The composition of magmas proposed as parental to the layered suite of the Bushveld Complex, and some models for the manner of their emplacement, are reviewed briefly. Included are some contributions published in South Africa, with which overseas readers might be less familiar. Emphasis is given to the broader features of the cumulates, and the contradictions raised by whole-rock compositional, Sr-isotopic, and trace-element data that cloud their correlation with proposed parental magmas. It is concluded that the Lower, Critical and Main Zones are the derivatives of only two primary magmatic lineages, while a third was added to residual liquids from which the layered rocks above the Pyroxenite Marker were formed. Excessive amounts of olivine and chromium in the cumulates of the Lower and Lower Critical Zones in the northern sector of the Western Limb can seemingly not be accounted for by the composition and volume of the putative magmas. This is attributed to (1) this sector being a proximal facies located close to the original feeder, and/or (2) crystal-charged magma batches, expelled from a lower magma chamber, being periodically injected into and dispersed within the liquids already in place in the Bushveld chamber. Thus, ongoing changes in the bulk composition of the liquids within the chamber would not be reflected in the rinds of earlier, chilled-facies rocks. The expulsion of significant volumes of liquid residua from the chamber during cumulate deposition is not ruled out.

The Main Zone of the Bushveld Complex, which is ˜3–3.5 km thick, comprises a sequence of gabbronorites with minor anorthosites and pyroxenites. The Pyroxenite Marker (PM), a thin orthopyroxenite layer occurring towards the top of the Main Zone in the eastern Bushveld, marks the change from an inverted pigeonite-bearing microgabbronorite below, to a primary orthopyroxene-bearing porphyritic gabbronorite above. In the western Bushveld the PM has only been observed in core material although its surface position can be inferred from the mineralogical and textural changes. Whole-rock geochemistry of surface and core samples from the Brits and Marikana areas, together with mineral compositional data, have been integrated with published analyses to elucidate the magmatic processes that occurred during the addition of new magma into the chamber, just below the level of the PM. Modal and major-element data show that most lithologies lie close to the plagioclase-two pyroxene cotectic. However, there are three units below the PM in which distinct modal layering is developed. Changes in geochemical trends for trace-element abundances in both whole-rock and mineral-separate data occur at ˜150 m below the PM, below a layered package, the Hexrivier Unit. However, there is a displacement of at least 40 m between the beginning of the reversals in An content in plagioclase and in Mg# in pyroxene which occur at 124 m and 80 m below the PM respectively. In addition, a gradual change occurs in the mineral parameter Mg# in opx minus An in plagioclase, with the plagioclase becoming more primitive relative to the pyroxene. This decrease and the decoupling of geochemical trends has not been noted in the Main Zone of the Bushveld Complex before. The decoupling between plagioclase and pyroxene compositional reversals is not easily explained by modal effects, an influx of phenocrysts in the new magma, or by infiltration metasomatism. Instead we propose that new magma pulses first entered the chamber some 150 m below the PM and that this magma had a composition that crystallized more primitive plagioclase but similar pyroxene to the magma residing in the chamber. The chamber was intruded by progressively more magma that had a more primitive composition particularly in terms of pyroxene composition. This model can explain the decoupling between plagioclase and pyroxene compositional trends. Continued mixing between resident magma and the new influxes occurred over an interval of ˜150–200 m. Above the PM fractional crystallization processes dominated and continued into the Upper Zone.

Spinels associated with discordant bodies of iron-rich ultramafic pegmatite are described from the Amandelbult Platinum mine in the northwestern part of the Bushveld Complex. The spinels are divided into three groups, disseminated Ti-magnetite, disseminated Fe-Ti-Cr spinel and massive Fe-Ti-Cr spinel. The Fe-Ti-Cr spinels show a range of unusual compositions intermediate between chromite and Ti-magnetite. A relationship was found between stratigraphic height and spinel-type, with the Fe-Ti-Cr spinels restricted to pegmatites from the Upper Critical zone and Ti-magnetite to pegmatites from the Lower Main zone. Ilmenite is a ubiquitous component of all of the pegmatites examined here. The massive Fe-Ti-Cr oxide pegmatites are found only where earlier-formed chromitite layers are juxtaposed with sheet-like bodies of olivine-clinopyroxene pegmatite. A distinct thickening of the original chromitite layers in this situation, and compositional gradients within them, points to accretion of Fe-Ti-Cr spinels onto them prior to partial sub-solidus re-equilibration. Analytical data are presented for these spinels and for the Ti-magnetite.

The composition of the Fe-Ti-Cr spinels is not duplicated by cumulus spinels in the Bushveld Complex, but the compositions and microtextures of the disseminated Ti-magnetite are very similar to cumulus Ti-magnetite from the Upper zone. Accordingly, it is deduced that the Ti-magnetite in the pegmatites from the Lower Main zone, together with the ilmenite, crystallized at magmatic temperatures from a suitable Fe-Ti-rich silicate-oxide melt. No evidence has been found to link the pegmatites to hydrothermal fluids. The Cr-rich nature of the disseminated spinels in pegmatites from the Upper Critical zone suggests that the pegmatite melt was richer in chromium at this stratigraphic height, although re-equilibration with earlier-formed cumulus chromite also occurred. Formation of the Fe-Ti-Cr oxide pegmatites reflects a complex process that is incompletely understood and why new oxides plate onto pre-existing chromitite layers that are juxtaposed with Fe-rich ultramafic pegmatites is a matter of conjecture.

The Upper Critical Zone of the Bushveld Complex consists of several cyclic units, each ranging from ultramafic to leucocratic. One view is that they were initiated by addition of, and crystallization from, relatively magnesian magma. Experimental studies on plausible parental magma compositions to this sequence show that plagioclase joins orthopyroxene in the crystallization sequence once the Mg# of the orthopyroxene has decreased below 83. Some norites from the Upper Critical Zone contain orthopyroxene with this Mg#, supporting the validity of the experimental studies. There are no orthopyroxene compositions with higher Mg# than 83 in the entire Upper Critical Zone in either the eastern or western limbs of the Bushveld Complex. This observation suggests that all these rocks formed from plagioclase-saturated magmas. If this interpretation is correct, pyroxenites in this succession are not the result of crystallization from a magnesian magma. Instead, they result from the mechanical separation of plagioclase and orthopyroxene, probably due to crystal sorting during settling, from a magma lying at the cotectic for these two minerals. Rocks exist in the Upper Critical Zone that contain non-cotectic proportions of cumulus plagioclase and orthopyroxene, again supporting models of crystal sorting during settling. In such a model, anorthosites result from the delayed accumulation of plagioclase relative to pyroxene, and not to formation from a magma saturated only in plagioclase.

When traced from northwest to southeast in the western limb, there is a change in the relative proportions of plagioclase to orthopyroxene in the Upper Group 2 chromitite cyclic unit. In the northwest the unit is dominated by ultramafic rocks. In the southeast the plagioclase to pyroxene ratio exceeds that of the cotectic proportion, relations that may result from the lateral transport of suspended plagioclase grains to the southeast.

Discordant veins, pipes and occasionally subconcordant sheets of iron-rich ultramafic pegmatite disrupt the layered cumulate sequence of the Upper Critical Zone, Rustenburg Layered Suite, Bushveld Complex. These pegmatite bodies have been studied where they replace the Merensky Reef footwall at Northam Platinum Mine, situated in the Swartklip Facies of the western lobe of the Rustenburg Layered Suite. Composed chiefly of ferroaugite and fayalitic olivine, the pegmatites appear to be formed by the preferential replacement of plagioclase-rich cumulates within the layered sequence. Fe-Ti oxides, sulphide (pyrrhotite and chalcopyrite) and plagioclase also occur in variable quantities. Differentiation within the pegmatite is observed where it has spread laterally beneath the impervious Merensky chromitite layer, with the development of subparallel cm-scale layers of massive magnetitite, massive sulphide and sulphide pegmatite. While some Fe-rich mobile phase must have been responsible for the pegmatites, it is concluded that the pegmatite bulk composition does not represent the original liquid. Furthermore the mode of occurrence precludes the injection of a crystal mush. Rather it is argued, mainlyon geochemical and isotopic grounds, that Fe-rich residual melts derived from the Upper Zone in the downward crosscutting gap areas migrated laterallyand upwards into the adjacent Upper Critical Zone. Variable reaction with the layered cumulates produced the anastomosing pegmatite bodies.

Granular aggregates of fine-grained graftonite (Fe,Mn,Ca)3(PO4)2 and intergrown wolfeite (Fe,Mn)2(PO4)(OH) occur in amphibolite-facies metamorphosed iron formations associated with the Gamsberg Zn-Pb deposit, South Africa. To date, these minerals were believed to have limited parageneses, being essentially restricted to granitic pegmatites and iron meteorites. This paper is the first report of the occurrence of graftonite and wolfeite in a regionally metamorphosed, iron formation-hosted setting. The aggregates are found together with Mn- and Pb-rich apatite and calcian pyromorphite in a pristine unit of almost pure chemical precipitates, the origin of which is intimately linked to the base-metal mineralizing process. Evidence from Gamsberg supports previous studies conducted on pegmatite-hosted graftonites that a simple host rock mineralogy and geochemical prerequisites, such as high activities of Fe, Mn, Ca and a deficiency in F, exert a dominant control on the stabilization of these minerals. However, in a marine sedimentary environment, significant concentrations of phosphorus have to be precipitated to prevent stabilization of all the phosphorus as fluorapatite. The paucity of graftonite in such settings suggests that the combination of these requirements is only rarely achieved.

The LGC (Leinster Granite Complex) in SE Ireland consists of five plutons or Units. Units 1 and 3 display complex magmatic disequilibrium textures described here for the first time. The textures indicative of disequilibrium include boxy cellular plagioclase, oscillatory zoning, calcic spike zones and combinations of these within individual plagioclase feldspar crystals. The textures are interpreted as having formed in a dynamic magmatic environment which facilitated acid-acid mixing on a variety of scales and at a variety of stages between source to emplacement. It is proposed that the restricted distribution of disequilibrium textures was a result of the complex strain regime operating within the crust during assembly of Units 1 and 3.

Reaction textures including corona structures in granulites from the Proterozoic Long Range Inlier of western Newfoundland are spatially associated with a Silurian (0.34 Ga) mafic intrusion, the Taylor Brook Gabbro Complex. They comprise, in metabasites and tonalitic gneiss, coronal orthopyroxene and plagioclase on garnet and, in metapelites, cordierite and spinel formed at the expense of sillimanite, garnet and quartz. Although generally interpreted to indicate near-isothermal decompression (ITD) following regional metamorphism, which in the inlier occurred at ˜1.10–1.03 Ga, these features appear to be absent elsewhere. Therefore they are interpreted to be products of contact metamorphism (near-isobaric heating – IBH) within the thermal aureole of the gabbro. Thus, there is a ˜0.7 Ga difference (i.e. mid-Proterozoic vs. mid-Silurian) between the age of the regional metamorphic mineral assemblages and the contact aureole assemblages. The observation that classic ITD features occur in this aureole environment underscores the fact that P-sensitive reactions can progress during IBH as well as by pressure release.

The Island of Linosa is a small part of the large submarine volcanic complex which is locatedat the SW edge of the Linosa Graben, Sicily Channel. The island was formed between 1.06±0.10 and 0.53±0.07 Ma, through three main stages of activity: Paleo-Linosa, Arena Bianca and Monte Bandiera. Major and trace element data show that the compositional variability of the three activity stages is limited, with most of the rocks showing basaltic to hawaiitic composition. Evolved benmoreites and trachytes are foundas lithics in some pyroclastic units of Paleolinosa. The mafic rocks of the three stages show porphyritic texture, with phenocryst assemblages characterized by olivine, clinopyroxene and plagioclase. The volume ratio of olivine vs. clinopyroxene decreases from early to late stages of activity in mafic rocks with comparable major element composition. Clinopyroxene phenocrysts from mafic rocks of the three stages have poorly variable composition, clustering in the augite field. Phenocrysts from the first activity stage (Paleo-Linosa), show a slight increase in TiO2, Al2O3 and CaO, and a decrease of Fe2O3 (total) with the increasing SiO2 content of the host rocks. Crystals from the second and the third stage (Arena Bianca and Monte Bandiera) display a slightly more restricted range of FeOtot, frequently with very high MgO, Al2O3 and TiO2 contents. Crystal chemical investigation of clinopyroxenes from rocks of the three stages with comparable degrees of evolution, revealed significant variation of structural parameters, in particular VM1 and Vcell. These show a consistent decrease, passing from clinopyroxenes of the early stage to crystals extracted from the mafic lavas of stages 2 and 3. Given the similar compositional ranges of the host rocks, structural variations of clinopyroxenes are interpreted to reveal modifications of crystallization pressure, which increased, passing from Paleo-Linosa to the Arena Bianca and Monte Bandiera stages. Given this information, the observed crystal-chemical variations provide information on the depth of magma reservoirs and on the evolution of the plumbing system of Linosa volcano.

The studied gabbroic nodules occurring in the Petrazza pyroclastic rocks consist mainly of plagioclase (An95–87), olivine (Fo83-73) and clinopyroxene (Mg# 90–77), with subordinate opaques (Ti-magnetite) and amphibole (Mg-hastingsite), which constitute the cumulate minerals. Interstitial material has a relatively high, but variable, degree of vesicularity and consists of variable amounts of glass and quenched crystals of plagioclase (An71–55), amphibole, clinopyroxene and rare biotite, olivine and opaques. Silicate melt inclusions are abundant in the cumulate minerals, but complete homogenization to melt has been observed only in the inclusions occurring in clinopyroxene, where the temperatures of homogenization vary from 1134 to 1190°C. Microthermometric investigations of fluid inclusions and of the shrinkage bubble of the melt inclusions suggest that the magma contained CO2. The apparent scarcity of H2O indicates that this component was strongly partitioned into the magma at the time of crystallization of the investigated minerals; this is further supported by the occurrence of (1) daughter biotite- and amphibole-bearing inclusions which show that the H2O activity in the magma was sufficiently high to allow their crystallization, and (2) calcic plagioclase (An95–87) which can be crystallized from a high-alumina basaltic magma at pressure ≤2 kbar, temperatures in the range 1050–1100°C and in the presence of 3–4 wt.% of water (MELTS software simulations).

The composition of the melt inclusions suggests that the hosting plagioclase, olivine and clinopyroxene crystallized from slightly different batches of magma. The S content in the melt inclusions of clinopyroxene and olivine is high (up to 0.41 wt.%). The presence of Fe-Cu(-Ni)-rich blebs of sulphide in plagioclase, olivine, amphibole, and locally in the melt inclusions too, further supports the important role of sulphur in the primitive magmas of the investigated gabbros. Small differences in redox conditions or in the Fe content of the melts favoured S mobilization as sulphide.

A deposit of chrysoberyl (BeAl2O4), including the variety alexandrite, occurs near Dowerin, in the southwestern region of the Archaean Yilgarn Craton, Western Australia. The deposit is situated in the northern part of the Lake Grace Terrain, a crustal component of the southwestern Yilgarn Craton, in granulite-facies gneisses (2640–2649 Ma; T = 700°C, P <6 kbar) adjacent to the margin of the Kellerberrin Batholith (2587±25 Ma). Beryllium mineralization at Dowerin occurs in plagioclase-quartz-biotite-garnet gneiss and cross-cutting tourmaline-plagioclase veins situated adjacent to lenses of actinolite-cummingtonite-phlogopite schist. Crystals of chrysoberyl (0.15–1.74 wt.% Cr2O3; 2.25–3.23 wt.% FeO; trace–0.13 wt.% ZnO; SiO2 <0.05 wt.%) are found embedded in almandine or plagioclase, and closely intergrown with biotite and/or zincian hercynite in the host-rock gneiss. Minor Cr and Fe in the alexandrite variety of chrysoberyl were possibly derived from associated zincian hercynite and/or almandine. Trace beryl (0.04–0.20 wt.% Cr2O3; 0.54–0.71 wt.% FeO; trace– 0.22 wt.% Na2O; 0.1–0.71 wt.% MgO) occurs as anhedral interstital grains between crystals of chrysoberyl, plagioclase and biotite, and as rare inclusions in chrysoberyl. Textural and mineral chemical evidence suggests that chrysoberyl and zincian spinels (chromite to hercynite containing from 2–8 wt.% ZnO) formed during granulite-facies regional metamorphism and probably pre-dated the formation of metamorphic tourmaline-plagioclase veins during the same metamorphic episode. The Be, B and Zn required to form chrysoberyl, beryl, tourmaline and zincian spinels may have been released by metamorphic reactions in host-rock metapelites during prograde granulite-facies metamorphism.

An orbicular diorite from Fisher Lake, California, USA, contains multi-shelled, magmatic orbicules with branching and budding orthopyroxene crystals as well as feather and acicular plagioclase crystals that are oriented perpendicular to the growth horizon. Plagioclase and orthopyroxene show gradual, reverse compositional zoning along the long axes and normal zoning along the short axes. The reverse zoning varies from An87 to An93 and Mg68 to Mg74 over distances of 4 mm and 8 mm respectively. The close proximity of these two minerals makes it likely that only one mechanism is responsible for the reverse zoning. This zoning can be explained by using relevant temperature-composition diagrams and Gibbs free energy-composition plots. Under sudden and moderate undercoolings, which produce high growth but low nucleation rates, the difference in Gibbs free energy (ΔG) between the crystals and liquid is not initially maximized, i.e. initial compositions are not near-to-equilibrium. This results in crystal compositions that are closer to that of the bulk liquid than expected for crystallization under near-to-equilibrium conditions (i.e. very small ΔT). Over time, and under isothermal crystallization conditions, ΔG gradually increases to a maximum producing crystal compositions that also gradually attain near-to-equilibrium compositions. Subsequent to attaining these conditions, normal zoning occurs perpendicular to the crystal growth axes.

Piston-cylinder cell assemblies experience inhomogeneous pressure distribution upon pressurization due to the variable compressibilities of the cell components. This results in the sample experiencing a pressure lower than expected, given the applied force of the piston. Although the effect is generally compensated for by applying a ‘friction’ correction, there have been wide variations in the corrections applied for some of the harder cell materials. We have determined friction correction factors for a range of cell assemblies commonly used in our laboratory relative to select well-characterized phase equilibria. Single-sleeve NaCl cells require, using the piston-in technique, very small corrections of the order −0.05 GPa for 12.7 mm diameter, and less for larger diameter assemblies. Four separate calibrations of the single sleeve 12.7 mm BaCO3 cell show that it requires a correction of −9%. This factor is entirely independent of temperature and pressure within the range 1000 to 1600°C and 1.5 to 3.2 GPa. This result is in contrast to the results of Fram and Longhi (1992) who claim that the correction for BaCO3 cells is highly dependent on pressure. For the assemblies included in this study there is an increase in the pressure correction required in the order of 12.7 mm diameter NaCl-pyrex −3%; 19 mm talc-pyrex −3.6%; 12.7 mm BaCO3 −9% and 12.7 mm BaCO3-silica glass −13%.

Massive chromitite, banded chromitite and disseminated chromite grains are found in a ˜3800 Ma layered ultrabasic body in West Greenland. The major part of the ultrabasite is dominated by dunite. In the upper exposed part, harzburgite and sheets of gabbro-anorthosite occur. Chromite grains in dunites, and in massive and banded chromitites are homogeneous, with increasing Fe contents upwards in the intrusion. In harzburgites chromites show unusual and very complex textural relationships, with two generations ofchromites one replacing the other, and both exhibiting exsolution textures. In harzburgites, an Fe-rich chromite crystallized first. This first chromite exsolved two spinel phases in a very fine-scale pattern and ilmenite lamellae in a trellis pattern. The Fe-rich chromite was later partly replaced by Al-rich chromite, which crystallized contemporaneously with formation of a late gabbro-anorthositic melt. Subsequently, the Al-rich chromite exsolved a very fine-scale magnetite-rich phase. The exsolutions in the first generation chromite were formed under magmatic conditions. Exsolution of ilmenite lamellae in Fe-rich spinel was caused by oxidation under magmatic conditions.

The sulphide ores from the Julcani mining district (Peru) display many retrograde reactions that may be attributed to solid-state processes accompanying cooling. Fahlores [˜(Cu,Ag)10(Zn,Fe)2(Sb,As)4S13] from the Herminia mine exhibit pronounced downstream enrichments in molar Ag/(Ag+Cu) ratios that are strongly correlated with the abundance of PbS-AgSbS2-AgBiS2 phases. These correlations, discontinuous core to rim Sb/(Sb+As) enrichments in bournonites, and prominent reaction textures involving fahlores, bournonites and galenas provide strong evidence that the fahlores in these ores have been enriched in Ag by the Ag–Cu exchange reaction

which occurred during cooling following mineralization of a PbS-AgSbS2-AgBiS2 galena and has been documented elsewhere. Secondary PbS-AgSbS2-AgBiS2 minerals aramayoite, bismuthian diaphorite [Pb2Ag3(Bi,Sb)3S8], and diaphorite were produced from primary galenas with cooling of ores with high Pb/Cu and Bi/Sb; pyrargyrite formed from the breakdown of the Ag10Zn2Sb4S13 component in the most Ag-rich fahlores, as an exsolution product of galena, and from replacement of aramayoite and galena with the evolution of semimetal sulphides. Based on mineral compositions, phase equilibria, a thermochemical database for sulphides and sulphosalts, and the reintegrated composition for primary grains of Ag-rich PbS-AgSbS2-AgBiS2 phases, we estimate that the primary temperature of hydrothermal mineralization was >320±10°C, that these reactions ceased to affect fahlore Ag/(Ag+Cu) ratios and Bi/(Bi+Sb) ratios of aramayoite and miargyrite after cooling through 220±10°C. Galenas, however, appear to have continued to adjust their compositions to reflect even lower temperatures by continued exsolution of semimetals and production a diverse suite of sulphosalts that occur in fine intergrowths with galena.

The molecular structure of the three vivianite-structure, compositionally related phosphate minerals vivianite, baricite and bobierrite of formula M32+(PO4)2.8H2O where M is Fe or Mg, has been assessed using a combination of Raman and infrared (IR) spectroscopy. The Raman spectra of the hydroxyl-stretching region are complex with overlapping broad bands. Hydroxyl stretching vibrations are identified at 3460, 3281, 3104 and 3012 cm−1 for vivianite. The high wavenumber band is attributed to the presence of FeOH groups. This complexity is reflected in the water HOH-bending modes where a strong IR band centred around 1660 cm−1 is found. Such a band reflects the strong hydrogen bonding of the water molecules to the phosphate anions in adjacent layers. Spectra show three distinct OH-bending bands fromstrongly hydrogen-bonded, weakly hydrogen bonded water and non-hydrogen bonded water. The Raman phosphate PO-stretching region shows strong similarity between the three minerals. In the IR spectra, complexity exists with multiple antisymmetric stretching vibrations observed, due to the reduced tetrahedral symmetry. This loss of degeneracy is also reflected in the bending modes. Strong IR bands around 800 cm−1 are attributed to water librational modes. The spectra of the three minerals display similarities due to their compositions and crystal structures, but sufficient subtle differences exist for the spectra to be useful in distinguishing the species.

The composition of chevkinite and perrierite, the most common members of the chevkinite group, is closely expressed by the formula A4BC2D2Si4O22, where A = (La,Ce,Ca,Sr,Th), B = Fe2+, C = (Fe2+,Fe3+,Ti,Al,Zr,Nb) and D = Ti. The A site is dominated by a strong negative correlation between (Ca+Sr) and the REE. Chondrite-normalized REE patterns are very variable, e.g. in LREE/HREE and Eu/Eu*. The C site is dominated by Ti, Al and Fe2+, in very variable proportions. Most chevkinites and perrierites are close to stoichiometric, with cation sums between 12.9 and 13.5, compared to the theoretical 13. There is no single, generally applicable charge balancing substitution scheme in the group; however, the general relationship

defines a linear array with r2 = 0.91. Chevkinite and perrierite are shown to be compositionally distinct on the basis of CaO, FeO*, Al2O3 and Ce2O3 abundances. Chevkinite forms mainly in chemically evolved parageneses, such as syenites, rhyolites and fenites associated with carbonatite complexes. Perrierite is more commonly recorded from igneous rocks of mafic to intermediate composition. The compositional characteristics and possible structural formulae of other members of the chevkinite group are reviewed briefly.