Fractional crystallization

The felsic lavas of the Bamenda Mountains, in the main part of the Cameroon Volcanic Line, are mainly represented by trachytes, with subordinated benmoreites and alkaline to peralkaline rhyolites. New K–Ar geochronological data define two main volcanic episodes, the first one between 18 and 22Ma, and the second one from 12.5 to 13.5Ma. Geochemical data indicate that these felsic rocks

New petrographic and chemical data from Upper Palaeozoic-Triassic granitoids from the Frontal Cordillera of Argentina (33"lO to 33"45) are presented. Five stocks with ages from Early Carboniferous to Late Permian were sampled. The rocks are all talc alkaline with the mineralogy of each stock comprising essential plagioclase, alkali-feldspar and quartz with monor biotite and hornblende. Chemically, the rocks are similar to granitoids from the-Frontal Cordillera of Chile, although the Argentinian stocks have generally higher WRb ratios. REE chondrite-normalized patterns and overall abundances, along with other petrological and geochemical similarities, suggest a common mode of origin for the granitoid stocks. The absence of systematic and wide variations in abundance of compatible trace elements suggest that the stocks represent magmas largely unaffected by extensive high-level fractional crystallization, and that their compositions are controlled primarily at source. Their age, tectonic setting and geochemical trends are consistent with magma pulses generated during crustal extension. The magma source region is considered to be relatively matic lower crust. Copyright 0 1996

White Island volcano comprises three main lava types: (1) silicic andesite, forming Western Cone and a tholoid at Troup Head, (2) dacite forming the Central Cone, and (3) mafic-silicic andesite erupted in March 1977 from the currently active crater. All lavas are calc-alkalic and medium-K orogenic types. The older andesites of Western Cone and Troup Head were probably formed from chemically dissimilar parental magmas by processes of assimilation and fractional crystallisation (AFC). Andesite blocks and bombs ejected during phreatomagmatic activity in March 1977 are geochemically primitive having high Mg-numbers, high Cr and Ni contents, and containing forsteritic olivine. They cannot be derived from magmatic compositions similar to known basaltic lavas of Taupo Volcanic Zone, and it is possible that the blocks are hybrid magmas resulting from mixing of a high-Mg basalt parent and Central Cone dacite. The bombs appear to be fractionated derivatives. Central Cone dacites may also be AFC derivatives of a common parent to the 1977 lavas, in which substantial chemical diffusion has occurred. Their earlier eruption might represent unloading of a zoned magma chamber.

The origin of igneous Fe–Ti oxide ores associated with massif-type anorthosites is investigated through a detailed study of the world-class Tellnes ilmenite deposit, part of the late-Proterozoic (930–920 Ma) AMC series of the Rogaland Anorthosite Province (SW Norway). More than ...

and-conditions-of-access.pdf This article may be used for research, teaching and private study purposes. Any substantial or systematic reproduction, redistribution , reselling , loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

The Whabouchi pegmatite is within a transpressional high-strain zone of a metamorphosed volcano-sedimentary belt. A well-developed subvertical transposition foliation and a subvertical stretching lineation are present in this zone. The main body of the Whabouchi pegmatite is parallel to the transposition foliation and branching veinlets from this body have been folded and boudinaged by shear zone deformation. Thus suggesting that the Whabouchi pegmatite intruded syn-kinematically during the deformation of the shear zone. Previously, the Whabouchi pegmatite gave a zircon age of 2577 + 14 Ma. The Whabouchi pegmatite is linked geochemically and geochronologically to a large pegmatite body to the south with a monazite age of 2594 + 11 Ma. This suggests that the Whabouchi pegmatite was likely formed by fractionation from a peraluminous granite generated by anatexis. These ages constrain the deformation within the shear zone responsible for the late-stage pegmatite intrusion event between 2594 to 2577 Ma during which, the Whabouchi pegmatite was emplaced.

This publication summarizes the results of petrologic studies aimed at the development and appli- cation of computer models simulating equilibrium and fractional crystallization in igneous systems of mafic to intermediate composition. The empiric and thermodynamic basis for the mineral-melt equilibria calculations is given in detail, including descriptions of the algorithm used in the COMAGMAT program to simulate both per- fect and partial magma fractionation. New techniques are presented for the genetic interpretation of igneous rocks from large layered intrusions and Siberian sills and volcanic rocks from the Mid-Atlantic Ridge and East- ern Kamchatka. Using the COMAGMAT crystallization model, a set of low-pressure phase equilibria calcula- tions (called Geochemical Thermometry ) was conducted for the most primitive rocks from the Marginal Series and the Lower Zones of differentiated intrusions (Partridge River and Talnakh) and large layered complexes, such as Skaergaard, Kiglapai...

The Canary plume is one of the few plumes that can be traced to the core-mantle boundary, making it an excellent location for studying mantle melting dynamics. We performed a comprehensive study of the subaerial shield stage lavas erupted on Gran Canaria, Tenerife, La Gomera and La Palma because these rocks are believed to have been formed by the greatest degree of partial melting of the source and thus can provide direct insights into the origin of plume material and deep crustal recycling. The most primitive picritic to alkali basaltic and basanitic shield stage lavas are moderately enriched in light rare earth elements ([La/Sm] n = 2.1-4.6), strongly enriched in Nb and Ta ([Nb/La] n = 0.8-1.5, [Ta/La] n = 1.3-1.8) and depleted in K and Pb ([K/ La] n = 0.2-0.7; [Pb/La] n = 0.2-0.3), resembling HIMU-type magmas. Fractional crystallization and phenocryst accumulation are processes affecting the compositions of parental magmas. The Sr-Nd-Pb isotope data (87 Sr/ 86 Sr = 0.702966-0.703312, 143 Nd/ 144

... determine the possible origin of the thermal anomaly that allows for partial melting of the ... after Turonian based on the timing of ophiolite obduction along the collision zone. ... Their differences are explained by a combination of fractionation, crustal contamination and differentiation ...

AssrRAcr In the northern sector of tle western Bushveld Complex, toward the top of tle upper Critical Zone, magnssian sli- vine is an unusually abundant cumulus phase in the cyclic units between tle UG-l chromitite layer and the Bastard Reef. Field relationships, whole-rock chemistry and electron-microprobe analyses of olivine (Foe-7) from different cyclic units provide evidence thatthe chamber was intruded by new inlluxes of magma. The new influxes, intruding the chamber at tle level of ttre UG-2 and Merensky Reefs, have not reached the "density-crossover point", thus intruding not as a jet or buoyantplume (Came beJl et al. 1983), but as a flow along the crystal-liquid inter- face (Sparks & Huppert 1984). Cyclic units consisting entirely of ultamafic cumrdates are attributed to fractional crystallization of a distinct stratified layer of new magma. The developing cyclic unit is then terminated by a further influx of magma. Hybrid cyclic units, however, consist of ultramaf...

The Puy Beaunit maar presents a large variety of mantle xenoliths (spinel peridotites, pyroxenites and layered rocks). A detailed study of the textures and mineral equilibria shows the unusual character of this occurrence and the local complexity of the upper mantle beneath the French Massif Central. Ultramafic nodules have a metamorphic, magmatic or pyrometamorphic origin; they display different stages of

Magmas erupted at the Kane Springs Wash volcanic center record the buildup and decay of a silicic magma chamber within the upper crust between 14.1 and 13.2 Ma ago. Intrusion of a variety of mantle-derived basaltic magmas into the crust sustained the system thermally, but only alkali basalts appear to be parental. Fractionation of alkali basalt, together with 10-20% contamination by partial melts of the lower crust, generated trachyandesite magmas. Marie trachytes, with magma temperatures of 1,000 ~ C, were initially generated from trachyandesites at depths greater than 15 kin. Continued fractionation combined with assimilation of upper crustal melts at a depth of 5-10 km produced more evolved trachytes and high-silica rhyolites. These silicic magmas erupted as the Kane Wash Tuff 14.1 Ma ago from a chamber zoned from fayalite-bearing alkali rhyolite near 820 ~ C at the roof to a trachytic "dominant volume". Initial ash flows of the Kane Wash Tuff, Member V 1, are metaluminous, whereas later cooling units, Members V2 and V3, are mildly peralkaline and have higher Fe, Zr, and Hf and lower Ca, Th/Ta, Rb/ Zr, and LREE/HREE. Less than 1% upper crustal component was involved in generation of Members V2 and V3 from trachytic magma. Eruption of ~ 130 km 3 of magma resulted in collapse of the Kane Springs Wash caldera. Trachytic magma from deeper levels of the system was extruded onto the caldera floor shortly afterward, forming a central trachyte/syenite complex. Replacement of this magma by hotter, more mafic magma may have induced additional melting of the already heated chamber walls, as high-silica rhyolites that erupted in the "moat" surrounding the central complex have a large crustal component. Early moat rhyolites had temperatures near 800~ and, in contrast to the Kane Wash Tuff, are ferroedenitebearing, have higher A1, K/Na, Th/Ta, and Ba, and have lower Fe, REE, and Zr. Fractional crystallization of this magma within the cooling and crystallizing magma chamber formed biotite-beating rhyolite in isolated pockets. The most evolved of these had temperatures near 700 ~ C, elevated F contents, H20 contents of ~5wt.%, Rb> 500 ppm, chondrite-normalized LREE/HREE < 1, and formed vapor-phase topaz. Declining temperatures and C1/ F from the Kane Wash Tuff through the moat rhyolites may reflect decreasing basalt input into the base of the system and increasing proportions of upper crustal melts in the silicic magmas.

A late-stage rift-related tholeiite-alkalic suite of igneous intrusions cut the Deccan Traps lavas at the western Indian continental margin. The suite comprises intrusives that can be grouped into ten lithotypes on the basis of their mutual relationships. Tholeiitic types predate the alkaline rocks and greatly predominate, however, the alkaline members exhibit more diversity in mineralogy and chemistry, and are amongst the rare magmatic rocks from the Deccan that host both mantle and lower crustal xenoliths. The mineralogy of most rock types is dominated by clinopyroxene. The diversity of the alkaline rocks could be mainly accounted for by fractional crystallization and mixing between evolved and primitive melts under varying P-T conditions. Sodic and potassic lamprophyres are amongst the most primitive samples with high Mg #, FeO/MgO < 1, high Cr and also with relatively high Ba, Sr, Zr and Nb. They are the most deeply derived magmas within the Deccan Traps as is evident from the mantle and lower crustal xenoliths entrained by them. They possibly represent low degree melts of incompatible elementenriched mantle source rocks. The nephelinites are strongly porphyritic and despite their high Mg #s can be regarded as evolved magmas that have been responsible for the formation of the tephriphonolite daughter. The nephelinites have undergone contamination by lower crustal granulites. The composite intrusions of microdiorites with their complexly zoned mineralogy dominated by plagioclase and amphiboles/micas represent hybrid rocks that have resulted from mixing between tholeiitic and trachytic melts partly at depth and partly at shallow crustal levels.

ABSTRACT The major and trace elements and Sr–Nd–Pb isotopes of the host rocks and the mafic microgranular enclaves (MME) gathered from the Dölek and Sariçiçek plutons, Eastern Turkey, were studied to understand the underlying petrogenesis and geodynamic setting. The plutons were emplaced at ∼ 43 Ma at shallow depths (∼ 5 to 9 km) as estimated from Al-in hornblende geobarometry. The host rocks consist of a variety of rock types ranging from diorite to granite (SiO2 = 56.98–72.67 wt.%; Mg# = 36.8–50.0) populated by MMEs of gabbroic diorite to monzodiorite in composition (SiO2 = 53.21–60.94 wt.%; Mg# = 44.4–53.5). All the rocks show a high-K calc-alkaline differentiation trend. Chondrite-normalized REE patterns are moderately fractionated and relatively flat [(La/Yb)N = 5.11 to 8.51]. They display small negative Eu anomalies (Eu/Eu⁎ = 0.62 to 0.88), with enrichment of LILE and depletion of HFSE. Initial Nd–Sr isotopic compositions for the host rocks are εNd(43 Ma) = − 0.6 to 0.8 and mostly ISr = 0.70482–0.70548. The Nd model ages (TDM) vary from 0.84 to 0.99 Ga. The Pb isotopic ratios are (206Pb/204Pb) = 18.60–18.65, (207Pb/204Pb) = 15.61–15.66 and (208Pb/204Pb) = 38.69–38.85. Compared with the host rocks, the MMEs are relatively homogeneous in isotopic composition, with ISr ranging from 0.70485 to 0.70517, εNd(43 Ma) − 0.1 to 0.8 and with Pb isotopic ratios of (206Pb/204Pb) = 18.58–18.64, (207Pb/204Pb) = 15.60–15.66 and (208Pb/204Pb) = 38.64–38.77. The MMEs have TDM ranging from 0.86 to 1.36 Ga. The geochemical and isotopic similarities between the MMEs and their host rocks indicate that the enclaves are of mixed origin and are most probably formed by the interaction between the lower crust- and mantle-derived magmas. All the geochemical data, in conjunction with the geodynamic evidence, suggest that a basic magma derived from an enriched subcontinental lithospheric mantle, probably triggered by the upwelling of the asthenophere, and interacted with a crustal melt that originated from the dehydration melting of the mafic lower crust at deep crustal levels. Modeling based on the Sr–Nd isotope data indicates that ∼ 77–83% of the subcontinental lithospheric mantle involved in the genesis. Consequently, the interaction process played an important role in the genesis of the hybrid granitoid bodies, which subsequently underwent a fractional crystallization process along with minor amounts of crustal assimilation, en route to the upper crustal levels generating a wide variety of rock types ranging from diorite to granite in an extensional regime.

The 14 Ma caldera-forming composite ignimbrite PI on Gran Canaria (Canary Islands) represents the first voluminous eruption of highly differentiated magmas on top of the basaltic Miocene shield volcano. Compositional zonation of the ignimbrite is the result of vertically changing proportions of four component magmas, which were intensely mixed during eruption: (I) Crystal-poor to highly phyric rhyolite (-10 km 3), (2) sodic trachyandesite through mafic to evolved trachyte (-6 km 3), (3) Nafoor trachyandesite (<I km 3), and (4) basalt zoned from 5.2 to 4.3 wt% MgO (-26 km). PI basalt is composed of two compositionally zoned magma batches, B2 basalt and B3 basalt. B3 basalt is derived from a mantle source depleted in incompatible trace elements compared to the shield basalt source. Basaltic magmas were stored in a reservoir probably underplating the crust, in which zoned B2 basaltic magma formed by mixing of "enriched" (shield) and "depleted" (B3) mafic melts and subsequent crystal fractionation. Evolved magmas formed in a shallow crustal chamber, whereas intermediate magmas formed at both levels. Abundant pyroxenitic to gabbroid cumulates in PI support crystal fractionation as the major differentiation process. On the basis of major and trace element modeling, we infer two contemporaneous fractional crystallization series: series I from "enriched" shield basalt through Na-poor trachyandesite to rhyolite, and series II from "depleted" PI basalt through sodic trachyandesite to trachyte. Series II rocks were significantly modified by selective contamination involving feldspar (Na, K, Ba, Eu, Sr), zircon (Zr) and apatite (P, Y, rare earth elements) components; apatite contamination also affected series I Na-poor trachyandesite. Substantial sodium introduction into sodic trachyandesite is the main reason for the different major element evolution of the two series, whereas their different parentage is mainly reflected in the high field strength trace elements. Selective element contamination involved not only rapidly but also slowly diffusing elements as well as different saturation conditions. Contamination processes thus variably involved differential diffusion, partial dissolution of minerals, partial melt migration, and trace mineral incorporation. Magma mixing between trachyte and rhyolite during their simultaneous crystallization in the PI magma chamber is documented by mutual mineral inclusions but had little effect on the compositional evolution of both magmas. Fe-Ti oxide thermometry yields magmatic temperatures of around 850°C for crystal-poor through crystal-rich rhyolite,-815°C for trachyte and-850°-900°C for the trachyandesitic magmas. High 1160°C for the basalt magma suggest its intrusion into the PI magma chamber only shortly before eruption. The lower temperature for trachyte compared to rhyolite and the strong crustal contamination of trachyte and sodic trachyandcsite support their residence along the walls of the vertically and laterally zoned PI magma chamber. The complex magmatic evolution of PI reflects the transient state of Gran Canaria's mantle source composition and magma plumbing system during the change from basaltic to silicic volcanism. Our results for PI characterize processes operating during this important transition, which also occurs on other volcanic ocean islands. Introduction and Geologic Setting During the major, Miocene magmatic cycle [Schmincke, t976, 1982, t990] on the volcanic ocean island of Gran Can aria (28'00'N, l5°35'W) in the Canarian archipelago (east central Atlantic), a major change from basaltic to silicic

The Oregon Coast Range underwent extension and basaltic magmatism accompanying tectonic rotation during the Tertiary. Basaltic magmatism was expressed initially in submarine eruptions, but mostly in younger subaerial flows. A tholeiitic basalt series, represented by Yachats units, evolved from a depleted, low K, Rb source through a process of fractional crystallization and magma mixing. An alkalic basalt series, represented by Cascade Head units, evolved primarily by fractional crystallization from parental magmas created by lower degrees of partial melting at greater depth than the Yachats series. Camptonite, found in sills and dikes, was formed by even smaller degrees of melting of sources similar to those of Cascade Head. Small nepheline syenite intrusions, emplaced as sills into middle Eocene Tyee Formation, appear to have formed through extensive fractional crystallization of parental basanite, which is sparsely represented in the Yachats and Cascade Head basaltic lava units. Individual syenite intrusions are not directly related to each other, as each intrusion represents a relatively small batch of magma separated at different evolutionary stages from underlying magma chambers. P-T segregation conditions were estimated for primitive Coast Range basalts. These include depths of~120-160 km and temperatures of 1600-1700°C for the alkalic series represented primarily by Cascade Head units and depths of~60-120 km with temperatures of~1500-1600°C for tholeiitic units. A few analyses from other Coast Range igneous units (Tillamook and Goble volcanics) suggest T-P segregations at shallower depths (~55-33 km) and lower temperatures (~1400°C). Paired Nd and Sr isotopic analyses of four Coast Range samples fall in the enriched MORB field. The depths and high temperatures indicated for the Central Oregon Coast ancestral basalts require magma generation in the garnet peridotite stability field, with the alkalic series probably originating below the subducting Juan de Fuca plate. This would suggest an origin for the magmas in a slab window or mantle plume environment. We further suggest that major extension documented for the Coast Range may have triggered upwelling of mantle asthenosphere giving birth to a mantle plume.

Eighteen basic rocks from Ascension Island (South Atlantic) give a mean S7Sr/86 Sr ratio of 0.70311 + 17 for both volcanics and plutonic inclusions. The late-stage differentiated rocks (rhyolites and granitic inclusions) have much higher 87Sr/86 Sr ratios, up to 0.712. All these rocks display the same range of Nd isotopic compositions (~ •a values from 6.9 to 11.1 with a mean on 12 samples of 8.4 + 0.6) implying a cogenetic relation between the two sequences. The D/H systematics lead to the same conclusion. In the Nd-Sr diagram, the data plot close to the mantle array and show a positive correlation. This suggests a mixing between a depleted MORB-type mantle, i.e. the upper mantle, and a hot-spot with less depleted geochemical characteristics, i.e. the OIB mantle source. The total range of 8180 values lies between 4.8%0 for plagioclase cumulates and 6.7%o for the most evolved rocks (peralkaline granites and comendites). The basic rocks have values around 5.3%0, typical of mantle-derived material. These oxygen data indicate that the high 875r/86 Sr ratios in the most evolved rocks (both volcanic and plutonic terms) result from the combination of two different processes: incorporation of slight amounts (< 1%) of high-temperature altered oceanic crust by the magma in the late stages of the differentiation process and then in-situ Rb decay since the time of formation of these rocks. Both processes were very effective because of the high Rb and low Sr contents of these evolved rocks. Oxygen isotope systematics in the Ascension Island granites and rhyolites indicate that a fractional crystallization process alone does not produce 8180 values higher than 6.7%~, i.e. that the ultimate 8180 enrichment, relative to the initial basic magma, is not greater than 1.5%o. * FNRS "' Chercheur Qualifi~".

One object of our study was to design a continuous slurry reactor capable of operating at high pressures (up to 5,000 psig and to determine the feasibility of its operation. At 180 C, 1,100 psigand 0.5 L/hr oil flow, the IV dropped 6 and 10 units for 0.5 and 1% catalyst; linolenic acid decreased to 4 and 2%. By increasing the temperature to 200 C, the IV drop increased to 12-16 units. Although this amount of IV drop is sufficient for salad oil production because the linolenate content dropped to 1%, further hydrogenation is necessary for the production of shortening and cooking oil. This was accomplished at the higher pressure. Iodine values of hydrogenated products were 1743 units less than the original oil, depending on temperature and catalyst concentration. Therefore, any product within this range could be prepared by adjusting the parameters. An attempt was made to compare continuous hydrogenation with previous results (3,7) of batch operation under similar conditions. Making exact comparisons is difficult because of slight differences in temperature, pressure and catalyst concentration and because of different lots of soybean oil used. The calculated residence times for the oilcatalyst slurry in the reactor were 4.5 min, 3.5 min and 2.4 min for 0.5 L/hr, 1.0 L/hr and 2.0 L/hr oil flow rates at 200 C and 1,100 psig. At 0.5 L/hr of oil flow, the decrease in IV was 11.7 units (Fig. 3), which compares favorably with batch hydrogenation. The residence times increased to 10, 6.2 and 3.5 min, respectively, for the 3 oil flow rates when pressure was increased to 4,500 psig. An IV drop of 23.5 units occurred at the higher pressure for a residence time of 4.5 min as read from the rate curve. Since this study was completed, another batch of oil was found to HYDROGENATION

Melting experiments carried out at 1-atm and at 2 kbar on mid-ocean ridge basalts dredged from the mid-Atlantic ridge near the Kane Fracture Zone (KFZ, 22 ~ to 25 ~ N. latitude) provide a basis for evaluating the role of crystal fractionation in generating compositional variability observed in "normal" mid-ocean ridge basalt. The 1-atm olivine-plagioclase-clinopyroxene saturation boundary for KFZ lavas defines a path in mineral projection schemes and in oxide-oxide diagrams that is displaced from the same experimentally determined boundaries in FAMOUS (Grove and Bryan 1983) and Oceanographer Fracture Zone (Walker et al. 1979) basalts. The glass margins of sparsely phyric KFZ lavas record small amounts of near surface, low pressure fractional crystallization, and their glass and bulk rock compositions are similar. An important signature of low pressure differentiation is recorded in the quenched glass margins of moderately phyric KFZ lavas compared to their bulk rock compositions, and the glass has evolved along low-pressure fractionation paths that are similar to those produced in the 1-arm experiments. Many of the lavas have retained phenocrysts in equilibrium proportions, so that their bulk rock compositions represent liquid compositions. When the effects of near-surface differentiation and crystal accumulation are removed from the Kane data set, and only liquid compositions are considered, a suite of basalt magmas can be identified that forms a trend in mineral component projection schemes parallel to the 1-atm olivplag-cpx multiple saturation boundary, but displaced from it toward olivine. These basalts have only olivine and plagioclase as phenocrysts, and are well removed from clinopyroxene saturation at low pressure, The compositional variation can not be generated by mixing any primary liquid composition with a low pressure liquid that has evolved along the oliv-plag-cpx multiple saturation boundary. Major and trace element models of this trend using olivine, plagioclase and clinopyroxene as fractionating phases match the compositional variability. This compositional trend is generated by fractionation at pressures greater than 2 kbar, but within the plagioclase stability field. A review of the data for other normal MORB suites from this part of the mid-Atlantic ridge reveals a similar elevated pressure fractionation signature which persists when the effects of low pressure magma mixing are removed from the data set.

The Miocene magmatism of the Apuseni Mountains in the Carpatho-Pannonian area hosts some of Europe's largest porphyry epithermal Cu-Au ore systems associated with shallow subvolcanic intrusions. Detailed field observations combined with K-Ar ages, geochemical analyses, Sr-Nd isotopes and paleomagnetic data constrain a model for the geotectonic evolution and processes of melt generation that may account for the exceptional mineralizing potential of the magmatic activity in this region. The magmatic activity developed mainly between 14.7 and 7.4 Ma and after a gap ceased at around 1.6 Ma. Geotectonic conditions do not support contemporaneous subduction processes, but were represented by transtensional and rotational tectonics, which generated horst and graben structures and favoured the generation and ascent of magmas. The "subduction signature" of the magmas emphasizes the significant involvement of fluids (mantle lithosphere and/or lower crust) inherited during previous geodynamic events. The mechanism of magmagenesis is considered to be related to decompressional melting (various degrees of) of a heterogeneous source situated at the crust-lithosphere mantle boundary. Mixing with asthenospheric melts generated during the extension-related attenuation of the lithosphere may also be implied. The evolution from normal to adakiticlike calc-alkaline and alkaline magmas generally is time-dependent as a consequence of variable fluid-present melting. Fractional crystallization-assimilation processes in shallow magma chambers are suggested for early magmatism but were almost absent from later magmatism, which related to an increasingly extensional regime. The youngest alkalic (shoshonitic) magmatism (1.6 Ma) is asthenosphere-derived, but in a different extensional event, being almost coeval with the OIB-like alkali-basaltic magmatism (2.5 Ma) occurring along the South Transylvanian fault. The fluid-present melting of the source seems to be the critical factor for the presence of the copper-gold-bearing mineralizing fluids.

Holocene Pelado, Guespalapa and Chichinautzin monogenetic scoria cones and associated lava flows located within the Sierra del Chichinautzin Volcanic Field (SCVF) at the southern margin of Mexico City were mapped and sampled for mineralogical and chemical analyses. With the exception of Par| ¤cutin volcano in western Mexico, few scoria cones in the Trans-Mexican Volcanic Belt have ever been sampled in greater detail. Chemical analyses of rocks indicate that mafic products (e.g. Guespalapa and Chichinautzin) from individual volcanoes in the Sierra del Chichinautzin are characterized by substantial chemical variability, whereas high-silica andesite volcanoes (e.g. Pelado) are very uniform in composition. These findings have important bearings for regional tephrochronology. As a whole, rock compositions form a continuous coherent calc^alkaline suite, explicable by polybaric fractional crystallization þ assimilation associated with successive stagnation at different depths along the ascent path. Trace element and Sr^Nd isotope analyses point toward a 6 1-km-scale heterogeneous (enriched/depleted) mantle wedge underneath the SCVF. The recently proposed plume-origin for these rocks is not in accord with our data. Instead, magma origin is discussed in relation to the tectonically complex subduction process of the oceanic Cocos Plate underneath the continental North American Plate.

At 39.5 ~ S in the southern volcanic zone of the Andes three Pleistocene-recent stratovolcanoes, Villarrica, Quetrupillan and Lanin, form a trend perpendicular to the strike of the Andes, 275 to 325 km from the Peru-Chile trench. Basalts from Villarrica and Lanin are geochemically distinct; the latter have higher incompatible element abundances and La/Sm but lower Ba/La and alkali metal/La ratios. These differences are consistent with our previously proposed models involving: a) a west to east decrease in an alkali metal-rich, high Ba/La slab-derived component which causes an across strike decrease in degree of melting; or b) a west to east increase in the contamination of subducdon-related magma by enriched subcontinental lithospheric mantle. Silicic and mafic lavas from the stratovolcanoes have overlapping Sr, Nd and O isotopic ratios. Silicic lavas also have geochemical differences that parallel those of their associated basalts, e.g., rhyolite from Villarrica has lower La/Sm and incompatible element contents than high-SiO2 andesite from Lanin. At each volcano the most silicic lavas can be modelled by dosed system fractional crystallization while andesites are best explained by magma mixing. Apparently crustal contamination was not an important process in deriving the evolved lavas. Basaltic flows from small scoria cones, 20-35 km from Villarrica volcano have high incompatible element contents and low Ba/La, like Lanin basalts, but trend to higher K/Rb (356-855) and lower 87Sr/ 86Sr (0.70361-0.70400) than basalts from either stratovolcano. However all basalts have similar Nd, Pb and O isotope ratios. The best explanation for the unique features of the cones is that the sources of SVZ magmas, e.g., slabderived fluids or melts of the subcontinental lithospheric mantle, have varying alkali metal and radiogenic Sr contents. These heterogeneities are not manifested in stratovolcano basalts because of extensive subcrustal pooling and mixing. This model is preferable to one involving crustal contamination because it can account for variable Sr isotope ratios and uniform Nd and Pb isotope ratios among the basalts, and the divergence of the cones from acrossstrike geochemical trends defined by the stratovolcanoes.

The average chemical compositions of the continental crust and the oceanic crust (represented by MORB), normalized to primitive mantle values and plotted as functions of the apparent bulk partition coefficient of each element, form surprisingly simple, ...

The Sr, Nd and Pb isotope compositions of ash-flow tuffs and lavas from the central caldera cluster of the San Juan volcanic field, Colorado, suggest that the silicic magmas were derived by fractional crystallization of mantle-derived basalts, coupled with extensive assimilation of both lower-and upper-crustal components. Temporal trends of increasing ENd values and decreasing 87Sr/86Sr ratios of the ash-flow tuffs suggest that extensive crustal hybridization of both upper-and lower-crustal reservoirs occurred as a result of magmatism. Mantle-derived basalts are envisioned to have initially crystallized and significantly interacted with crust near the crust-mantle boundary, creating a hybrid crust that is a mixture of mantle and lower-crustal components. Evolved magmas ascended into the upper crust, where they continued to assimilate and crystallize, modifying the bulk uppercrustal composition through transfer of both lower-crustal and mantle components into the upper crust, strongly affecting the isotopic compositions of the lower and upper crust. Lower-crustal ENd values and 87Sr/86Sr ratios are calculated to have shifted * Corresponding author Elsevier Science B.V.

Spectral interpretation that silicic rocks are widespread on Mars implies that Earth's differentiated crust is not unique. Evaluation of observations bearing on the composition of the Martian crust (Martian meteorite petrology and a possible crustal assimilant, analysis of Mars Pathfinder rocks, composition of Martian fines, interpretation of spacecraft thermal emission spectra, and inferred crustal densities) indicates that the crust can be either basalt plus andesite or basalt plus weathering products. New calculated chemical compositions for Thermal Emission Spectrometer (TES) global surface units indicate that surface type 1 has basaltic andesite composition and surface type 2 has the composition of andesite. If these materials represent volcanic rocks, their calc-alkaline compositions on a FeO*/MgO versus silica diagram suggest formation by hydrous melting and fractional crystallization. On Earth, this petrogenesis requires subduction, and it may suggest an early period of plate tectonics on Mars. However, anorogenic production of andesite might have been possible if the primitive Martian mantle was wet. Alternatively, chemical weathering diagrams suggest that surface type 2 materials could have formed by partial weathering of surface type 1 rocks, leading to depletion in soluble cations and mobility of silica. A weathered crust model is consistent with the occurrence of surface type 2 materials as sediments in a depocenter and with the alpha proton X-ray spectrometer (APXS) analysis of excess oxygen suggesting weathering rinds on Pathfinder rocks. If surface type 1 materials are also weathered or mixed with weathered materials, this might eliminate the need for hydrous melting, consistent with a relatively dry Martian mantle without tectonics.

During its 1800-year-long persistent activity the Stromboli volcano has erupted a highly porphyritic (HP) volatile-poor scoriaceous magma and a low porphyritic (LP) volatile-rich pumiceous magma. The HP magma is erupted during normal Strombolian explosions and lava effusions, while the LP one is related to more energetic paroxysms. During the March-April 2003 explosive activity, Stromboli ejected two typologies of juvenile glassy ashes, namely highly vesicular LP shards and volatile-poor HP shards. Their textural and in situ chemical characteristics are used to unravel mutual Keywords Stromboli Á Volcanic ash Á Lithium isotopes Á Degassing-induced crystallization Á Petrologic monitoring Communicated by J. Hoefs.

Upper Triassic volcanic and sedimentary rocks of the Mamonia Complex in southwestern Cyprus are exposed in erosional windows through the post-Cretaceous cover, where the Mamonia Complex is tectonically imbricated with the Troodos and Akamas ophiolitic suites. Most of these Upper Triassic volcanic rocks have been considered to represent remnants of Triassic oceanic crust and its associated seamounts. New Nd and Pb isotopic data show that the whole Mamonia volcanic suite exhibits features of oceanic island basalts (OIB). Four rock types have been distinguished on the basis of the petrology and chemistry of the rocks. Volcanism began with the eruption of depleted olivine tholeiites (Type 1) and oceanic island tholeiites (Type 2) associated with deep basin siliceous and/or calcareous sediments. The tholeiites were followed by highly phyric alkali basalts (Type 3) interbedded with pelagic Halobia-bearing limestones or white reefal limestones. Strongly LREE-enriched trachytes (Type 4) wer...

Archean migmatites in the vicinity of Gwenoro Dam in Zimbabwe-Rhodesia are composed chiefly of trondhjemite gneiss (TR), mafic tonalite (MT), amphibolite (AM), leuco-trondhjemite veins (LTR), and pegmatites. The gneiss is intruded in nearby areas with small tonalite plutons (TN). Geochemical model studies together with field relationships are consistent with the following model for migmatite production: AM is produced by partial melting of a partly depleted ultramafic parent in which neither garnet nor amphibole remain in the residue; TR and TN are produced by partial melting of undepleted to variably depleted amphibolite in which garnet does not remain in the residue; MT is produced by mixing of plagioclase-rich TR with AM ; and LTR represents the solid residue after fractional crystallization of TR.

The Late Palaeozoic voluminous magmatism in Transbaikalia, Russia (a territory of N 600,000 km 2 to the east of Lake Baikal) is highly diverse and complex. Of special interest are (1) the significant overlap in time between magmatic suites commonly ascribed to post-collisional and within-plate settings and (2) the provenance of the coeval, but distinct, granitoid magmas that are closely spaced within a large region. Magmatic activity lasted almost continuously from~330 Ma to~275 Ma and included five igneous suites occupying a total area of~200,000 km 2 : (1) the Barguzin suite of high-K calc-alkaline granite (330-310 Ma); (2 and 3) the coeval Chivyrkui suite of low-silica calc-alkaline granitoids and the Zaza suite of high-K calcalkaline to alkaline granite and quartz syenite which were emplaced between 305 and 285 Ma; and (4 and 5) the partially overlapped in time Lower-Selenga monzonite-syenite suite (285-278 Ma) and the Early-Kunalei suite of alkali-feldspar and peralkaline quartz syenite and granite (281-275 Ma). The overall increase in alkalinity of the granitoids with time reflects the progress from post-collisional to within-plate settings. However, a~20 m.y. long transitional period during which both calc-alkaline and alkaline granitoids were emplaced indicates the coexistence of thickened (batholiths) and thinned (rift) crustal tracts. Sr-Nd-O isotope and elemental geochemical data suggest that the relative contribution of mantle-derived components to the generation of silicic magmas progressively increased with time. The high-K calc-alkaline granite magmas that formed the Angara-Vitim batholith were generated by high degree melting of supracrustal metamorphic rocks [εNd(t) = − 5.7 to − 7.7; δ 18 O(Qtz) = 12‰], with minor contribution of H 2 O and K from the underplated mafic magma (the convective diffusion model). The coeval calc-alkaline Chivyrkui suite and the transitional to alkaline Zaza suite formed as a result of mixing of crustal silicic and mantle-derived basic melts in roughly equal proportions. In so doing, the former crystallized immediately from the hybrid magmas, whereas the latter (Zaza suite) formed by fractional crystallization of the hybrid melts following magma mixing. Finally the partly overlapping in time monzonite-syenite (Lower-Selenga) suite and highly alkaline syenite-granite (Early-Kunalei) suite were produced from the similar K-rich basalt source. For the former suite, the magma generation process was dominated by fractional crystallization of Krich basalt magma. In contrast, the highly alkaline granitoid magmas were likely generated by partial melting (~20%) of K-rich mafic rocks in the lower crust.

The study of the distribution of PCE, Au, Ni, Cu, Co, and S in the Proterozoic NiquelAndia mafic-ultramafic layered intru- sion (Brazil), 65 km long and 35 km wide, shows the existence of two distinct stratigraphic zones of contrasting PGE content. The Ultramafic Unit and the lowermost part of the Central Mafic Unit have relatively high PGE contents, whereas all

Petrographic and structural observations on the calc-alkalic plutonism of western Corsica revealed the existence of several successively emplaced units associated with large basic bodies. The present mineralogical and geochemical study deals with the genesis, evolution and relationships of these different units. Basic plutonism is represented by three genetically linked types of rock: nontes and troctolites with cumulate textures characterized by low REE contents and either no Eu anomaly or a positive Eu anomaly; gabbros with enriched LREE relatively to HREE patterns, probably close to an initial basaltic liquid; and diorites ranging up to charnockites which represent liquids evolved to varying degrees, mainly by fractional crystallization. Trace element data and studies on the evolution of pyroxene pairs demonstrate the consanguinity of these calc-alkaline basic rocks which are derived from a high alumina basaltic melt. The various granitoids (granodiontes, monzogranites and leucocratic monzogranites, i.e., adamellites) have distinct evolution trends as shown by the composition of their mafic minerals and by trace element distributions. They cannot be considered as being derivatives of the basic suite and they cannot be related by a common fractionation sequence. Rather, they represent distinctive batches of crustal anatexis. In addition, hybridization phenomena with the basic melt are noticed in granodiorites. The particular problem of the low La/Yb, EU/Eu* and the high U, Th, Cs leucocratic monzogranites is discussed in detail. In addition to more conventional trace element diagrams, the simultaneous statistical treatment of all the geochemical data by correspondence factor analysis is shown to be a very useful tool in distinguishing between the different units and to classify the elements according to their geochemical properties.

The Ivrea-Verbano and adjacent Strona-Ceneri zones have been described collectively as a section through the continental crust. While resident in the lower crust, amphibolite-to granulite-facies paragneiss of the Ivrea-Verbano Zone was intruded by huge volumes of mafic to intermediate plutonic rocks grouped as the Mafic Complex. Growth of the Mafic Complex involved hypersolidus deformation in an extensional environment. Isotopic and trace element variations close to the axis of this structure indicate crystallization from mantle-derived melts that were extensively contaminated by crustal material. Previous investigations determined that the contaminant was fingerprinted by 87Sr/86Sr > 0.71, •5180 = 10-12.5%0, and a positive Eu anomaly. In the present study, the contaminant is also shown to have been enriched in Ba with respect to Rb and K. Charnockites associated with paragneiss septa in the lower part of the Mafic Complex have the appropriate chemistry to be samples of the contaminating material. These chemical features can be explained by melting of granulitefacies paragneiss, which had previously been depleted in K and Rb by an earlier melting event. The Ba enrichment in the core of the Mafic Complex can be modeled by a replenishmenttapping-fractional-crystallization (RTF) process operating within a small magma chamber is repeatedly replenished by mantle melts and contaminated by Ba-rich charnockite. Very high Ba/K in the lower part of the complex are tentatively attributed to chemical exchange between the cumulate framework and infiltrating anatectic melts from underlying paragneiss septa. In contrast to the Mafic Complex, the chemistry of coeval granites in the adjacent Strona-Ceneri zone reflect a component derived from crustal rocks that had not been significantly depleted by a previous melting event. Significantly, the incompatible trace element abundances in the Mafic Complex and Strona-Ceneri granites are similar to model compositions for the lower and upper crust, respectively. 2U.S. Geological Survey, Denver, Colorado.

ABSTRACT Plagioclase phenocryts in volcanic rocks commonly record various differentiation processes in their magma chambers. A fundamental assumption is that the normal zoning followed by a reverse zoning or oscillatory zoning are either due to the changes in temperature (T) during mixing of two magmas or due to the fluctuations in water pressure (PH2O) and total pressure (PTotal) during decompressional crystallization. However, this assumption is not always true as demonstrated here for a parental basaltic composition from Small Hasandag Volcano, Central Anatolia, Turkey. Our hypothesis is that the changes recorded in the plagioclase zones of the Small Hasandag Volcano can be explained by the differentiation of magma under isochoric conditions. Once the magma fills the magma chamber, system is isochoric between two successive eruptions provided that wall rocks are rigid. Under these conditions, volume (V) and T are independent variables, whereas P is a dependent variable which changes with subtle differences in V of crystals and melt during crystallization. In order to test this hypothesis, a parental composition (Mg#=61.98 with 3% H2O) is allowed to differentiate under isochoric fractional crystallization conditions. First, a rapid decrease in P has been recorded due to the bulk compressibility 10-6bar-1. Later in the crystallization history of the magma, the water in the original magma, exsolves causing a P increase. As denser minerals crystallize in response to the decreasing T, P decreases significantly from 1500 bars to a minimum of 900 bars at about 7% crystallization and then increases to 1500 bars at about 75% crystallization and continues to increase to 1600 bars at 85% of crystallization and then decreases again to 1540 bars at 90% of crystallization. These fluctuations in P affect the anorthite content (XAn) resulting in normal, reverse and oscillatory zoning in plagioclase phenocrysts.

Northwest Africa 479 (NWA 479) is a lunar meteorite recovered in 2000 from Morocco. This unbrecciated low-Ti basalt is paired with NWA 032. The texture of NWA 032/479 indicates a simple crystallization history and a fast cooling, followed by an impact event. The occurrence of high-pressure polymorphs of olivine (ringwoodite and wadsleyite) in shock-melt veins indicates shock-pressures of at least 20 GPa. Lithium abundances and isotopic compositions were measured by ion microprobe in pyroxene, olivine crystals, and magmatic inclusions. The ␦ 7 Li values in the magmatic inclusions indicate that the NWA 479 parental melt was enriched in 7 Li (␦ 7 Li ϭ ϩ15‰). The behavior of Li depicted by the phenocrysts is complex and is not controlled by their major element compositions. Li abundances and ␦ 7 Li values range respectively from 3.2 to 11.8 g/g and ϩ2.4 to ϩ15.1‰ in olivine and from 2.8 to 18.4 g/g and Ϫ0.2 to ϩ 16.1‰ in pyroxene phenocrysts. Neither hot desert weathering, closed-system fractional crystallization, involvement of a low-␦ 7 Li reservoir, degassing of NWA 479 parental melt, nor shock metamorphism correctly explain the Li distribution in the phenocrysts. We propose that the wide range of ␦ 7 Li values displayed by the phenocrysts results from the large diffusivity differences between 6 Li and 7 Li. It is shown that this difference is able to produce large isotopic heterogeneities in a very short time.

Lavas from Akutan Island, located in the eastern Aleutian arc at the transition between continental and oceanic crust, show a gradual change in their petrologic and chemical characteristics over the last 4 million years. The oldest lavas exposed on the island, the Hot Springs Bay Volcanics (HSBV), range from magnesian basalt to dacite (45%-62% SiO2). The most mafic basalts contain salitic clinopyroxene, Cr-and Al-rich spinel, and pargasitic amphibole suggesting that they were derived from relatively hydrous magmas at greater pressures than lavas from the younger Akutan Volcanics (AKV) and the modern volcano (MOD). AKV lavas also range between basalt and dacite (46%-63% SiO2) , but contain no hydrous phenocrysts and seem to have fractionated within a shallow level magma chamber. Lavas from the modern volcano are andesitic (52%-57% SiO2) and have a mineral assemblage similar to that of AKV lavas of similar composition. With the exception of clinopyroxene and spinel in the most mafic lavas, the compositions of plagioclase (An9a_45), olivine (Fo88-sa), orthopyroxene (En69-56), and titanomagnerite (15%-21% TiOz) phenocrysts found in these lavas are within the range observed in lavas from other Aleutian volcanoes. Variations in the major element chemistry of the older lavas can be reproduced by fractional crystallization of the observed mineral assemblages, however closed system crystal fractionation models are inadequate to explain the trace element variations. During the last 4 million years, La/Yb ratios have decreased (6.5-3.3 for HSBV lavas and 2.9-1.9 for MOD lavas) whereas Ba/La ratios appear to have increased slightly (37-43 for HSBV and AKV, and 41-45 of MOD). The lower La/Yb ratios of MOD lavas correspond with lower total abundances of the REE and slightly higher Sr and Pb isotopic ratios. The increased 87Sr/86Sr ratios and Pb isotopic ratios in the MOD lavas, the less en

Lower Jurassic volcanic-sedimentary successions of the Algarve Basin (∼198 Ma) are associated with the Central Atlantic Magmatic Province (CAMP), that was one aspect of Pangean intracontinental rifting. Volcanic products of the Algarve CAMP include subaerial lava flows, pyroclastic deposits and peperites, and contemporaneous sedimentation is dominated by mudstones and conglomerates, often containing volcanic fragments, intercalated in the volcanic sequence. These lithological characteristics and associations are compatible with a facies model typical of continental basaltic magmatism. Volcanic rocks are fractionated low-Ti tholeiites enriched in large ion lithophile elements relative to high field strength elements. The compositional spectrum (e.g., Ba/Nb-Zr/Nb) is consistent with derivation of primary magmas by partial melting of heterogeneous sources, initially continental lithospheric mantle progressively replaced by an asthenospheric source. Fractional crystallization processes occurred in magma chambers located at depths close to the crust/mantle boundary from density and geobarometric constraints, reflecting magma underplating during the initial stages of continental rifting. Sporadic occurrence of crustal xenoliths, widespread evidence for basement carbonate assimilation, as well as Sr ((87 Sr/ 86 Sr) 0 N 0.7050) and oxygen (δ 18 O N + 7.40‰) isotope ratios that are higher than those typical of continental lithospheric mantle, indicate that crustal contamination also contributed to the observed geochemical variations. The Carnian-Norian (228.0 to 203.6 Ma) or older age estimated for the basal Silves Sandstones suggests that the inception of rift related sedimentation preceded volcanism by 20 to 30Ma. These results lead us to propose a passive rifting model for the Algarve sector of CAMP magmatism, the orientation of which was conditioned by strain localization along pre-existing lithospheric mechanical anisotropies.

The Izera Block in the West Sudetes, which is composed of granites, gneisses (and transitional granitegneisses) and minor mica schists, is one of the largest outcrops of Early Palaeozoic (ca. 500 Ma) metagranitoid rocks in the basement units of the Variscides of Central Europe. The Izera granites show S-type features: magmatic cordierite, relict garnet and sillimanite, lack of mafic enclaves, and absence of coexisting tonalites and diorites. The paucity of pegmatites indicates that the granitic magma was relatively dry. The S-type character of these granites is further supported by their peraluminous character (A/CNK 1.0-1.63), high content of normative corundum (up to 3.5%) and relatively high 87 Sr / 86 Sr initial ratio. The chemical variation of these rocks was controlled by the fractional crystallization of plagioclase (CaO, Sr, Eu/Eu*), biotite and cordierite (Al 2 O 3 , MgO, FeO), zircon (Zr, Hf) and monazite (REE). Initial Nd values range from À5.2 to À6.9 (mean: À5.9, SD=0.6). These largely negative Nd values imply that the granitic magmas emplaced ca. 500 Ma were extracted from a source reservoir that was strongly enriched in LREE (i.e., with low Sm/Nd ratio) on a time-integrated basis. The relatively consistent depleted mantle model ages (1,730-2,175 Ma; mean: 1,890 Ma) is in agreement with the earlier reported presence of ca. 2.1 Ga old inherited Pb component in zircon from the closely related Rumburk granite. This points to an old (Early Proterozoic) crustal residence age of the inferred metasedimentary protoliths of the Izera granitoids, with only minor contribution to their protoliths of juvenile components of Late Proterozoic/Early Palaeozoic age. Although the Izera granites show some trace element features reminiscent of syn-collisional or post-collisional granitoids, they more likely belong to the broad anorogenic class. Our data corroborate some previous interpretations that granite generation was connected with the Early Palaeozoic rifting of the passive margin of the Saxothuringian block, well documented in the region by bimodal volcanic suites of similar age (Kaczawa Unit, eastern and southern envelope of the Karkonosze-Izera Block). In this scenario, granite magmatism and bimodal volcanism would represent two broadly concomitant effects of a single major event of lithospheric break-up at the northern edge of Gondwana.

The Shillong Plateau of northeastern India hosts four Early Cretaceous (105-107 Ma) ultramafic-alkaline-carbonatite complexes (UACC), which have been associated with the Kerguelen plume igneous activity. Petrological and geochemical characteristics of one of these UACC, the Sung Valley, are presented. The Sung Valley UACC was emplaced in to the Proterozoic Shillong Group of rocks and consists of ultramafics (serpentinized peridotite, pyroxenite, and melilitolite), alkaline rocks (ijolite and nepheline syenite), and carbonatites. Serpentinized peridotite, pyroxenite, and ijolitic rocks form the major part of the complex, the others constitute less than 5% of the total volume. Ijolite and melilitolite intrude peridotite and pyroxenite, while nepheline syenite and carbonatite intrude the ultramafic rocks as well as ijolite. Mineralogically, the carbonatites are classified as calcite carbonatite with minor apatite, phlogopite, pyrochlore and ilmenite. The serpentinized peridotites are wehrlitic. Chemical compositions of the silicate rocks do not show a distinct co-genetic relationship amongst them, nor do they show any geochemical relationships with the carbonatites. No noticeable fractionation trend is observed on the chemical variation diagrams of these rocks. It is difficult to establish the genetic evolution of the Sung Valley UACC through fractional crystallization of nephelinitic magma or through immiscible liquids. On the basis of petrological and geochemical data and previously published isotopic results from these rocks, it is suggested that they have been derived from a primary carbonate magma generated by the low-degree melting of a metasomatized mantle peridotite.

We present a new dataset for whole-rock major, trace, isotopic, and phenocryst compositions indicating a genetic link between andesites of the Holocene eruptions of the Bezymianny stratovolcano (the Bezymianny stage), the andesitic to dacitic Late Pleistocene lava dome complex (the pre-Bezymianny stage), and the magnesian to high-alumina basalts of the adjacent Kliuchevskoi Volcano. We demonstrate that volcanic products from the Bezymianny stage of volcano evolution are most likely the products of magma mixing between silicic products of the earliest stages of magma fractionation and the less evolved basaltic andesite parental melts periodically injected into the magma reservoir. In contrast, the intermediate and silicic magmas of the pre-Bezymianny stage together with basalts from Kliuchevskoi much more closely resemble the liquid line of descent and may represent a unique prolonged and continuous calc-alkaline trend of magma evolution from high-magnesian basalt to dacite. As a result of the geothermobarometry, we recognize variable conditions of magma fractionation and magma storage beneath Bezymianny for different magma types during its evolution since the Late Pleistocene: (1) 1100-1150°C, 500-640 MPa, 1-2.5 wt.% H 2 O for parental basaltic andesite; (2) 1130-1050°C, 700-600 MPa, 2.5-5 wt.% H 2 O for two-pyroxene andesites; (3) 1040-990°C, 560-470 MPa, 5-6.5 wt.% H 2 O for orthopyroxene-bearing andesites; (4) 950-1000°C, 450-150 MPa, 3.5-5.5 wt.% H 2 O for hornblende-bearing andesites; and (5) 950-900°C, 410-250 MPa, 6-7 wt.% H 2 O for dacites. Repeated basalt injections and magma fractionation combined with internal mixing in the magma chamber are the main processes responsible for both the complex petrography and the geochemical trends observed in the lavas of Bezymianny Volcano.

The analysis of micro-inclusions in fibrous diamonds from the Diavik mine, Canada revealed the presence of high density fluids (HDFs) that span a continuous compositional range between carbonatitic and saline end-members. The carbonatitic end-member is rich in Na, Ca, Mg, Fe, Ba and carbonate; the saline one is rich in K, Cl and water. In molar proportions, the composition of the saline end-member is: K 38 Na 7.

The magma generation at Unzen volcano may be considered as the product of crustal material mixed with mantle magma Ž. accompanied by fractional crystallization AFC. The magma in the Unzen volcano is estimated to consist of about 50-80% Ž. Ž. of residual magma F and about 30-70% assimilated crustal material A relative to the original magma. Concerning the 1991-1995 eruption, it is estimated that the magma formed as the result of mixing of about 50-60% crustal material and about 55-65% of residual magma. An alternative magma eruption model for the 1991-1995 eruption is proposed here. In the early stage, the isotopic characteristics of 1991 eruption are defined by AFC process in the deeper magma chamber. Later, the magma ascended through the conduit and quiescently stayed for a long time in a shallow reservoir before eruption. The minerals continuously crystallized as phenocrysts especially at the chilled top and outer margin in the shallow chamber. The crystallized phenocryst mush was reworked into the central part of the magma chamber by means of magma convection and rapid magma ascent. Therefore, the reaction between phenocrysts and melt occurs only in internal chemical disequilibrium in the magma chamber. In contrast, the isotopic compositions of the original magma shall be little influenced by the above processes throughout its eruptive history. The 1991-1995 eruptive rocks of the Unzen volcano show their characteristics in Sr and Nd isotopic values independent of their two previous eruptions. However, the isotopic values of early eruptive product could represent the original magma value. This result also supports the previous work of Chen et al.

Along-axis variations for 70 km from the northern propagating tip at Wadi Fizh to the southern margin of the Hilti block in the northern Oman ophiolite are studied on the basis of bulk rock and pyroxene compositions of the sheeted dike complex. Less evolved dikes with tight compositions predominate at the segment center (Wadi Thuqbah-Ays area), while dikes with much broader compositions occur at both northern and southern ends. The appearance of highly evolved rocks at the segment margin suggests more intensive fractional crystallization at the margin. Paradoxically, primitive rocks occur at both ends, indicating processes other than simple crystallization. These features are interpreted as follows: a more steady state melt lens at the segment center would result in less evolved melts with tight compositions, and a smaller melt lens at the segment margin due to cooler conditions would result in two melts, evolved melts ascribed to a more closed system fractional crystallization, and primitive melts emplaced from depths without stagnancy in the melt lens. Zr/Nb and Y/Nb ratios of the bulk analyses are constant over the whole segment, indicating a comparatively uniform source along the segment. For quantitative considerations, 26 analyses were selected from 125 analyses. Na 8.0 and Ti 8.0 values are lower at the segment center than at the north and south parts. Partial melting degrees are estimated to be about 20% at the center and 16% at the segment boundary, respectively. This is consistent with clinopyroxene compositions of the sheeted dike complex and layered gabbros. The pyroxenes are lower in Na 2 O at the segment center than those at the segment margin as compared with the same Mg#, suggesting independently a higher degree of partial melting at the segment center. The trace element contents of the sheeted dike complex show MORB signatures. However, the Ti 8.0 and Si 8.0 of the sheeted dike complex are lower and slightly higher than modern N-MORBs, respectively. This implies that the source mantle beneath paleo-ridges which produced the Oman ophiolite was different from those of modern mid-ocean ridges.