Eight-phase alkali feldspars: low-temperature cryptoperthite, peristerite and multiple replacement reactions in the Klokken intrusion

Eight feldspar phases have been distinguished within individual alkali feldspar primocrysts in laminated syenite members of the layered syenite series of the Klokken intrusion. The processes leading to the formation of the first four phases have been described previously. The feldspars crystallized as homogeneous sodian sanidine and exsolved by spinodal decomposition, between 750 and 600 °C, depending on bulk composition, to give fully coherent, strain-controlled braid cryptoperthites with sub-μm periodicities. Below ~500 °C, in the microcline field, these underwent a process of partial mutual replacement in a deuteric fluid, producing coarse (up to mm scale), turbid, incoherent patch perthites. We here describe exsolution and replacement processes that occurred after patch perthite formation. Both Or- and Ab-rich patches underwent a new phase of coherent exsolution by volume diffusion. Or-rich patches began to exsolve albite lamellae by coherent nucleation in the range 460–340 °C, depending on patch composition, leading to film perthite with ≤1 μm periodicities. Below ~300 °C, misfit dislocation loops formed, which were subsequently enlarged to nanotunnels. Ab-rich patches (bulk composition ~Ab₉₁Or₁An₈), in one sample, exsolved giving peristerite, with one strong modulation with a periodicity of ~17 nm and a pervasive tweed microtexture. The Ab-rich patches formed with metastable disorder below the peristerite solvus and intersected the peristerite conditional spinodal at ~450 °C. This is the first time peristerite has been imaged using TEM within any perthite, and the first time peristerite has been found in a relatively rapidly cooled geological environment. The lamellar periodicities of film perthite and peristerite are consistent with experimentally determined diffusion coefficients and a calculated cooling history of the intrusion. All the preceding textures were in places affected by a phase of replacement correlating with regions of extreme optical turbidity. We term this material ultra porous late feldspar (UPLF). It is composed predominantly of regions of microporous very Or-rich feldspar (mean Ab_2.5Or_97.4An_0.1) associated with very pure porous albite (Ab_97.0Or_1.6An_1.4) implying replacement below 170–90 °C, depending on degree of order. In TEM, UPLF has complex, irregular diffraction contrast similar to that previously associated with low-temperature albitization and diagenetic overgrowths. Replacement by UPLF seems to have been piecemeal in character. Ghost-like textural pseudomorphs of both braid and film parents occur. Formation of patch perthite, film perthite and peristerite occurred 10⁴–10⁵ year after emplacement, but there are no microtextural constraints on the age of UPLF formation.     

Feldspar microtextures and multistage thermal history of syenites from the Coldwell Complex,Ontario

Optical and TEM (transmission electron microscopy) observations of perthites from augite syenites in the Coldwell Complex (Ontario) reveal a complex set of microtextures that outline a multistage thermal history. Regular microtextures (linear or braid texture, straincontrolled, coherent intergrowths) show a progressive evolution from the margin of the intrusion inwards with lamellar spacings in the range 40–100 nm. The textures evolve in a manner similar to those for the Klokken intrusion and reflect differences in cooling rates and bulk composition. Superimposed upon the regular microtexture are 10 m scale compositional fluctuations which we call ripples. The boundary relationships and bulk composition of ripples, which are themselves Ab-rich and Or-rich linear coherent cryptoperthites, suggest that they formed by coarsening during a phase of high-temperature (530°C) fluid-feldspar interaction. This was followed by a return to coherent exsolution in which fluid was not involved. Coarse, irregular, patch microperthite cross-cuts all other microtextures. These final deuteric intergrowths are believed to result from a further low-temperature (< 380° C) fluid-feldspar interaction and are associated with subgrain formation and the presence of micropores. The outermost syenite sample, against a gabbro ring structure, has distinctive, modified microtextures, indicating that the gabbro is, at least in part, a later intrusion. Our findings show that TEM work on alkali feldspar microtextures can identify discrete thermal events in the cooling history of igneous plutons and illustrates the potential of such microtextures for establishing the relative ages of intrusive rocks.     

Cryptic microtextures and geological histories of K-rich alkali feldspars revealed by charge contrast imaging

Charge contrast imaging in the scanning electron microscope can provide new insights into the scale and composition of alkali feldspar microtextures, and such information helps considerably with the interpretation of their geological histories and results of argon isotope thermochronological analyses. The effectiveness of this technique has been illustrated using potassium-rich alkali feldspars from the Dartmoor granite (UK). These feldspars contain strain-controlled lamellar crypto- and microperthites that are cross-cut by strain-free deuteric microperthites. The constituent albite- and orthoclase-rich phases of both microperthite generations can be readily distinguished by atomic number contrast imaging. The charge contrast results additionally show that sub-micrometre-sized albite ‘platelets’ are commonplace between coarser exsolution lamellae and occur together to make cryptoperthites. Furthermore, charge contrast imaging reveals that the orthoclase-rich feldspar is an intergrowth of two phases, one that is featureless with uniform contrast and another that occurs as cross-cutting veins and grains with the {110} adularia habit. Transmission electron microscopy shows that the featureless feldspar is tweed orthoclase, whereas the veins and euhedral grains are composed of irregular microcline that has formed from orthoclase by ‘unzipping’ during deuteric or hydrothermal alteration. The charge contrast imaging results are especially important in demonstrating that deuteric perthites are far more abundant in alkali feldspars than would be concluded from investigations using conventional microscopy techniques. The unexpected presence of such a high volume of replacement products has significant implications for understanding the origins and geological histories of crustal rocks and the use of alkali feldspars in geo- and thermochronology. Whilst the precise properties of feldspars that generate contrast remain unclear, the similarity between charge contrast images and corresponding cathodoluminescence images of deuteric microperthites indicates that trace element chemistry and possibly also elastic strain within the crystal play a major role.     

Exsolution and coarsening mechanisms and kinetics in an ordered cryptoperthite series

Cryptoperthites from the Klokken layered syenite intrusion were examined by TEM to determine the role of exsolution, ordering and twinning in the development of the coherent microtextures during slow cooling, the stratigraphic position of the samples in the layered series giving an independent variable in determining their evolution. Both periodicity (primary and secondary) and morphology change with distance from the top of the series. Most samples contain low microcline in the diagonal association.Partial ordering occurred before exsolution, which was followed by Albite-twin formation in the albite lamellae. The twin periodicity depends on the average lamellar thickness (or on the primary lamellar periodicity, ₁) and no longer changes during subsequent morphological evolution. In the Or-rich lamellae long-period Albite twins develop before waves form in the lamellar interface. The interfaces rotate with increasing order to give parallel-sided zig-zag lamellae of low microcline with Albite twinned lamellae of low albite, which may pinch and swell. Where the albite lamellae are discontinuous, adjacent microcline lamellae coalesce giving oblique lamellae and Pericline or M-type twins. Thickening of some oblique lamellae gives a distinct secondary periodicity, ₂, which outlines lozenge-shaped areas with relics of the primary periodicity and, if coarse enough, is responsible for optically-visible braid microperthite. Coherency, demonstrated by high resolution images, is maintained through all stages of the coarsening.A time-temperature-transformation diagram for continuous cooling is presented and can be used to interpret the kinetics and morphological evolution of cryptoperthites from rocks with very different cooling rates (dykes and lavas to very large plutons), which have, however, similar primary lamellar periodicities. The finest periodicities are only slightly larger than the supposed initial periodicities ( _o) for spinodal decomposition and little coarsening can have occurred. Coarsening at cooling rates slow enough to produce significant ordering may be much slower than coarsening in disordered feldspars. Primary coarsening may be stopped by the development of Albite twins in the Abrich phase, which will require reversal of the order-antiorder sense of parts of the framework. Coarsening may also be slowed if the phases at intermediate temperatures order at different rates or have different equilibrium degrees of Al-Si order. Secondary coarsening can develop at much lower temperatures (<400° C) on the formation of low microcline, when both phases have the same framework order.     

Sidewall crystallization in the Klokken intrusion: zoned ternary feldspars and coexisting minerals

The syenitic layered series in the Klokken intrusion is surrounded by a zone (500 m thick) of nearly structureless unlaminated syenite followed outwards by a zone of vertically banded gabbro (200 m thick) at the outer rim. The unlaminated syenite is intrusive into the gabbro and develops a thin (2 m) transition zone of syenodiorite at the contact. A traverse across the vertical transition zone and inwards towards the layered series was sampled with a portable drill. Mafic silicates (olivine, clinopyroxene, biotite) show inward evolution in Fe/(Fe+Mg) across the syenodiorite-unlaminated syenite zones. Feldspars change rapidly across the syenodiorite zone from rocks dominated by plagioclase, in some cases together with two alkali feldspars, one a mesoperthite or cryptomesoperthite, the other a cryptoperthite, to rocks in which plagioclase is seen only rarely as cores to cryptomesoperthitic alkali feldspar crystals. Plagioclase is absent from the layered series.Alkali feldspars occurring in pairs have bulk compositions on solvus isotherms in the Or-Ab-An ternary system, estimated at 950° C in a syenogabbro and 910° C in a syenodiorite, at 1 kbar. The more calcic liquids from which they crystallized fractionated on paths that intersected the two- feldspar surface, whereas the more syenitic members crystallized from liquids which terminated crystallization in the one- feldspar field at 900° C. Plagioclases evolve from calcic andesine in syenodiorites, to very rare sodic oligoclase in the most evolved unlaminated syenites. The boundaries between plagioclase cores and alkali feldspar rims, which are usually optically abrupt, involve complex mixed zones on the m -scale, consistent with arrested reaction between plagioclase primocrysts and crystallizing syenitic liquid. Ternary liquidus-solidus relationships are in qualitative agreement with this interpretation. The syenodiorites are cumulates produced during sidewall crystallization of a trachytic magma against a gabbroic chamberlining. This magma changed little in bulk composition as it evolved, giving rise to the unlaminated syenites by further sidewall crystallization. Water build- up in this liquid probably caused a change in style of chamber filling, giving rise to the layered series by bottom accumulation. Microtextures in the zoned feldspars are described in an accompanying paper.CRPG contribution 729     

Trace and Minor Element Chemistry of Alkali Feldspars in the Klokken Layered Syenite Series

Trace and minor element concentrations in alkali feldspars fromthe layered syenite series in the Klokken gabbro-syenite complex,South Greenland have been measured using an ion microprobe.The technique has high precision, avoids problems of mineralseparation, and has allowed investigation of zoning and theeffects of deuteric alteration. Li, Be, B, Cs and Pb occur at< 1 p.p.m. levels, but Ba, Sr, Rb, Mg, Fe, Ti and P are presentin the 1–7000 p.p.m. range. Provided unaltered strain-controlledcrypto- or microperthitic feldspar is sampled and deutericallycoarsened material is avoided, all of the latter elements exceptMg show systematic variation. This confirms the evidence forin situ fractionation of a single syenitic magma-pulse affordedby major element variation in coexisting mafic minerals, particularlypyroxenes and olivines. The distinction between granular andlaminated syenites in the series is supported by the trace elements,which confirms the view that the granular syenites are a condensedroof-chill series with inverted cryptic variation, while thelaminated syenites are a bottom accumulated sequence. Using mainly experimentally determined partition coefficientsfrom the literature and a simple Rayleigh fractionation modelbased on the variation in Ba, Sr and Rb (elements whose concentrationin the melt is largely controlled by feldspar) we calculategeologically reasonable minimum thicknesses for the entire layeredseries of 1370–3260 m. Ti variation in the feldspars wascontrolled by Fe-Ti oxide fractionation, and Fe contents alsowere controlled by mafic phases. Controls of Mg, and particularlythe irregular P levels, are not clear. Behaviour of Ca, whichis present at < 6000 p.p.m. levels in the feldspars, is qualitativelyexplained in terms of the ternary feldspar phase diagram.     

Zoned ternary feldspars in the Klokken intrusion: exsolution microtextures and mechanisms

The microtextures developed during relatively slow cooling as a function of bulk composition in zoned ternary feldspars from syenodiorites and syenites in the Klokken intrusion, described in the preceding paper, were determined by TEM and their origin and evolution deduced. The feldspars normally have a plagioclase core and an alkali feldspar rim; cores become smaller and rims larger and the An content of both decrease with distance from the contact of the intrusion. The following microtextural sequence was observed. The inner plagioclase cores are homogeneous oligoclase-andesine with Albite growth twins only, but are crypto-antiperthitic towards the outer core. At first small platelets of low sanidine a few nanometres thick and up to 10 nm long occur sporadically only on Albite-twin composition planes. With further increase in bulk Or they are homogeneously distributed in the plagioclase. Thicker, through-going plates in platelet-free areas are found, which induce Albite twins in the surrounding plagioclase. The microtextures in the rims are regular cryptomesoperthitic, with (¯601) lenses or lamellae, depending on the bulk Or-content, of low sanidine in Albite-twinned low oligoclase-andesine. Albite and Pericline twins in plagioclase in an M-twin relationship, together with lenticular low sanidine, were found in only one small area. The overall diffraction symmetry of the mesoperthites is monoclinic, showing that exsolution started in a monoclinic feldspar, whereas that of the antiperthites is triclinic. The intermediate zone between the core and rim is more complex and microtextures vary over distances of a few micrometres.The cryptomesoperthites are very regular where Or-rich and probably arose by spinodal decomposition. The platelets in the outer cores arose by heterogeneous nucleation on twin composition planes and by homogeneous nucleation elsewhere. Near the intermediate zone they coarsened to give larger plates which induced Albite-twins in the plagioclase. Because of the zoning, microtextures that were initiated in areas of given composition, can propagate laterally into zones of different composition. A diagram is given showing the relationship between ternary bulk composition and the microtexture developed in coherent perthitic alkali feldspars and plagioclases from slowly-cooled rocks.CRPG contribution 730     

Development of microporosity,diffusion channels and deuteric coarsening in perthitic alkali feldspars

Turbidity is an almost universal feature of alkali feldspars in plutonic rocks and has been investigated by us in alkali feldspars from the Klokken syenite using SEM and TEM. It is caused by the presence of myriads of tubular micro-inclusions, either fluid-filled micropores or sites of previous fluid inclusions, and is associated with coarsening of microperthite and development of sub-grains. Micropores are abundant in coarsened areas, in which porosities may reach 4.5%, but are almost absent from uncoarsened, pristine braind-microperthite areas. The coarsening is patchy, and involves a scale increase of up to 10³ without change in the composition of the phases, low albite and low microcline, or in the bulk composition of the crystal. It occurs abruptly along an irregular front within individual crystals, which retain their original shapes. The coherent braid microperthite gives way across the front to an irregular semi-coherent film perthite over a few m and then to a highly coarsened irregular patch perthite containing numerous small sub-grains on scales of a few hundred nm, in both phases. The coarsening and micropore formation occured at a T400°–450° C and it is inferred to have been driven by the release of coherent strain energy, low-angle grain-boundary migration being favoured by a fluid. The patchy nature of the coarsening and the absence of a relationship with initial grain boundaries suggest that the fluid was of local origin, possibly arising in part through exsolution of water from the feldspar. The sub-grain texture and microporosity modify profoundly the permeability of the rock, and greatly enhance the subsequent reactivity of the feldspars.     

Partial homogenization of cryptoperthites in an ignimbrite cooling unit

A single-crystal x-ray study of alkali feldspars of bulk composition Or_39–43 and Or_62–64, from a single cooling unit of Battleship Rock Tuff, northern New Mexico, reveals a trend of decreasing degree of exsolution, from the non-welded zone toward the densely welded center of the cooling unit. Crystals of bulk composition Or_62–64 range from cryptoperthite with both phases monoclinic in the nonwelded zone to virtually unexsolved crystals in the welded center of the cooling unit. Crystals of bulk composition Or_39–43 include crytpoperthites with both phases monoclinic, and cryptoperthites with Pericline-twinned sodic lamellae, with ^* of the sodic phase increasing systematically from 87.3° in the nonwelded zone to 90° in the densely welded zone. Composition estimates based on unitcell parameters show decreasing compositional differences between coexisting lamellae toward the welded zone. The feldspar crystals studied are interpreted to be xenocrysts, which had undergone exsolution prior to incorporation in the erupting magma, and which were then partially homogenized during emplacement and post-emplacement cooling. The data indicate a maximum re-equilibration temperature of the feldspars of about 500° C, and a more rapid cooling of the tuff than calculated for simple conduction in a uniform slab.     

Cooling process of a granitic body deduced from the extents of exsolution and deuteric sub-solidus reactions: Case study of the Okueyama granitic body, Kyushu, Japan

The variation in cooling processes with depth in a magma body is evaluated quantitatively by analysis of the extent of exsolution coarsening and deuteric coarsening as sub-solidus reactions. This method is applied to evaluation of the Okueyama granitic body of central Kyushu, Japan. Exsolution coarsening has produced microperthite textures in this body, while deuteric coarsening has resulted in patchperthite, myrmekite, and reaction rims, respectively. Through measurement of six textural parameters, including the width and spacing of lamellae and the thickness of myrmekite and reaction rims, the extent of these sub-solidus reactions is shown to increase systematically with depth in the granite body, indicating that the Okueyama cooled gradually from the roof. The hornblende–plagioclase and ternary feldspar thermometers indicate temperature a range of 710 to 620 °C for volume diffusion associated with exsolution coarsening, while deuteric coarsening is found to have occurred at temperatures below 500 °C on the basis of the ternary feldspar thermometer. The cooling period corresponding to exsolution coarsening is estimated using a one-dimensional heat transfer model, yielding periods of 820 y at the roof and 1390–1890 y at the base of the exposure (1000 m below the roof) depending on total depth of the original magma body.     

The nature of potassium feldspar,exsolution microtextures and development of dislocations as a function of composition in perthitic alkali feldspars

Transmission electron microscope data on the morphology of exsolution lamellae, the nature of the potassium feldspar and the development of dislocations at lamellar interfaces in coherent cryptoperthites and fine microperthites are reviewed. Dislocations have been reported previously in only two crystals, and periodic dislocations noted in only one, an Or-rich microperthite. Periodic dislocations (spacing 100–150 nm) are here described from a ternary mesoperthite (Or₂₆ Ab₅₂ An₂₂). Small crystallites (<30 nm) of other phases have sometimes nucleated on the dislocations. The 020 lattice fringes of the feldspar phases have been imaged; the difference in 020 spacings can be almost entirely accommodated by the regular dislocations, so that the boundaries may be termed nearlyperfectly semicoherent.Dislocations have been found so far only in cryptoperthites with lens-shaped or straight lamellae, either in Or-rich feldspars or in Ab-rich ternary ones. In intermediate compositions with wavy or zig-zag albite lamellae, or lozengeshaped albite areas (braid microperthites) dislocations have not been observed. Strain reduction in intermediate compositions occurs by migration of lamellar interfaces from (¯601) to near (¯6¯61) as microcline forms in the diagonal association. In Ab-rich ternary feldspars the relatively high Ancontent blocks interface migration, and strain reduction occurs by nucleation of dislocations; the Or-rich feldspar phase is tweed orthoclase. In Or-rich bulk compositions the low volume of albite exerts insufficient stress to promote microcline formation, and tweed orthoclase develops. Interfaces do not migrate, and dislocations again develop. Fields in which different potassium feldspar polymorphs occur and in which the different exsolution textures are developed are summarized on a ternary diagram.     

Biotite Equilibria and Fluid Circulation in the Klokken Intrusion

Chemical variation in biotite from the KJokken gabbro-syeniteintrusion in the Proterozoic Gardar province in South Greenlandhas been investigated by electron probe and, for F and Li, ionmicroprobe. Most mica occurs in small amounts as fringes onilmenomagnetite or fayalite, rarely as an intercumulus or poikiliticphase. The micas range continuously from Phlog₇₀Ann₃₀ in a gabbro,to Phlog₄Ann₉₆ in the most evolved (slightly persilicic andperalkaline) syenite. In the syenites Fe-Mg partitioning betweenbiotite and olivine can be described by a single distributioncoefficient, K_d = XF X^Biot_Ms/X^Biot_Fe 3, suggesting that thesereactant phases mix ideally at the reaction T. Experimentaldata for Fe-Mg exchange via aqueous chloride solutions (Schulien,1980) imply low T (32Q?C). F was absent in the experiments andmay significantly affect the exchange equilibrium. K_d in thegabbros is 1, consistent with equilibrium via a fluid depletedin F because of crystallization of large amounts of amphibole. Al, Mn, and Ti vary regularly throughout the series and canbe used as markers of cryptic variation in the layered syenites.(Al + Si): 22 O is always in the range 7.7–7.85. A1/(A1+ Si) decreases from 0.31 in gabbros to 0.25 in the most Fe-richmicas. Li is always < 260 ppm w. In the syenite series, Fshows a near-linear inverse relationship with Fe/(Fe + Mg) whichpasses close to OF at Ann₁₀₀ with l.4 wt% F(0–7 F to 44positive charges) at Ann₄₄. Biotites in the gabbro unit (whichforms an outer sheath to the intrusion) have relatively lessF, probably because it was consumed by coexisting amphibole.^I8O is similar for both gabbros and syenites, and it is unlikelythat an envelope fluid was involved in the reactions. G reachesa maximum of 0.3 wt. % in biotite except for that in one syenitesample with 0–7 wt. %. Calculation of relative F-OH fugacitiesfrom the reaction OH-phlogopite + F-annite = F-phlogopite +OH-annite, as calibrated by Munoz (1984), appears to suggestthat each horizon in the layered series was in equilibrium witha slightly different fluid. In view of the intimate interleavingof these lithologies, this is improbable. The equilibrium constantof the exchange reaction, obtained from the experimental data,seems not to be appropriate to the Klokken assemblage, or toother examples of regular F-Fe avoidance. Explanations may includeshort-range Fe-Mg ordering in the natural examples or the effectof additional components in the fluid. F contents are high incomparison with biotites from calc-alkaline complexes; highmagmatic F may account for the igneous layering common in theGardar. Temperatures calculated from reactions involving fayaliteand magnetite show that most biotites grew subsolidus. The F-poorannites grew > 300 ?C subsolidus even when texturally intercumulus.Stable isotope data are consistent with the separation and retentionof a deuteric fluid during the final stage of magmatic crystallization.Klokken was not generally subject to the pervasive, long-range(in both distance and time) dydrothermal interactions demonstratedin calcalkaline and theleiitic intrusions, although more extensivefluid flow is indicated for the more permeable laminated syenites.The biotites preserve chemical variation indicating local equilibriumwith other mafic phases, and halogens provide a useful markerof subsolidus fluid flow.     

Nucleation on perthite-perthite boundaries and exsolution mechanisms in alkali feldspars

A transmission electron microscope study of intracrystalline boundaries between two perthites of markedly different composition in composite crystals, one a tenary mesoperthite (Or₂₆Ab₅₂An₂₂, initially a homogeneous potassian monalbite) the other a more potassic cryptoperthite (Or₆₁Ab₃₃An₆, initially a homogeneous sodian sanidine), shows that the two perthites are in nearly parallel intergrowth. Most boundaries examined were of (hkO) type; (010) boundaries are straight, whereas other (hkO) boundaries are curved or stepped. Exsolution occurred first in the potassian monalbite (mesoperthite) and was unaffected by the boundary. Subsequent exsolution in the sodian sanidine (cryptoperthite) was affected by the boundary, but for up to only a few micrometers. Exsolution occurred by heterogeneous nucleation and growth of oligoclase on and from the intracrystalline boundary. At almost the same time the rest of the volume of sanidine exsolved by spinodal decomposition. 1–2 μm from the boundary in the intervening K-rich matrix of the sodian sanidine, further exsolution occurred by homogeneous nucleation. Time — temperature — transition curves for continuous cooling have been devised to account for the unusual complexity of the exsolution texture. Except in such exceptional circumstances as the example studied, the initial exsolution in high-temperature alkali feldspars of intermediate composition, unlike other minerals, probably does not occur by nucleation, but only by spinodal decomposition.     

The Klokken Gabbro--Syenite Complex, South Greenland: Cryptic Variation and Origin of Inversely Graded Layering

The Klokken stock has an outer sheath of vertically banded alkaligabbro which passes through syenogabbros into unlaminated syenitessurrounding a central layered syenite core. The layered seriescomprises sheets of granular syenite interleaved with drusylaminated syenite which shows repeated inversely graded minerallayering in which normal leucosyenite becomes progressivelymore hedenbergite-rich upwards. Tops to layers are nearly feldspar-free,fayalite—magnetite horizons. Model cumulate textures areexhibited, and cross-bedding and normally graded channel structuresare encountered. Granular layers often rest on ultramafic layersand load structures occur at the junctions. The series is cutby a sheet of biotite-syenodiorite and finally by quartzsyeniteaplites. Plagioclase (An₅₇ in gabbros) is progressively mantled by ternaryalkali feldspar and is absent only in the layered series inwhich alkali feldspars cluster around Ab₆₅ Or₃₅. Pyroxene showsunbroken evolution with respect to Fe/Fe + Mg, from augite toferrohedenbergite and sodie hedenbergite, with acmite in aplites,but Al and Ti decrease suddenly between syenogabbro and unlaminatedsyenite stages. Olivines show parallel evolution but there isa jump in Fe: Mg between gabbros and unlaminated syenites fromFo₅₂ to Fo₃₄, with subsequent steady evolution to Fo₃Te₅Fa₉₂in laminated syenites. Biotites show steady increase in Fe/Fe+ Mg throughout the syenites, approaching pure annite, but thegabbro-syenogabbro series show a reversal, interpreted as indicatingcrystallization under conditions of increasing f_o2. Ilmenomagnetitesshow systematically varying minor element ratios, with a suddendrop in MgAl₂O₄ content between gabbros and syenites. Amphiboles(ferroedenites and sodic hastingsites) appear only in the mostfractionated syenites in the layered series. The Klokken chamber was lined by gabbro with only limited insitu fractionation to syenogabbro. The termination of gabbrocrystallization was marked by ingress of water into the conduit,perhaps causing the liquid to embark on its fractionation trendtowards oversaturated residua. The mineralogical breaks areconsistent with a drop in temperature, and increase in a_SIO2at this point. The chamber was subsequently filled by a drytrachytic liquid which fractionated in situ, initially by congelationagainst the walls as unlaminated syenite. The layered seriesdeveloped within this chamber. The granular syenites show invertedcryptic and phase-layering (downward appearance of amphibole),and downward increase in grain size. They are sheets stopedoff the roof of the intrusion, representing chilled equivsalentsof the fractionating trachytic liquid of which the laminatedsyenites are the bottom accumulated products. In the inverselygraded layers the cumulus phases have slightly more advancedcompositions than the same phases in either normal rock or whenpresent as an intercumulus phase in adjacent parts of layers.The layering is attributed to crystallization at varying degreesof undercooling in a magma in which all phases exhibit a narrowcrystallization interval and which was subject to rhythmic P-build-upand sudden release. Crystal accumulation took place under nearstagnant to rhythmie P-build-up and sudden release. Crystalaccumulation took place under near stagnant conditions in athin chamber immediately beneath the roof of the intrusion.     

Sm−Nd dating of multiple garnet growth events in an arc-continent collision zone,northwestern U.S. Cordillera

Integrated petrologic and Sm–Nd isotopic studies in garnet amphibolites along the Salmon River suture zone, western Idaho, delineate two periods of amphibolite grade metamorphism separated by at least 16 million years. In one amphibolite,P–T studies indicate a single stage of metamorphism with final equilibration at 600°C and 8–9 kbar. The Sm–Nd isotopic compositions of plagioclase, apatite, hornblende, and garnet define a precise, 8-point isochron of 128±3 Ma (MSWD=1.2) interpreted as mineral growth at the metamorphic peak. A⁴⁰Ar/³⁹Ar age for this hornblende indicates cooling through 525°C at 119±2 Ma. In a nearby amphibolite, garnets with a two-stage growth history consist of inclusion-rich cores surrounded by discontinuous, inclusion-free overgrowths. Temporal constraints for core and overgrowth development were derived from Sm–Nd garnet — whole rock pairs in which the garnet fractions consist of varying proportions of inclusion-free to inclusion-bearing fragments. Three garnet fractions with apparent ages of 144, 141, and 136 Ma are thought to represent mixtures between late Jurassic (pre-144 Ma) inherited radiogenic components preserved within garnet cores and early Cretaceous (128 Ma) garnet overgrowths. These observations confirm the resilience of garnet to diffusive exchange of trace elements during polymetamorphism at amphibolite facies conditions. Our geochronologic results show that metamorphism of arc-derived rocks in western Idaho was episodic and significantly older than in arc rocks along the eastern margin of the Wrangellian Superterrane in British Columbia and Alaska. The pre-144 Ma event may be an expression of the late Jurassic amalgamation of marginal oceanic arc-related terranes (e.g., Olds Ferry, Baker, Wallowa) during the initial phases of their collision with North American rocks. Peak metamorphism at 128 Ma reflects tectonic burial along the leading edge of the Wallowa arc terrane during its final penetration and suturing to cratonic North America.     

The alteration of olivine in basaltic and associated lavas

Olivines which cool under oxidizing conditions exsolve iron oxides at high temperature, and at low temperatures break down to essentially chloritic materials. Olivines which cool under non-oxidizing conditions alter at intermediate temperatures to complex assemblages of chlorite and interstratified phyllosilicates containing a smectite. Alteration under oxidizing conditions at low temperature, probably below 140° C, produces iddingsite, an orientated assemblage of goethite and interstratified phyllosilicates also containing a smectite.Post-deuteric alteration extends or initiates the breakdown of olivine to phyllosilicate mixtures often resulting in widespread movement of materials through the flow. Post-deuteric alteration of iddingsite produces strongly pleochroic, highly ordered varieties which eventually break down to green phyllosilicate assemblages.Weathering of olivine may produce orientated assemblages similar to deuteric forms of iddingsite. Weathering of green alteration products results in oxidation and the liberation of discrete iron hydroxides.Alteration in all cases requires exchange of material between interstitial components and olivine. During deuteric alteration, plagioclase and pyroxene are usually unaffected.     

The Feldspathic Syenites of the Loch Ailsh Intrusion, Assynt, Scotland

The highly feldspathic syenites are divided on field evidenceinto three phases, S1, S2, and S3. S3 overlies S1 and S2, andencloses xenoliths and xenocrysts from the underlying and earliersyenites. The feldspars show a range of forms and coarseness of perthite,and show enrichment in potassium from Or₂₀ to Or₄₀, and increasein 2V, in successive syenites. X-ray diffractometer and single-crystalmethods show that SI has a dominantly monoclinic K-phase, S2a microcline with some monoclinic material, and S3 a highlyoblique microcline, exclusively. The earlier feldspars preservethese features even when enclosed by later S3 material. Thereis a suggestion of systematic regional distribution of obliquityvalues. The compositional range of the feldspars is related to the experimentallydetermined minimum in the Ab-Or-An system. The feldspars crystallizedas sodium-rich sanidines. Temperature-pressure limits are deduced,and the unmixing history of the perthites discussed. A mechanismfor the preservation of orthoclase in S1 and S2 is suggested.     

Exsolution and coarsening in iron-free clinopyroxene during isothermal annealing

Exsolution and coarsening in Fe-free clinopyroxene of composition En_45.1Di_54.9 was studied at temperatures of 1300, 1200, and 1100°C, and annealing times between 10 min and 4320 h. Based on the wavelength of the exsolution lamellae, the exsolution process can be divided into exsolution sensu stricto and coarsening. During exsolution sensu stricto, the average wavelength of the “001” and “100” lamellae remains constant in contrast to the subsequent coarsening process. A progressive development of the microstructure is observed before coarsening which includes formation of island-like regions of dark contrast without phase separation, occurrence of pigeonite and diopside domains predominantly elongated along (100), formation of “100” exsolution lamellae, dissolution of the “100” lamellae, and predominant occurrence of “001” exsolution lamellae. All observations are in accordance with the exsolution mechanism of spinodal decomposition. After termination of the exsolution process, coarsening of the exsolution lamellae is observed. The “001” lamellae coarsen according to the rate law

     

Nepheline-bearing rocks from the Poohbah Lake complex,Ontario: Malignites and Malignites

Malignites from the Poohbah Lake complex of northwestern Ontario, Canada are melanocratic cumulates. Cumulus pyroxene and apatite are poikilitically enclosed in a groundmass of large plates of intercumulus orthoclase and nepheline. Nepheline-feldspar fingerprint-like intergrowths occur. Nephelines are commonly zeolitized and pyroxenes altered to aggregates of biotite and/or garnet by deuteric alteration. Pyroxenes are weakly zoned from Di₇₁ Hd¹⁸Ac₁₁ to Di₆₃Hd₂₂Ac₁₅, and are similar to the least evolved pyroxenes of other alkaline rocks. Nephelines all have compositions within the Morozewicz-Buerger convergence field and feldspars have a limited compositional range from Or⁸⁸ to Or⁹⁵. Perthites are absent.Inconsistancies in the usage of the terms malignite and juvite are discussed and it is considered that a non-genetic petrographic classification of nepheline syenites leads to the obscuration of a group of potassic nepheline syenites, characterized by the presence of nepheline plus orthoclase which are typically associated with saturated to over-saturated alkaline rocks, contain pseudo-leucite or nepheline-orthoclase intergrowths, are emplaced in mobile belts and are not associated with rocks of the ijolite-carbonatite suite.A genetic classification of nepheline syenites is suggested and it is proposed that; (1) mafic-rich nepheline syenites be referred to as mela-nepheline syenites (sensu lato) rather than as malignites; (2) the term malignite be used for magmatic potassic nepheline syenites characterised by the presence of nepheline plus a single potassium-rich feldspar (orthoclase or microcline) and devoid of exsolution perthite under subsolvus conditions; (3) the metasomatic malignites and juvites of ijolite-carbonatite complexes be referred to as varieties of fenites.     

Origin and differentiation of picritic arc magmas,Ambae (Aoba), Vanuatu

Key aspects of magma generation and magma evolution in subduction zones are addressed in a study of Ambae (Aoba) volcano, Vanuatu. Two major lava suites (a low-Ti suite and high-Ti suite) are recognised on the basis of phenocryst mineralogy, geochemistry, and stratigraphy. Phenocryst assemblages in the more primitive low-Ti suite are dominated by magnesian olivine (mg 80 to 93.4) and clinopyroxene (mg 80 to 92), and include accessory Cr-rich spinel (cr 50 to 84). Calcic plagioclase and titanomagnetite are important additional phenocryst phases in the high-Ti suite lavas and the most evolved low-Ti suite lavas. The low-Ti suite lavas span a continuous compositional range, from picritic (up to 20 wt% MgO) to high-alumina basalts (<5 wt% MgO), and are consistent with differentiation involving observed phenocrysts. Melt compositions (aphyric lavas and groundmasses) in the low-Ti suite form a liquid-line of descent which corresponds with the petrographically-determined order of crystallisation: olivine + Cr-spinel, followed by clinopyroxene + olivine + titanomagnetite, and then plagioclase + clinopyroxene + olivine + titanomagnetite. A primary melt for the low-Ti suite has been estimated by correcting the most magnesian melt composition (an aphyric lava with 10.5 wt% MgO) for crystal fractionation, at the oxidising conditions determined from olivine-spinel pairs (fo₂ FMQ + 2.5 log units), until in equilibrium with the most magnesian olivine phenocrysts. The resultant composition has 15 wt% MgO and an mg ^Fe2 value of 81. It requires deep (3 GPa) melting of the peridotitic mantle wedge at a potential temperature consistent with current estimates for the convecting upper mantle (T _p 1300°C). At least three geochemically-distinct source components are necessary to account for geochemical differences between, and geochemical heterogeneity within, the major lava suites. Two components, one LILE-rich and the other LILE- and LREE-rich, may both derive from the subducting ocean crust, possibly as an aqueous fluid and a silicate melt respeetively. A third component is attributed to either differnt degrees of melting, or extents of incompatible-element depletion, of the peridotitic mantle wedge.