Geochemistry and petrogenesis of Biabanak-Bafq mafic magmatism: Implication for the evolution of central Iranian terrane

Precambrian magmatism in the Biabanak-Bafq district represents an extensive sequence of mafic magmatic rocks. Major, trace and rare earth elements reveal that the low-Ti basement mafic rocks are magnesium tholeiite and low-Ti cover a mafic rock belongs to Fe-tholeiite, whereas, the high-Ti alkaline mafic rocks, as well as dolerites, show much more Fe–Ti enrichment. Primitive mantle normalized trace element patterns show a relative enrichment of LREE and LILE and depletion of HFSE, but have an equally distinct continental signature reflected by marked negative Nb, Sr, P, and Ti anomalies. The composition of the intrusive rocks is consistent with fractional crystallization of olivine ± clinopyroxene ± plagioclase, whereas variations in the Sr and Nd isotope compositions suggest heterogeneous sources and crustal contamination. Low-Ti group samples contain a crustal signature in the form of high La/Yb, Zr/Nb, and negative \(\varepsilon \hbox {Nd}\) values. In contrast, high-Ti mafic magmatic rocks display an increase in La/Yb with a decrease in Proterozoic alkaline rocks recognized across the central Iran. The presence of diverse mafic magmatic rocks probably reflects heterogeneous nature of sub-continental lithospheric mantle (SCLM) source. The mafic magmatism largely represents magmatic arc or rift tectonic setting. It is suggested that the SCLM sources were enriched by subduction processes and asthenospheric upwelling.     

Cenozoic intra-plate magmatism in the Darfur volcanic province: mantle source, phonolite-trachyte genesis and relation to other volcanic provinces in NE Africa

Chemical and Sr, Nd and Pb isotopic compositions of Late Cenozoic to Quaternary small-volume phonolite, trachyte and related mafic rocks from the Darfur volcanic province/NW-Sudan have been investigated. Isotope signatures indicate variable but minor crustal contributions. Some phonolitic and trachytic rocks show the same isotopic composition as their primitive mantle-derived parents, and no crustal contributions are visible in the trace element patterns of these samples. The magmatic evolution of the evolved rocks is dominated by crystal fractionation. The Si-undersaturated strongly alkaline phonolite and the Si-saturated mildly alkaline trachyte can be modelled by fractionation of basanite and basalt, respectively. The suite of basanite–basalt–phonolite–trachyte with characteristic isotope signatures from the Darfur volcanic province fits the compositional features of other Cenozoic intra-plate magmatism scattered in North and Central Africa (e.g., Tibesti, Maghreb, Cameroon line), which evolved on a lithosphere that was reworked or formed during the Neoproterozoic.     

Plio－Quaternary magmatism in Italy

Plio-Quaternary magmatism in Italy exhibits an extremely variable composition, which spans almost entirely the spectrum of magmatic rocks occurring worldwide. Petrological and geochemical data pro-vide a basis for distinguishing various magmatic provinces, which show different maior element and/or trace element and/or isotopic compositions. The Tus-cany province (14-0.2 Ma) consists of silicic magmas generated through crustal anatexis, and of mantle-derived calcalkaline to ultrapotassic mafic rocks. The Roman, Umbria, Ernici-Roccamonfina and Neapolitan provinces (0.8 Ma to present) are formed by mantle-derived potassic to ultrapotassic rocks having variable trace element and isotopic compositions. The Aeolian arc (?1 Ma to present) mainly consists of calcalkaline to shoshonitic rocks. The Sicily province contains young to active centers (notably Etna) with a tholeiitic to Na-alkaline affinity. Finally, volcanoes of variable composition occur in Sardinia and, as seamounts, on the Tyrrhenian Sea floor. Magmas in the Aeolian arcand along the Italian peninsula have a subduction-related geochemical character, whereas the Sicily and Sardinia provinces display intraplate signatures.Intraplate and orogenic volcanics coexist on theTvrrhenian Sea floor.The geochemical and isotopic complexities of Plio-Quaternary magmatism reveal that the upper mantle beneath Italy consists of various domains, spanning both orogenic and anorogenic compositions. Isotopic data suggest that compositional heterogeneity originated from mixing between various mantle reservoirs, and between these and subduction-related crustal material.This probably occurred during the Cenozoic-Quaternary geodynamic evolution of the western Mediterranean.     

The evolution of a silicic magma system: isotopic and chemical evidence from the Woods Mountains volcanic center,eastern California

The isotopic compositions of Nd and Sr and concentrations of major and trace elements were measured in flows and tuffs of the Woods Mountains volcanic center of eastern California to assess the relative roles of mantle versus crustal magma sources and of fractional crystallization in the evolution of silicic magmatic systems. This site was chosen because the contrast in isotopic composition between Precambrian-to-Mesozoic country rocks and the underlying mantle make the isotope ratios sensitive indicators of the proportions of crustal- and mantle-derived magma. The major eruptive unit is the Wild Horse Mesa tuff (15.8 m.y. old), a compositionally zoned rhyolite ignimbrite. Trachyte pumice fragments in the ash-flow deposits provide information on intermediate composition magma types. Crustal xenoliths and younger flows of basalt and andesite (10 m.y. old) provide opportunities to confirm the isotopic compositions of potential mantle and crustal magma sources inferred from regional patterns. The trachyte and rhyolite have _Nd values of -6.2 to -7.5 and initial ⁸⁷Sr/⁸⁶Sr ratios mostly between 0.7086 and 0.7113. These magmas cannot have been melted directly from the continental basement because the _Nd values are too high. They also cannot have formed by closed system fractional crystallization of basalt because the ⁸⁷Sr/⁸⁶Sr ratios are higher than likely values for parental basalt. Both major and trace element variations indicate that crystal fractionation was an important process. These results require that the silicic magmas are end products of the evolution of mantle-derived basalt that underwent extensive fractional crystallization accompanied by assimilation of crustal rock. The mass fraction of crustal components in the trachyte and rhyolite is estimated to be between 10% and 40%, with the lower end of the range considered more likely. The generation of magmas with SiO₂ contents greater than 60% appears to be dominated by crystal fractionation with minimal assimilation of upper crustal rocks.     

Cenozoic continental arc magmatism and associated mineralization in Ecuador

Most of the economic ore deposits of Ecuador are porphyry-Cu and epithermal style gold deposits associated with Tertiary continental arc magmatism. This study presents major and trace element geochemistry, as well as radiogenic isotope (Pb, Sr) signatures, of continental arc magmatic rocks of Ecuador of Eocene to Late Miocene (~50–9 Ma, ELM) and Late Miocene to Recent (~8–0 Ma, LMR) ages. The most primitive ELM and LMR rocks analyzed consistently display similar trace element and isotopic signatures suggesting a common origin, most likely an enriched MORB-type mantle. In contrast, major and trace element geochemistry, as well as radiogenic isotope systematics of the whole sets of ELM and LMR samples, indicate strikingly different evolutions between ELM and LMR rocks. The ELM rocks have consistently low Sr/Y, increasing Rb/Sr, and decreasing Eu/Gd with SiO₂, suggesting an evolution through plagioclase-dominated fractional crystallization at shallow crustal levels (<20 km). The LMR rocks display features of adakite-type magmas (high Sr/Y, low Yb, low Rb/Sr) and increasing Eu/Gd and Gd/Lu ratios with SiO₂. We explain the adakite-type geochemistry of LMR rocks, rather than by slab melting, by a model in which mantle-derived melts partially melt and assimilate residual garnet-bearing mafic lithologies at deeper levels than those of plagioclase stability (i.e., >20 km), and most likely at sub-crustal levels (>40–50 km). The change in geochemical signatures of Tertiary magmatic rocks of Ecuador from the ELM- to the LMR-type coincides chronologically with the transition from a transpressional to a compressional regime that occurred at ~9 Ma and has been attributed by other investigations to the onset of subduction of the aseismic Carnegie ridge.The major districts of porphyry-Cu and epithermal deposits of Ecuador (which have a small size, <<200 Mt, when compared to their Central Andean counterparts) are spatially and temporally associated with ELM magmatic rocks. No significant porphyry-Cu and epithermal deposits (except the epithermal high-sulfidation mineralization of Quimsacocha) appear to be associated with Late Miocene-Recent (LMR, ~8–0 Ma) magmatic rocks. The apparent infertility of LMR magmas seems to be at odds with the association of major porphyry-Cu/epithermal deposits of the Central Andes with magmatic rocks having adakite-type geochemical signatures similar to LMR rocks. The paucity of porphyry-Cu/epithermal deposits associated with LMR rocks might be only apparent and bound to exposure level, or real and bound (among other possibilities) to the lack of development of shallow crustal magmatic chambers since ~9 Ma as a result of a prolonged compressional regime in the Ecuadorian crust. More work is needed to understand the actual metallogenic potential of LMR rocks in Ecuador.Editorial handling: J. Richards     

The Geochemistry of Ultramafic to Mafic Volcanics from the Belingwe Greenstone Belt, Zimbabwe: Magmatism in an Archean Continental Large Igneous Province

The evolution of the late Archean Belingwe greenstone belt,Zimbabwe, is discussed in relation to the geochemistry of theultramafic to mafic volcanic rocks. Four volcanic types (komatiite,komatiitic basalt, D-basalt and E-basalt) are distinguishedin the 2·7 Ga Ngezi volcanic sequence using a combinationof petrography and geochemistry. The komatiites and D-basaltsare rocks in which isotopic systems and trace elements are depleted.Chemical variations in komatiites and D-basalts can be explainedby fractional crystallization from the parental komatiite. Incontrast, komatiitic basalts and E-basalts are siliceous anddisplay enriched isotopic and trace element compositions. Theirchemical trends are best explained by assimilation with fractionalcrystallization (AFC) from the primary komatiite. AFC calculationsindicate that the komatiitic basalts and E-basalts are derivedfrom komatiites contaminated with 20% and 30% crustal material,respectively. The volcanic stratigraphy of the Ngezi sequence,which is based on field relationships and the trace elementcompositions of relict clinopyroxenes, shows that the leastcontaminated komatiite lies between highly contaminated komatiiticbasalt flows, and has limited exposure near the base of thesuccession. Above these flows, D- and E-basalts alternate. Thekomatiite appears to have erupted on the surface only in theearly stages, when plume activity was high. As activity decreasedwith time, komatiite magmas may have stagnated to form magmachambers within the continental crust. Subsequent komatiiticmagmas underwent fractional crystallization and were contaminatedwith crust to form D-basalts or E-basalts. KEY WORDS: komatiite; crustal assimilation; Belingwe greenstone belt; continental flood basalt; plume magmatism     

Evidence for Multiple Mechanisms of Crustal Contamination of Magma from Compositionally Zoned Plutons and Associated Ultramafic Intrusions of the Alaska Range

Models of continental crustal magmagenesis commonly invoke theinteraction of mafic mantle-derived magma and continental crustto explain geochemical and petrologic characteristics of crustalvolcanic and plutonic rocks. This interaction and the specificmechanisms of crustal contamination associated with it are poorlyunderstood. An excellent opportunity to study the progressiveeffects of crustal contamination is offered by the compositeplutons of the Alaska Range, a series of nine early Tertiary,multiply intruded, compositionally zoned (Peridotite to granite)plutons. Large initial Sr and Nd isotopic contrasts betweenthe crustal country rock and likely parental magmas allow evaluationof the mechanisms and extents of crustal contamination thataccompanied the crystallization of these ultra-mafic throughgranitic rocks. Three contamination processes are distinguishedin these plutons. The most obvious of these is assimilationof crustal country rock concurrent with magmatic fractionalcrystallization (AFC), as indicated by a general trend towardcrustal-like isotopic signatures with increasing differentiation.Second, many ultramafic and mafic rocks have late-stage phenocrystreaction and orthocumulate textures that suggest interactionwith felsic melt. These rocks also have variable and enrichedisotopic compositions that suggest that this felsic melt wasisotopically enriched and probably derived from crustal countryrock. Partial melt from the flysch country rock may have reactedwith and contaminated these partly crystalline magmas followingthe precipitation and accumulation of the cumulus phenocrystsbut before complete solidification of the magma. This suggeststhat in magmatic mush (especially of ultramafic composition)crystallizing in continental crust, a second distinct processof crustal contamination may be super imposed on AFC or magmamixing involving the main magma body. Finally, nearly all rocks,including mafic and ultramafic rocks, have (⁸⁷Sr/⁸⁶Sr)_i thatare too high, and (T) _Nd that are too low, to represent theexpected isotopic composition of typical depleted mantle. However,gabbro xenoliths with typical depicted-mantle isotopic compositionsare found in the plutons. This situation requires either anadditional enriched mantle component to provide the parentalmagma for these plutons, or some mechanism of crustal contaminationof the parent magma that did not cause significant crystallizationand differentiation of the magma to more felsic compositions.Thermodynamic modeling indicates that assimilation of alkali-andwater-rich partial melt of the metapelite country rock by fractionating,near-liquidus basaltic magma could cause significant contaminationwhile suppressing significant crystallization and differentiation. KEY WORDS: crustal contamination; Alaska Range; isotope geochemistry; zoned plutons; assimilation ^*Corresponding author. e-mail: preiners{at}u.washington.edu; fax: (206) 543-3836.     

Rocks mineralogy controls the metasomatic activities: an indication from water/rock ratio modeling

The final ratio equation of an isotopic element in a rock, derived from water/rock formula of McCulloch et al. Earth Planet Sci Lett 46:201-211, 1980, McCulloch et al. J Geophys Res 86:B4 2721-2735, 1981 is used to assess the behavior of diverse suites of rocks towards the alteration effect, and what implications can give about hydrothermal alteration in terms of isotopic compositions. Due to their higher Sr and lower Nd initial ratios than seawater, rocks of metamorphic and sedimentary signatures such as carbonates and Precambrian basement rocks show similar but inverse mixing curves compared with igneous rocks. Sr composition of rocks immediately alters by seawater, while Nd composition keeps unchanged until large volumes of water are added. Although, this can be attributed to the very low Nd concentration in seawater, it indicates that Nd-exchange may only take place under seawater, possibly hydrothermally by circulated seawater, and Nd-concentration of less altered crustal rocks are apparently primary. The isotopic composition and rock mineralogy seem to be the main factors controlling the volume of water required to cause isotopic alteration in rocks. Crustal rocks require higher water volumes due to their relatively low temperature minerals, whereas, mantle peridotites mainly consist of residual olivine minerals that are highly susceptible to alteration and lack of Sr and Nd compositions, and so need less amount of water for metasomatism. This property reduces the limited penetration effect as the mafic affinity increases at depth in the oceanic crust, and enables modified (probably acidified) circulated fluids to maintain ion exchanging and leaching throughout their passageway.     

Variations in the Nd isotopic ratios and canonical ratios of concentrations of incompatible elements as an indication of mixing sources of alkali granitoids and basites in the Khaldzan-Buregtei massif and the Khaldzan-Buregtei rare-metal deposit in western Mongolia

The paper reports data on the Nd isotopic composition and the evaluated composition of the sources of magmatism that produced massifs of alkali and basic rocks of the Khaldzan-Buregtei group. The massifs were emplaced in the terminal Devonian at 392–395 Ma in the Ozernaya zone of western Mongolia. The host rocks of the massifs are ophiolites of the early Caledonian Ozernaya zone, which were dated at 545–522 Ma. The massifs were emplaced in the following succession (listed in order from older to younger): (1) nordmarkites and dolerites syngenetic with them; (2) alkali granites and syngenetic dolerites; (3) dike ekerites; (4) dike pantellerites; (5) rare-metal granitoids; (6) alkali and intermediate basites and quartz syenites; and (7) miarolitic rare-metal alkali granites. Our data on the Nd isotopic composition [?_Nd(T)] and conventionally used (canonical) ratios of incompatible elements (Nb/U, Zr/Nb, and La/Yb) in rocks from the alkaline massifs and their host ophiolites indicate that all of these rocks were derived mostly from mantle and mantle-crustal enriched sources like OIB, E-MORB, and IAB with a subordinate contribution of N-MORB (DM) and upper continental crustal material. The variations in the ?_Nd(T) values in rocks of these massifs suggest multiple mixing of the sources or magmas derived from them when the massifs composing the Khaldzan-Buregtei group were produced. The OIB and E-MORB sources were mixed when the rocks with mantle signatures were formed. The occurrence of nordmarkites, alkali granites, and other rocks whose isotopic and geochemical signatures are intermediate between the values for mantle and crustal sources testifies to the mixing of mantle and crustal magmas. The crustal source itself, which consisted of rocks of the ophiolite complex, was obviously isotopically and geochemically heterogeneous, as also were the magmas derived from it. The model proposed for the genesis of alkali rocks of the Khaldzan-Buregtei massifs implies that the magmas were derived at two major depth levels: (1) mantle, at which the plume source mixed with an E-MORB source, and (2) crustal, at which the ophiolites were melted, and this gave rise to the parental magmas of the nordmarkites and alkali granites. The basites were derived immediately from the mantle. The mantle syenites, pantellerites, and rare-metal granitoids were produced either by the deep crystallization differentiation of basite magma or by the partial melting of the parental basites and the subsequent crystallization differentiation of the generated magmas. Differentiation likely took place in an intermediate chamber at depth levels close to the crustal (ophiolite) level of magma generation. Only such conditions could ensure the intense mixing of mantle and crustal magmas. The principal factor initiating magma generation in the region was the mantle plume that controlled within-plate magmatism in the Altai-Sayan area and the basite magmas related to this plume, which gave rise to small dikes and magmatic bodies in the group of intrusive massifs.     

Geochemical and isotopic characteristics and evolution of the Jurassic volcanic arc between Arica (18°30′S) and Tocopilla (22°S), North Chilean Coastal Cordillera

The present-day North Chilean Coastal Cordillera between 18°30′S and 22°S records an important part of the magmatic evolution of the Central Andes during the Jurassic. Calc-alkaline to subordinate tholeiitic members from four rock groups with biostratigraphically constrained age display incompatible element pattern characteristic of convergent plate-margin volcanism, whereas alkaline basalts of one group occurring in the Precordillera show OIB-type trace element signatures. The correlation of biostratigraphic ages, regional distribution, and composition of the volcanic rocks provides a basis for the discussion on geochemical evolution and isotope ratios.Major and trace element distributions of the volcanic rocks indicate their derivation from mantle-derived melts. LILE and LREE enrichments in calc-alkaline basaltic andesites to dacites and some of the tholeiites hint at the involvement of hydrous fluids during melting and mobile element transport processes. A part of the Early Bajocian to ?Lower Jurassic and Oxfordian andesites and dacites are adakite-like rocks with a substantial participation of slab melt and are characterized by high Sr/Y ratios and low HREE contents. The Middle Jurassic tholeiitic and calc-alkaline basalts and basaltic andesites have been transported and partly stored within a system of deep-seated feeder fissures and crustal strike-slip faults before eruption.The isotopic composition of Sr (⁸⁷Sr/⁸⁶Sr_i=0.7032-0.7056) and Nd (εNd_i=2.2-7.1) of the Jurassic volcanic rocks mostly fall in the range characteristic for mantle melts although some crustal components may have been involved. A few samples show slightly more radiogenic Sr isotopic composition, which is probably due to interaction with ancient sea-water. The Pb isotopic composition of the arc rocks is uncoupled from the isotopic composition of Sr and Nd and is dominated by the crustal component. Since the Cretaceous and Modern arc volcanic rocks show Pb isotopic compositions that can be largely explained by in situ Pb isotope growth of Jurassic arc volcanic rocks, we argue that the various Andean arc systems between 18°30′S and 22°S formed on the same type of basement.Most of the investigated samples have high Ba, Zr, and Th concentrations compared to island arc mafic volcanic rocks. About 20% of the Jurassic arc volcanics comprise of dacitic to rhyolitic rocks. These characteristics combined with the Pb isotopic composition that shows the influence of a Palaeozoic (or partly older) basement point to a continental margin setting for the North Chilean Jurassic arc. The distribution of the magmatic rocks throughout time, their textures, and the character of intercalated sedimentary rocks reflect westward movement of the magma sources and of the arc/back-arc boundary relative to the current coast line during the Early Bajocian on a broad front between 19°30′ and 21°S.     

Paleoproterozoic (1.90–1.86 Ga) arc volcanism in the Flin Flon Belt,Trans-Hudson Orogen,Canada

Geochemical and isotopic (Nd, Sr) data are reported on Paleoproterozoic (1904–1864 Ma), maficintermediate (<63% SiO₂), arc metavolcanic rocks from the Flin Flon greenstone belt, Manitoba and Saskatchewan. Major element criteria permit subdivision of the rocks into tholeiitic (TH), calc-alkaline (CA), alkaline, and boninitic (BO) magma series. Subaqueously erupted, TH and related CA basalt-basaltic andesite, and rare high-Ca boninites dominated between 1904 Ma and 1890 Ma. The TH rocks are similar to modern island are tholeiites, having low high-field-strength element (HFSE) and rare earth element (REE) abundances, and chondrite-normalized light REE depletion to slight enrichment. The boninites have even lower HFSE and REE abundances (1–2X chondrites). Along with their extreme ratios of refractory incompatible elements (e.g., high Al/Ti, Ti/Zr, low Ti/V, Zr/Y), these features indicate that the arc mantle source was strongly depleted, probably residual after MORB or back-arc basin basalt extraction. Elevated Th/Yb, Ba/La, La/Nb values, and the spread in Nd isotopic compositions (initial _Nd=–0.4 to +4.8) suggest recycling of small amounts (0–8%) of Archean and possibly older Proterozoic crust via sediment subduction and, locally, intracrustal contamination. Calcalkaline andesite-rhyolite and rare shoshonite and trachyandesite, erupted between 1890 Ma and 1864 Ma, are more strongly light REE enriched and have comparatively higher HFSE abundances, and higher Zr/Y and Nb/Y values. The rocks have strong arc trace element signatures (e.g., high Th/Nb, La/Nb), and initial _Nd values (+2.3 to +4.6) indicate that depleted mantle contributions to the magmas continued to be dominant. The geochemistry and geology of these younger volcanic rocks suggest a mature island arc setting in which the arc lithosphere was thicker than in the previous period, and a more fertile sub-arc mantle source was tapped. The pre-1890 Ma volcanism occurred in one or more separate arcs, probably characterized by rapid subduction of oceanic lithosphere, relatively thin, tholeiitic arc crust, and extensive backarc basin formation. In contrast, post-1890 Ma volcanism is dominantly calc-alkaline to (rarely) alkaline, and is interpreted to reflect crustal thickening due to longterm growth of arc edifice(s) and tectonic thickening associated with intraoceanic arc-arc (>1870 Ma) collision and subsequent intra-arc deformation.     

Isotopic constraints on the origin of AMCG-suite rocks on the Lofoten Islands, N Norway

Summary ¶Sm–Nd, Rb–Sr and Pb–Pb isotopic compositions of 34 intrusive AMCG (anorthosite–mangerite–charnockite–granite) suite rocks and spatially related ferrodiorites and gabbros from the Lofoten Islands, northern Norway, suggest that almost all Lofoten intrusive rocks can be modelled as a mixture of mainly two components: Archean lower crustal material and an about 1.8Ga mantle-derived component. Isotopically, the gabbros and anorthosites overlap the mangeritic rocks in all three isotope systems. Hence, the isotopic data are in agreement with a model that relates the mangeritic rocks to the fractionation of crustally contaminated mantle-derived basaltic melts. Overlap in all three isotopic systems indicates that at least part of the assimilation process predates or accompanies fractionation. Whole rock geochemistry supports this model: based on major, minor, and trace element data, primitive 1.79–1.86Ga mangerites could have formed from anorthosite residual liquids – the ferrodiorites – by fractionation processes involving Fe–Ti oxides, Fe–Mg silicates and apatite combined with contamination of the ferrodiorites with about 50wt.% crustal anatectic melt.Received May 7, 2002; revised version accepted November 26, 2002 Published online: April 14, 2003     

Accreted fragments of the Late Cretaceous Caribbean–Colombian Plateau in Ecuador

The eastern part of the Western Cordillera of Ecuador includes fragments of an Early Cretaceous (≈123 Ma) oceanic plateau accreted around 85–80 Ma (San Juan–unit). West of this unit and in fault contact with it, another oceanic plateau sequence (Guaranda unit) is marked by the occurrence of picrites, ankaramites, basalts, dolerites and shallow level gabbros. A comparable unit is also exposed in northwestern coastal Ecuador (Pedernales unit).

Picrites have LREE-depleted patterns, high Nd_i and very low Pb isotopic ratios, suggesting that they were derived from an extremely depleted source. In contrast, the ankaramites and Mg-rich basalts are LREE-enriched and have radiogenic Pb isotopic compositions similar to the Galápagos HIMU component; their Nd_i are slightly lower than those of the picrites. Basalts, dolerites and gabbros differ from the picrites and ankaramites by flat rare earth element (REE) patterns and lower Nd; their Pb isotopic compositions are intermediate between those of the picrites and ankaramites. The ankaramites, Mg-rich basalts, and picrites differ from the lavas from the San Juan–Multitud Unit by higher Pb ratios and lower Nd_i.

The Ecuadorian and Gorgona 88–86 Ma picrites are geochemically similar. The Ecuadorian ankaramites and Mg-rich basalts share with the 92–86 Ma Mg-rich basalts of the Caribbean–Colombian Oceanic Plateau (CCOP) similar trace element and Nd and Pb isotopic chemistry. This suggests that the Pedernales and Guaranda units belong to the Late Cretaceous CCOP. The geochemical diversity of the Guaranda and Pedernales rocks illustrates the heterogeneity of the CCOP plume source and suggests a multi-stage model for the emplacement of these rocks. Stratigraphic and geological relations strongly suggest that the Guaranda unit was accreted in the late Maastrichtian (≈68–65 Ma).     

Petrogenesis of the magmatic complex at Mount Ascutney,Vermont, USA

The Ascutney Mountain igneous complex in eastern Vermont, USA, is composed of three principal units with compositions ranging from gabbro to granite. Sr and O isotopic and major element relationships for mafic rocks, granites, and nearby gneissic and schistose country rock have been investigated in order to describe the petrogenesis of the mafic suite which ranges from gabbro to diorite. The entire complex appears to have been formed within a short interval 122.2±1.2 m.y. ago. The granites with ¹⁸O near +7.8 had an initial ⁸⁷Sr/⁸⁶Sr of 0.70395(±6) which is indistinguishable from the initial ratio of the most primitive gabbro. Initial ⁸⁷Sr/⁸⁶Sr ratios and ¹⁸O values for the mafic rocks range from 0.7039 to 0.7057 and +6.1 to +8.6, respectively. The isotopic ratios are highly correlated with major element trends and reflect considerable crustal contamination of a mantle-derived basaltic parent magma. The likely contaminant was Precambrian gneiss similar to exposed bedrock into which the basic rocks were emplaced. A new approach to modelling of assimilation during the formation of a cogenetic igneous rock suite is illustrated. Chemical and isotopic modelling indicate that the mafic rocks were produced by simultaneous assimilation and fractional crystallization. The relative amounts of fractionation and assimilation varied considerably. The mafic suite was not produced by a single batch of magma undergoing progressive contamination; rather, the various rocks probably were derived from separate batches of magma each of which followed a separate course of evolution. The late stage granite was apparently derived from basaltic magma by fractionation with little or no crustal assimilation. The early intrusive phases are much more highly contaminated than the final one. The observed relationships have important implications for the formation of comagmatic complexes and for isotopic modelling of crustal contamination.     

Trace element and isotopic variations in a zoned pluton and associated volcanic rocks,Unalaska Island,Alaska: A model for fractionation in the Aleutian calcalkaline suite

Trace elements, including rare earth elements (REE), exhibit systematic variations in plutonic rocks from the Captains Bay pluton which is zoned from a narrow gabbroic rim to a core of quartz monzodiorite and granodiorite. The chemical variations parallel those in the associated Aleutian calcalkaline volcanic suite. Concentrations of Rb, Y, Zr and Ba increase as Sr and Ti decrease with progressive differentiation. Intermediate plutonic rocks are slightly enriched in light REE (La/Yb=3.45–9.22), and show increasing light REE fractionation and negative Eu anomalies (Eu/Eu^*=1.03–0.584). Two border-zone gabbros have similar REE patterns but are relatively depleted in total REE and have positive Eu anomalies; indicative of their cumulate nature. Initial ⁸⁷Sr/⁸⁶Sr ratios in 8 samples (0.70299 to 0.70377) are comparable to those of volcanic rocks throughout the arc and suggest a mantle source for the magmas. Oxygen isotopic ratios indicate that many of the intermediate plutonic rocks have undergone oxygen isotopic exchange with large volumes of meteoric water during the late stages of crystallization; however no trace element or Sr isotopic alteration is evident.Major and trace element variations are consistent with a model of inward fractional crystallization of a parental high-alumina basaltic magma at low pressures (6 kb). Least-squares approximations and trace element fractionation calculations suggest that differentiation in the plutonic suite was initially controlled by the removal of calcic plagioclase, lesser pyroxene, olivine and Fe-Ti oxides but that with increasing differentiation and water fugacity the removal of sub-equal amounts of sodic plagioclase and hornblende with lesser Fe-Ti oxides effectively drove residual liquids toward dacitic compositions. Major and trace element compositions of aplites which intrude the pluton are not adequately explained by fractional crystallization. They may represent partial melts derived from the island arc crust. Similarities in Sr isotopes, chemical compositions and differentiation trends between the plutonic series and some Aleutian volcanic suites indicates that shallow-level fractional crystallization is a viable mechanism for generating the Aleutian calcalkaline rock series.LDGO Contribution no. 2964     

The Mesoproterozoic Midsommersø dolerites and associated high-silica intrusions, North Greenland: crustal melting, contamination and hydrothermal alteration

The Midsommersø dolerites and the flood basalts of the Zig-Zag Dal Basalt Formation, eastern North Greenland, represent a major Mesoproterozoic (∼1,380 Ma) igneous event. The intrusive rocks form large sheets within a succession of feldspathic sandstones which underlie the basalts. The geochemistry of the basalts has recently been re-investigated and reported elsewhere in this Journal; here we present new trace element and Nd-, Sr- and Pb-isotopic data for the intrusive rocks. Unlike the basalts, the intrusions yield evidence of considerable interaction and contamination with upper crustal rocks, especially the sandstones. High-silica rocks (80–90 wt% SiO₂) occur in sheets, up to 60 m thick. They were formed by mobilisation of sandstones, and indicate a very high rate of emplacement of hot basic magma into the sandstones at depth. These mobilised sandstones (‘rheopsammites’) are among the most SiO₂-rich intrusive rocks on earth. Sheets of remobilised granitoid rocks from the crystalline basement (∼70% SiO₂) are also present. Hydrothermal activity, associated with the igneous event, significantly changed the compositions of the silicic rocks as well as that of many dolerites. Sheets of hydrothermally altered (‘red’) dolerites and silicic rocks invariably have borders of dark, fresh dolerite; this is interpreted to be the result of intrusion from zoned magma chambers. Nd isotope data confirm the crustal origin of the silicic rocks as well as the contamination of some dolerites by components derived from crustal sources, while Sr- and Pb-isotopic systems are strongly affected by the hydrothermal alteration, and give little information on the petrogenesis of the rocks. Recent loss of Sr from hydrothermally altered rocks further affected the Sr isotope systems, and earlier age determinations by the Rb–Sr whole-rock isochron method (1,230 Ma) have proved to be in error. The dolerites and the basalts are geochemically very similar, but most dolerites have moderately negative Eu anomalies that are not observed in the basalts. Eu anomalies in the dolerites could be related to contamination by sandstone at depth, but it is not clear why the basalts escaped a similar contamination.     

On the origin of Tasmanian dolerites

The Tasmanian dolerites, part of the Ferrar Province of Australia and Antarctica, have some trace‐element and isotopic compositions that suggest continental contamination of mantle‐derived magmas. The debate has centred on whether the contamination occurred during intrusion into the crust, or if the mantle source itself was contaminated. The behaviour of Sr and O isotopes suggests that the mantle source had a δ¹⁸O composition of +6‰ and an initial ⁸⁷Sr/⁸⁶Sr ratio of 0.709, which supports the latter contention. Recently published Re–Os data likewise dismiss upper crustal contamination: Re–Os isotopic compositions of magnetite‐rich mineral separates from seven Tasmanian dolerites yield an isochron that gives the same age, within uncertainties, as other dating techniques, namely 175 ± 5 Ma. Moreover, Re–Os data from a study of Antarctic Ferrar Province samples lie on the same isochron and the data together give an age of 177.3 ± 3.5 Ma. The initial ¹⁸⁷Os/¹⁸⁸Os of 0.125 ± 0.033 is the calculated mantle composition at the time. These results support previous models that attribute chemical features of the Ferrar magmas to re‐enrichment of a depleted mantle source region rather than processes involving assimilation of crust by basaltic magma.     

<Superscript>10</Superscript>Be, <Superscript>18</Superscript>O and radiogenic isotopic constraints on the origin of adakitic signatures: a case study from Solander and Little Solander Islands,New Zealand

Subduction-related Quaternary volcanic rocks from Solander and Little Solander Islands, south of mainland New Zealand, are porphyritic trachyandesites and andesites (58.20–62.19 wt% SiO₂) with phenocrysts of amphibole, plagioclase and biotite. The Solander and Little Solander rocks are incompatible element enriched (e.g. Sr ~931–2,270 ppm, Ba ~619–798 ppm, Th ~8.7–21.4 ppm and La ~24.3–97.2 ppm) with MORB-like Sr and Nd isotopic signatures. Isotopically similar quench-textured enclaves reflect mixing with intermediate (basaltic-andesite) magmas. The Solander rocks have geochemical affinities with adakites (e.g. high Sr/Y and low Y), whose origin is often attributed to partial melting of subducted oceanic crust. Solander sits on isotopically distinct continental crust, thus excluding partial melting of the lower crust in the genesis of the magmas. Furthermore, the incompatible element enrichments of the Solander rocks are inconsistent with partial melting of newly underplated mafic lower crust; reproduction of their major element compositions would require unrealistically high degrees of partial melting. A similar argument precludes partial melting of the subducting oceanic crust and the inability to match the observed trace element patterns in the presence of residual garnet or plagioclase. Alternatively, an enriched end member of depleted MORB mantle source is inferred from Sr, Nd and Pb isotopic compositions, trace element enrichments and εHf ? 0 CHUR in detrital zircons, sourced from the volcanics. ¹⁰Be and Sr, Nd and Pb isotopic systematics are inconsistent with significant sediment involvement in the source region. The trace element enrichments and MORB-like Sr and Nd isotopic characteristics of the Solander rocks require a strong fractionation mechanism to impart the high incompatible element concentrations and subduction-related (e.g. high LILE/HFSE) geochemical signatures of the Solander magmas. Trace element modelling shows that this can be achieved by very low degrees of melting of a peridotitic source enriched by the addition of a slab-derived melt. Subsequent open-system fractionation, involving a key role for mafic magma recharge, resulted in the evolved andesitic adakites.     

Petrogenesis of the Tertiary Anorogenic Volcanic Series of Southern Queensland, Australia, in the Light of Trace Element Geochemistry and O, Sr and Pb Isotopes

The magmas of the Tertiary volcanic province of S. Queenslandare chemically bimodal, and occur in numerous volcanic centres,at least three representing original shield volcanoes. The maficlavas are dominantly hawaiites and tholeiitic andesites, whereasthe silicic magmas comprise mainly trachytes, rhyolites, andcomendites. The silicic rocks exhibit variable trace element abundance patterns.There is a progressive depletion of Sr, Ba, V, Mg, Ni, Cr, Mn,and P, through the trachytes to the rhyolites and comenditeswhile the behaviour of Zr, Nb, LREE, Y and Zn is very variable.Rb, Th, and to a lesser extent Pb exhibit a more regular behaviour,becoming most generally concentrated in the comendites and rhyolites.These trace element patterns are modelled by application ofthe Rayleigh distillation model, using partition coefficientsbased on analysed phenocrysts from the S. Queensland siliciclavas. Trace mineral phases, namely zircon, chevkinite, andallanite, are shown to be important in the probable controlof LREE, Zr, and Th abundances, while Nb and Zn are probablycontrolled during fractionation by magnetite. Trace elementdata for the hawaiites and tholeiitic andesites also indicateextensive although variable levels of fractional crystallizationof these magmas. The Sr and O isotopic compositions of the mafic lavas, trachytes,comendites and rhyolites are as follows: initial ⁸⁷Sr/⁸⁶Sr ratios;0.70357–0.70456, 0.70432–0.70589, 0.70495–0.70917,and 0.70708–0.70863 respectively. ¹⁸O range between 5.6–7.0(mafic lavas), 4.9–8.7 (trachytes), 5.0–7.6 (comendites)and 8.1–10.4 per mil (rhyolites). Pb isotopic compositionsare variable, showing a variation of 6.7 per cent for ²⁰⁶Pb/²⁰⁴Pbratios through the range of volcanic compositions. The rhyolitesexhibit a much greater divergence in their O, Sr, and Pb isotopiccompositions compared with those of associated mafic lavas,than is found in the trachytes and comendites. Within the silicicvolcanics, positive correlations exist between ¹⁸O and initialSr ratios, and between Pb isotopic compositions and initialSr ratios (with one group of trachytes providing a noteworthyexception). These correlations are not so clearly defined forthe mafic lavas, although these do exhibit positive correlationsbetween differentiation index, ¹⁸O, and initial Sr isotope ratios. The development of the silicic magmas, excepting two groups,is interpreted in terms of a model in which assimilation andfractional crystallization occur concurrently, involving a basaltor hawaiite magma component and a crustal component (modelledon the analysed Carboniferous basement greywackes outeroppingin the region); the data indicate, however, that differentiationcontinued in isotopically closed systems (i. e. isolated fromthe wallrocks). The highly depleted Sr and Ba abundances ofthe rhyolites and comendites suggest that contamination didnot occur after differentiation had ceased. The rhyolites havethe highest isotopic input of the crustal components and areinterpreted as crustal anatectic melts, produced locally withinthe crust in response to basalt/hawaiite magma intrusion, whereasmost of the trachytes and comendites are interpreted as primarilythe differentiated products from original mafic parental magmas,with variable assimilation of crustal wallrock components. Theisotopic data suggest that only the Minerva Hills trachyticlavas, and a Glass House comendite, have not been significantlymodified by wallrock assimilation processes. The erpted maficmagmas were also evidently modified by isotopic crustal wallrockinteractions, which independent petrological data suggest hasoccurred at intermediate to lower crustal depths.     

Progressive crustal contamination of the Bushveld Complex: evidence from Nd isotopic analyses of the cumulate rocks

We report the first Nd isotopic data on the cumulate rocks of the Bushveld Complex, South Africa. We analysed 17 whole-rock samples covering 4700 m of stratigraphy through the Lower, Critical and Main Zones of the intrusion at Union Section, north-western Bushveld Complex. The basal ultramafic portions of the complex have markedly higher ɛNd(T) (−5.3 to −6.0) than the gabbronoritic Main Zone (ɛNd(T) −6.4 to −7.9). The rocks of the Upper Critical Zone have intermediate values. These results are in agreement with new Nd isotope data on marginal rocks and sills in the floor of the complex that are generally interpreted as representing chilled parental magmas, and with published Sr isotopic data, all of which show a larger crustal component in the upper part of the intrusion. In contrast, the concentrations of many highly incompatible trace elements are decoupled from the isotopic signatures. The basal portions of the complex have higher ratios of incompatible to compatible trace elements than the upper portions. The variations of isotopic and trace-element compositions are interpreted in terms of a change in the nature of the crustal material that contaminated Bushveld magmas. Those magmas that fed into the lower part of the complex had assimilated a relatively small amount of incompatible trace-element-rich partial melt of upper crust, whereas magmas parental to the upper part of the complex had assimilated a higher proportion of the incompatible trace-element-poor residue of partial melting. Received: 5 October 1999 / Accepted: 7 July 2000