India-Asia collision. Associated deformation has a N-S-oriented stretching lineation, implying that deformation was not driven by topographic gradients in the Tibetan Plateau. The observed moderate amounts of distributed postmetamorphic E-W shortening can probably explain the present thickness of the continental crust in the area. These results do not support models of crustal thickening caused by solid-state lateral flow of midcrustal metamorphic rocks.

Ophiolites and high-pressure (HP) metamorphic rocks are studied to test continuation of Paleozoic and early Mesozoic geological units from Japan to Primorye over the Japan Sea. The early Paleozoic ophiolites are present on both sides, and the late Paleozoic ophiolite of south-western Japan may also have its counterpart in Primorye. The Shaiginskiy HP schist and the associated Avdakimov gneiss in Primorye, both tectonically underlying the early Paleozoic ophiolitic complex, yield a 250-Ma phengite and hornblende K-Ar age, which is intermediate between those of the Renge (280-330 Ma) and Suo (170-220 Ma) blueschists in south-western Japan. This age also coincides with that of the coesite-bearing eclogites in the Sulu-Dabie suture in China and several medium-pressure metamorphic rocks in East Asia. On the basis of these results and other geological data, the authors propose the 'Yaeyama promontory' model for an eastward extension of the Sulu-Dabie suture. The collision suture warps southward into the Yellow Sea and detours around Korea, turns to the north at Ishigaki Island in the Yaeyama Archipelago of Ryukyu, where it changes into a subduction zone and further continues toward south-western Japan and Primorye. Most ophiolites from this area represent crust-mantle fragments of an island arc-back-arc basin system, and the repeated formation of ophiolite-blueschist associations may be due to the repetition of the Mariana-type non-accreting subduction and Nankai-type accreting subduction.

Late Paleozoic glaucophane eclogite and garnet glaucophane schist ate intercalated with pelitic schist from the Omi eclogitic unit in the Renge metamorphic belt, Hida Mountains, southwestern Japan. Eclogite and garnet glaucophane schist have MORB-like bulk compositions (major and trace), but garnet glaucophane schist is free from omphacite, reflecting a higher Mg/(Mg+Fe) ratio bulk-rock composition. In the eclogite, three different. metamorphic stages (stages I to III) are defined on the basis of microstructures and mineral zoning. In particular, prograde-zoned porphyro-blastic garnet preserves the transition from blueschist to eclogite facies. The inclusion trails in-the rims of garnet are parallel to a penetrative foliation S, in the matrix, and consist of eclogite-facies minerals (omphacite, glaucophane, epidote, rutile, and quartz). Its core contains the preceding epidote blueschist-facies minerals (glaucophane, epidote, titanite, albite, and quartz) instead of eclogite-facies minerals, and has an internal fabric So at a high angle to the surrounding S-1 foliation. In the foliated matrix, the eclogite-facies mineral assemblage garnet + omphacite + glaucophane + epidote + rutile + quartz + phengite is partly replaced by secondary minerals such as chlorite, titanite, albite, and calcite, which coexist with recrystallized glaucophane. These petrographic features show a "hair-pin"-like P-T path that passes from the epidote blueschist facies (stage 1) through the eclogite facies (stage II), and then is retraced on a nearly coincident path to epidote blueschist facies (stage III). The systematic decrease from core to rim in K-D value between garnet and omphacite, and a significant increase in jadeite component of omphacite suggest a rise in both temperature and pressure during prograde eclogitization. The host pelitic schist contains the primary mineral assemblage of quartz + paragonite + phengite + garnet + ferroglaucophane + clinozoisite + rutile, and is strongly replaced by secondary albite, chlorite, and titanite. Pseudomorphs after omphacite and possible kyanite are also found in host pelitic schist. Using geothermolyarometry for eclogite and a petrogenetic grid for the host pelitic schist, conditions of peak eclogite-facies metamorphism are estimated to be around 550-600degreesC and at least 1.8 GPa, indicating an apparent paleo-geothermal gradient of similar to10degreesC/km. The eclogite-bearing Renge metamorphic belt in southwestern Japan may be a new candidate for the eastern extension of a suture zone in east-central China. Late Paleozoic blueschist and eclogite metamorphism may be related to subduction of oceanic crust between the Sino-Korean and Yangtze blocks, prior to their collision.

Granulite facies relics are found in the early Paleozoic kyanite-bearing melanocratic metagabbro from the Fuko Pass metacumulate mass of the Oeyama belt, southwestern Japan. The granulite facies assemblage consists of relict Al-rich clinopyroxene (up to 8.5 wt.% Al2O3) and pseudomorphs of spinel and plagioclase. The spinel pseudomorphs consist mainly of symplectic intergrowth of corundum and magnetite with minor gahnitic spinel. The plagioclase pseudomorphs are composed mainly of clinozoisite with minor kyanite. The symplectite suggests oxidation and Mg-depletion of the original spinel: hercynite (3FeAl2O4)+(O) = corundum (3Al2O3)+magnetite (Fe3O4). This oxidation reaction may have taken place at 700-900°C temperature. The melanocratic metagabbro has later been hydrated to form the epidote amphibolite assemblage represented by clinozoisite+kyanite+paragonite. The clinozoisite+kyanite assemblage has further reacted to form margarite at a lower temperature. The first granulite facies assemblage implies that the metacumulate has originally constituted a basal part of thick oceanic crust, and then has experienced the high-P/T type metamorphism in a subduction zone. This indicates that the thick oceanic crust has been formed and accreted to the Circum-Pacific orogenic belt in the early Paleozoic time.

Eclogitic glaucophane schist has been discovered as a boulder (about 4m diameter) from the Yunotani valley in the western Omi area of the late Paleozoic Renge metamorphic belt (Fig.1). The eclogitic glaucophane schist (Fig.2) occurs as a mafic layer (1.2m wide) intercalated within pelitic schist (garnetparagonite-phengite schist), and is characterized by the mineral assemblage garnet (modal volume: 21%)+omphacite (19%)+ glaucophane (37%)+epidote (19%)+rutile+phengite+albite+quartz (Fig.3). This is the first finding of the late Paleozoic eclogite facies metabasite, which is almost devoid of retrogression and preserving textural evolution (Fig.4) and mineral zoning (Fig.5) during progressive metamorphism. This rock provides an evidence for the eclogite facies metamorphism in the late Paleozoic Western-Pacific margin. More detailed description will appear in Tsujimori et al. (in press).

平均標高5,000m,「世界の屋根」チベット(Fig.1)はインドーアジア両大陸の衝突現場である(Fig.2). そして, その南縁のヒマラヤ山脈までを合わせた地域は面積・標高ともに地球上で比類のない規模を持つ. この異常にぶ厚い地殻は造山運動における重力の役割を際立たせる. 造山運動に対して重力がいかなる影響を与えるかを知るのに, チベットーヒマラヤは最適の地域なのである. この重要性にも関わらず, チベット高原とヒマラヤ山脈という2地域の関連性や相違点については未だ不明な点が多い. 特にチベットとヒマラヤの境界である南チベットは両者の成因上の関係を知る鍵となる地域の1つであるが, その地質はほとんどわかっていない. この夏, 私達はこの関係を知るべく, 南チベットのTethyan Himalayan Sequenceを3週間に渡って調査した(Figs. 2and 3). この調査でわれわれはチベット調査の難しさを実感した. 一つの問題は高山病, もうひとつは道路の悪さである(Fig.4). とはいえ, 大陸衝突を研究する上で, チベット高原は世界でも最高級にやりがいがあり, かつ潜在的に成果を期待できる地域である. さまざまな困難を押してでも挑戦する価値がある.

金沢大学大学院自然科学研究科The Fuko Pass metacumulate is a tectonic block (4.5×1.5 km in size) within the Oeyama peridotite body in the Inner Zone of southwestern Japan. It was metamorphosed under the high-pressure and moderate-temperature condition to form a mineral assemblage hornblende+clinozoisite+kyanite+paragonite+albite+rutile. K-Ar ages were determined on hornblende separates from four epidote amphibolite samples, yielding 443-403 Ma. This suggests a Siluro-Ordovician subduction-related metamorphic event for the Fuko Pass metacumulate, which is distinct from the Late Paleozoic high-P/T type metamorphic events of the Renge metamorphic belt in the Inner Zone of southwestern Japan. The Fuko Pass metacumulate is correlated, in age, with the Early Paleozoic high-P/T type schists (445-402 Ma) of the Kurosegawa klippe in the Outer Zone.

蓮華変成帯, 新潟県青海町上路(あげろ)地区湯ノ谷にエクロジャイト質藍閃石片岩が産する.この岩石は主として藍閃石(37%), ざくろ石(21%), オンファス輝石(19%), 緑れん石(19%)と少量の石英, 曹長石, フェンジャイト, 緑泥石, ルチル, チタン石から構成される.エクロジャイト相鉱物組み合わせ 'ざくろ石+オンファス輝石+藍閃石+緑れん石+石英+ルチル' はマトリクスの片理(S_1)を構成し, わずかに二次的な緑泥石, 曹長石, 方解石に置換され, S_1片理形成以前の緑れん石青色片岩相の鉱物組み合わせ '藍閃石+緑れん石+チタン石+石英+曹長石' の包有物列(S_0)がざくろ石のコアに観察される.蓮華変成帯では, これまでにも残存エクロジャイト相鉱物の報告はあったが, 今回, 岩石組織と鉱物化学組成・累帯構造から, 初めて, 緑れん石青色片岩相からエクロジャイト相への累進変成作用を読みとることができた.

Blueschist-bearing Osayama serpentinite melange develops beneath a peridotite body of the Oeyama ophiolite which occupies the highest position structurally in the central Chugoku Mountains. The blueschist-facies tectonic blocks within the serpentinite melange are divided into the lawsonite-pumpellyite grade, lower epidote grade and higher epidote grade by the mineral assemblages of basic schists. The higher epidote-grade block is a garnet-glaucophane schist including eclogite-facies relic minerals and retrogressive lawsonite-pumpellyite-grade minerals. Gabbroic blocks derived from the Oeyama ophiolite are also enclosed as tectonic blocks in the serpentinite matrix and have experienced a blueschist metamorphism together with the other blueschist blocks. The mineralogic and paragenetic features of the Osayama bluesehists are compatible with a hypothesis that they were derived from a coherent blueschist-facies metamorphic sequence, formed in a subduction zone with a low geothermal gradient (similar to 10 degrees C/km). Phengite K-Ar ages of 16 pelitic and one basic schists yield 289-327 Ma and concentrate around 320 Ma regardless of protolith and metamorphic grade, suggesting quick exhumation of the schists at ca 320Ma. These petrologic and geochronologic features suggest that the Osayama blueschists comprise a low-grade portion of the Carboniferous Renge metamorphic belt. The Osayama blueschists indicate that the &apos;cold&apos; subduction type (Franciscan type) metamorphism to reach eclogite-facies and subsequent quick exhumat;ion took place in the northwestern Pacific margin in Carboniferous time, like some other circum-Pacific orogenic belts (western USA and eastern Australia), where such subduction metamorphism already started as early as the Ordovician.

日本列島は古生代から新生代にかけて形成された典型的な付加型造山帯であり, その基本構造は衝上断層で境されたナップの重なりであって, ナップに含まれる地層や岩石の形成年代は一般に構造的上位のナップほど古い. ナップには基盤岩(火成岩・変成岩)を含むものと, 堆積岩のみよりなるものがあるが, 日本列島の付加体の場合, 含まれる基盤岩の多くはオフィオライトや海山の断片である. 大陸地殻の断片と考えられる隠岐帯・飛帯を除くと, 西南日本内帯の古生代付加体には, 古生代前期以前に形成された大江山オフィオライトが構造的最上位を占め, その下に三郡－蓮華ナップ, 秋吉ナップ, 舞鶴ナップ(古生代後期以前の夜久野オフィオライトを含む), 超丹波ナップ, 三郡－周防ナップ(一部は中生代前期の付加体)があり, これら全体がジュラ紀付加体(美濃・丹波ナップ)を構造的に覆っている(Fig. 1)<BR>最近20年間に, これらのナップを境する衝上断層の露頭が次々に発見されてきた. ここにそれらの代表例をまとめて示す. Fig. 2は若桜地域において, 大江山オフィオライトの蛇紋岩化したマントルかんらん岩が蓮華ナップ(志谷層)の泥質片岩に衝上する露頭である(上村ほか, 1979, p.28). Fig.3 は新見地域で, 大江山オフィオライトの同岩が秋吉帯の石灰岩に衝上する露頭である(早坂ほか, 1995, Stop 1). この断層は3km以上にわたって追跡できる(Fig.4). Fig.5 は大屋地域において大江山オフィオライト(関宮岩体)の同岩が夜久野オフィオライトの玄武岩質凝灰角礫岩に衝上する露頭で, ここでは時代も岩石学的性質も異なる2つのオフィオライトが重なっている(Fig.6; Ishiwatari and Hayasaka, 1992のStop 14(中瀬地窓)の2km南方). Fig, 7は高浜地域で夜久野オフィオライトが超丹波帯の砂岩, 泥岩, 凝灰岩に衝上する露頭である(Ishiwatari and Hayasaka, 1992;Stop 6). Fig.8 とFig.9 は, 小浜地域および青垣地域において, 超丹波帯の二畳系氷上砂岩が丹波帯II型地層群のジュラ系泥質岩に衝上する露頭である(石賀ほか, 1987;Stop 2, 3).

Omphacite (Jd(46.1-52.0)Ae(0-8.4)Aug(48.0-51.2)) and diopside (Jd(4.3-6.3)Ae(0.04)Aug(93.6-95.6)) coexist in a vein cutting an omphacitite block in a serpentinite melange of the Oeyama ophiolite, central Chugoku Mountains. The compositional gap between omphacite and diopside is significantly wider than for other omphacite-diopside pairs reported in the literature. The intergrowth texture of the omphacite-diopside vein suggests that the clinopyroxene pair was contemporaneously crystallized in the fracture from a Ca-bearing, alkali-rich fluid in a sub-solvus condition. Such a fluid may have been introduced from the surrounding serpentinized clinopyroxene-bearing harzburgite. The stability of omphacite and Al-rich pumpellyite in the matrix and the omphacite-diopside vein indicate that the earlier matrix recrystallization and later fracture filling may have both taken place under high-P-T condition during the melange-forming stage.