Multiphase inclusions in plagioclase from anorthosites in the Stillwater Complex,Montana: implications for the origin of the anorthosites

Multiphase inclusions, consisting of clinopyroxene+ilmenite+apatite, occur within cumulus plagioclase grains from anorthosites in the Stillwater Complex, Montana, and in other rocks from the Middle Banded series of the intrusion. The textures and constant modal mineralogy of the inclusions indicate that they were incorporated in the plagioclase as liquid droplets that later crystallized rather than as solid aggregates. Their unusual assemblage, including a distinctive manganiferous ilmenite and the presence of baddeleyite (ZrO₂), indicates formation from an unusual liquid. A process involving silicater liquid immiscibility is proposed, whereby small globules of a liquid enriched in Mg, Fe, Ca, Ti, P, REE, Zr and Mn exsolved from the main liquid that gave rise to the anorthosites, became trapped in the plagioclase, and later crystallized to form the inclusions. The immiscibility could have occurred locally within compositional boundaries around crystallizing plagioclase grains or it could have occurred pervasively throughout the liquid. It is proposed that the two immiscible liquids were analogous, n terms of their melt structures, to immiscible liquid pairs reported in the literature both in experiments and in natural basalts. For the previously reported pairs, immiscibility is between a highly polymerized liquid, typically granitic in composition, and a depolymerized liquid, typically ferrobasaltic in composition. In the case of the anorthosites, the depolymerized liquid is represented by the inclusions, and the other liquid was a highly polymerized aluminosilicate melt with a high normative plagioclase content from which the bulk of the anorthosites crystallized. Crystallization of the anorthosites from this highly polymerized liquid accounts for various distinctive textural and chemical features of the anorthosites compared to other rocks in the Stillwater Complex. A lack of correlation between P contents and chondrite-normalized rare earth element (REE) ratios of plagioclase separates indicates that the amount of apatite in the inclusions is too low to affect the REE signature of the plagioclase separates. Nevertheless, workers should use caution when attempting REE modelling studies of cumulates having low REE contents, because apatite-bearing inclusions can potentially cause problems.     

An evaluation of models of apatite compositional variability using apatite from the Middle Banded series of the Stillwater Complex,Montana

 Apatite is a relatively common accessory mineral in the olivine-bearing zones (OB-III and OB-IV) of the Middle Banded series of the Stillwater complex, occurring interstitial to cumulus grains, as monomineralic inclusions in cumulus grains, and in polymineralic clusters which may or may not be included in the cumulus grains. Eighty-nine of 185 samples examined were found to contain some apatite. The F-Cl-OH content of the apatite show distinct stratigraphic variations. XClAp (mole fraction Cl in apatite) of interstitial apatite increases upsection in both OB-III and OB-IV with average values ranging from 0.15 to 0.85 in OB-III and from 0.03 to 0.60 in OB-IV. XFAp varies inversely with XClApwith average values ranging from 0.00 to 0.70, while XOHAp remains relatively constant near 0.40 or decreases slightly with height. These variations are remarkable given that no appreciable stratigraphic variations in either the major or trace element compositions of any of the cumulus minerals are found in the 800 m of section that comprise OB-III and OB-IV. Within-sample variation of XClAp for samples containing Cl-rich apatite is substantially larger (up to 0.65 XClAp) than for samples with more F-rich apatite (XClAp varies by approximately 0.15). Although interstitial apatite is found throughout OB-III and OB-IV, apatite occurring as monomineralic inclusion in cumulus grains or in polymineralic clusters is almost exclusively found in samples with Cl-rich apatite. The data are best explained by a model involving the degassing of a Cl-rich volatile phase from the crystallizing interstitial liquid. The up-section migration of this fluid resulted in the crystallization of F-rich apatite in the lower portion and progressive Cl-enrichment in the apatite with height. The presence of hornblende-bearing dikes, veins and pegmatoids at the level of maximum Cl-enrichment is consistent with a fluid migration model. Received: 5 October 1995 / Accepted: 19 March 1996     

Discordant bodies from olivine-bearing zones III and IV of the Stillwater Complex, Montana – evidence for postcumulus fluid migration and reaction in layered intrusions

Mineralogically zoned and unzoned discordant bodies composed predominately of plagioclase with up to 35% olivine, occur at three different levels in Olivine-Bearing zones III and IV of the Middle Banded series of the Stillwater complex. The discordant bodies are elongate perpendicular to the layering of the host cumulates with slender concordant apophyses. Although the host olivine-gabbros are foliated with tabular plagioclase, the discordant bodies lack a discernible fabric and have blocky plagioclase. Average olivine in the host rocks is slightly more magnesian than that of the discordant bodies (Mg#_{75.8 ± 0.7} versus Mg#_{74.6 ± 0.3} respectively) but plagioclase compositions are indistinguishable (An_{77.6 ± 2.0} versus An_{76.6 ± 4.3}– average host and discordant bodies respectively). Whole-rock major- and trace-element compositions of the discordant bodies are generally indistinguishable from cumulates with similar modal abundance. However, bulk compositions of anorthositic cores from the discordant bodies are enriched in K, Na, Ba, Sr and P. We conclude that the discordant bodies formed when cooler volatile fluids or fluid-rich silicate liquids moved upward and encountered a hotter undersaturated solid-plus-liquid assemblage. Continued liquid/fluid fluxing increased the permeability along the flow path and focused the flow, allowing the original bulk compositions to be modified and leaving plagioclase-rich troctolites and anorthosites. The shapes of the discordant bodies suggest that the cumulus pile had anisotropic permeability during late-stage liquid/fluid flow. Chemical and mineralogical evidence from other parts of Olivine-Bearing zones III and IV suggests that the processes that formed the discordant bodies may have influenced other cumulates. In fact, it appears that the same processes that formed the discordant bodies operated within an anorthositic layer, strongly modifying the chemistry of the rock but leaving no mineralogical or textural evidence. Received: 10 December 1996 / Accepted: 12 August 1997     

Rare Earth Element Evidence for the Petrogenesis of the Banded Series of the Stillwater Complex, Montana, and its Anorthosites

A rare earth element (REE) study was made by isotope-dilutionmass spectrometry of plagioclase separates from a variety ofcumulates stratigraphically spanning the Banded series of theStillwater Complex, Montana. Evaluation of parent liquid REEpatterns, calculated on the basis of published plagioclase-liquidpartition coefficients, shows that the range of REE ratios istoo large to be attributable to fractionation of a single magmatype. At least two different parental melts were present throughoutthe Banded series. This finding supports hypotheses of previousworkers that the Stillwater Complex formed from two differentparent magma types, designated the anorthositic- or A-type liquidand the ultramafic- or U-type liquid. On the basis of our data,one melt has a REE pattern with a distinctive shallow slopeand is represented by samples from the thick, massive Anorthositezones I and II (AN I and AN II) of the Middle Banded series.Although samples from AN I and AN II are separated by as muchas 1400 m stratigraphically, they have remarkably similar calculatedparent liquid characteristics, with (Ce/Sm)n = 1.7–1.9,(Nd/Sm)n = 1.3–1.4 and (Ce/Yb)n = 2.9–4.6 (wheren denotes chondrite-normalized). These calculated liquids areprobably close to representing A-type magma. In addition, plagioclase-bronzitecumulates from Norite zones I and II (N I and N II), althoughthought to be U-type cumulates, contain plagioclase that hasA-type REE characteristics, implying that A-type magmas wereinjected into the magma chamber during formation of those zones.In contrast, calculated parent liquids of cumulus augite-bearingrocks have REE patterns that display distinctly steeper slopesthan the A-type REE pattern. The extreme is the calculated parentliquid of a plagioclase-bronzite-augite cumulated with (Ce/Sm)n= 2.9, (Nd/Sm)n = 1.7, and (Ce/Yb)n = 10.1. Analysis of published REE and Nd isotopic data for Stillwatercumulates reveals similarities between AN I, AN II, and otherthin plagioclase cumulate layers in the Lower and Upper Bandedseries, which supports the notion that they were all derivedfrom similar (A-type) parent melts. In contrast, plagioclaseseparates from cumulus augite-bearing rocks display light REEand Nd isotopic characteristics that are similar to U-type cumulatesfrom the Ultramafic series as described by previous studies.Thus far, the only cumulates from the Banded series that displayU-type REE and Nd isotopic characteristics are those that containcumulus augite. Therefore, cumulus augite appears to be an importantindicator of magmatic parentage. The REE and Nd isotopic ratios show erratic variation with stratigraphicposition, indicating that the magmas from which the Banded seriescrystallized were injected at various levels into the magmachamber. Different cumulate types crystallized from discreteliquids, as indicated by the correlation between REE signatureand cumulate type. Samples from Olivine-bearing zones III andIV (stratigraphically between AN I and AN II) display a rangein REE ratios; e.g., (Ce/Sm)n = 1.8–2.8 and (Ce/Yb)n =3.9–6.1, results that rule out the crystallization ofthe Middle Banded series from a single magma type. Furthermore,the possibility that AN I and AN II are directly related tothe underlying Ultramafic series, either as flotation cumulatesor as crystallization products of expelled liquids, is not substantiatedby the REE data because the calculated parent magma of AN Iand AN II was different from that of the Ultramafic series asdefined by previous studies. The REE data of this study further constrain interpretationsof published Pb isotopic data (Wooden et al., 1991) and indicatethat the magmas from which the Stillwater Complex formed werederived from two sources that had only small differences inPb isotopic composition. The REE and isotopic data, as wellas crystallization sequences of the two main parental magmas,indicate that the magmas were probably derived from two closelyrelated upper-mantle sources, one harzburgitic and the otherlherzolitic in composition, resulting in the U-type and A-typemagmas, from which orthopyroxene crystallized before and afterclinopyroxene, respectively. Both sources had been enrichedin large-ion lithophile elements, probably owing to mantle metasomatism.     

An Investigation of the Oxygen Isotope Geochemistry of the Stillwater Complex

Oxygen isotopic analyses of samples spanning the entire stratigraphicheight of the Stillwater Complex show that the intrusion hasretained its magmatic isotopic composition. The melt had a calculated¹⁸O value of 5?9 per mille which agrees well with the reportedisotopic compositions of the Kiglapait Intrusion and other mantle-derivedbasic melts. While there has been alteration, the isotopic dataconfirm the petrograhpic observation that the alteration tookplace at low temperature (most likely in the weathering environment). Isotopic thermometry yields magmatic temperatures. The temperaturesdefine two trends: a high temperature trend extending from thebase of the Ultramafic Series to the top of Norite-2 in theBanded Series and a lower temperature trend extending from thetop of Norite-2 to the top of the intrusion. The high temperaturetrend is interpreted as indicating an ultramafic magma whilethe low temperature trend suggests a more gabbroic melt. However,the temperature break does not coincide with any significantpetrologic break. Fractional crystallization modelling of the isotopic variationof appropriate compositions shows that most of the isotopicvariation within the complex can be accounted for by simplefractional crystallization. ¹⁸O depletions observed in the UltramaficSeries and in the olivine-bearing zones of the Banded Seriesindicate the presence of magmatically derived late stage hydrousfluids in those zones.     

Petrology of the Proterozoic Potato River Layered Intrusion, Northern Wisconsin, USA

The Potato River intrusion is a Keweenawan (1100 Ma) mafic plutonemplaced in Keweenawan volcanics and earlier Proterozoic metasedimentaryrocks along the southeastern flank of the Lake Superior syncline.It comprises the following lithostratigraphic zones: a thinto absent Border zone of altered olivine gabbro; a Lower zoneof olivine gabbro; a Picritic zone of picrite and troctolite;a Middle zone of olivine gabbro and leucogabbro; an Upper zoneof quartz leucogabbro and ferrogabbro; and a Roof zone of granophyricand granitic rocks. Fractional crystallization is evident fromcompositional changes in the rocks and cumulus minerals withstratigraphic height. Elements concentrated in the cumulus mineralsolivine and plagioclase (Mg, Fe²⁺, Al, Ca, Ni, Co, Cr, Sr) decreasewith height; elements concentrated in the trapped liquid (Na,K, La, Y, Zr, Nb, Rb, Ba) increase with height; and other elements(Ti, Fe³⁺, P, Ga, V, Sc, Cu, Zn) show complicated behavior relatedto the appearance of additional cumulus phases such as clinopyroxene,Fe-Ti oxides, and apatite. Lower zone rocks contain some sulfide,probably from sulfur derived from the country rock, and theUpper zone has sulfides probably precipitated from an immisciblesulfide liquid. The sulfide-bearing rocks have similaritiesto those of other intrusions, such as Bushveld, Stillwater,and Skaergaard. The picritic and troctolitic rocks of the Picritic zone indicatethat the intrusion was open to additional injections of maficmagma. Roof zone granophyric rocks are residual liquids intrudedalong the upper margin of the intrusion during regional tilting,but Roof zone granitic rocks are probably melted country rock.An attempt is made to estimate by reverse stratigraphic summationthe compositional path of the magma that solidified above thePicritic zone. The first compositions are highly aluminous,which suggests that the upper part of the intrusion has beenenriched in plagioclase by convection-aided crystal sorting.A complementary unit of mafic rocks is not exposed, but it couldbe present down dip. Some of the later compositions are similarto typical Keweenawan high-Al tholeiites. The magma did notundergo extreme iron enrichment, probably because of oxygenfugacity buffering.     

Platinum mineralisation potential of the Longwood Igneous Complex,New Zealand

Summary The Longwood Igneous Complex is situated at the southern tip of the South Island of New Zealand. The Complex is 32 km long and up to 12 km wide and is part of a more extensive north-trending belt of late Paleozoic volcanic, volcaniclastic and intrusive rocks.The complex is comprised of an extensive series of layered gabbros (Pahia Layered Series) together with diorites, trondjhemite and granitoids.Exploration by Sigma Resources has located significant platinum-in-drainage anomalies which are derived from a 10 km x 1 km area of largely plagioclase-rich, olivinebearing cumulate rocks within the Pahia Layered series. Grains of platinum and palladium sulphide, arsenide and alloy minerals have been identified in placer deposits derived from this area. A total of 1,500 ounces of platinum was reported to have been recovered as a by-product of historic placer gold mining of these deposits. Platinumgroup element mineralisation in float rocks (up to 3.0 ppm Pt and 3.3 ppm Pd) and significant platinum-in-soil anomalies (peak value 0.47 ppm Pt) have been found within this area.The mineralised troctolite and olivine-gabbro float and adjacent exposures are broadly similar to the anorthosite-troctolite-olivine gabbro zones within the Banded Series of the Stillwater Complex. The mineralised rocks are leucocratic plagioclaseolivine adcumulates with minor clinopyroxene oikocrysts, primary amphibole and sparse (<0.5%) sulphide. Sulphide minerals are pyrrhotite, chalcopyrite, pyrite and secondary copper sulphides. The similarity of this mineralisation style and its host rocks to the JM Reef of the Stillwater Complex and AP and PV mineralisation of the Penikat Intrusion highlights the excellent potential for discovery of a large stratiform platinumgroup element deposit in the Pahia Layered Series.

---

Das Platin Mineralisations Potential des Longwood Komplexes, Neuseeland

Zusammenfassung Der Longwood Igneous Complex liegt au der Südküste der der Südinsel Neuseelands. Er ist 32 km lang und 12 km breit und ist Teil eines Nord-Süd verlaufenden Gesteinsgürtels paläozoischer vulkanischer, vulkano-klastischer und intrusiver Gesteine. Der Komplex beinhaltet geschichtete Gabbros (Pahia Layered Series) gemeinsam mit Dioriten, Trondjhemiten und Granitoiden.Bedeutende Platinanomalien in Bachsedimenten, die aus einem ca. 10 x 1 km großen Gebiet mit Plagioklas-reichen, Olivin-führenden Kumulat-Gesteinen stammen, wurden von Sigma Resources lokalisiert. Platin- und Palladiumsulfide, -arsenide und -Legierungen wurden in Seifenlager-stätten, in diesem Gebiet nachgewiesen. Insgesamt wurden 1500 Unzen Platin während der historischen Bergbautätigkeit auf Gold in diesen Lagerstätten als Nebenprodukt gewonnen. PGE-Mineralisation (bis 3.0 ppm Pt und 3.3 ppm Pd) in Geschieben und bedeutende Pt-Bodenanomalien (max. 0.47 ppm Pt) wurden in diesem Gebiet entdeckt.Die mineralisierten Troktolit- und Olivin-Gabbro-Gerölle und benachbarte Aufschlüsse zeigen weitgehende Ähnlichkeit mit der Anorthosit-Troktolit-Olivin-GabbroAbfolge der Banded Series des Stillwater Komplexes.Die mineralisierten leukokraten Gesteine führen Plagioklas und Olivin als Kumulusphasen sowie untergeordnet Klinopyroxene, primäre Amphibole und geringe Mengen (< 0.5 %) an Sulfiden. Als Sulfide kommen Magnetkies, Kupferkies, Pyrit und sekundäre Kupfersulfide vor. Die Ähnlichkeit dieser Mineralisation und ihrer Nebengesteine mit dem JM Reef des Stillwater Komplexes und der AP und PV Mineralisation in der Penikat Intrusion läßt die Pahia Layered Series als geeignetes Gebiet für die Entdeckung großer stratiformer PGE-Lagerstätten erscheinen.

---

---

With 5 Figures     

Serpentinization and alteration in an olivine cumulate from the Stillwater Complex,Southwestern Montana

Some of the olivine cumulates of the Ultramafic zone of the Stillwater Complex, Montana, are progressively altered to serpentine minerals and thompsonite. Lizardite and chrysotile developed in the cumulus olivine and postcumulus pyroxenes; thompsonite developed in postcumulus plagioclase. The detailed mineralogy, petrology, and chemistry indicate that olivine and plagioclase react to form the alteration products, except for H₂O, without changes in the bulk composition of the rocks.     

Platinum potential of mafic–ultramafic massifs in the western part of the Dambuka ore district (Upper Amur Region,Russia)

New data on the Pt potential of mafic–ultramafic massifs of the Khani–Maya, Uldegit, and Dzhalta complexes in the western part of the Dambuka ore district are discussed. The Khani–Maya Complex is represented by metamorphosed gabbro, gabbronorites, gabbro anorthosites, subordinate pyroxenites, hornblendites, and peridotites. The Uldegit Complex is composed of pyroxenites, hornblendites, gabbro, gabbronorites, norites, troctolites, peridotites, dunites, actinolite–tremolites, serpentinites, anthophyllites, and tremolite–plagioclase rocks. The Dzhalta Complex is formed of peridotites, gabbro, eclogitized gabbro, hornblendites, cortlandites, and pyroxenites. All these complexes differ from each other by the concentrations of Ni, Cu, Co, Au, and platinoids depending on the composition of the constituting rocks and the presence of sulfide minerals.     

Composition and geodynamic conditions of formation of the intrusive rocks of the Gromadnen-Vurguveem peridotite-gabbro massif,western Chukotka

The Gromadnen-Vurguveem peridotite-gabbro massif is confined to one of the largest ophiolite complex of western Chukotka and composed mainly of intrusive rocks. This paper reports the first comprehensive compositional data for its plutonic rocks (petrochemistry, geochemistry, and compositions of minerals). In terms of petrography, two groups of rocks can be distinguished in the Gromadnen-Vurguveem peridotite-gabbro massif. The first group includes leucocratic gabbroids (mostly gabbronorites), composing most of the massif. The second group includes olivine-bearing cumulate rocks: olivine gabbros, troctolites, plagioclase-bearing dunites, and amphibolized wehrlites. The major element variations in these rocks suggest their affiliation to low-titanium, low-potassium, and high-alumina plutonic derivatives of island-arc magmatism. According to geochemical characteristics (distribution of REEs and indicator incompatible elements), the gabbroids of the first group are akin to both island-arc tholeiites and boninites. The olivine-bearing rocks of the second groups show boninitic affinity. Based on these observations, it was concluded that the intrusive complex of the Gromadnen-Vurguveem massif was formed during an early stage of the development of an ensimatic island arc.     

Chemical assembly of Archean anorthosites from amphibolite- and granulite-facies terranes, SW Greenland

We report whole-rock, major- and trace-element compositions (obtained by XRF and INA methods) for the amphibolite-facies Buksefjorden and granulite-facies Nordland anorthosites, SW Greenland. In a previous petrologic study on the same sample suite, we documented differences in texture, mineralogy, and mineral compositions between these two anorthosite bodies. Chemical analyses confirm differences in composition between the two bodies, but these differences cannot be explained by variations in metamorphic conditions, and point towards differences in the nature of their protoliths. Analyzed Nordland samples are anorthosites and leucogabbros with 88-98% normative plagioclase, whereas those from Buksefjorden include anorthosite, leucogabbro, and gabbro with ~55-95% normative plagioclase. Two or more compositional groupings can be recognized at each site, which correspond to differences in color and mineralogy of the hand samples. Samples from Buksefjorden are mainly quartz-normative, whereas those from Nordland are olivine (- nepheline) normative. Other differences include higher Ni/Co ratio and REE contents in the granulite-facies anorthosites from Nordland. REE pattern shapes are similar, however, being moderately fractionated at ~0.5-102 chondrites with positive Eu-anomalies. Calculated equilibrium melt patterns are similar for both anorthosites, being relatively flat at ~50-1502 chondrites, suggesting unfractionated (but evolved) parental magmas. Olivine must have been present in the protoliths of the Nordland rocks compared with Buksefjorden. Otherwise, the protoliths contained plagioclase with variable An-content (~An₆₂-An₉₂) and a mafic component with variable Fe/Mg (mg ~0.3-0.8). This mafic component was either hornblende or a combination of ortho- and clinopyroxene in fixed proportions, plus a small amount of magnetite. Mixing calculations demonstrate that some Buksefjorden anorthosites contain two varieties of plagioclase: a calcic type that may correspond to cumulus crystals, and a sodic-type that may correspond to a trapped-melt component. On plots of normative whole-rock An versus mg, compositions of the Buksefjorden and Nordland anorthosites form crude negative arrays that differ from the generally positive trends of mafic layered intrusions (Kiglapait, Skaergaard) and from the generally flats trends of plagioclase-rich cumulate rocks (St. Urbain and Stillwater anorthosites). This difference provides further evidence for the distinctive nature of Archean calcic-anorthosite complexes compared with other types of mafic intrusions. Moreover, this distribution of data points is consistent with the assembly of the protolith of the SW Greenland anorthosites mainly as mixtures of plagioclase and hornblende. Finally, the field for the Buksefjorden and Nordland anorthosites overlaps only slightly with that for the Fiskenaesset Complex, thus extending the known range of compositions for Archean anorthosites in West Greenland.     

Petrology of the G and H Chromitite Zones in the Mountain View Area of the Stillwater Complex, Montana

The Ultramafic Series of the Stillwater Complex in the MountainView area of the intrusion consists of 17 cyclic units thathave been numbered stratigraphically. A typical unit has olivinecumulates at the base, olivine–bronzite cumulates at intermediatelevels, and bronzite cumulates at the top. Most cyclic unitsalso have chromite-rich layers near their base, the thickestbeing the G and H chromitite zones in units 10 and 11. The Gand H zones are each separated from the top of the underlyingcyclic unit by 1–3 m of coarse-grained olivine cumulateand pegmatite; and they are both succeeded by thinner chromititezones, respectively called the hanging wall G (HWG) and thehanging wall H (HWH) zones, situated {small tilde}20 m and 5m above them. The G and H chromitite zones feature rhythmicsequences of thin layers that tend to progress upward from massivechromitite through chromite–olivine cumulate to olivine–chromitecumulate (the last with the minerals in approximately cotecticproportions of about 98:2). In cyclic units 10 and 11, variationsof Mg/Fe in the olivine and bronzite, and of Ni in the olivine,are small and show no clear stratigraphic fractionation trends.The abundance of Cr in the chromite in unit 10 does have a fractionationtrend, however, being generally highest at the bottom of theunit and lowest at the top, with a regression at the HWG zone.In general, Cr in chromite is highest at the base of a rhythmicunit and decreases upward, but it shows no overall decline throughsuccessive rhythmic units; Fe³ exhibits opposite variation,being lowest in the massive chromite, and highest in the disseminatedgrains. The G and H chromitite zones, in the Mountain View area, eachcontain enough chromite to form a single layer of massive chromitite{small tilde} 1 m thick. If their formation involved removalof only 30% of the Cr in the parental magmatic liquid (estimatedconcentration, 600 ppm), then this liquid could have amountedvolumetrically to an areally equivalent layer at least 2000m thick. Model calculations demonstrate that such a large volumeof liquid is consistent with the small variations of Mg/Fe inthe pyroxenes and olivines in the Stillwater cyclic units. We postulate that the G and H chromitite zones and cyclic unitsthat host them formed in response to the entry of new pulsesof primitive magmatic liquid into the Stillwater chamber. Fromexperimental observations, we infer that these pulses producedfountains in which the primitive liquid mixed with residualfractionated liquids, yielding hybrids that were compositionallywithin the chromite liquidus field (or volume) and that weresupercooled (supersaturated ) with respect to the oxide mineral.These effects may have been enhanced by low fO₂ (oxygen fugacity)in the primitive liquid and(or) by high fO₂ of the fractionatedliquid. The hybrid liquids probably collected at the bottomof the chamber in a zoned layer that then divided into double-diffusiveconvecting layers. In these circumstances, the lowest chromite-richlayer in a rhythmic sequence could have formed from the lowestdouble-diffusive liquid layer, and the next could then haveformed when this liquid mixed with the liquid layer above it—andso on up the sequence. We argue that the thick G and H chromititezones are situated toward the top of the Ultramafic Series becausethat level marks when the compositional contrasts between theinjected primitive liquid and the residual fractionated liquidsin the chamber were greatest.     

Anorthosite formation by plagioclase flotation in ferrobasalt and implications for the lunar crust

The Sept Iles layered intrusion (Quebec, Canada) is dominated by a basal Layered Series made up of troctolites and gabbros, and by anorthosites occurring (1) at the roof of the magma chamber (100-500 m-thick) and (2) as cm- to m-size blocks in gabbros of the Layered Series. Anorthosite rocks are made up of plagioclase, with minor clinopyroxene, olivine and Fe-Ti oxide minerals. Plagioclase displays a very restricted range of compositions for major elements (An₆₈-An₆₀), trace elements (Sr: 1023-1071 ppm; Ba: 132-172 ppm) and Sr isotopic ratios (⁸⁷Sr/⁸⁶Sr_i: 0.70356-0.70379). This compositional range is identical to that observed in troctolites, the most primitive cumulates of the Layered Series, whereas plagioclase in layered gabbros is more evolved (An₆₀-An₃₈). The origin of Sept Iles anorthosites has been investigated by calculating the density of plagioclase and that of the evolving melts. The density of the FeO-rich tholeiitic basalt parent magma first increased from 2.70 to 2.75 g/cm³ during early fractionation of troctolites and then decreased continuously to 2.16 g/cm³ with fractionation of Fe-Ti oxide-bearing gabbros. Plagioclase (An₆₉-An₆₀) was initially positively buoyant and partly accumulated at the top of the magma chamber to form the roof anorthosite. With further differentiation, plagioclase (<An₆₀) became negatively buoyant and anorthosite stopped forming. Blocks of anorthosite (autoliths) even fell downward to the basal cumulate pile. The presence of positively buoyant plagioclase in basal troctolites is explained by the low efficiency of plagioclase flotation due to crystallization at the floor and/or minor plagioclase nucleation within the main magma body. Dense mafic minerals of the roof anorthosite are shown to have crystallized from the interstitial liquid.The processes related to floating and sinking of plagioclase in a large and shallow layered intrusion serve as a proxy to refine the crystallization model of the lunar magma ocean and explain the vertically stratified structure of the lunar crust, with (gabbro-)noritic rocks at the base and anorthositic rocks at the top. We propose that the lunar crust mainly crystallized bottom-up. This basal crystallization formed a mafic lower crust that might have a geochemical signature similar to the magnesian-suite without KREEP contamination, while flotation of some plagioclase grains produced ferroan anorthosites in the upper crust.     

Halogen Geochemistry of the Stillwater and Bushveld Complexes: Evidence for Transport of the Platinum-Group Elements by Cl-Rich Fluids

Compositional data on apatite, phlogopite, and amphibole indicatethat the high-temperature hydrothermal fluids which affectedthe lower portions of the Stillwater and Bushveld Complexeswere Cl-rich. Apatites from the platinum-group element (PGE)ore zones from both complexes are enriched in Cl relative toother cumulus and noncumulus apatites in these intrusions andto apatites from the Skaergaard and Kiglapait Intrusions andthe Great Dyke. Apatites from all five intrusions can be groupedinto three distinct compositional fields: (a) Cumulus apatitesare essentially fluorapatites with molar Cl/(Cl+OH+F) <0?03;(b) noncumulus apatites, with the exception of those from thePGE ore zones of the Stillwater and Bushveld Complexes, haveCl/(Cl+OH+F) <0?20; (c) Cl-rich apatites associated withPGE-rich zones have Cl/(Cl+OH+F) between 0?45 and 1?0. The REEcontent of noncumulus and Cl-rich apatites also show a positivecorrelation with Cl concentration. It is argued that becauseCl is less soluble in silicate melts than F and because meltswith extremely high Cl/F ratios are unknown, the Cl-rich apatitesequilibrated with Cl-rich hydrothermal fluids exsolved duringsolidification of the cumulate sequence. The Cl, F, and OH contents of phlogopites and amphiboles aremore variable. Compositional heterogeneity is due to crystal-chemicalcontrols on halogen contents, variation in the halogen contentof the original melt/fluid phase and subsolidus re-equilibrationduring cooling with both surrounding mineral phases and lowtemperature fluids. However, both the Stillwater and Bushveldphlogopites are enriched in Cl compared to those from the Skaergaardand Kiglapait Intrusions. The compositions of coexisting minerals from the platinum depositof Olivine-Bearing Subzone I of the Stillwater Complex are usedto compute a fluid composition. The fluid is rich in alkalisand iron as well as HCl, and the solution composition is consistentwith fluid compositions deduced for the PGE-bearing secondaryhortonolite pipes of the Bushveld Complex. The high (Pt+Pd)/Irratios of these deposits are also consistent with a hydrothermalorigin, as both Pt and Pd are more soluble in Cl-complexingfluids than Ir.     

Investigations of the Stillwater Complex: Part V. Apatites as indicators of evolving fluid composition

Variations in the F, Cl and OH contents of apatite are not constrained by crystal-chemical factors (in contrast to micas and amphiboles), and thus changes in the abundance of these components provide an indicator of halogen fugacity variations and insights into the degassing history of igneous rocks. Microprobe analysis of intercumulus apatites from the Stillwater Complex reveal that Cl-rich apatites, typically containing <0.4 wt % F and >6.0 wt % Cl, occur throughout the lower 1/3 of the complex excluding the Basal series. A change from Cl-rich to more F-rich apatite occurs within olivine-bearing zone I (OB I) of the Banded series, the host zone of the platiniferous J-M Reef. Although apatite compositions are somewhat variable above the J-M Reef, more F-rich apatites predominante and typically contain >1.2 wt % F and <3.0 wt % Cl. The most F-rich apatites occur in the uppermost exposed cumulates. Pristine apatites from coeval sills and dikes from below the complex and from the Basal series are similarly F-rich. In all apatites, the Cl and F contents are lower in rocks affected by later metamorphic fluids. Rare earth element (REE) concentrations in chlorapatites show a marked peak in the olivine-rich rocks of the J-M Reef, and contain up to 2 wt % Ce₂O₃ + La₂O₃. The trend of first increasing, then decreasing Cl/F ratios with stratigraphic height is modeled by a vapor-driven zone refining process occurring within the cumulate pile causing Cl-enrichment in the interstitial melt accompanied by degassing at the top of the magma chamber causing overall loss of Cl from the magma as crystallization proceeded. The abrupt change from Cl-rich to more F-rich apatites within OB I is interpreted as the result of a breakdown of the Cl-rich zone refining front and mixing with Cl-poor supernatant melt. Any high temperature fluids that exsolved and circulated through the lower 1/3 of the complex must have been enriched in Cl and could have transported REE and trace metals.     

The Petrology and Geochemistry of Mafic Igneous Rocks in the Anorthosite-Bearing Adirondack Highlands, New York

Igneous rocks of broadly basaltic composition are widely distributedin the anorthosite-bearing Adirondack Highlands of New York.They constitute a mafic series of rocks that occurs on the marginsof the anorthosite series and up to 50 km away from the anorthosite.On an Sr reference diagram, the mafic series has an apparentinitial ⁸⁷Sr/⁸⁶Sr value of 0.7036, which is consistent with1100 Ma subcontinental mantle. The series was probably emplacedin the interval 1150–1100 Ma, during which time anorthositeand granitic magmas were also emplaced. The rocks were metamorphosedshortly thereafter ({small tilde} 1088 Ma) under upper-amphiboliteand granulite facies conditions. Individual bodies of maficrocks range in thickness from small enclaves and layers >>1 m thick to large lensoid masses several tens of meters thick.Despite deformation and metamorphic recrystallization, manyof the rocks retain chiHed margins, cross-cutting relationships,and relict igneous textures. Selected samples of the mafic series have been analyzed fortheir major and trace element compositions. Metamorphism didnot significantly alter the igneous geochemical relationships,and the rocks retain mantle-like values for Zr/Nb, K/Zr, K/Rb,Rb/Sr, and ⁸⁷Sr/⁸⁶Sr. The most primitive rocks of the seriesare silica-undersaturated gabbroic troctolites, and the moreevolved rocks are basaltic in composition. The mafic seriesas a whole has high abundances of A1₂O₃, FeO, the light rareearth elements (LREE), and other incompatible trace elements.Even the most geochemically primitive compositions have highFeO contents. The Fe enrichment and Si depletion that are shownby chemically evolved compositions are consistent with a Fennertrend of fractionation. Low levels of normative di indicatethat high Fe is not a result of the extensive fractionationof cpx. The geochemical trends that are defined by the traceelements, including the REE, suggest the basaltic rocks maybe differentiates of a parental magma of gabbroic troctolitecomposition. The main compositional trend of the mafic seriescan be simulated by 61% crystallization of olivine and plagioclaseof a gabbroic troctolite, followed by 13% crystallization ofolivine, plagioclase, clinopyroxene, and titanomagnetite atthe final stages. The modelled ratio of olivine to plagioclasecrystallization changes from 1–8: 1 to 0–64: 1.These non-cotectic ratios may reflect a delay in the crystallizationof plagioclase relative to olivine, possibly as a result oflow nucleation rates. At later stages of differentiation, plagioclasewas more important in the crystallization of the series. Delayedcrystallization of plagioclase may also have resulted in thehigh A1₂O₃ contents and enhancement of Eu and Sr relative toother trace elements at early stages of differentiation. The mafic series and the silica-saturated anorthosite seriestogether form an anorthosite-norite-troctolite (ANT) suite.More than one mantle composition may have been involved in generatingthe Adirondack mafic magmas. The rocks retain geochemical evidenceof a source that was depleted in basaltic components (cpx) butenriched in Fe, Ti, K, and the LREE. Previously documented evidenceof anticlockwise cooling paths (Bohlen, 1987) and of a regionalgravity high centered beneath the Adirondack region (Simmons,1964) suggests that of the continental crust by basaltic magmasmay have underplating of the continental crust by basaltic magmasmay have been an important feature of the tectonic evolutionofthe region. A model of mantle upwelling beneath thinning continentalcrust explains the geochemically hybrid nature of the maficseries magmas. It is also consistent with a tectonic settingof incipient or failed continental rifting, to which the generationof the anorthosites is commonly attributed.     

A lead isotopic study of the Stillwater Complex,Montana: constraints on crustal contamination and source regions

Analyses of the Pb isotopic compositions of plagioclase from 23 samples covering the stratigraphic thickness of the Stillwater Complex indicate a narrow range of apparent initial isotopic compositions (²⁰⁶Pb/ ²⁰⁴Pb=13.95; ²⁰⁷Pb/²⁰⁴Pb=14.95–15.01; ²⁰⁸Pb/²⁰⁴Pb=33.6). The uniformity of our data is in contrast to, but not necessarily contradictory to, other recent investigations which give indications that the complex formed by repeated injection of magmas with at least two distinct compositions that were presumably derived from different source regions. Samples from the Basal series of the complex have consistently higher ²⁰⁷Pb/²⁰⁴Pb ratios, suggesting either minor contamination from adjacent country rocks or a slight distinction between parental magmas. Apparent initial Pb isotopic compositions of the complex are very radiogenic compared to Late Archean model-mantle values, but are nearly identical to initial Pb isotopic compositions found for the the adjacent, slightly older (2.73–2.79 Ga), Late Archean crustal suite in the Beartooth Mountains. Contamination of magmas parental to the Stillwater Complex by the Late Archean crustal suite is rejected for two reasons: (1) Th and U concentrations in Stillwater rocks and plagioclase are very low (about 0.08 and 0.02 ppm respectively), yet Th/U ratios are uniform at about 4, in contrast to the highly variable (2–26) but often high Th/U ratios found for the Late Archean crustal complex; (2) it seems improbable that any contamination process would have adjusted the isotopic compositions of the diverse magmas entering the Stillwater chamber to near-identical values. The preferred hypothesis to explain the Pb isotopic data for the Stillwater Complex and the associated Late Archean crustal suite involves a major Late Archean crust-forming event that resulted in a compositionally complex crust/mantle system with relatively homogeneous and unusual Pb isotopic compositions. The parental magmas of the Stillwater Complex were generated at different levels within this crust/mantle system, before isotopic contrasts could develop by radioactive decay within compositionally discrete reservoirs. This situation limits the utility of all isotopic tracer systems in discriminating among the various mantle and crustal reservoirs that may have affected the final isotopic character of the Stillwater magmas. The late Archean crustal complex and the Stillwater Complex melts were ultimately derived from the same distinct mantle without obvious direct interaction with the Middle to Early Archean crust present in the region.     

High-temperature carbonate minerals in the Stillwater Complex, Montana, USA

High-temperature carbonate minerals have been observed in association with sulfide minerals below the platiniferous Johns-Manville (J-M) reef of the Stillwater Complex in a stratigraphic section that has been previously shown to be characterized by unusually Cl-rich apatite. The carbonate assemblage consists of dolomite with exsolved calcite in contact with sulfide minerals: chalcopyrite and pyrrhotite in the Peridotite Zone; and pyrrhotite with pentlandite, pyrite and chalcopyrite in Gabbronorite Zone I of the Lower Banded Series. A reaction rim surrounds the carbonate–sulfide assemblages, showing an alteration of the host orthopyroxene to a more calcium-enriched, Fe-depleted pyroxene. The calcite–dolomite geothermometer yields a minimum formation temperature as high as 950 °C for the unmixed assemblages. Iron and manganese concentrations exceed the range seen in carbonatite and mantle xenolith carbonates and are distinctly different from the nearly pure end-member carbonates associated with greenschist-grade (and lower) assemblages (e.g., carbonate veins in serpentinite) that occur locally throughout the complex. The association of high-temperature carbonates with sulfides beneath the J-M reef supports the hydromagmatic theory which involves a late-stage chloride–carbonate fluid percolating upwards, dissolving PGE and sulfides and redepositing them at a higher stratigraphic level. Characterization of the processes which form strategically important metal deposits, such as the J-M reef of the Stillwater Complex and the analogous Merensky reef of the Bushveld Complex in South Africa, could potentially lead to better exploration models and, more broadly, a deeper understanding of the cooling and compositional evolution of large bodies of ultramafic and mafic magma and of carbonatites, on both a local and a regional scale.     

Geochemical Evidence for Mantle Origin and Crustal Processes in Volcanic Rocks from Popocatepetl and Surrounding Monogenetic Volcanoes, Central Mexico

Elemental, isotopic, and mineral compositions as well as rocktextures were examined in samples from Popocatépetl volcanoand immediately surrounding monogenetic scoria cones of theSierra Chichinautzin Volcanic Field, central Mexico. Magma generationis strongly linked to the active subduction regime to the south.Rocks range in composition from basalt to dacite, but Popocatépetlsamples are generally more evolved and have mineral compositionsand textures consistent with more complicated, multi-stage evolutionaryprocesses. High-Mg calc-alkaline and more alkaline primitivemagmas are present in the monogenetic cones. Systematic variationsin major and trace element compositions within the monogeneticsuite can mostly be explained by polybaric fractional crystallizationprocesses in small and short-lived magmatic systems. In contrast,Popocatépetl stratovolcano has produced homogeneous magmacompositions from a shallow, long-lived magma chamber that isperiodically replenished by primitive basaltic magmas. The currenteruption (1994–present) has produced silicic dome lavasand pumice clasts that display mingling of an evolved daciticcomponent with an olivine-bearing mafic component. The longevityof the magma chamber hosted in Cretaceous limestones has fosteredinteraction with these rocks as evidenced by the chemical andisotopic compositions of the different eruptive products, contact-metamorphosedxenoliths, and fumarolic gases. Popocatépetl volcanicproducts display a considerable range of ⁸⁷Sr/⁸⁶Sr (0·70397–0·70463)and _Nd (+6·2 to +3·0) whereas Pb isotope ratiosare relatively homogeneous (²⁰⁶Pb/²⁰⁴Pb 18·61–18·70;²⁰⁷Pb/²⁰⁴Pb 15·56–15·60). KEY WORDS: Popocatépetl; Sierra Chichinautzin Volcanic Field; arc petrogenesis; radiogenic isotopes     

Slow,dense replenishments of a basic magma chamber: the layered series of the Newark Island layered intrusion,Nain, Labrador

The Newark Island layered intrusion is a composite layered intrusion within the Nain anorthosite complex, Labrador. The intrusion comprises a lower layered series (LS) dominated by troctolites, olivine gabbros and oxide-rich cumulates and an upper hybrid series (HS) characterized by a wide range of mafic, granitic and hybrid cumulates and discontinuous layers of chilled mafic rocks (Wiebe 1988). The HS crystallized from a series of replenishments of both silicic and basic magmas. The LS crystallized from periodically replenished basic magmas. The LS has a lower zone that consists mainly of olivine-plagioclase cumulates and contains minor cryptic reversals in mineral compositions that resulted from replenishments of relatively primitive magma. An upper zone is dominated by olivine-plagioclaseaugite-ilmenite cumulates. Cumulus titanomagnetite and pyrrhotite occur within some oxide-rich cumulates, and the stratigraphically highest layers contain cumulus apatite. At intermediate levels in the sequence, cumulus inverted pigeonite occurs in place of olivine. Several prominent regressions in the stratigraphy of the upper zone are marked by fine-grained troctolitic layers with much higher Mg no. [100 MgO/(MgO+FeO)] and anorthite than underlying cumulates. These layers coarsen upward and grade back to oxide-bearing olivine gabbros within thicknesses ranging from 10 cm to 15 m. Dikes that cut the LS have major- and trace-element compositions that strongly suggest that they are feeders for the replenishments. In the lower zone when olivine and plagioclase were the only cumulus phases, replenishments were less dense than the resident magma and rose as plumes and mixed with it. Precipitation of cumulus oxides in the upper zone lowered the density of resident magma so that subsequent replenishments were more dense than resident magma. Replenishments that occurred after oxides began to precipitate had small injection velocities. These post-oxide injections flowed along the interface between resident magma and the cumulate pile and precipitated flow-banded, fine-grained troctolites.