Showing papers in "Journal of Petrology in 1998"

Melting of Garnet Peridotite and the Origin of Komatiite and Depleted Lithosphere

Abstract: However, it can be a residue of komatiite melt extraction at >7 Melting experiments on fertile peridotite KR4003, a ‘pyrolitic’ GPa from a mantle enriched in SiO2 relative to pyrolite. composition, were made from 3 to 7 GPa in piston-cylinder and multi-anvil apparatus. Temperature gradients across the sample were minimized (<25°C), and the compositions of all phases were determined. Modal abundances of coexisting phases were calculated by mass balance, and the results were used to determine phase

Experimental Constraints on Himalayan Anatexis

Abstract: We have melted metapelitic rocks from the High Himalayan Crystalline Sequence that are likely sources of leucogranite magmas. Starting materials were a muscovite schist and a tourmaline-bearing muscovite–biotite schist. Both are kyanite-zone rocks from the hanging wall of the Main Central Thrust. Experiments were conducted at 6, 8 and 10 kbar and 700–900°C, both without added H 2 O (dehydration–melting) and with 1–4 wt % added H 2 O. Dehydration-melting begins at 750–800°C, and produces melts that are virtually identical in composition to the Himalayan leucogranites. Adding H 2 O lowers the solidus by promoting plagioclase + quartz melting. Melts produced from these starting materials at T ≤ 750°C by H 2 O-fluxing are trondhjemitic, and different in composition from most Himalayan leucogranites. Leucogranite magmas in the Himalaya formed by dehydration-melting of metapelites during adiabatic decompression, at 6–8 kbar and 750–770°C. The dehydration-melting solidus for muscovite schist has a smaller dP/dT slope than that for biotite schist. In consequence, muscovite schist undergoes decompression-melting more readily than does biotite schist. The two solidi probably cross over at ∼10 kbar, so that muscovite may be a more important deep crustal H 2 O reservoir than biotite.

Peridotites from the Izu-Bonin-Mariana Forearc (ODP Leg 125): Evidence for Mantle Melting and Melt-Mantle Interaction in a Supra-Subduction Zone Setting

Abstract: Ocean Drilling Program Leg 125 recovered serpentinized harzburgites and dunites from a total of five sites on the crests and flanks of two serpentinite seamounts, Conical Seamount in the Mariana forearc and Torishima Forearc Seamount in the Izu–Bonin forearc. These are some of the first extant forearc peridotites reported in the literature and they provide a window into oceanic, supra-subduction zone (SSZ) mantle processes. Harzburgites from both seamounts are very refractory with low modal clinopyroxene (<4%), chrome-rich spinels (cr-number = 0.40–0.80), very low incompatible element contents, and (with the exception of amphibole-bearing samples) U-shaped rare earth element (REE) profiles with positive Eu anomalies. Both sets of peridotites have olivine–spinel equilibration temperatures that are low compared with abyssal peridotites, possibly because of water-assisted diffusional equilibration in the SSZ environment. However, other features indicate that the harzburgites from the two seamounts have very different origins. Harzburgites from Conical Seamount are characterized by calculated oxygen fugacities between FMQ (fayalite–magnetite–quartz) – 1.1 (log units) and FMQ + 0.4 which overlap those of mid-ocean ridge basalt (MORB) peridotites. Dunites from Conical Seamount contain small amounts of clinopyroxene, orthopyroxene and amphibole and are light REE (LREE) enriched. Moreover, they are considerably more oxidized than the harzburgites to which they are spatially related, with calculated oxygen fugacities of FMQ – 0.2 to FMQ + 1.2. Using textural and geochemical evidence, we interpret these harzburgites as residual MORB mantle (from 15 to 20% fractional melting) which has subsequently been modified by interaction with boninitic melt within the mantle wedge, and these dunites as zones of focusing of this melt in which pyroxene has preferentially been dissolved from the harzburgite protolith. In contrast, harzburgites from Torishima Forearc Seamount give calculated oxygen fugacities between FMQ + 0.8 and FMQ + 1.6, similar to those calculated for other subduction-zone related peridotites and similar to those calculated for the dunites (FMQ + 1.2 to FMQ + 1.8) from the same seamount. In this case, we interpret both the harzburgites and dunites as linked to mantle melting (20–25% fractional melting) in a supra-subduction zone environment. The results thus indicate that the forearc is underlain by at least two types of mantle lithosphere, one being trapped or accreted oceanic lithosphere, the other being lithosphere formed by subduction-related melting. They also demonstrate that both types of mantle lithosphere may have undergone extensive interaction with subduction-derived magmas.

Carbonatite metasomatism in the southeastern Australian lithosphere

Abstract: New mineralogical and geochemical data from a suite of glass +/- apatite +/- amphibole +/- phlogopite +/- carbonate-bearing spinel wehrlite, lherzolite and harzburgite xenoliths from the Newer Volcanics, southeastern Australia, are consistent with metasomatic interactions between harzburgitic or refractory lherzolitic lithosphere, and penetrative sodic dolomitic carbonatite melts. Metasomatism occurred when ascending dolomitic carbonatites crossed the reaction enstatite + dolomite = forsterite + diopside + CO2 at similar to 1.5-2.0 GPa, resulting in partial to complete replacement of primary orthopyroxene by sodic clinopyroxene, together with crystallization of apatite, amphibole and phlogopite, and release of CO2-rich fluid. In the sample suite examined, the minimum amount of carbonatite melt may be estimated on the assumption that metasomatism occurred in a closed system, and that the precursor lithology was clinopyroxene-poor harzburgite. The derivative wehrlite compositions require 6-12% carbonatite addition, the lherzolites require similar to 8% or less, and the harzburgites require minimal addition of carbonatite. However, metasomatism probably also involved an open system component, during which by partitioning relationships with the reacting carbonatite, resulting in loss from the metasomatized volume of a fugitive, siliceous, aluminous, alkali- and LILE-enriched silicate melt.

On the Interpretation of Crystal Size Distributions in Magmatic Systems

Abstract: holocrystalline under a wide spectrum of cooling regimes implies batch system. Instead, the CSDs of each system reflect a combination that cooling and crystallization can be uncoupled and considered of kinetic and dynamic influences on crystallization. Heterogeneous separately. This is tantamount to realizing that the Avrami number nucleation and annexation of small crystals by larger ones, enis large in most igneous systems. Crystallization automatically trainment of earlier grown and ripened crystals, rate of solidification adjusts through nucleation and growth to the cooling regime, and front advance, and protracted transit of a well-established mush all aspects of the ensuing crystal population reflect the relative roles column are some of the eVects revealed in the observed CSDs. There of nucleation and growth, which reflect the cooling regime. The may be an overall CSD evolution, reflecting the maturity of characteristic scales of crystal size, crystal number, and crys- the magmatic system, from simple straight nonkinked CSDs in tallization time are intimately tied to the characteristic rates of monogenetic systems to multiply kinked, piecewise continuous CSDs nucleation and growth, but it is the crystal size distributions (CSDs) in well-established systems such as Hawaii and Mount Etna. This that provide fundamental insight on the time variations of nucleation is not unlike the evolution of CSDs in some industrial systems. and growth and also on the dynamics of magmatic systems. Crystal Finally, the fact that comagmatic CSDs are not often captured size distributions for batch systems are calculated by employing evolving systematically through large changes in nucleation rates, the Johnson‐Mehl‐Avrami equation for crystallinity related to even in low crystallinity systems, may suggest that magma is always exponential variations in time of both nucleation and growth. The laced with high population densities of nuclei, supernuclei, and slope of the CSD is set by the diVerence a ‐ b, where a and b crystallites or clusters that together set the initial CSD at high are exponential constants describing, respectively, nucleation and characteristic population densities. Further evolution of the CSD growth. The batch CSD has constant slope and systematically occurs through sustained heterogeneous nucleation and rapid anmigrates to larger crystal size (L) with increasing crystallinity. The nealing at all crystallinities beginning at the liquidus itself and diminution in nucleation with loss of melt is reflected in the CSD operating under more or less steady (not exponentially increasing) at late times by a strong decrease in population density at small rates of nucleation. crystal sizes, which is rarely seen in igneous rocks themselves. Observed CSDs suggest that a ‐ b ~6‐10 and that b ~0. That is, growth rate is approximately constant and nucleation rate apparently increases exponentially with time. Correlations among CSD slope, intercept, and maximum crystal size for both batch and open systems suggest that certain diagnostic relations may be useful in interpreting the CSD of comagmatic sequences. These systematics are explored heuristically and through the detailed

The Continuum of Primary Carbonatitic-Kimberlitic Melt Compositions in Equilibrium with Lherzolite: Data from the System CaO-MgO-Al2O3-SiO2-CO2 at 6 GPa

Abstract: -free CMAS solidus. The aimof this study was to trace the compositional path of thisline and determine the compositions of coexisting phases.The advantage of this approach is that bulk compositionscan be constructed that maximize the amount of liquidto facilitate analysis. As long as all five phases are present,the system remains invariant at constant temperatureand pressure; and adjustments of the starting composition

Magma Genesis in the New Britain Island Arc: Further Insights into Melting and Mass Transfer Processes

Abstract: Quaternary volcanic rocks from the New Britain island arc display INTRODUCTION a wide range in chemical compositions. The source of the lavas The New Britain region of Papua New Guinea represents shares isotopic characteristics with Indian Ocean type mid-ocean an outstanding opportunity to reach an understanding ridge basalt (MORB). In contrast, the high field strength elements of the processes of magma genesis in an oceanic island (HFSE) are extremely depleted in the volcanic front rocks compared arc. The Quaternary volcanoes found there define the with MORB. We propose that this results from a previous melteastern part of the Bismarck volcanic arc, and have extraction event—hypotheses invoking residual phases in either the formed in response to northward subduction of the small mantle wedge or subducting slab cannot account for the depletion Solomon plate beneath the Bismarck plate (Fig. 1). The relative to MORB. In addition, elements other than the HFSE are New Britain arc is outstanding for two main reasons: also affected. Chemical signatures in parts of the New Britain arc (1) Arc volcanism in the central sector of the island and Manus Basin may relate to a previous subduction episode has taken place over an exceptionally wide range of along the now inactive Vitiaz–West Melanesian trench. Ultradepths to the Wadati–Benioff zone—from ~100 km deep depleted volcanic front basalts invariably have strong ‘fluid’-related at the ‘volcanic front’ (closest to the submarine trench) trace element signatures, including high Sr/Nd and U/Th (and down to ~600 km in the northwest beneath the Witu U disequilibrium), together with positive Eu anomalies that can Islands. The reasons for the correspondingly large width be related to the mobility of Eu in the slab-derived flux. Negative of the volcanic arc are still unknown, but isotopic and Ce anomalies are attributed to a minor sedimentary component. elemental differences between the rocks are systematic Across-arc geochemical profiles record a decrease in the degree of as depths to the Wadati–Benioff zone increase. We believe partial melting and diminishing influence of a slab-derived fluid that these differences provide unparalleled insights into with depth, superimposed upon the depleted mantle composition the geochemical architecture of subduction-zone systems. beneath the volcanic front. Element partitioning into (and not (2) Rocks of the New Britain volcanic front are low in necessarily the source of) the fluid is considered to exert strong control potassium, range from basalt to rhyolite, and have (as on the chemistry of volcanic front magmas, a feature that may go illustrated below) exceptional depletions in high field some way to explaining the contradictory estimates of the slab flux strength elements (HFSE). Indeed, they may represent derived from isotope vs trace element data in many subduction suites. the most HFSE-depleted arc rocks known. These rocks

Chemical and Isotopic Composition of Lavas from the Northern Mariana Trough: Implications for Magmagenesis in Back-arc Basins

Abstract: We report the results of a geochemical and isotopic study of mostly basaltic glasses recovered from 25 dredge stations along the northernmost 500 km of the Mariana Trough extension axis. The distribution of samples links regions of seafloor spreading to the south with regions farther north where a progression of rifting styles accompanies the earliest stages of back-arc basin extension. Petrographic, chemical and isotopic compositions of igneous rocks reflect the changing styles of extension, with typical back-arc basin basalts in the south which become increasingly similar to arc lavas to the north. Felsic lavas also appear along the extensional axis in the north. Glassy, sparsely phyric basalts characterize regions of seafloor spreading. Felsic lavas and porphyritic basalts occur in the northern, rifting portion. Geochemical and isotopic compositions distinguish between mature arc portions (Ce/Pb 20; ^(206)Pb/^(204)Pb >18.5, ^(87)Sr/^(86)Sr >0.7032, ^eNd 10, Ba/La +7). Samples from along the extensional axis of the northern Mariana Trough show progressive changes in chemical and isotopic compositions, from back-arc basin basalts that formed by seafloor spreading northward through increasingly arc–like basalts, until lavas that are indistinguishable from arc lavas are encountered in the northernmost portion of the rift. Batch-melting models indicate that northernmost rift lavas reflect higher degrees of melting, with 13 ± 5% melting where seafloor spreading occurs, doubling to 28 ± 8% for the northernmost part of the rift axis. The greater degree of melting in the north reflects the greater amount of water added to the mantle source, reflecting the arc-like nature of the source region and melt generation style characteristic of the initial stages of back-arc basin formation. Our data indicate that F = 0.44W + 0.07, where F is the degree of mantle melting and W is the percent water in the mantle. ‘True’ back-arc basin basalts are generated by adiabatic decompression associated with mantle upwelling in mature extensional settings. Eruption of ‘true’ back-arc basin basalts accompanies seafloor spreading, which begins when the basin is 100–150 km wide. The arc is disrupted during early rift formation, because arc magmatism is captured by the extension axis, but the generation of arc melts by hydrous melting of the mantle wedge continues whether or not back-arc extension is occurring. Back-arc basin seafloor spreading requires development of an upwelling mantle flow regime, allowing melting by adiabatic decompression, similar to that responsible for mid-ocean ridge basalt (MORB). The arc begins to re-form once extension progresses nearly to the point of seafloor spreading.

Tracing the Indian Ocean Mantle Domain Through Time: Isotopic Results from Old West Indian, East Tethyan, and South Pacific Seafloor

Abstract: The isotopic difference between modern Indian Ocean and Pacific or North Atlantic Ocean ridge mantle (e.g. variably lower 206Pb/204Pb for a given eNd and 208Pb/204Pb) could reflect processes that occurred within a few tens of millions of years preceding the initial breakup of Gondwana. Alternatively, the Indian Ocean isotopic signature could be a much more ancient upper-mantle feature inherited from the asthenosphere of the eastern Tethyan Ocean, which formerly occupied much of the present Indian Ocean region. Age-corrected Nd, Pb, and Sr isotopic data for 46–150 Ma seafloor lavas from sites in the western Indian Ocean and ocean-ridge-type Tethyan ophiolites (Masirah, Yarlung–Zangpo) reveal the presence of both Indian-Ocean-type compositions and essentially Pacific–North Atlantic-type signatures. In comparison, Jurassic South Pacific ridge basalts from Alexander Island, Antarctica, possess normal Pacific–North Atlantic-type isotopic ratios. Despite the very sparse sampling of old seafloor, the age-corrected eNd(t) values of the old Indian Ocean basalts cover a greater range than seen for the much more thoroughly sampled present-day spreading axes and islands within the Indian Ocean (e.g. 18 eNd units for basalts in the 60–80 Ma range vs 15 eNd units for 0–10 Ma ones). The implications of these results are that the upper mantle in the Indian Ocean region is becoming increasingly well mixed through time, and that the Indian Ocean mantle domain may not greatly pre-date the age of earliest spreading in the Indian Ocean.

Geochemical Evolution within the Tonga–Kermadec–Lau Arc–Back-arc Systems: the Role of Varying Mantle Wedge Composition in Space and Time

Abstract: New trace element and Sr, Nd, and Pb isotope data for lavas from the active Tonga-Kermadec arc in the southwest Pacific, the volcano of Niua fo'ou in the back-arc Lau Basin, and Pacific Ocean sediments from DSDP Sites 204 and 275, and ODP Site 596, are integrated with existing geochemical data for lavas from the Lau Basin, Samoa, the Louisville Ridge Seamount Chain (LR-SMC) and the extinct Lau Ridge arc, giving new insights into the petrogenesis of lavas in an active arc - back-arc system. Geochemical variations in Tonga-Kermadec arc lavas are the result of (1) differences in the amount and composition of the material being subducted along the arc, and (2) pre-existing heterogeneities in the upper mantle. Differences in the material being subducted beneath the arc have an important influence on the chemistry of the arc lavas. At the Kermadec Trench, ∼1 km thick layer of sediment is being subducted beneath the arc, compared with ∼200 m at the Tonga Trench. This results in the high Th/U and more radiogenic Pb isotope compositions of Kermadec lavas compared with Tonga lavas. The latter have Pb isotope compositions intermediate between those of Pacific sediments and Pacific mid-ocean ridge basalt (MORB), suggesting that much of the Pb in these lavas is derived from subducting Pacific Ocean crust. This is supported by the Pb isotope signatures of the subducting LR-SMC, which are also observed in lavas from the northern Tongan islands of Tafahi and Niuatoputapu. High field strength element (HFSE) and heavy rare earth element (HREE) concentrations are generally lower in Tongan lavas (particularly those from northern Tongan islands) than in Kermadec lavas. The Tonga Ridge basement, the proto-Tonga arc lavas (ODP Site 839) and the older Lau Ridge arc lavas are generally less depleted than the modern arc lavas. In the back-arc region, upper-mantle depletion as inferred from HFSE and HREE contents of the lavas broadly increases eastwards across the Lau Basin, whereas the subduction signature and volatile (CO and F) contents increase eastwards towards the modern arc. These observations suggest thai depletion is due to melt extraction during back-arc extension and vokanism, together with a long 'residence time' of mantle material within the mantle wedge. The upper mantle beneath the northernmost end of the Tonga arc and Lau Basin contains an ocean-island basalt (OIB) component derived from the Samoa plume to the north. This is reflected in high concentrations of Nb relative to other HFSE in lavas from Niua fo'ou, and Tafahi and Niuatoputapu islands at the northern end of the Tonga arc. Pb isotopes also suggest an LR-SMC contribution into Tafahi and Niuataputapu. Trace element and isotope modelling is used to investigate the combined effects of varying mantle source depletion and subduction on the geochemistry of the arc lavas. The results suggest that the arc lava geochemistry can be explained largely by the balance between a relatively constant subduction input of Pb, Th, U, Cs, Ba, Sr, Rb, K and Sc [corresponding to 0.001-0.005 weight fraction of the Stolper & Newman (1994, Earth and Planetary Science Letters, 121, 293-325] 'HO-rich component' composition), into the overlying, but variably depleted mantle wedge.

Emplacement and Crystallization Time for the Bushveld Complex

Abstract: The Bushveld Complex formed by the crystallization of successive basaltic magmatism may be of the order of several million years. For example, the Columbia River Basalts (Hooper, injections of magma, which were sufficiently closely spaced in time that each previous magma had not cooled and differentiated 1988) were erupted in the period 17–12 Ma, with minor eruptions for a further 5 my, although most outpouring significantly before the addition of the next one. To constrain the emplacement and crystallization times, a thermal model is presented occurred within the first 2 my. It is now recognized that large intrusions were not emplaced in a single pulse, but which permits the investigation of the rate of cooling of magma in an intrusion repeatedly subjected to magma addition (and subresult from multiple magma injection. The question is how rapidly were magma chambers, such as the BC, traction). Such modelling indicates that magmas injected into the Bushveld Complex were emplaced within 75 000 years. At that filled and how much magma was involved. Answers are relevant to the dynamics of melt production, storage and time injection into the Complex ceased. The volume of rock in the Eastern and Western limbs is 370 000–600 000 km. However, transport in the mantle and crust. This paper describes a thermal modelling technique (not previously applied a quantitative evaluation of the Cr budget in the formation of chromitite layers indicates that large volumes of magma cannot be to magma chambers) which can be used to analyse this process, and to obtain an estimate of the emplacement accounted for in the preserved rock sequence. Similarly, an evaluation of the incompatible trace-element abundances, such as those for Zr time. To present this model it is necessary to discuss the stratigraphy, size, and connectivity of the different limbs and K, suggests that the chamber was open and that large volumes of differentiated magma escaped. The volume of magma therefore of the Bushveld Complex, and to consider the extent of tapping of the magma chamber as well as its filling. greatly exceeded the preserved volume of cumulate rocks, giving an estimated magma volume of over 1× 10 km. An average The term ‘Bushveld Complex’ has been given several meanings in the literature, and according to the South emplacement rate of 13 km/year is indicated by these calculations. African Commission on Stratigraphy (1980) includes not only the ultramafic–mafic layered rocks, but also the sills beneath the intrusion, volcanic rocks which pre-date the

Trace Element Partitioning in Immiscible Silicate–Carbonate Liquid Systems: an Initial Experimental Study Using a Centrifuge Autoclave

Abstract: The origin of carbonatites remains a contentious topic. However, INTRODUCTION an important role for liquid immiscibility between silicate and Silicate–carbonate liquid immiscibility has often been carbonate liquids has often been proposed. To understand and proposed as a possible mechanism for the origin of constrain the role this process may play, it is important to have carbonatites associated with ultramafic alkaline rocks in trace element partitioning data available. Few experimental studies composite volcanic and subvolcanic complexes. Field on trace element partitioning between silicate and carbonate liquids observations (Ferguson & Currie, 1971; Kjarsgaard & have been undertaken, reflecting both analytical and experimental Peterson, 1991; Church & Jones, 1995) and data on difficulties. To achieve better phase separation new two-liquid melt inclusions in minerals (Rankin & Le Bas, 1974; experiments have been performed utilizing the rotating centrifuge Romanchev & Sokolov, 1979) support the role of liquid autoclave. Trace elements in the run products were analysed in situ immiscibility in carbonatite petrogenesis. Since the piusing laser ablation microprobe–inductively coupled plasma mass oneering work of Koster van Groos & Wyllie (1963) spectrometry. Partition coefficients (D) have been determined for the separation of carbonatitic immiscible liquids from selected rare earth elements (La, Nd, Sm, Tb, Er, Tm), high field carbonated silicate melts has been demonstrated exstrength elements (Zr, Hf, Nb, Ta), and for Sr, Ba and Y. Most perimentally in many synthetic silicate–carbonate systems of the rare earth elements partition preferentially into the silicate and for natural rock compositions [e.g. see review by liquid. La, Sr and Ba, however, strongly partition into the carbonate Kjarsgaard & Hamilton (1989)]. However, there is no liquid. The high field strength elements, although all preferentially consensus about the role of liquid immiscibility in the partitioning into the silicate liquid, are characterized by a wide development of carbonatite magmas (see Gittins, 1989). range of D values. Zr and Hf have similar D values, which are In this connection, quantitative experimental data on one to two orders of magnitude lower than those of Nb, Ta and trace element partitioning between the immiscible liquids Ti. Ti and Nb behave similarly, whereas Ta demonstrates behaviour are important for constraining the origin of carbonatites. intermediate to that of Zr and Hf. Nb/Ta ratios are strongly Experimental studies in silicate–carbonate systems are fractionated by two-liquid partitioning. hampered by the rapid crystallization of carbonate-rich

The Geochemistry of Volcanic Rocks from Pantelleria Island, Sicily Channel: Petrogenesis and Characteristics of the Mantle Source Region

Abstract: Major and trace element, Sr–Nd–Pb isotope and mineral chemical at least two distinct geochemical components: a mid-ocean ridge basalt (MORB) source, relatively depleted component, and a data are presented for mafic and felsic volcanic rocks from the island of Pantelleria. The mafic rocks, mostly basalts, range from hyHIMU-like enriched component. A further enriched component, normative transitional basalts, through alkali basalts, to basanites. similar to the Enriched Mantle 1 (EM 1) component, could also Clinopyroxene in the mafic rocks varies in composition from Al, Tihave been involved. According to geophysical data, the lithosphere poor diopside to Al, Ti-rich augite. These two populations can be is thinned beneath the island, and the asthenospheric mantle rises present simultaneously in the same sample and even in the same to a depth of 60 km. Rare earth element data require residual garnet crystal, suggesting polybaric fractionation in the pressure range 0–4 in the source and constrain the melting process to a depth of kbar, or mixing between basaltic magmas with different degrees of 70–80 km. The petrological and geochemical data suggest that the alkalinity. On the basis of their major and trace element and mafic magmas are generated within the asthenospheric mantle, from Sr–Nd–Pb isotope composition and age of eruption, two groups of a deep plume bringing the HIMU–EM 1 isotopic and trace element basalts are distinguished: a high TiO2–P2O5 group, erupted before signatures. Interaction of these OIB-like magmas with the shallower 50 ka BP, and a low TiO2–P2O5 group, erupted after 50 ka BP, asthenospheric mantle, providing a depleted MORB signature, separated by a caldera collapse. The felsic volcanic rocks have gives rise to magmas with the observed isotopic and geochemical compositions ranging from comenditic trachyte to comendite and characteristics. pantelleritic trachyte to pantellerite, with progressively increasing peralkalinity. The Sr–Nd isotope compositions of most of the felsic volcanic rocks are similar to those of the mafic volcanic rocks, except for some very Sr-poor pantellerites, which show post-depositional

Migrating Cretaceous-Eocene Magmatismin the Serra do Mar Alkaline Province,SE Brazil: Melts from the Deflected Trindade Mantle Plume?

Abstract: DEPARTMENT OF GEOLOGICAL SCIENCES, UNIVERSITY OF DURHAM, SOUTH ROAD, DURHAM DH1 3LE, UK DEPARTMENT OF EARTH SCIENCES, UNIVERSITY OF CAMBRIDGE, DOWNING STREET, CAMBRIDGE CB2 3EQ, UK DEPARTMENT OF PHYSICS, THE UNIVERSITY, NEWCASTLE UPON TYNE NE1 7RU, UK DEPARTMENT OF GEOLOGY, McMASTER UNIVERSITY, 1280 MAIN STREET WEST, HAMILTON, ONT., CANADA L8S 4M1 DEPARTMENTO DE GEOQUIMICA E RECURSOS MINERAIS, INSTITUTO DE GEOCIENCIAS, UNIVERSIDADE DE BRASILIA,

Formation and Evolution of Granite Magmas During Crustal Reworking: the Significance of Diatexites

Abstract: (Taylor & McLennan, 1985), the physical processes inA continuous section through reworked Archaean crust records the volved, particularly how granite magma is formed, remain generation of granitic magma and its subsequent development in the controversial (Miller et al., 1988; White & Chappell, Opatica subprovince in the Canadian shield. There, the transition 1990). A partially melted source can form a granitic from palaeosome to granite was a closed-system process through magma only if large volumes of melt are separated from intermediate stages of patch migmatite and diatexite. The average most of their residuum. Understanding this process is degree of partial melting was less than 30%, but the melt crucial to determining how crustal recycling and differfraction was redistributed within individual diatexite layers during entiation occurs. For example, melt–residuum separation deformation. Regions that lost melt became residual diatexites could be a nearly perfect, single-stage process at the enriched in TiO2, FeOT, MgO, CaO, Sc, Cr, Co, Sr, rare earth site of partial melting, as described for leucosomes in elements (REE) and high field strength elements (HSFE). Melt metatexite migmatites (Wickham, 1987a; Sawyer, 1991, accumulated to create diatexite magmas enriched in large ion 1994; Brown et al., 1995). Alternatively, the separation lithophile elements (LILE), but contaminated with residuum macould be an imperfect, multi-stage process that yields a terial. Such diatexite magmas are parental to granites found at magma with a large residual component, as is observed higher crustal levels in the terrane. Flow of the diatexite magma in in diatexite migmatites (Bea, 1991; Greenfield et al., response to deformation separated some of its residuum into schlieren. 1996; Sawyer, 1996). The metatexite model necessarily Parautochthonous plutons were created where ascending granitic produces leucocratic, residuum-free granites, such as magma locally ponded below impermeable layers and structures. those described by Le Fort (1981) and Montel et al. Magma left the anatectic region in dykes and lost its remaining residuum as it crystallized. Consequently, the allochthonous granite (1991), and maximizes the geochemical signature of magmas that rose through 20 km of crust to feed the highest level crustal differentiation. In contrast, the diatexite model plutons in this region are highly fractionated and essentially free of can produce residuum-rich granites, and so reduce the residuum. geochemical effect of crustal differentiation, as Chappell (1996) noted. How anatectic magmas formed in the deeper crust evolve to the granitic magmas emplaced in the upper crust is also poorly known, because few crustal sections

The Case for Primary, Mantle-derived Carbonatite Magma

Abstract: magmas discussed in this paper are best considered as being derived There is much debate about whether carbonatite magmas are derived from primary carbonatite magmas generated in the mantle by partial in ‘secondary’ fashion through the advent of liquid immiscibility melting of carbonated peridotite. operating in the crust on evolved nephelinitic magma, or whether they are derived in the mantle by direct partial melting of a carbonated peridotite. This paper briefly summarizes the eSr–eNd data for carbonatites in general and evaluates the isotopic relationships between carbonatites and alkaline silicate rocks in several well-studied complexes from Africa. Available data for carbonatites

Chromite in Komatiites, 1. Magmatic Controls on Crystallization and Composition

Abstract: Chromite is a widespread accessory mineral in komatiites, ranging Zhou & Kerrich, 1992). This is the first in a series of papers which addresses the occurrence of chromite in from skeletal through euhedral and equant to lobate and rarely poikilitic habits depending on cooling regime and the composition komatiites and its chemical variability. This paper looks at the factors controlling the crystallization of chromite of the coexisting olivine. Chromite is least abundant in highly magnesian chrome-undersaturated lavas, and most abundant in from komatiites, and at the textural and compositional variation within chromites which have apparently restrongly differentiated layered cumulate bodies. Abundances are typically lower than the expected cotectic proportions. Chromite tained some or all of their igneous characteristics. Textural and compositional features of chromites are compositions reflect variations in lava composition and oxygen fugacity; reaction with trapped intercumulus liquid; and sub-solidus interrelated in complex and interesting ways, and show systematic relationships to host rock chemistry and crysFe–Mg exchange with olivine. Primary chromites in thick dunitic channels and sheet flows have very low Fe contents, whereas tallization regime. Future papers in the series will deal with superimposed metamorphic effects. comparable thin flow chromites contain higher Fe. This is attributed to an initial reduced state of the magma, thin flow lavas A large body of new data on chromite compositions, including high-precision wavelength-dispersive analyses being subsequently oxidized because of post-eruption processes. Extensive reaction of chromite with trapped liquid causes decreasing of minor and trace elements Ti, V, Mn, Zn, Ni and Co, is presented from the Norseman–Wiluna Greenstone Belt Mg/(Mg+ Fe), and enrichment in Fe, TiO2, V2O5, MnO, CoO and ZnO. Nickel trends depend on sulphide content. These in Western Australia, and synthesized with a compilation of literature data taken mainly from Roeder (1994) and trends are best developed in rocks combining presence of trapped intercumulus liquid with prolonged cooling histories. The deficiency various unpublished sources. Whole-rock geochemical data are similarly compiled from the literature and some of chromite in channelized environments is partly a consequence of more primitive lavas being found in these environments, but is unpublished sources, and include a large body of data collected over the last 15 years at CSIRO. Details of mainly due to the difficulty in nucleating chromite crystals under low degrees of supercooling. data sources are listed in Appendix A, and details of electron microprobe analytical techniques in Appendix B.

Etendeka Volcanism of the Goboboseb Mountains and Messum Igneous Complex, Namibia. Part I: Geochemical Evidence of Early Cretaceous Tristan Plume Melts and the Role of Crustal Contamination in the Paraná–Etendeka CFB

Abstract: The Goboboseb Mountains and Messum Complex represent a major Cretaceous (132 Ma) bimodal eruptive centre in the southern Etendeka continental flood basalt (CFB) province. The eruptives compris the Awahab Formation and are represented by a lower sequence of mafic lavas, followed by the Goboboseb quartz latite members, the Messum Mountain Basalts, and finally the Springbok quartz latite. The sequence is cut by numerous dolerite dykes, sills and plugs, rare rhyolite, and carbonatite. The mafic lavas comprise two distinct series, which although corresponding broadly to the Etendeka regional low Ti and Zr basalts (LTZ type), are distinguished by Ti/Zr ratios into the LTZ.H (higher Ti/Zr) and LTZ.L (lower Ti/Zr) series. The LTZ.H basalts have no previously described extrusive equivalent in the Etendeka (or Parana) CFB, and consist of magnesian, mildly alkaline to tholeiitic lavas, dominated by oliv + cpx phenocryst assemblages which fractionate (near the surface) to phono-tephrite. They are identified as predominantly mantle plume melts (Tristan-Walvis plume). The LTZ.L lavas are less magnesian, extending to icelandites, are tholeiitic, with cpx ± oliv + pl + Fe-Ti oxide phenocryst assemblages and groundmass pigeonite and subcalcic augite. Stratigraphically, the LTZ.H lavas are interbedded with LTZ.L types in the lower part of the sequence and also occur as dykes. Within the Messum Complex, a remnant early sequence of basalts (Messum Carter Basalts) are in part transitional between the LTZ.L and LTZ.H series. The LTZ.H, and at least some of the LTZ.L lavas are inferred to have been erupted from the Goboboseb-Messum Centre. Chemically, the LTZ.H melts are broadly intermediate between E-MORB and OIB magmas, with higher Ti/Zr Sm/ γb and Ti/γ ratios than the LTZ.L types, which suggest segregation depths between the garnet and spinel peridotite stability fields. The Pb-Nd-Sr isotopic compositions of the LTZ.H eruptives are similar to, but not identical with the modern Tristan plume composition, and the observed variability is attributable to limited lower-crust assimilation and/or Atlantic MORB source mixing. The LTZ.L lavas show evidence for crystal fractionation, have 'arc-like' trace element signatures, correlated e-SiO, e-Ti/γ and -Ti/Zr, e-1/Sr and 1/Nd-e variations, and relatively radiogenic Pb, evolved Sr(e, 58-174) and low Nd(e -6·1 to -9·5) isotopic compositions. Their geochemistry is inferred to be AFC (assimilation-fractional crystallization) controlled, and is modelled by three-component mixing involving mantle plume derived melt, mafic lower crust and silicic mid-upper crust. The voluminous quartz latites (Part II, Ewart et al., 1978) extend these geochemical trends.

Petrogenesis of Cenozoic Basalts from Vietnam: Implication for Origins of a 'Diffuse Igneous Province'

Abstract: after the early Tertiary India‐Asia collision. This activity does not conform to the ‘Large Igneous Province’ model in view of lower eruption and melt production rates, wide dispersal of centres and the apparent absence of deep mantle upwelling. Age data for Vietnamese plateau basalts reflect spatial‐temporal KEY WORDS: Vietnam; basalt; Cenozoic; geochemistry patterns consistent with a rotating stress field rather than suprahotspot lithosphere migration. For most of the volcanic centres there are two eruptive episodes: an early series formed by highSiO2, low-FeO* quartz and olivine tholeiites—large melt fractions of refractory (lithosphere-like) mantle—and a later series made INTRODUCTION up of low-SiO2, high-FeO* olivine tholeiites, alkali basalts and Neogene‐Quaternary intraplate volcanism is widespread basanites—smaller melt fractions of more fertile (asthenosphere- in east and southeast Asia (Fig. 1a) forming basalt plateaux like) mantle. Comparison of Mg-15 normalized basalt com- associated with pull-apart, extensional rifts (Barr & positions with parameterized anhydrous and hydrous experimental McDonald, 1981; Whitford-Stark, 1987). Although melt compositions allowed calculation of melt segregation pressures widely dispersed the activity shares common source isoand temperatures. Computed for anhydrous conditions these range topic and lithosphere structural character with intraplate from <4 GPa and ~1470∞C (for alkali basalts) to <0·5 GPa and back-arc volcanism in the western Pacific and has and ~1400∞C (quartz tholeiites), and for H2O-undersaturated been referred to as a ‘diVuse’ igneous province (Hoang conditions, from <3·5 GPa and ~1450∞C to ~1·5 GPa and et al., 1996). The activity post-dates the early Tertiary 1350‐1400∞C, respectively. Hydrous conditions are more realistic India‐Asia collision and may be related to asthenospheric in view of high measured basalt H2O + contents, pressure and lithospheric tectonic extrusion processes (Tapponnier estimates consistent with melting below a thinned mechanical et al., 1982, 1986). The province is bounded to the east boundary layer (MBL) and interpolated mantle adiabats of and southeast by active subduction at the Izu‐Bonin, 2‐3∞C/km (compared with <1∞C/km for anhydrous conditions), Mariana and Indonesian archipelagos, and to the northconsistent with fluid dynamic constraints and a 1440∞C potential west by the collision-thickened Tibet plateau (Flower et temperature. After collision-induced ‘extrusion’ of east and al., 1998a). The Indochina and China plates appear to southeast Asia, the lithosphere was probably thinned during have been tectonically extruded along regional strike-slip faults, with concomitant opening of the South China

Petrogenesis of the Toba Tuffs, Sumatra, Indonesia

Abstract: Earth (Smith & Bailey, 1968). It is elongated in a NW–SE During the past 1·2 my, at least 3400 km of magma have been direction parallel to the active volcanic front of Sumatra, erupted in four ash flow tuff units from the Toba Caldera Complex. and it measures 100 km by 30 km (Fig. 1). Over the past This activity culminated at 74 ka with the fourth eruption, which 1·2 my, there have been four ash flow tuff eruptions from produced 2800 km of magma and formed the 100 km× 30 km the caldera complex (Chesner & Rose, 1991; Chesner et caldera visible today. A relatively homogeneous two-pyroxene dacite al., 1991). The youngest tuff was erupted at 74 ka was erupted during the first phase of activity. Magma erupted during (Ninkovich et al., 1978; Chesner et al., 1991) and has a each successive eruption was compositionally zoned, generally ranging minimum volume of 2800 km (Rose & Chesner, 1987). from rhyodacite to rhyolite. The youngest three tuff deposits contain Eruption of the Youngest Toba Tuff (YTT) is responsible up to 40 wt % crystals of quartz, sanidine, plagioclase, biotite, for the collapse structure visible today (Van Bemmelen, and amphibole. Minor minerals are magnetite, ilmenite, allanite, 1949). It consists of an extensive, mostly non-welded zircon, fayalite, and orthopyroxene; inclusions of apatite and pyrrhooutflow sheet with abundant pumice blocks ( PH2O; thus volatile units are the Middle Toba Tuff (MTT) ( Ar/Ar age oversaturation probably did not initiate the eruptions. Individual 0·50 Ma, Chesner et al., 1991), the Oldest Toba Tuff pumice blocks and fiamme collected from the youngest three units (OTT) (Ar/Ar age 0·84 Ma, Diehl et al., 1987), and record simultaneous eruption across compositional boundaries. Lowthe Haranggoal Dacite Tuff (HDT) (fission track age 1·2 energy ring fracture eruptions resulted in dense welding of all units Ma, Nishimura et al., 1977). They were erupted alexcept for the top of the youngest unit, and thick accumulations of ternately from north and south vent areas in the present rhyodacitic magma in the collapsing calderas. caldera, but they are generally exposed only in the steep caldera walls, and are densely welded (Chesner & Rose, 1991). Collectively, the YTT, MTT, and OTT are referred to as the quartz-bearing Toba tuffs. They typically contain

Calculation of peridotite partial melting from thermodynamic models of minerals and melts I. Review of methods and comparison with experiments

Abstract: Thermodynamic calculation of partial melting of peridotite using the MELTS algorithm has the potential to aid understanding of a wide range of problems related to mantle melting. We review the methodology of MELTS calculations with special emphasis on the features that are relevant for evaluating the suitability of this thermodynamic model for simulations of mantle melting. Comparison of MELTS calculations with well–characterized peridotite partial melting experiments allows detailed evaluation of the strengths and weaknesses of the algorithm for application to peridotite melting problems. Calculated liquid compositions for partial melting of fertile and depleted peridotite show good agreement with experimental trends for all oxides; for some oxides the agreement between the calculated and experimental concentrations is almost perfect, whereas for others, the trends with melt fraction are comparable, but there is a systematic offset in absolute concentration. Of particular interest is the prediction by MELTS that at 1 GPa, near–solidus partial melts of fertile peridotite have markedly higher SiO_2 than higher melt fraction liquids, but that at similar melt fractions, calculated partial melts of depleted peridotites are not SiO_2 enriched. Similarly, MELTS calculations suggest that near–solidus partial melts of fertile peridotite, but not those of depleted peridotite, have less TiO_2 than would be anticipated from higher temperature experiments. Because both experiments and calculations suggest that these unusual near–solidus melt compositions occur for fertile peridotite but not for depleted peridotite, it is highly unlikely that these effects are the consequence of experimental or model artifacts. Despite these successes of the results of calculations of peridotite melting using MELTS, there are a number of shortcomings to application of this thermodynamic model to calculations of mantle melting. In particular, calculated compositions of liquids produced by partial melting of peridotite have more MgO and less SiO_2 than equivalent experimentally derived liquids. This mismatch, which is caused by overprediction of the stability of orthopyroxene relative to olivine, causes a number of other problems, including calculated temperatures of melting that are too high. Secondarily, the calculated distribution of Na between pyroxenes and liquid does not match experimentally observed values, which leads to exaggerated calculated Na concentrations for near–solidus partial melts of peridotite. Calculations of small increments of batch melting followed by melt removal predict that fractional melting is less productive than batch melting near the solidus, where the composition of the liquid is changing rapidly, but that once the composition of the liquid ceases to change rapidly, fractional and batch melting produce liquid at similar rates per increment of temperature increase until the exhaustion of clinopyroxene. This predicted effect is corroborated by sequential incremental batch melting experiments (Hirose & Kawamura, 1994, Geophysical Research Letters, 21, 2139–2142). For melting of peridotite in response to fluxing with water, the calculated effect is that melt fraction increases linearly with the amount of water added until exhaustion of clinopyroxene (cpx), at which point the proportion of melt created per increment of water added decreases. Between the solidus and exhaustion of cpx, the amount of melt generated per increment of water added increases with temperature. These trends are similar to those documented experimentally by Hirose & Kawamoto (1995, Earth and Planetary Science Letters, 133, 463–473).

Origin of Anorthosite by Textural Coarsening: Quantitative Measurements of a Natural Sequence of Textural Development

Abstract: The textures of plagioclase crystals within large olivine oikocrysts preserve the sequence of the formation of anorthosite. Such textures have been quantified in a troctolite‐anorthosite from the Proterozoic Lac-St-Jean anorthosite complex, Canada. Crystal size distributions (CSDs) indicate that initially plagioclase nucleated and grew in an INTRODUCTION environment of linearly increasing undercooling, producing a straight- The textures of plutonic rocks are one of their most line CSD. During this phase, latent heat of crystallization was striking aspects, but they have been little used to elucidate largely removed by circulation of magma through the porous crystal petrological processes as compared with chemical and mush. By about 25% solidification the crystallinity was such that isotopic methods. Where used, discussion has tended it reduced, but did not eliminate, the circulation of magma, resulting to be qualitative, in contrast to quantitative chemical in the retention of more latent heat within the crystal pile. The methods, hence reducing the weight of this approach. temperature rose until it was buVered by the solution of plagioclase Indeed, it is not possible to verify physical models if close to the liquidus temperature of plagioclase. Nucleation of quantitative data are not available. Quantitative textural plagioclase was inhibited and conditions were suitable for textural studies of volcanic rocks are becoming more common coarsening of both plagioclase and olivine to occur (Ostwald and these methods are applied here to the development ripening). In this process small crystals were resorbed, whereas of anorthosite. larger crystals grew from both material recycled by the resorption of Quantitative textural studies of plutonic rocks really crystals smaller than the critical size and new material brought in started with Jackson’s studies of the Stillwater complex by the circulating magmatic fluid. The results of this process resemble in 1961 ( Jackson, 1961). He measured crystal size disthose of high-temperature metasomatism but there is no necessity for tributions, but did not record the area that he measured, a magmatic fluid with a diVerent composition. The shapes of the hence his data cannot be completely compared with plagioclase CSDs fit better the communicating neighbours equation of modern data. His pioneering work was not really followed textural coarsening, rather than the classic Lifshitz‐Slyozov‐Wagner up, and since then most studies have looked only at equation, as do other examples drawn from the literature. If olivine ‘average’ or ‘typical’ crystal sizes. These data are much started to nucleate at a higher temperature than plagioclase, then less revealing of petrological processes than crystal size during the textural coarsening phase olivine would have been more distributions, partly because they only give one parameter undercooled than plagioclase, and would have had a higher maximum for each sample. The few existing studies of crystal size growth rate. In these conditions olivine would coarsen more rapidly distributions in plutonic rocks have not treated than plagioclase and engulf it. Hence the order of crystallization plagioclase. determined from the textures would be the reverse of the order of One problem especially pertinent to quantitative texfirst nucleation of the two phases, from equilibrium phase diagrams. tural studies is the way that many igneous petrologists Maintenance of the temperature near the plagioclase liquidus may look at rocks: they commonly try to reconstruct how a rock has formed by examining the final, fully crystallized also inhibit the nucleation and growth of other phases.

Mineralogy of Crystallized Melt Inclusions from Gardiner and Kovdor Ultramafic Alkaline Complexes: Implications for Carbonatite Genesis

Abstract: Cumulus olivine, clinopyroxene, melilite and perovskite from silicate INTRODUCTION rocks and carbonatites of the Gardiner (East Greenland) and Kovdor Carbonatite-bearing ultramafic alkaline complexes gen(Kola Peninsula) ultramafic alkaline complexes contain primary erally exhibit chaotic field relations because of explosive melt inclusions crystallized into aggregates of daughter minerals. magmatism, metasomatic alteration and high activity of The petrography and homogenization temperatures of the inclusions volatiles. In these complexes the primary characteristics constrain the fractionation paths and the formation of carbonatites of the parental melts and the primary liquidus asin the complexes. Carbonated melanephelinite was parental to both semblages may be obscured by post-magmatic reactions, complexes and early cumulates (dunite, olivinite, peridotite and and volatile components are easily lost during crysmelteigite) are comparable. The common occurrence of phlogopite tallization together with components soluble in fluids. and amphibole in the inclusions and in the host rocks indicates Melt and fluid inclusions in liquidus minerals constitute that these were important liquidus phases. In both complexes the microscopic natural autoclaves and their study is espefractionation of phlogopite and amphibole drove the melts towards cially powerful in volatile-rich systems. The petrography Ca-rich (larnite-normative) compositions. At the ijolite stage the and composition of inclusions constrain petrogenetic evolutionary trends are believed to separate and the evolved larnitemodels for ultramafic alkaline complexes and carbonatite normative melt produced calcite-bearing ijolite in Kovdor and formation. Carbonatites and related silicate rocks have melilitolite in Gardiner. The two assemblages are related by debeen the subject of melt and fluid inclusion studies for carbonation reactions. A fractional crystallization origin is suggested several decades (e.g. Roedder, 1984, pp. 406–411), and for the Kovdor carbonatite, whereas the Gardiner carbonatite formed such studies have revealed traces of strongly alkaline by liquid immiscibility in the course of melilite fractionation. carbonatite melt associated with alkali-poor plutonic carNa–K–Ca and Na–Mg carbonates are common daughter phases bonatite (Aspden, 1981; Le Bas & Aspden, 1981; Kogarko and especially abundant in late-stage inclusions. Thus, all caret al., 1991). bonatite melts in the two complexes are alkaline. Calcite carbonatites Here we present results of an investigation of crysappear to be cumulates. tallized melt inclusions in rock-forming minerals from

Crustal Age Domains and the Evolution of the Continental Crust in the Mozambique Belt of Tanzania: Combined Sm–Nd, Rb–Sr, and Pb–Pb Isotopic Evidence

Abstract: African reworking. Eclogite-facies metapelites of the Early Prot- Mozambique Belt; Sr‐Nd‐Pb isotopes erozoic Usagaran Belt likewise exhibit Archean Nd model ages, but higher Pb isotopic ratios are consistent with last recrystallization of feldspar at 2 Ga. Granulites with Nd model ages from 1 to 1·5 Ga only occur in NE Tanzania; because of their restricted range INTRODUCTION in Pb isotopic composition they are interpreted as juvenile additions The study of ancient high-grade gneiss belts provides during late Proterozoic time. Granulites of the W Uluguru Mts important insights into the dynamics of deep-seated orohave Nd model ages between 2·1 and 2·6 Ga, and highly variable genic processes that often cannot be observed in modern feldspar Pb isotope composition indicating possible derivation from active orogenic belts because most of these expose only cratonic and/or Usagaran material, reworked and mixed with a the upper brittle parts of the continental crust. In such old small proportion of younger Proterozoic material during the Pan- and eroded belts, Pb and Nd isotopes supply particularly African orogeny. This could indicate the suture zone between a valuable information on crustal genesis, evolution and western Archean‐Proterozoic continental mass and juvenile arc- terrane amalgamation and can be used to distinguish terranes docking on from the east during subduction of the Mo- between old reworked and juvenile crust. zambique Ocean. The combined isotope data provide strong evidence This study combines Nd and Sr isotope systematics of that parts of the East African crust grew by lateral accretion of whole rocks and Pb isotopes from leached feldspars Early and Mid Proterozoic segments onto an Archean nucleus. to investigate the assembly and crustal history of the However, the Neoproterozoic (Pan-African) orogeny not only led to Proterozoic, polymetamorphic Mozambique Belt. The diVerent chemical properties of the elements determine addition of new crust in the NE of Tanzania, but also reworked

The Generation of Potassic Lavas from the Eastern Virunga Province, Rwanda

Abstract: Lavas from the eastern Virunga province, Rwanda, are dominated to lie within the continental mantle lithosphere (e.g. Fraser et al., 1986; Nelson et al., 1986; Dudas, 1991; by K-hawaiites, K-basanites and latites. All lavas are shoshonitic with 1 < K2O/Na2O < 2 and strongly enriched in incompatible Rogers et al., 1992; Gibson et al., 1995) of which they are a complementary sample to that provided by peridotite elements. Sr/Sr varies from 0·70586 in the K-basanites to 0·70990 in the latites, Nd/Nd from 0·51254 to 0·51206, xenoliths (e.g. Menzies et al., 1987; Jochum et al., 1989). In addition, their trace element and isotopic charand Pb isotopes define sub-vertical trends on isotope diagrams ( Pb/Pb 19·30–19·51, Pb/Pb 15·69–15·93 and acteristics can be used to infer the temporal evolution of the mantle lithosphere and the processes involved in Pb/Pb 40·28–41·5). Ar/Ar ages of leucite and phlogopite separates suggest that the latites are between 100 and 200 ka and lithosphere stabilization. Potassic magmas are also among the most compositionally extreme products of processes the K-basanites <100 ka. The latites are hybrid magmas produced by mixing between a K-basanite melt with a silicic melt from the that scavenge and fractionate incompatible elements in the upper mantle. Hence their geochemical variations deep crust. The low-silica K-basanites reflect interaction between a mafic K-basanitic melt with Nd/Nd ~0·51204, Sr/Sr can also be used to investigate the time-integrated effects of these processes on the radiogenic isotope evolution of ~0·707, and a nephelinite with Nd/Nd ~0·51267 and Sr/Sr ~0·7045. Both are derived from the mantle lithosphere the mantle lithosphere and to explore links between their lithospheric source regions and those of ocean-island with source ages of 1 Ga and 0·5 Ga, respectively, and the youngest ages correspond to the deepest magma sources. The magma production basalt (OIB) (McKenzie & O’Nions, 1983, 1995; Turner et al., 1996). rate in the Virunga is low (~0·04 km/yr), and reflects prolonged (10–15 My) heating of the lithosphere by the East African mantle The Virunga province of the western branch of the African Rift is a classic example of intra-plate potassic plume. magmatism, and the Ugandan part of the province was first described in remarkable detail by Holmes & Harwood (1937). They presented both excellent field and petrographic descriptions of mineralogically and

Carbonatites—Into the Twenty-First Century

Abstract: s, p. A-139. ephemeral carbonatite melts. Nature 336, 459–462. Gittins, J., Beckett, M. F. & Jago, B. C. (1990). Composition of the Tilton, G. R. & Bell, K. (1994). Sr–Nd–Pb isotope relationships in Late Archean carbonatites and alkaline complexes: applications to fluid phase accompanying carbonatite magma: a critical evaluation. American Mineralogist 75, 1106–1109. the geochemical evolution of the Archean mantle. Geochimica et Cosmochimica Acta 58, 3145–3154. Hamilton, D. L., Bedson, P. J. & Esson, J. (1989). The behaviour of trace elements in the evolution of carbonatites. In: Bell, K. (ed.) Tuttle, O. F. & Gittins, J. (1966). Carbonatites. London: John Wiley. Twyman, J. D. & Gittins, J. (1987). Alkalic carbonatite magmas: Carbonatites: Genesis and Evolution. London: Unwin Hyman, pp. 405– 427. parental or derivative? In: Fitton, J. G. & Upton, B. G. J. (eds) Alkaline Igneous Rocks. Geological Society, London, Special Publication 30, Harmer, R. E. & Gittins, J. (1997). The origin of dolomitic carbonatites: field and experimental constraints. Journal of African Earth Sciences 25, 85–94. Wyllie, P. J. & Huang, W.-L. (1975). Peridotite, kimberlite and car5–28. Hart, S. R. (1988). Heterogeneous mantle domains: signatures, genesis bonatite explained in the system CaO–MgO–SiO2–CO2. Geology 3, 621–624. and mixing chronologies. Earth and Planetary Science Letters 90, 273–296.

Petrogenesis of Carbonatite Magmas from Mantle to Crust, Constrained by the System CaO–(MgO + FeO*)–(Na2O + K2O)–(SiO2 + Al2O3 + TiO2)-CO2

Abstract: Experimental results from the systems CaO–MgO–SiO_2–CO_2, Na_2O–CaO–Al_2O_3–SiO_2–CO_2, and a primitive magnesian nephelinite mixed with carbonates have been combined for construction of phase diagrams for the pseudoquaternary system CaO–(MgO + FeO*)–(Na_2O + K_2O)–(SiO_2 + Al_2O_3 + TiO_2) with CO_2 at 1.0 and 2.5 GPa pressure. These diagrams provide a petrogenetic framework for magmatic processes from mantle to deep crust, with particular reference to the melting products of carbonate peridotite and the paths of crystallization of carbonated silicate magmas toward carbonatite magmas, with or without the intervention of silicate–carbonate liquid immiscibility. Three key features control these processes: (1) the liquidus surface bounding the silicate–carbonate liquid miscibility gap, (2) the silicate–carbonate liquidus boundary surface which separates the liquidus volume for primary silicates from that for primary carbonates, and (3) the curve of intersection of these two surfaces (1 and 2) which defines the coprecipitation of silicates and calcite with coexisting immiscible silicate- and carbonate-rich liquids. The geometrical arrangement of the two surfaces varies as a function of both pressure and bulk composition (e.g. with Si/Al, Na/K, Mg/Fe). Surface (2) is the locus of initial liquids from partial melting of carbonate–silicate assemblages, and the limit for residual liquid compositions derived from silicate–CO_2 liquids. The carbonate liquidus volume is a forbidden region for carbonate-rich magmas derived from silicate magmas at the pressures investigated. The immiscible liquids dissolve no more than 80 wt % CaCO_3, and the miscibility gap (MG) becomes smaller with increasing Mg/Ca. Extrapolation of experimental data indicates that the MG disappears with more than ∼50 wt % (MgO + FeO*) at 1.0 GPa for the compositions investigated. The distance between the miscibility gap, surface (1), and the silicate–carbonate liquidus surface, surface (2), increases significantly with increasing (MgO + FeO*). This observation, coupled with knowledge of the phase boundaries in the system, allows comparisons with projected rock compositions, and this permits the following conclusions. Calciocarbonatites and natrocarbonatites are excluded as candidates for primary magmas from the mantle, which must have compositions dominated by calcic dolomite. The formation of (equilibrium) carbonate-rich liquids immiscible with silicate magmas in the mantle is unlikely, which denies the formation of CaCO_3 ocelli in mantle xenoliths as immiscible liquids. Immiscible carbonate-rich magmas separated from many silicate magmas may tend to be concentrated near calciocarbonatite compositions, with maximum CaCO_3 75–80 wt %, low (MgO + FeO*), and (Na,K)_2CO_3 near 15 wt %. Silicate parents with higher Na/Ca and peralkalinity may yield immiscible magmas approaching natrocarbonatite compositions. Exsolution of immiscible carbonate-rich magma occurs without the coprecipitation of calcite except along the limiting field boundary (3). Only after the carbonate-rich magma is physically separated from the parent magma, and cooled with the precipitation of silicates, does it reach the silicate–carbonate field boundary and precipitate cumulate carbonatites, with inevitable enrichment of residual liquids in alkalis. Calciocarbonatite magmas cannot be derived from natrocarbonatite magmas. Dolomitic carbonatite magmas cannot be formed by liquid immiscibility, but only by fractionation of calciocarbonatites (according to CaCO_3–MgCO_3), or as primary magmas.