| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Petrology Volume 41 Number 5 Pages 627-649 2000
© Oxford University Press 2000
The Plagioclase–Magma Density Paradox Re-examined and the Crystallization of Proterozoic Anorthosites
DEPARTMENT OF EARTH AND ENVIRONMENTAL SCIENCES (DSTE), UNIVERSITÉ LIBRE DE BRUXELLES, CP160/02, AVENUE F. D. ROOSEVELT 50, B-1050, BRUSSELS, BELGIUM
Received February 4, 1999; Revised typescript accepted October 5, 1999
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGEOLOGIC SETTING OF THE...BLOCKS IN THE POE...DENSITY CALCULATIONSBLOCK, PLAGIOCLASE AND MAGMA...CONCLUSIONSREFERENCES |
|---|
Intermediate-composition plagioclase (An40–60) is typically less dense than the relatively evolved basaltic magmas from which it crystallizes and the crystallization of plagioclase produces a dense residual liquid, thus plagioclase should have a tendency to float in these magmatic systems. There is, however, little direct evidence for plagioclase flotation cumulates either in layered intrusions or in Proterozoic anorthosite complexes. The layered series of the Poe Mountain anorthosite, southeast Wyoming, contains numerous anorthosite–leucogabbro blocks that constrain density relations during differentiation. All blocks are more mafic than their hosting anorthositic cumulates, their plagioclase compositions are more calcic, and each block is in strong Sr isotopic disequilibrium with its host cumulate. Associated structures—disrupted and deformed layering—indicate that (1) a floor was present during crystallization and that plagioclase was accumulating and/or crystallizing on the floor, (2) compositional layering and plagioclase lamination formed directly at the magma–crystal pile interface, and (3) the upper portions of the crystal pile contained significant amounts of interstitial melt. Liquid densities are calculated for proposed high-Al olivine gabbroic parental magmas and Fe-enriched ferrodioritic and monzodioritic residual magmas of the anorthosites taking into account pressure, oxygen fugacity, P2O5, estimated volatile contents, and variable temperatures of crystallization. For all reasonable conditions, calculated block densities are greater than those of the associated melt. The liquid densities, however, are greater than those for An40–60 plagioclase, which cannot have settled to the floor. Plagioclase must either have been carried to the floor in relatively dense packets of cooled liquid plus crystals or have crystallized in situ. A sloping floor, possibly produced by diapiric ascent of relatively light plagioclase-rich cumulates, is required to allow for draining and removal of the dense interstitial liquid produced in the crystal pile and may be a characteristic feature during the crystallization of many Proterozoic anorthosites and layered intrusions.
KEY WORDS: magma; density; Proterozoic anorthosites; blocks; plagioclase
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGIC SETTING OF THE...BLOCKS IN THE POE...DENSITY CALCULATIONSBLOCK, PLAGIOCLASE AND MAGMA...CONCLUSIONSREFERENCES |
|---|
During the low-pressure crystallization of basaltic magma, the densities of the common ferromagnesian silicates that crystallize—olivine, orthopyroxene, clinopyroxene, pigeonite—increase progressively with decreasing mg-number (Fig. 1), allowing for relatively easy separation of crystals and melt. The density of plagioclase, however, decreases progressively with fractionation, such that plagioclase of intermediate composition (An40–60) is less dense than the relatively evolved basaltic magma from which it crystallizes (Fig. 1). The crystallization of plagioclase alone results in depletion of the relatively light oxide components SiO2, Al2O3, CaO and Na2O, and enrichment in the heavier components TiO2, FeO* and MgO, resulting in progressively denser residual magmas. The combination of these processes implies that intermediate-composition plagioclase should have a strong tendency to float in evolved basaltic magmas. However, there is abundant evidence for the crystallization of plagioclase on the floor of magma chambers—layering, lamination, scours—and very little demonstrable evidence for plagioclase flotation as an effective mechanism of differentiation. This contradiction is known as the plagioclase–magma density paradox (Morse, 1973
) and has resulted in a major re-evaluation of how layered intrusions crystallize (Campbell, 1978; Campbell et al., 1978; McBirney & Noyes, 1979; Chen & Turner, 1980; Irvine et al., 1983; Huppert & Sparks, 1984), and is the source of continued discussion (McBirney, 1985; Morse, 1986a; Irvine, 1987). The plagioclase–magma density paradox has major implications for the formation of anorthosite horizons in layered intrusions (e.g. Stillwater intrusion, Montana: Raedeke & McCallum, 1980; Salpas et al., 1983; Czamanske & Bohlen, 1990; Haskin & Salpas, 1992) and especially for Proterozoic anorthosite complexes.
|
Much of the debate over where plagioclase crystallizes in basaltic magma chambers has focused on layered mafic intrusions, in particular the Skaergaard intrusion (e.g. McBirney & Noyes, 1979; Irvine, 1987; McBirney, 1995; Irvine et al., 1998), which are typically multi-saturated in ferromagnesian silicates and plagioclase ± Fe–Ti oxides. Proterozoic anorthosites represent the worst-case scenario for the plagioclase–magma density paradox as they consist of enormous volumes (1000s to 10 000s of km3) of cumulate anorthosite–leuconorite–leucotroctolite (75–95% plagioclase of intermediate composition, An40–60), associated with minor intrusions of troctolite and Fe-enriched dioritic rocks, and large granitic batholiths (Morse, 1982; Emslie, 1985; Wiebe, 1992; Ashwal, 1993). Each complex contains a number of different plagioclase-rich plutons that span a structural range from massive to layered to diapiric (Wiebe, 1992), the last being characterized by strongly deformed marginal zones. Individual plutons typically show little variation in plagioclase compositions, although extensive Fe–Mg fractionation in the ferromagnesian silicates may be present (Wiebe, 1992; Scoates, 1994). Individual plagioclase grains commonly show only minor zoning, although prominent oscillatory zoning may be found locally (Wiebe, 1992; see also the cover of the Journal of Petrology for 1998).
If plagioclase were to float in evolved basaltic magmas, then Proterozoic anorthosites should be the ideal type-example of this process. This presumption has been exploited by numerous workers and forms the basis for the general model for the formation of Proterozoic anorthosites involving polybaric crystallization of mantle-derived basaltic magma (Morse, 1968; Emslie, 1985; Longhi & Ashwal, 1985; Wiebe, 1992; Ashwal, 1993; Longhi et al., 1993). Fractional crystallization in chambers at the base of the crust produces ultramafic cumulates on the floor and relatively evolved resident magma. Plagioclase eventually saturates and, because of the enhanced density contrast between basaltic magma and plagioclase at higher pressures (Kushiro, 1980), floats to the top of the chambers, where it may be locally remelted as a result of periodic replenishment of hotter, less evolved magma, thus enriching the interstitial liquid in plagioclase components (Wiebe, 1992). Buoyancy-driven ascent of plagioclase-rich diapirs (50–70 vol. %) into the crust is then proposed for the formation of individual plutons (Emslie, 1985; Longhi & Ashwal, 1985; Longhi et al., 1993) that consolidate essentially as flotation cumulates with progressive crystallization of the interstitial melt and removal of the dense Fe–Ti–P-rich components. Unambiguous evidence of flotation cumulates, however, has been difficult to identify in the field. Some layered leucotroctolites grade upwards into massive, undeformed leuconorite to anorthosite that could be interpreted as stagnant accumulations of suspended plagioclase beneath the roofs of magma chambers (Emslie, 1970; Wiebe, 1992). In contrast, numerous layered anorthosites, leuconorites and troctolites from many Proterozoic anorthosite complexes show abundant evidence for bottom accumulation of plagioclase (Kiglapait, Nain: Morse, 1969, 1979; Harp Lake: Emslie, 1980; Paul Island, Nain: Wiebe, 1990; Bjerkreim–Sokndal, Rogaland: Wilson et al., 1996).
Resolution of the plagioclase–magma density paradox is clearly critical to better understanding the magmatic evolution of Proterozoic anorthosites, and will certainly have applications to many layered mafic intrusions where plagioclase is an important cumulus mineral. In this paper, I present new field observations and geochemistry from leucogabbroic to anorthositic blocks in the layered series of the Poe Mountain anorthosite, southeastern Wyoming. The presence of deformational structures beneath these blocks demonstrates the existence of a floor to the magma chamber where plagioclase was accumulating and crystallizing. Layering and plagioclase lamination formed directly at the magma–crystal pile interface. Density calculations involving appropriate parental and residual magma compositions are consistent with observed block impact structures, but clearly indicate that intermediate-composition plagioclase was not capable of sinking in any of the associated magmas; thus alternative mechanisms must be sought to explain bottom accumulation. The density contrast between plagioclase and dense residual magma, which becomes even greater as crystallization of intermediate-composition plagioclase progresses, poses a major problem for the formation of ‘pure’ anorthositic cumulates—sloping floors are required for drainage and removal of the interstitial liquid (Morse, 1986a, 1988). Diapiric ascent of hot, relatively light, consolidated anorthosite deeper in the crystal pile is responsible for progressive rotation of the magma chamber floor.
|
GEOLOGIC SETTING OF THE POE MOUNTAIN ANORTHOSITE |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGIC SETTING OF THE...BLOCKS IN THE POE...DENSITY CALCULATIONSBLOCK, PLAGIOCLASE AND MAGMA...CONCLUSIONSREFERENCES |
|---|
The Poe Mountain anorthosite is one of three composite intrusions dominated by anorthositic cumulates in the unmetamorphosed 725 km2 Laramie anorthosite complex (LAC), southeast Wyoming (Fig. 2). The LAC is the larger of two Proterozoic anorthosite complexes exposed in the core of one of the Late Cretaceous to early Jurassic Laramide uplifts, the Laramie Mountains, that form the present-day Rocky Mountains in the western USA. The 1·43 Ga LAC (Scoates & Chamberlain, 1995
), and the earlier 1·76 Ga Horse Creek anorthosite complex to the south (Scoates & Chamberlain, 1997), intruded along the Cheyenne belt, the 
1·78 Ga collisional zone that separates Archaean rocks of the Wyoming Province to the north from accreted Proterozoic island arc terranes to the south (Karlstrom & Houston, 1984) (Fig. 2). Only a portion of the LAC is exposed at present; west-dipping Laramide thrust and high-angle reverse faults truncate the eastern margins, and shallowly dipping early Palaeozoic sediments unconformably overlie the western margins. The LAC consists of a central mass of anorthositic rocks (550 km2) intruded by dioritic and troctolitic rocks of the Strong Creek complex (Mitchell et al., 1995, 1996), monzonitic rocks of the Maloin Ranch pluton (Kolker & Lindsley, 1989; Kolker et al., 1990, 1991), the Sybille intrusion (Fuhrman et al., 1988; Frost & Touret, 1989; Scoates et al., 1996) and the Red Mountain pluton (Anderson, 1995), and granitic rocks of the Sherman batholith (Geist et al., 1989; Edwards, 1993) (Fig. 2). Emplacement pressures for the monzonitic rocks were 3 kbar in the north (Anderson et al., 1987; Fuhrman et al., 1988) and 4 kbar in the south (Kolker & Lindsley, 1989) (Fig. 2).
|
The Poe Mountain anorthosite, exposed over 200 km2 in the northern LAC (Fig. 3), is structurally a north-plunging antiform. The core of the antiform is occupied by pervasively recrystallized anorthosites containing few relict magmatic features. Along the western and northern margins of the intrusion, the core anorthosites grade outward into a 5–7 km thick marginal layered series (Fig. 3) that displays abundant igneous layering and a prominent, pervasive plagioclase lamination. The layered series is composed of two distinct zones, the lower or inner anorthositic layered zone (ALZ) with shallow to moderate dips of 30–60° and the upper or outer leucogabbroic layered zone (LLZ) with moderate to steep dips of 60–90°. Both layered zones contain distinct, laterally continuous layered sections (lower, middle and upper in the ALZ, and lower and upper in the LLZ) that are defined by changes in mineral assemblages, layer characteristics, mineral compositional variation, plagioclase abundance and Sr isotopic variation (Scoates, 1994; Scoates & Frost, 1996) (Fig. 4). The uppermost levels of the LLZ are not preserved, because of subsequent emplacement of monzonitic rocks of the 1·43 Ga Sybille intrusion (Scoates et al., 1996). Anorthosite xenoliths, including both strongly laminated and pervasively recrystallized types, occur within the Sybille intrusion. The southern limit of the Poe Mountain anorthosite is more difficult to define as the majority of anorthosites in the central part of the LAC are pervasively recrystallized and primary magmatic structures are rare. The southernmost extension of the LLZ is clearly truncated by troctolitic rocks that may be related to the larger Strong Creek troctolite to the south (Mitchell, 1993). Recent mapping has shown that the upper and middle ALZ in the same area are cut by coarse-grained megacrystic anorthosite, similar to that of the Snow Creek anorthosite (D. H. Lindsley & B. R. Frost, personal communication, 1998). Where recrystallization is less pronounced, the central core area of the Poe Mountain anorthosite is characterized by megacrystic anorthosite (iridescent and strongly zoned) of the Snow Creek anorthosite, more calcic (>An50) than plagioclase from the core of the Poe Mountain anorthosite (An46).
|
|
|
BLOCKS IN THE POE MOUNTAIN ANORTHOSITE |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGIC SETTING OF THE...BLOCKS IN THE POE...DENSITY CALCULATIONSBLOCK, PLAGIOCLASE AND MAGMA...CONCLUSIONSREFERENCES |
|---|
Blocks of igneous origin occur throughout the layered series of the Poe Mountain anorthosite—the major localities are noted in Fig. 3—and range in composition from anorthosite to olivine leucogabbronorite (Table 1). It is likely that a significant number of additional blocks occur, but unfavourable exposures (sections perpendicular to layering show the best relations) and locally extensive weathering limit the number of occurrences that can confidently be identified. A number of blocks have been identified in outcrop along Wyoming State Highway 34 (WY34) and correspond to nearly the same relative stratigraphic position in the Poe Mountain anorthosite (-2400 to -2240 m) near the upper contact of the lower ALZ with the middle ALZ. A single block of 3 m width located in the middle ALZ to the south of the Sybille Fe–Ti oxide deposit contains very coarse-grained plagioclase (50–100 cm) and high-Al clinopyroxene megacrysts, and will be discussed separately in more detail below.
|
Outcrop relations between blocks and layered anorthosites
The blocks are typically elongate and sub-rounded with the longest dimension ranging from 1 to 50 m (Fig. 5). Where the exposures are perpendicular to compositional layering in the anorthosites, an asymmetric set of structures in the hosting anorthositic cumulates consistently envelopes each block (Figs 6 and 7). Stratigraphically below the blocks, anorthosite typically shows strongly disrupted or deformed igneous layering. The prominent plagioclase lamination that parallels compositional layering in the undisturbed sections of the Poe Mountain anorthosite wraps around individual blocks (Figs 6 and 7). In addition, several localities contain abundant irregularly distributed mafic pegmatoids within the zone of deformation–disruption. Stratigraphically above each block, compositional layering and plagioclase lamination are coplanar or drape slightly over individual blocks (Fig. 6). There is no structural evidence of disruption above the blocks. There is also no evidence that these blocks represent the products of late-stage metasomatism as has been proposed for many of the anorthosite inclusions in the Skaergaard intrusion (McBirney, 1996; Sonnenthal & McBirney, 1998)—contacts are sharp, structures cannot be traced through the blocks, and the bulk composition of the blocks is more mafic than that of their host cumulates (Figs 6 and 7).
|
|
|
I interpret the structural and textural features to record the impact of settled blocks onto a pre-existing floor of plagioclase cumulates containing interstitial melt, similar to those described in the Duke Island and Skaergaard intrusions (Irvine, 1987; Irvine et al., 1998). The disrupted layering and curvilinear plagioclase lamination may be attributed to the force of impact. The presence of mafic pegmatoids in the zones of disruption probably reflects the migration of interstitial liquid produced during compaction of the plagioclase-rich crystal pile by the blocks. Small block fragments associated with larger blocks suggest that some may have broken apart on impact. Many blocks contain a structural fabric defined by the preferred orientation of ferromagnesian silicates and this foliation is in each case oblique to the general foliation (compositional layering plus plagioclase lamination) of the enveloping anorthositic cumulates. After impact, progressive accumulation–crystallization of plagioclase on the chamber floor continued, resulting in the coherent, planar compositional layering and lamination that drapes over the tops of the blocks.
The outcrop relations have important implications for the crystallization of plagioclase-rich cumulates in the Poe Mountain anorthosite:
- a floor to the Poe Mountain anorthosite magma chamber was present during crystallization. The presence of this interface between the anorthositic parent magma and the crystal pile indicates that intermediate-composition plagioclase (An45–55) was capable of accumulating on the floor of the chamber and forming thick piles of plagioclase cumulates.
- Compositional layering and plagioclase lamination formed directly at the magma–crystal pile interface, as both planar features are extensively disrupted or deformed by the impact of the blocks.
- The upper portions (5 m) of the crystal pile must have contained, at least locally, significant proportions of interstitial liquid whereby compaction of the plagioclase network occurred and interstitial liquid was mobilized to form mafic pegmatoids.
- The consistent orientation of the asymmetric deformation structures beneath the blocks throughout the Poe Mountain anorthosite indicates that the floor (magma–crystal pile interface) also maintained a consistent orientation and that crystallization proceeded from the inner portions of the ALZ towards the margins.
Blocks containing high-Al clinopyroxene megacrysts
High-Al clinopyroxene megacrysts, characterized by fine exsolution lamellae of calcic plagioclase, have been found at two localities in the Poe Mountain anorthosite, and in each case, the megacrysts occur within blocks. The first locality is in the middle ALZ (Fig. 3), on the road leading up to the Sybille Fe–Ti oxide deposit (Bolsover, 1986; Frost & Simons, 1991). Large clinopyroxene megacrysts (20 cm across) with visible plagioclase exsolution lamellae are associated with very coarse-grained plagioclase (individual crystals exceed 1 m in diameter), olivine and Fe–Ti oxides. The contacts with the surrounding layered and laminated anorthositic cumulates cannot be observed, but given the obvious difference in grain-size, this occurrence is interpreted as a block. Coarse-grained mafic pegmatoids do occur in the Poe Mountain anorthosite, although they are readily identifiable by the presence of fayalitic olivine and abundant apatite (Scoates, 1994). Much smaller clinopyroxene megacrysts, 1–2 cm diameter, with visible plagioclase exsolution lamellae also occur in the large central block in Fig. 6.
Geochemistry of the blocks
Six blocks in the Poe Mountain anorthosite were analysed for major, trace and rare earth element (REE) concentrations to compare with the observed compositional variation in the composite stratigraphic section of the Poe Mountain anorthosite (Scoates, 1994) and for use in density calculations (Table 1; see also sample locations in Fig. 3). Four of the samples were collected using a geological hammer (PM466, PM555, SR248 and SR246) and the remaining two samples (GR289 and GR258) were retrieved with a diamond drill coring device to ensure fresh material (5–10 individual cores, 2·5 cm x 20 cm long). In addition, a single 4 cm x 10 cm high-Al clinopyroxene megacryst from the coarse-grained block south of the Sybille Fe–Ti oxide deposit was analysed. Tungsten carbide (WC) sandpaper was used to remove potential contaminants from the core tubing on each core. Coarse-crushing was done using a WC-plated hydraulic press, and a single homogenized aliquot (75–100 g) of the finely crushed material from each sample was powdered in a WC shatterbox for 3 min. Major element compositions were determined using X-ray fluorescence spectrometry (XRF) at XRAL Laboratories, Toronto, Canada, and standardized using University of Wyoming standards and replicates (Scoates, 1994). Trace element and REE concentrations were determined by inductively coupled plasma mass spectrometry at the University of Nebraska (analysts A. Kolker, M. Ghazi and J. S. Scoates). Most of the blocks were also analysed for Sr and Nd isotopic compositions (Table 1). Dissolution procedures, analytical techniques and associated errors have been described by Scoates, (1994) and Scoates & Frost (1996).
The analysed blocks range in cation normative plagioclase content (An + Ab + Or) from 72 to 91% (Table 1; Fig. 4b). The majority of the blocks contains less total normative plagioclase than the anorthositic cumulates that they are contained within. Major and trace element compositions correspond to the range of compositions observed in the Poe Mountain anorthosite: An* = 55–49, mg-number = 0·52–0·38, Sr = 676–874 and Ba = 338–610, where An* = [An/(An +Ab + Or)] x 100 and mg-number = Mg/(Mg + Fe2+) (Table 1; Scoates, 1994). There is no apparent correlation of composition with stratigraphic position. However, each block contains higher An* than the hosting cumulate (Fig. 8a) suggesting that they are not locally derived. The high-Al clinopyroxene megacryst is compositionally distinct. It contains bleb-like exsolutions of calcic plagioclase and olivine, 6·2 wt % Al2O3 and 510 ppm Cr with mg-number of 0·68 (Table 1). In contrast, oikocrystic clinopyroxene in the ALZ typically contains 1–2 wt % Al2O3 with mg-number in the range 0·64–0·38 (J. S. Scoates, unpublished data, 1994).
|
The blocks show a restricted range of REE concentrations with total REE = 20–25 ppm (Fig. 8b), and uniformly high positive Eu anomalies (Eu/Eu* = 5·1–11·3) similar to samples from the lower ALZ and the recrystallized core of the Poe Mountain anorthosite (Table 1). The one exception is SR246, a very large (>20 m diameter as exposed in a roadcut) block in the upper LLZ, which is strongly enriched in total REE (76 ppm) and has a correspondingly lower positive Eu anomaly (1·8) (Fig. 8b). With 0·29 wt % P2O5, SR246 probably contains a much greater proportion of components crystallized from the interstitial melt than do the other blocks (0·04–0·09 wt % P2O5), thus accounting for the elevated REE abundances. The high-Al clinopyroxene megacryst also has unique REE characteristics. It has relatively high concentrations (total REE = 71 ppm), prominent La and Ce depletion relative to the rest of the light REE (LREE), overall heavy REE (HREE) depletion and no Eu anomaly (Fig. 8b).
The initial Sr and Nd isotopic compositions (calculated at 1434 Ma) of the analysed blocks indicate that they crystallized from magmas distinctive in composition from those that formed the layered series of the Poe Mountain anorthosite (Scoates & Frost, 1996). Their isotopic compositions are relatively restricted compared with those of the layered cumulates, initial 87Sr/86Sr = 0·7043–0·7046 and initial 
Nd = -1·5 to -2·6, but each block displays marked Sr isotopic contrasts with its host cumulate or equivalent relative stratigraphic position (RSP); 
ISr = 0·0002–0·0007 (Fig. 8c). The three blocks from the ALZ where Sr isotopic compositions are available are significantly less radiogenic than the surrounding cumulates. The range in initial Nd for the blocks is limited (-1·5 to -2·6) and within the range for the majority of the surrounding cumulates, although GR289 does show Nd isotopic disequilibrium outside analytical error (Fig. 8d). The isotopic composition of the high-Al clinopyroxene megacryst, ISr = 0·7041 and Nd = -0·2, also demonstrates that it is not in isotopic equilibrium with the equivalent RSP host cumulate (ISr 0·7051 and Nd -2·0) or any other rock in the ALZ (Fig. 8c and d).
The combined An* and ISr disequilibrium shown by the blocks and the high-Al clinopyroxene megacryst indicate that they cannot represent fragments of the layered series that were transported to the chamber floor, unless different parts of the chamber were crystallizing from magmas of different isotopic compositions that are recorded only in the block compositions. The observed isotopic disequilibrium between the high-Al clinopyroxene megacryst and the surrounding ALZ cumulates is also consistent with a high-pressure origin for this megacryst and, by inference, for the block within which it resides (Scoates & Frost, 1996).
Age relations between blocks and layered anorthosites
Block compositions (major, trace and rare earth element, and isotopic) are similar to those of the Poe Mountain anorthosite in general, thus they either represent fragments of a now-eroded roof zone to the same chamber, possibly formed by disruption during magma replenishment, or perhaps a distinct earlier phase of anorthositic magmatism. This can be evaluated, as the ages of the Poe Mountain anorthositic cumulates and one of the blocks are known precisely (Scoates & Chamberlain, 1995). Concordant to very slightly discordant (<1% discordance) U–Pb isotopic data from zircon and baddeleyite from three samples—a laminated oikocrystic anorthosite in the ALZ (xenolith PM466 is found in this outcrop), a strongly layered and laminated olivine leucogabbro in the LLZ and a cross-cutting oxide-rich leucotroctolite in the ALZ—give identical weighted average 207Pb/206Pb ages of 1434 ± 1 Ma. Zircon and baddeleyite were also separated and analysed from the leucogabbroic block SR246 and yield a U–Pb age of 1435·4 ± 0·5 Ma, statistically older than the LLZ cumulates it resides in. At least one of the blocks, and perhaps some of the others, represent earlier phases of anorthositic magmatism, and some may have crystallized at depth, as is supported by the presence of high-Al clinopyroxene megacrysts. Similar Sr and Nd isotopic compositions indicate that they probably crystallized from magmas derived from the same source and/or with similar contamination histories.
|
DENSITY CALCULATIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGIC SETTING OF THE...BLOCKS IN THE POE...DENSITY CALCULATIONSBLOCK, PLAGIOCLASE AND MAGMA...CONCLUSIONSREFERENCES |
|---|
The presence of deformational structures in layered and laminated anorthositic cumulates beneath the blocks in the Poe Mountain anorthosite is clear evidence that the blocks were denser than the resident magmas and that plagioclase was accumulating on the chamber floor. The density of silicate liquids can be calculated from the major element oxide composition of a whole rock if the partial molar volume of the major element oxides and thermal expansion and compressibility coefficients are known (Lange & Carmichael, 1990
; Lange, 1994). Calculation of magma–block density variations during evolution of the Poe Mountain anorthosite magmatic system requires the identification of appropriate candidates for parental and residual liquids, and it must be demonstrated or assumed that the major element compositions of the whole rocks represent those of silicate liquids—they must represent magma compositions without accumulated crystals. Calculated magma densities are also strongly dependent on relative oxygen fugacity conditions during crystallization, the P2O5 content of the magmas, as well as their predicted volatile contents (H2O and CO2), as discussed below.
Parental and residual magma compositions to the Poe Mountain anorthosites
One of the major goals in the study of Proterozoic anorthosites is the determination of appropriate parental and residual magma compositions. This matter is complicated by the fact that (1) the anorthosites themselves represent cumulate rocks, typically extreme adcumulates with little remaining evidence of interstitial melt, and this makes simple geochemical inversion difficult (e.g. Haskin & Salpas, 1992; McBirney & Hunter, 1995; Meurer & Boudreau, 1998), and (2) fine-grained marginal gabbros, or dykes injected into surrounding country rock, that could preserve magma compositions are rare in Proterozoic anorthosite complexes. However, mafic dykes, covering a wide range of compositions from high-Al gabbro to ferrodiorite to monzodiorite, occur throughout nearly every examined complex. The LAC contains hundreds of fine-grained mafic dykes that have been the focus of several recent petrographic, geochemical and isotopic studies (Mitchell et al., 1995, 1996). Their compositions are consistent with magmas similar to those that the anorthosites may have crystallized from (high-Al olivine gabbros) or magma compositions expected to result from the removal of large quantities of intermediate-composition plagioclase (ferrodiorites and olivine ferrodiorites) and their subsequent fractionates (monzodiorites). The chemistry of some of the samples considered by Mitchell et al. (1995, 1996) does reflect varying degrees of mineral accumulation (e.g. anomalously high Sr reflecting plagioclase accumulation, high Sc reflecting pyroxene accumulation, etc.) and these samples have been omitted from the general dataset for density calculations. The high-Al olivine gabbroic (HAG) dykes have been subdivided into three distinct geochemical groups following Scoates & Mitchell (2000): the HAG1 group represents the majority of the dykes characterized by increasing silica and decreasing iron with decreasing MgO, the HAG2 group is distinguished by increasing iron with decreasing MgO, and relatively high potassium contents (0·82–1·11 wt % K2O) are diagnostic of the HAG3 group.
Oxygen fugacity constraints
The calculated densities of the high-Al gabbros, ferrodiorites and monzodiorites in the LAC depend strongly on the assumed Fe3+/Fe2+ during crystallization, which is a function of the oxygen fugacity conditions of the system of interest. The relative oxygen fugacities during the crystallization of the dykes can be determined by considering the QUILF relations between ferromagnesian silicates and Fe–Ti oxides (Frost & Lindsley, 1992; Lindsley & Frost, 1992; Andersen et al., 1993). After determining the relative oxygen fugacity at a specified temperature and pressure, the calculation of Fe2O3 and FeO abundances in the whole-rock analyses, assuming that they represent melt compositions, is model dependent. In this case, the thermodynamic model for chemical mass transfer in magmatic systems, MELTS (Ghiorso & Sack, 1995), which incorporates the redox relation of Kress & Carmichael (1991), has been used by inputting the major element composition of the rock of interest, the pressure (3 kbar), and the temperature and oxygen fugacity determined from the QUILF relations.
The majority of the gabbroic and dioritic dykes in the LAC are characterized by the coexistence of olivine and pyroxenes with both ilmenite and magnetite, which allows for the oxygen fugacity to be closely constrained. Figure 9, adapted from Frost et al. (1996), shows the variation in relative oxygen fugacity, where log fO2 = log fO2 (actual) - log fO2 (FMQ; fayalite–magnetite–quartz), with respect to the Fe/Mg of the system, as expressed by XFe in olivine. The diagram is polythermal [see Frost & Lindsley (1992) for similar usage], with a linear distribution of temperature from T = 1200°C at XFe = 0 to T = 1000°C at XFe = 1, and calculated for a pressure of 3 kbar. The temperature range is consistent with the results of thermometry studies in rocks of the LAC (Fuhrman et al., 1988; Kolker & Lindsley, 1989). Nearly all the rocks under discussion crystallized along the olivine-saturated QUILF surface [reaction (13) of Lindsley & Frost (1992)—OpAUIlO: opx/pig + ulvöspinel = olivine + augite + ilmenite), and underscore the proposed petrogenetic relationship between the mafic dykes in the LAC. Some ferrodiorite dykes lack olivine and thus crystallized at conditions slightly more oxidized than the olivine-saturated QUILF surface and higher silica activity, whereas some of the olivine ferrodiorites lack orthopyroxene or inverted pigeonite, and thus crystallized at slightly more reduced conditions and lower silica activity. The total range of oxygen fugacity conditions during crystallization of the LAC gabbroic and dioritic dykes is from conditions at or slightly above the FMQ buffer in the high-Al olivine gabbros to one log unit below FMQ for the majority of the ferrodiorites to about two log units below FMQ for the monzodiorites. Average calculated Fe2O3/FeO is 0·16 for the high-Al gabbros, 0·10 for the ferrodiorites and 0·10 for the monzodiorites.
|
Significance of P2O5 contents
The addition of phosphorus to a ferrobasaltic melt reduces magma densities because of the very large partial molar volume of P2O5 (64·5 cm3/mol at 1300°C) (Toplis et al., 1994). Phosphorus enrichment will tend to counteract to a certain extent the increase in density caused by iron enrichment, which implies that during progressive differentiation the actual densities of the residual liquids may be lower than those calculated without considering the role of P2O5. The reduction in density is about 0·02 g/cm3 per 2 wt % added P2O5 (Toplis et al., 1994), a significant decrease when considering silicate melt densities.
The LAC dykes are ferrobasaltic in composition, and those with near-liquid compositions can show significant enrichment in P2O5: 1·63–2·49 wt % in all ferrodiorites and 0·69–1·45 wt % in monzodiorites (Mitchell et al., 1996). The calculated reduction in melt density for all dykes in the study is 0–0·003 g/cm3 for the high-Al olivine gabbros, 0·01–0·02 g/cm3 for all ferrodiorites and 0·003–0·01 g/cm3 for the monzodiorites. Although important in the variation of melt density shown by the LAC dykes, the effect of P2O5 on melt density is smaller than that produced when assuming the presence of very small amounts of volatile constituents as discussed next.
Realistic volatile contents
Addition of even small amounts of H2O can significantly reduce the densities of multicomponent silicate melts because of the low molecular weight (18·016 g/mol) and relatively small partial molar volume (17 cm3/mol) of H2O (Lange, 1994). The effect of adding CO2 is less pronounced because of its higher molecular weight (44·0098 g/mol) and higher partial molar volume (24 cm3/mol in basaltic systems) (Lange, 1994), but is still important. Nearly all rocks associated with Proterozoic anorthosite complexes are notoriously dry, and the LAC is no exception to the rule. Strikingly few studies, however, have addressed the absolute values of fluid contents in these magmas, with the notable exceptions of Huntington (1979) for the Kiglapait intrusion and Frost & Touret (1989) for rocks of the LAC. The LAC magmas were hot and very dry: minimum crystallization temperatures, even for the most fractionated magmas, are typically above 1000°C (Fuhrman et al., 1988; Kolker & Lindsley, 1989), primary hydrous minerals such as amphibole and biotite are extremely rare, grain-boundary graphite occurs in samples of anorthosite and monzonite (Frost et al., 1989), and fluid inclusions in monzonitic rocks from the Sybille intrusion (Frost & Touret, 1989) and from the Red Mountain pluton (Anderson, 1995) are CO2 rich.
All natural silicate melts, even those considered to be ‘dry’, contain small amounts of volatiles, and their effects need to be incorporated into the density calculations presented here. Volatile contents are estimated based on measured contents in other basaltic systems. The volatile contents of mid-ocean ridge basalts are too low to cause crystallization of hydrous minerals; they contain between 0·05 and 0·60 wt % H2O with the bulk of the measured data between 0·20 and 0·25 wt % H2O, typically less than the contents associated with ocean island basalts (Johnson et al., 1994). For the purposes of this study, a value of 0·25 wt % H2O has been assumed in the density calculations for the least-fractionated group, the high-Al olivine gabbros. As volatiles are incompatible in the fractionating crystals, H2O (and CO2) contents must become enriched as crystallization proceeds and will reduce magma densities in the evolved residual liquids. To test the evolution of melt H2O contents in high-Al olivine gabbroic magmas from the LAC, a series of compositions were run in MELTS under fractional crystallization conditions and assuming an initial water content of 0·25 wt % H2O, a pressure of 3 kbar and oxygen fugacities corresponding to the FMQ buffer. After 50% crystallization, the evolved melts are broadly similar to Fe-enriched ferrodiorites and have 0·50 wt % H2O; after 75% crystallization, the predicted melt compositions share some compositional similarities with the monzodiorites and have 0·75 wt % H2O. Thus, for this study, density calculations were carried out at 3 kbar for the high-Al olivine gabbros at 1200°C assuming 0·25 wt % H2O, for the ferrodiorites at 1150°C assuming 0·50 wt % H2O, and for the monzodiorites at 1100°C assuming 0·75 wt % H2O. The LAC magmas may have been relatively CO2 enriched as discussed above. Although somewhat controversial, the CO2 content in submarine MORB glasses is typically <400 ppm (Johnson et al., 1994) and an initial value of 200 ppm has been used in the density calculations for the high-Al olivine gabbros. The effect of the evolution of melt CO2 contents on magma density during fractionation has been simulated by doubling the starting value for the ferrodiorites (400 ppm) and tripling it for the monzodiorites (600 ppm), paralleling the evolution of the model melt H2O contents. The effect of the addition of water is obviously much more significant than that of CO2, and the combined H2O + CO2 effect reduces calculated densities by 0·011–0·012 g/cm3 for the high-Al olivine gabbros, 0·024–0·027 g/cm3 for the ferrodiorites and 0·032–0·037 g/cm3 for the monzodiorites. Given the inherent uncertainties involved in assigning volatile contents to magmas associated with Proterozoic anorthosite complexes, these changes probably represent the maximum possible reduction in magma density.
|
BLOCK, PLAGIOCLASE AND MAGMA DENSITY RELATIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGIC SETTING OF THE...BLOCKS IN THE POE...DENSITY CALCULATIONSBLOCK, PLAGIOCLASE AND MAGMA...CONCLUSIONSREFERENCES |
|---|
The plagioclase–magma density paradox is well illustrated by the block–layering relations described in previous sections of this paper. As the anorthositic–leucogabbroic blocks arrived and struck the floor of the crystallizing Poe Mountain magma chamber, the blocks must have been denser than the resident magma. This constraint, and the evidence for deformation of pre-existing layering and lamination, requires that intermediate-composition plagioclase be on the floor. Of fundamental importance then is the following question: can An40–60 plagioclase settle through any of the proposed parental and residual magma compositions?
Density variation during crystallization of anorthosite-related magmas
The results of the density calculations for the proposed parental and residual magma compositions in the LAC [omitting the high-Zr and oxide-rich ferrodiorites of Mitchell et al. (1996)], taking into consideration oxygen fugacity constraints, P2O5 contents and evolving volatile contents, can be effectively shown in a diagram of density as a function of the whole-rock mg-number, where mg-number = Mg/(Mg + Fe2+) (Fig. 10). Figure 10 is isobaric (3 kbar) and polythermal, and the relative oxygen fugacity decreases from that equivalent to the FMQ buffer at mg-number = 0·6–0·7 to 2 log units below FMQ at mg-number = 0·15–0·20. The solid densities of the blocks calculated at 1000°C are shown in histogram form on the left side of the diagram for comparison; their whole-rock mg-number is not relevant to this discussion. Also shown is the field for An40–60 plagioclase (shaded field at bottom of diagram independent of mg-number).
|
The density variation shown by the mafic dykes of the LAC in Fig. 10 is remarkably coherent and similar to that shown for the progressive differentiation of natural anhydrous basalts (Sparks et al., 1980; Stolper & Walker, 1980; Sparks & Huppert, 1984). High-Al olivine gabbro magma densities of group 1 (HAG1) decrease progressively to a minimum of 2·68 g/cm3 at mg-number = 0·57. The other two high-Al olivine gabbro groups, HAG2 and HAG3, depart from the HAG1 trend with increasing density as mg-number decreases and trend towards the ferrodiorites. Calculated densities for the ferrodiorites show a maximum in the region of 2·85–2·90 g/cm3 at mg-number 0·3. The monzodiorites show a striking trend of rapidly decreasing magma density with a small decrease in mg-number, down to 2·65 g/cm3 at mg-number = 0·13 (Fig. 10). The regular density variation shown by this diverse group of dykes in the LAC further supports the proposition that they may be related through fractionation, albeit open system (Mitchell et al., 1995, 1996; Scoates et al., 1996).
The density relations for the blocks clearly show that they are nearly all capable of sinking through the majority of the proposed high-Al olivine gabbroic parental magmas, consistent with previously described field relations. Blocks GR258 and SR246 are denser than all high-Al olivine gabbros, whereas blocks PM466, PM555 and GR289 fall within the calculated range (Fig. 10). The capability of these blocks to sink is evidently controlled by the relative degree of fractionation of the high-Al olivine gabbros. The one possible exception is SR248, which is less dense than all calculated magma densities, except the most fractionated monzodiorite (Fig. 10). The physical evidence for impact on the chamber floor appears incontrovertible (Fig. 5a). It is possible that the sample collected is not representative of the bulk composition of the block, but instead represents that of a more plagioclase-rich, and thus less dense, portion.
In sharp contrast to the blocks, all of the plagioclase compositions associated with the Poe Mountain anorthosite (An45–55) are less dense than those of the calculated melt densities (Fig. 10), ranging from the relatively MgO-rich high-Al olivine gabbros to the evolved ferrodiorites and monzodiorites. Similar conclusions with respect to relative plagioclase–magma densities have been reached in experimental studies of ferrodioritic or jotunitic melts from other Proterozoic anorthosite complexes. In both the Newark Island layered intrusion within the Nain Plutonic Suite, Labrador (Snyder et al., 1993) and the Bjerkreim–Sokndal layered intrusion within the Rogaland anorthosite complex (Vander Auwera & Longhi, 1994), melt densities from a range of fractionated compositions (6–3 wt % MgO) remain persistently well above that of the equilibrium plagioclase. Thus, evidence from several different Proterozoic anorthosite complexes consistently shows that plagioclase cannot settle in the associated melts, and in situ or other methods of crystallization–accumulation must be invoked (see discussion below).
Crystallizing mineral assemblages and fractionation densities
The mineral assemblage that crystallizes from a magma controls the resultant density variations in the melt. At 3 kbar, the LAC high-Al olivine gabbros are co-saturated in plagioclase, olivine and high-Ca pyroxene ± low-Ca pyroxene, the ferrodiorites are co-saturated in plagioclase, high-Ca pyroxene, pigeonite ± olivine and Fe–Ti oxides, and the monzodiorites are co-saturated in sodic plagioclase (nearing ternary feldspar), high-Ca pyroxene, ilmenite, magnetite, olivine and apatite (Mitchell et al., 1995, 1996). The density maximum shown in the ferrodiorites and the subsequent dramatic decrease in magma density clearly marks the onset of Fe–Ti oxide crystallization (Fig. 10). The existence of a density minimum in basaltic systems is commonly inferred to mark the onset of plagioclase (± high-Ca pyroxene) crystallization, and the initial decrease in density at relatively high mg-number is generally considered to result from the crystallization of olivine ± pyroxene (Stolper & Walker, 1980). However, all the LAC high-Al olivine gabbroic dykes are saturated in plagioclase (Mitchell et al., 1995).
The influence of the fractionating assemblage on melt density can be examined considering fractionation densities, the ratio of the gram formula weight to molar volume of the chemical components in the liquid phase that are being removed by fractional crystallization (Sparks & Huppert, 1984). Fractionation densities of the major minerals that crystallize from basaltic magmas—plagioclase, clinopyroxene, pigeonite, orthopyroxene and olivine—have been calculated after the method of Sparks & Huppert (1984) at 3 kbar incorporating the partial molar volumes and associated compressibilities of Lange (1994) (Fig. 11). Because differentiation is associated with decreasing temperatures, the effect of thermal expansion on the fractionation densities has been considered. Figure 11 is polythermal with linearly decreasing temperature, like Fig. 9, from 1200°C at mg-number or An* = 1 to 1000°C at mg-number or An* = 0. For the high-Al olivine gabbro densities to systematically decrease with decreasing mg-number, a fractionating assemblage of >50% plagioclase and subequal amounts of olivine and clinopyroxene is initially required (mg-number = 0·65). The relative amount of plagioclase that crystallizes can increase during fractionation as the density of plagioclase decreases and those of the ferromagnesian silicates progressively increase. To account for increasing densities with decreasing mg-number as observed in the HAG2 and HAG3 dykes, the crystallizing assemblage must become more rich in plagioclase and/or much poorer in olivine (Fig. 11). Pigeonite becomes the stable low-Ca pyroxene at about mg-number = 0·6 (Lindsley & Frost, 1992). In conjunction with increased silica activity, olivine is probably replaced by pigeonite as the low-Ca ferromagnesian phase, which is reflected in the relative density relations of the evolved magmas.
|
Plagioclase crystallization and the formation of plagioclase-rich layered rocks
Densities for plagioclase (An40–60) in the anorthositic rocks range from 2·61 to 2·65 g/cm3, significantly below those for calculated parental magma compositions (Fig. 10) and yet plagioclase was clearly ‘accumulating’ on the chamber floor before the arrival of the blocks (Figs 5, 6 and 7). Considering the simple end-member case where the resident magma contained few suspended crystals, for plagioclase to have ended up on the intrusion floor beneath the blocks, it must either have crystallized in a boundary layer at the crystal pile–magma interface—in situ crystallization (Jackson, 1961; Campbell, 1978; McBirney & Noyes, 1979; Morse, 1986a; Langmuir, 1989)—or it must have been carried to the floor in relatively dense packets of cooled liquid plus crystals from the roof or walls of the intrusion—two-phase convection (Grout, 1918; Morse, 1986b). Once at the floor, plagioclase will probably remain there if there is negligible density contrast (0·1 g/cm3) and if there is the slightest amount of yield strength to the magma (Irvine, 1987). Crystallization of interstitial ferromagnesian silicates and Fe–Ti oxides, much of which is oikocrystic (Scoates, 1994), will consolidate the crystal pile and allow it to remain in place.
The above description of plagioclase crystallization is considerably at odds with the prevailing model for the formation of Proterozoic anorthosites. As discussed at the start of this paper, buoyancy-driven ascent of crystal-rich diapirs containing 50–70 vol. % plagioclase is a mechanism favoured by many workers to account for the characteristic large masses of anorthosite–leucotroctolite–leuconorite (Emslie, 1985; Longhi & Ashwal, 1985; Ashwal, 1993; Longhi et al., 1993). This mechanism leaves little room for the existence of dynamic magma chambers at the final level of emplacement, as the high proportion of crystals to liquid approaches (or exceeds) the limit of critical crystallinity, where viscosities increase so dramatically that the rheology of the magma becomes essentially that of a solid (Marsh, 1981). The field evidence from the Poe Mountain anorthosite in the LAC, and from layered anorthosites from other Proterozoic anorthosite complexes (Wiebe, 1992), suggests that a significant component of melt was present in the mid- to upper-crustal magma chambers after emplacement. Assuming that the resident magmas contained some amount of suspended plagioclase crystals upon emplacement, both two-phase convection and in situ crystallization may still be viable processes. If two-phase convection was operating, then the descending packets of cooler melt may have entrained some of the suspended plagioclase and transported it to the floor. Conversely, if plagioclase was crystallizing directly on the floor of the intrusion then the suspended plagioclase would be progressively incorporated in the advancing solidification front from the floor. Both of these processes may explain the rather common occurrence of randomly distributed blocky megacrysts of plagioclase within cumulates consisting of tabular, laminated plagioclase of the Poe Mountain anorthosite layered series (Scoates, 1994)—the megacrysts may represent plagioclase crystallized at depth and transported in suspension in a feldspathic magma.
The effects of plagioclase remelting on plagioclase–magma density contrasts
The plagioclase–magma density paradox obviously ceases to be a problem if the resident magma density is less than that of the crystallizing–accumulating plagioclase. Is there evidence for parental magmas that are less dense than intermediate-composition plagioclase in Proterozoic anorthosite complexes? The mafic dykes in the LAC, and those found in other anorthosites, are multiply saturated in plagioclase + ferromagnesian silicates (olivine + pyroxenes) ± Fe–Ti oxides at their level of emplacement, typically in the middle to upper crust. The formation of anorthosites, which are defined by the presence of abundant excess plagioclase compared with established cotectic proportions with ferromagnesian silicates, requires that the parental magmas lie well within the plagioclase-only field and thus are richer in the lighter oxide components Al2O3, SiO2, CaO and Na2O, and poorer in the denser oxide components MgO, FeO* and TiO2. Could the addition of plagioclase–melt components sufficiently reduce magma densities such that intermediate-composition plagioclase would actually sink?
The polybaric model for anorthosite formation discussed in previous sections involves extensive fractionation at depth of basaltic magmas that are multiply saturated at high pressures, perhaps corresponding to the base of the crust. Plagioclase that has accumulated by flotation, as a result of the increased compressibility of silicate melts at high pressure, at the roof of these staging chambers may be partially remelted and resorbed during periodic replenishments of higher temperature, less evolved magma (Wiebe, 1992). Diapirs containing suspended plagioclase are then assumed to rise through the crust from these deeper staging chambers. Because of the expansion of the plagioclase stability field with decreasing pressure (Morse, 1982), cotectic magmas at depth lie in the plagioclase field on ascent and thus additional suspended plagioclase could be remelted, provided that little heat is lost to melting of the crust, and reduce the overall magma density. A simplified test of this hypothesis is shown in Fig. 12, which shows magma density as a function of the percentage of added remelted plagioclase, the composition of the remelted plagioclase and pressure. The initial magma composition (filled squares) is taken as the average of the high-Al olivine gabbroic group 1 (HAG1) from the LAC: 2·73 g/cm3 at 3 kbar (1200°C, FMQ, 0·25 wt % H2O and 200 ppm CO2) and 2·83 g/cm3 at 10 kbar (1250°C, FMQ, 0·25 wt % H2O and 200 ppm CO2). Use of a ferrodioritic composition results in a much greater initial plagioclase–magma density contrast. For both pressures, the effect of adding remelted plagioclase components ranging from An40 to An60 and taking into account the effect of compressibility is shown (equivalent melt densities—3 kbar: An40–60 = 2·49–2·54 g/cm3; 10 kbar: An40–60 = 2·60–2·65 g/cm3). At 10 kbar, the magma densities remain higher than the crystallizing plagioclase for all reasonable values of percent remelted plagioclase. At 3 kbar, 40–50% remelted plagioclase is required to eliminate the plagioclase–magma density contrast, and >50% remelted plagioclase would be necessary to allow the plagioclase to effectively sink. This simplified test does not take into account the heat necessary to remelt plagioclase, but demonstrates only the density requirements needed to allow plagioclase to sink in appropriate parental magmas. By incorporating thermal expansion and heat capacity terms, Longhi et al. (1999) calculated that pressure release from 13 to 4 kbar of an ascending anorthositic diapir could remelt as much as 4% of the suspended plagioclase, producing only minor changes in magma density with respect to the relations shown in Fig. 12. Thus, if plagioclase remelting is to be an effective mechanism for decreasing magma density, it appears that extensive remelting must be done at depth before ascent. Additionally, anorthositic magmas (melt + crystals) emplaced at pressures >3 kbar will be subjected to progressively larger plagioclase–magma density contrasts as a result of the effects of silicate melt compressibility.
|
The fate of the residual liquid and the significance of sloping floors
The accumulation and crystallization of plagioclase on the floor of the Poe Mountain anorthosite magma chamber leads to another major problem encountered when considering the crystallization of anorthosites: the rejected interstitial liquid is denser than the plagioclase in the crystal pile (Figs 10 and 11) and will remain so until well after saturation in Fe–Ti oxides. As a result, this process should lead to stagnation of the dense residual liquid among the cumulus plagioclase on the floor with the added result that no anorthosite can form. Some dense residual liquid could infiltrate downward eventually forming small dykes or pods of ferrodiorite that are so characteristic of Proterozoic anorthosite complexes. However, to form an anorthosite the majority of this dense liquid must be removed from the crystal pile. To overcome this problem, Morse (1986a, 1988) proposed the general case of a sloping floor for the crystallization of felsic adcumulates. A sloping or inclined floor allows for the progressive downslope migration of dense, rejected interstitial liquid, which may eventually rejoin the main mass of magma above the crystal pile, ensuring the magma resident in the chamber will always be denser than the cumulates. An important consequence of this process is that compositionally evolved liquids can be redistributed downslope, thus influencing the bulk composition of the resultant cumulates as evolved liquids interact with less evolved liquids. This would occur before significant compaction, upward expulsion of interstitial liquid and associated compositional modification (Meurer & Boudreau, 1998), and should be incorporated into current models for the evolution of solidifying crystal piles. A sloping floor certainly appears to be required for the crystallization of layered anorthosites, leuconorites and leucotroctolites in Proterozoic anorthosite complexes, and may be a general characteristic of all feldspathic cumulates in layered intrusions where the rejected interstitial liquid is relatively dense.
The crystallization regimes for the layered series cumulates in the Poe Mountain anorthosite involving a sloping floor can be envisaged as consisting of (1) a resident magma composed of high-Al olivine basalt plus suspended plagioclase that crystallized at depth or during ascent through the crust, (2) an upper mush zone to the crystal pile that contains cumulus plagioclase and dense Fe–Ti–P-rich interstitial liquid (ferrodiorite), and (3) a deeper zone in the crystal pile that is below the solidus and consists of solidified anorthosite (Fig. 13). Plagioclase accumulates and crystallizes on the floor through combined in situ crystallization, the arrival of dense two-phase packets composed mainly of cooler liquid and some crystals, and progressive incorporation of suspended megacrysts. Anorthositic to leucogabbroic blocks periodically struck the floor causing extensive disruption and deformation beneath them, and demonstrating that most layering and lamination forms directly at the magma–crystal pile interface. Crystallization of intermediate-composition plagioclase produces a dense residual liquid enriched in Fe–Ti–P and somewhat depleted in silica, essentially ferrodioritic in composition. The dense liquid is gravitationally unstable in the upper part of the inclined crystal pile and migrates downslope through a permeable network of plagioclase crystals, perhaps similar to the way interstitial liquid drains from crystal networks in partially melted basalts (Philpotts & Carroll, 1996; Philpotts et al., 1998). The ferrodioritic liquid may mix with other less evolved and more evolved liquids during percolation and may react with cumulus plagioclase. The extent to which the dense liquid can seep downwards is probably limited by lithologic variations—layers of nearly pure anorthositic adcumulate will act as impermeable barriers to both downward and upward migrating liquid (Fig. 13). The much modified dense residual liquid must eventually rejoin the main mass of resident magma.
|
The presence of a sloping floor may be a requirement for the evolution of Proterozoic anorthosites. The overall density of the anorthositic cumulates on the floor will always be less than that of the resident magmas above them, producing an inherently unstable situation. Large volumes of relatively light anorthositic cumulates may rise diapirically because of density contrasts with the underlying material, progressively tilting the floor during crystallization and allowing for the downward escape of dense residual liquid. Small amounts of deformation related to slow, diapiric rise could be responsible for the recrystallized texture of many Proterozoic anorthosites that formed through high-temperature ‘fast’ grain boundary migration (Lafrance et al., 1996). The resultant domical structure would also be consistent with the forms of many separate plutons in Proterozoic anorthosite complexes (Emslie, 1980; Frost et al., 1993).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGIC SETTING OF THE...BLOCKS IN THE POE...DENSITY CALCULATIONSBLOCK, PLAGIOCLASE AND MAGMA...CONCLUSIONSREFERENCES |
|---|
The age, compositions and structures associated with anorthositic to leucogabbroic blocks in the layered series of the Poe Mountain anorthosite, combined with calculations of proposed parental and residual magma densities, place important constraints on the plagioclase–magma density paradox in Proterozoic anorthosites and plagioclase-rich layered intrusions. The blocks may represent fragments of a now-eroded roof zone, a distinct earlier phase of anorthositic magmatism or material crystallized at high pressures. Disrupted and deformed layered anorthosites beneath the blocks provide consistent stratigraphic tops indicators. The blocks fell through resident magma and struck a floor that was present during crystallization of the layered series, where intermediate-composition plagioclase (An45–55) was accumulating and/or crystallizing, and where compositional layering and lamination were forming. The upper parts of the crystal pile contained significant amounts of relatively dense interstitial melt that was remobilized by the block impacts.
Block densities are greater than, or nearly identical to, the calculated magma densities. The density variation of mafic dykes in the LAC, which have compositions appropriate for magmas that produced the anorthositic cumulates (high-Al olivine gabbros) or Fe–Ti–P-rich residual magmas (ferrodiorites and monzodiorites), is remarkably coherent and similar to that for the progressive differentiation of natural anhydrous basalts. Densities for intermediate-composition plagioclase are significantly below those for the calculated magma compositions. Plagioclase cannot have settled to the chamber floor; it must either have crystallized in situ in a boundary zone at the magma–crystal pile interface or arrived at the floor in dense two-phase packets composed mainly of liquid and some crystals from the roof zone of the chamber. Plagioclase suspended in the resident high-Al olivine gabbroic magma was incorporated by progressive growth of the crystal pile.
The rejected interstitial liquid in the crystal pile of layered anorthosites is denser than the network of cumulus plagioclase. As a result, the liquid should stagnate and prevent anorthosite from forming. However, if the floor is inclined, the dense liquid will infiltrate downslope and eventually rejoin the main mass of magma. A sloping floor is required to form layered anorthosites in Proterozoic anorthosite complexes and may be a general characteristic of all plagioclase-rich cumulates in layered intrusions where the rejected interstitial liquid is denser than the combined density of the cumulus crystals. Sloping floors may form during the crystallization of Proterozoic anorthosites as a result of the slow diapiric rise of the relatively light cumulates deep in the crystal pile, and may explain the common association of pervasive recrystallization and domical structures in intrusions within individual complexes.
|
ACKNOWLEDGEMENTS |
|---|
I would like to thank B. R. Frost, D. H. Lindsley, J. N. Mitchell and W. P. Meurer for invaluable assistance and discussions during the field part of this study. R. F. J. Scoates, O. R. Eckstrand, M. Zeintek, R. A. Wiebe, R. F. Emslie and S. A. Morse provided insightful comments during various field trip visits to the outcrops. Field work was supported by two Geological Society of America Summer Research Grants, and field and laboratory work were both funded by National Science Foundation (NSF) grants EAR8618480 and EAR8816040 to D. H. Lindsley and EAR9017465 and EAR9218360 to B. R. Frost and C. D. Frost. The ideas in this paper benefited greatly from interactions with geoscientists world-wide at the Penrose Conference on the Origin and Evolution of Proterozoic Anorthosites and at several IGCP290 Origin of Anorthosites Conferences. The mapping involved in this project would not have been possible without the co-operation of the numerous landowners in the southern Laramie Mountains, past and present, and the staff of the Wyoming Game and Fish Experimental Station in Sybille Canyon. Many thanks are due also to A. Kolker and M. Ghazi for providing high-quality trace element and REE analyses. External reviews by D. H. Lindsley, B. R. Frost, R. F. J. Scoates and J. N. Mitchell, and journal reviews and comments by S. A. Morse, R. A. Wiebe and S. R. Tait have materially improved the presentation of arguments in the manuscript.
|
FOOTNOTES |
|---|
*Telephone: +322-650-4714. Fax: +322-650-3748; e-mail: jscoates{at}ulb.ac.be
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGIC SETTING OF THE...BLOCKS IN THE POE...DENSITY CALCULATIONSBLOCK, PLAGIOCLASE AND MAGMA...CONCLUSIONSREFERENCES |
|---|
Andersen, D. J., Lindsley, D. H. & Davidson, P. M. (1993). QUILF: a PASCAL program to assess equilibria among Fe–Mg–Mn–Ti oxides, pyroxenes, olivine, and quartz. Computers and Geosciences 19, 1333–1350.
Anderson, I. C. (1995). Petrology and geochemistry of the Red Mountain pluton, Laramie anorthosite complex, Wyoming. Ph.D. Dissertation, University of Wyoming, Laramie, 164 pp.
Anderson, I. C., Frost, B. R. & Lindsley, D.H. (1987). Crystallization conditions of the Red Mountain syenite. EOS Transactions, American Geophysical Union 68, 442–443.
Ashwal, L. D. (1993). Anorthosites. Berlin: Springer-Verlag, 422 pp.
Bolsover, L. R. (1986). Petrogenesis of the Sybille iron–titanium oxide deposit, Laramie anorthosite complex, Laramie Mountains, Wyoming. M.Sc. Thesis, State University of New York at Stony Brook, 78 pp.
Campbell, I. H. (1978). Some problems with the cumulus theory. Lithos 11, 311–323.[Web of Science]
Campbell, I. H., Roeder, P. L. & Dixon, J. M. (1978). Plagioclase buoyancy in basaltic liquids as determined with a centrifuge furnace. Contributions to Mineralogy and Petrology 67, 369–377.
Chen, C. F. & Turner, J. S. (1980). Crystallization in a double diffusive system. Journal of Geophysical Research 85, 2573–2593.[Web of Science]
Czamanske, G. K. & Bohlen, S. R. (1990). The Stillwater complex and its anorthosites: an accident of magmatic underplating? American Mineralogist 75, 37–45.[Abstract]
Deer, W. A., Howie, R. A. & Zussman, J. (1992). An Introduction to the Rock-Forming Minerals, 2nd edn. Harlow, UK: Longman, 696 pp.
Edwards, B. R. (1993). A field, geochemical, and isotopic investigation of the igneous rocks in the Pole Mountain area of the Sherman batholith, southern Laramie Mountains, Wyoming, USA. M.S. Thesis, University of Wyoming, Laramie.
Emslie, R. F. (1970). The geology of the Michikamau intrusion, Labrador. Geological Survey of Canada Paper 68-57, 85 pp.
Emslie, R. F. (1980). Geology and petrology of the Harp Lake complex, central Labrador: an example of Elsonian magmatism. Geological Survey of Canada Bulletin 293, 136 pp.
Emslie, R. F. (1985). Proterozoic anorthosite massifs. In: Tobi, A. C. & Touret, J. L. R. (eds) The Deep Proterozoic Crust in the North Atlantic Provinces. Dordrecht: D. Reidel, pp. 39–60.
Frost, B. R. & Lindsley, D. H. (1992). Equilibria among Fe–Ti oxides, pyroxenes, olivine, and quartz: Part II. Application. American Mineralogist 77, 1004–1020.[Abstract]
Frost, B. R. & Simons, J. P. (1991). Fe–Ti oxide deposits of the Laramie anorthosite complex: their geology and proposed economic utilization. In: Frost, B. R. & Roberts, S. (eds) Wyoming Geological Association 42nd Annual Field Conference Guidebook. Laramie: Geological Survey of Wyoming, pp. 41–48.
Frost, B. R. & Touret, J. L. R. (1989). Magmatic CO2 and saline melts from the Sybille monzosyenite, Laramie anorthosite complex, Wyoming. Contributions to Mineralogy and Petrology 103, 178–186.
Frost, B. R., Fyfe, W. S., Tazali, K. & Chan, T. (1989). Grain-boundary graphite in rocks and implications for high electrical conductivity in the lower crust. Nature 340, 134–136.
Frost, B. R., Frost, C. D., Lindsley, D. H., Scoates, J. S. & Mitchell, J. N. (1993). The Laramie anorthosite complex and Sherman batholith: geology, evolution, and theories of origin. In: Snoke, A. W., Steidtmann, J. R. & Roberts, S. M. (eds) Geology of Wyoming. Geological Survey of Wyoming Memoir 5, 119–161.
Frost, B. R., Frost, C. D., Chamberlain, K. R., Scoates, J. S. & Lindsley, D. H. (1996). A field guide to the Proterozoic anorthositic, monzonitic, and granitic plutons, Laramie Range, southeastern Wyoming. In: Thompson, R. A., Hudson, M. R. & Pillmore, C. L. (eds) Geologic Excursions to the Rocky Mountains and Beyond: Guidebook to the 1996 Annual Meeting, Geological Society of America, Denver, CO, 28–31 October. Denver: Colorado Geological Survey, pp. 1–28.
Fuhrman, M. L., Frost, B. R. & Lindsley, D. H. (1988). Crystallization conditions of the Sybille monzosyenite, Laramie anorthosite complex, Wyoming. Journal of Petrology 29, 699–729.
Geist, D. J., Frost, C. D., Kolker, A. & Frost, B. R. (1989). A geochemical study of magmatism across a major terrane boundary: Sr and Nd isotopes in Proterozoic granitoids of the southern Laramie Range, Wyoming. Journal of Geology 97, 331–342.
Ghiorso, M. S. & Sack, R. O. (1995). Chemical mass transfer in magmatic systems IV. A revised and internally consistent thermodynamic model for the interpolation and extrapolation of liquid–solid equilibria in magmatic systems at elevated temperatures and pressures. Contributions to Mineralogy and Petrology 119, 197–212.[Web of Science]
Grout, F. F. (1918). Two phase convection in igneous magmas. Journal of Geology 26, 481–499.
Hanson, G. N. (1980). Rare earth elements in petrogenetic studies of igneous systems. Annual Review of Earth and Planetary Sciences 8, 371–406.[Web of Science]
Haskin, L. A. & Salpas, P. A. (1992). Genesis of compositional characteristics of Stillwater AN-I and AN-II thick anorthosite units. Geochimica et Cosmochimica Acta 56, 1187–1212.
Huntington, H. D. (1979). Kiglapait mineralogy I: Apatite, biotite, and volatiles. Journal of Petrology 20, 625–652.
Huppert, H. E. & Sparks, R. S. J. (1984). Double diffusive convection due to crystallization in magmas. Annual Review of Earth and Planetary Sciences 12, 11–37.[Web of Science]
Irvine, T. N. (1987). Layering and related structures in the Duke Island and Skaergaard intrusions: similarities, differences, and origins. In: Parsons, I. (ed.) Origins of Igneous Layering. Dordrecht: D. Reidel, pp. 185–245.
Irvine, T. N., Keith, D. W. & Todd, S. G. (1983). The J-M platinum–palladium reef of the Stillwater complex, Montana: II. Origin by double diffusive convective magma mixing and implications for the Bushveld complex. Economic Geology 78, 1287–1334.
Irvine, T. N., Andersen, J. C. Ø. & Brooks, C. K. (1998). Included blocks (and blocks within blocks) in the Skaergaard intrusion: geologic relations and the origins of rhythmic modally graded layers. Geological Society of America Bulletin 110, 1398–1447.
Jackson, E. D. (1961). Primary textures and mineral associations in the Ultramafic zone of the Stillwater complex, Montana. US Geological Survey Professional Paper 358, 106 pp.
Johnson, M. C., Anderson, A. T., Jr & Rutherford, M. J. (1994). Pre-eruptive volatile contents of magmas. In: Carroll, M. R. & Holloway, J. R. (eds) Volatiles in Magmas. Mineralogical Society of America, Reviews in Mineralogy 30, 281–330.
Karlstrom, K. E. & Houston, R. S. (1984). The Cheyenne Belt: analysis of a Proterozoic suture in southern Wyoming. Precambrian Research 25, 415–446.[Web of Science]
Kolker, A. & Lindsley, D. H. (1989). Geochemical evolution of the Maloin Ranch pluton, Laramie anorthosite complex, Wyoming: petrology and mixing relations. American Mineralogist 74, 307–324.[Abstract]
Kolker, A., Lindsley, D. H. & Hanson, G. N. (1990). Geochemical evolution of the Maloin Ranch pluton, Laramie anorthosite complex, Wyoming: trace elements and petrogenetic models. American Mineralogist 75, 572–588.[Abstract]
Kolker, A., Frost, C. D., Hanson, G. N. & Geist, D. J. (1991). Neodymium, strontium, and lead isotopes in the Maloin Ranch pluton, Wyoming: implications for the origin of evolved rocks at anorthosite margins. Geochimica et Cosmochimica Acta 55, 2285–2297.
Kress, V. & Carmichael, I. S. E. (1991). The compressibility of silicate liquids containing Fe2O3 and the effect of composition, temperature, oxygen fugacity and pressure on their redox states. Contributions to Mineralogy and Petrology 108, 82–92.
Kushiro, I. (1980). Viscosity, density, and structure of silicate melts at high pressures, and their petrological applications. In: Hargraves, R. B. (ed.) Physics of Magmatic Processes. Princeton, NJ: Princeton University Press, pp. 93–120.
Lafrance, B., John, B. E. & Scoates, J. S. (1996). Syn-emplacement recrystallization and deformation microstructures in the Poe Mountain anorthosite, Wyoming. Contributions to Mineralogy and Petrology 122, 431–440.
Lange, R. A. (1994). The effect of H2O, CO2 and F on the density and viscosity of silicate melts. In: Carroll, M. R. & Holloway, J. R. (eds) Volatiles in Magmas. Mineralogical Society of America, Reviews in Mineralogy 30, 331–369.
Lange, R. A. & Carmichael, I. S. E. (1990). Thermodynamic properties of silicate liquids with emphasis on density, thermal expansion and compressibility. In: Nicholls, J. & Russell, J. (eds) Modern Methods of Igneous Petrology: Understanding Magmatic Processes. Mineralogical Society of America, Reviews in Mineralogy 24, 25–64.
Langmuir, C. H. (1989). Geochemical consequences of in situ crystallization. Nature 340, 199–205.
Lindsley, D. H. & Frost, B. R. (1992). Equilibria among Fe–Ti oxides, pyroxenes, olivine, and quartz: Part I. Theory. American Mineralogist 77, 987–1003.[Abstract]
Longhi, J. & Ashwal, L. D. (1985). Two-stage models for lunar and terrestrial anorthosites: petrogenesis without a magma ocean. Journal of Geophysical Research 90, C571–C584.
Longhi, J., Fram, M. S., Vander Auwera, J. & Montieth, J. N. (1993). Pressure effects, kinetics, and rheology of anorthositic and related magmas. American Mineralogist 78, 1016–1030.[Abstract]
Longhi, J., Vander Auwera, J., Fram, M. S. & Duchesne, J. C. (1999). Some phase equilibrium constraints on the origin of Proterozoic (massif) anorthosites and related rocks. Journal of Petrology 40, 339–362.
Marsh, B. D. (1981). On the crystallinity, probability of occurrence, and rheology of lava and magma. Contributions to Mineralogy and Petrology 78, 85–98.
McBirney, A. R. (1985). Further considerations of double-diffusive stratification and layering in the Skaergaard intrusion. Journal of Petrology 26, 993–1001.
McBirney, A. R. (1995). Mechanisms of differentiation in the Skaergaard intrusion. Journal of the Geological Society, London 152, 421–435.
McBirney, A. R. (1996). The Skaergaard intrusion. In: Cawthorn, R. G. (ed.) Layered Intrusions. Amsterdam: Elsevier, pp. 147–180.
McBirney, A. R. & Hunter, R. H. (1995). The cumulate paradigm reconsidered. Journal of Geology 103, 114–122.
McBirney, A. R. & Noyes, R. M. (1979). Crystallization and layering in the Skaergaard intrusion. Journal of Petrology 20, 487–554.
Meurer, W. P. & Boudreau, A. E. (1998). Compaction of igneous cumulates Part I: Geochemical consequences for cumulates and liquid fractionation trends. Journal of Geology 106, 281–292.[Web of Science]
Mitchell, J. N. (1993). Petrology and geochemistry of dioritic and gabbroic rocks in the Laramie anorthosite complex, Wyoming: implications for the evolution of Proterozoic anorthosite. Ph.D. Dissertation, University of Wyoming, Laramie, 213 pp.
Mitchell, J. N., Scoates, J. S. & Frost, C. D. (1995). High-Al gabbros in the Laramie anorthosite complex, Wyoming: implications for the composition of melts parental to Proterozoic anorthosite. Contributions to Mineralogy and Petrology 119, 166–180.
Mitchell, J. N., Scoates, J. S., Frost, C. D. & Kolker, A. (1996). The geochemical evolution of anorthosite residual magmas in the Laramie anorthosite complex, Wyoming. Journal of Petrology 37, 637–660.
Morse, S. A. (1968). Layered intrusions and anorthosite genesis. In: Isachsen, Y. W. (ed.) Origin of Anorthosite and Related Rocks. New York State Museum and Science Service Memoir 18, 175–187.
Morse, S. A. (1969). The Kiglapait Layered Intrusion, Labrador. Geological Society of America, Memoir 112, 204 pp.
Morse, S. A. (1973). The feldspar/magma density paradox. In Morse, S. A. (ed.) The Nain Anorthosite Project, Labrador: Field Report 1972. Amherst, MA: University of Massachusetts at Amherst, pp. 113–116.
Morse, S. A. (1979). Kiglapait geochemistry I: Systematics, sampling, and density. Journal of Petrology 20, 555–590.
Morse, S. A. (1982). A partisan review of Proterozoic anorthosites. American Mineralogist 67, 1087–1100.[Abstract]
Morse, S. A. (1986a). Convection in aid of adcumulus growth. Journal of Petrology 27, 1183–1214.
Morse, S. A. (1986b). Thermal structure of crystallizing magma with two-phase convection. Geological Magazine 123, 205–214.[Abstract]
Morse, S. A. (1988). Motion of crystals, solute, and heat in layered intrusions. Canadian Mineralogist 26, 209–224.
Philpotts, A. R. & Carroll, M. (1996). Physical properties of partly melted tholeiitic basalt. Geology 24, 1029–1032.
Philpotts, A. R., Shi, J. & Brustman, C. (1998). Role of plagioclase crystal chains in the differentiation of partly crystallized basaltic magma. Nature 395, 343–346.
Raedeke, L. D. & McCallum, I. S. (1980). A comparison of fractionation trends in the lunar crust and the Stillwater complex. In: Papike, J. J. & Merrill, R. B. (eds) Proceedings of the Conference on Lunar Highlands Crust. New York: Pergamon, pp. 133–153.
Salpas, P. A., Haskin, L. A. & McCallum, I. S. (1983). Stillwater anorthosites: a lunar analog? Proceedings of the 14th Lunar and Planetary Science Conference, Part 1. Journal of Geophysical Research 88, B27–B39.
Scoates, J. S. (1994). Magmatic evolution of anorthositic and monzonitic rocks in the mid-Proterozoic Laramie anorthosite complex, Wyoming, USA. Ph.D. Dissertation, University of Wyoming, Laramie, 277 pp.
Scoates, J. S. & Chamberlain, K. R. (1995). Baddeleyite (ZrO2) and zircon (ZrSiO4) from anorthositic rocks of the Laramie anorthosite complex, Wyoming: petrologic consequences and U–Pb ages. American Mineralogist 80, 1319–1329.
Scoates, J. S. & Chamberlain, K. R. (1997). Orogenic to post-orogenic origin for the 1·76 Ga Horse Creek anorthosite complex, Wyoming, USA. Journal of Geology 105, 331–343.
Scoates, J. S. & Frost, C. D. (1996). A strontium and neodymium isotopic investigation of the Laramie anorthosites, Wyoming, USA: implications for magma chamber processes and the evolution of magma conduits in Proterozoic anorthosites. Geochimica et Cosmochimica Acta 60, 95–107.
Scoates, J. S. & Mitchell, J. N. (2000). The evolution of troctolitic and related magmas in Proterozoic anorthosite complexes and implications for Voisey’s Bay-type massive Ni–Cu sulfide deposits. Economic Geology (in press).
Scoates, J. S., Frost, C. D., Mitchell, J. N., Lindsley, D. H. & Frost, B. R. (1996). Residual-liquid origin for a monzonitic intrusion in a mid-Proterozoic anorthosite complex: the Sybille intrusion, Laramie anorthosite complex, Wyoming. Geological Society of America Bulletin 108, 1357–1371.
Snyder, D., Carmichael, I. S. E. & Wiebe, R. A. (1993). Experimental study of liquid evolution in an Fe-rich, layered mafic intrusion: constraints of Fe–Ti oxide precipitation on the T–fO2 and T–
of tholeiitic magmas. Contributions to Mineralogy and Petrology 113, 73–86.
Sonnenthal, E. L. & McBirney, A. R. (1998). The Skaergaard Layered Series. Part IV. Reaction–transport simulations of foundered blocks. Journal of Petrology 39, 633–661.
Sparks, R. & Huppert, H. (1984). Density changes during the fractional crystallization of basaltic magmas: fluid dynamic implications. Contributions to Mineralogy and Petrology 85, 300–309.
Sparks, R. S. J., Meyer, P. & Sigurdsson, H. (1980). Density variation amongst mid-ocean ridge basalts: implications for magma mixing and the scarcity of primitive lavas. Earth and Planetary Science Letters 46, 419–430.
Stolper, E. & Walker, D. (1980). Melt density and the average composition of basalt. Contributions to Mineralogy and Petrology 74, 7–12.
Toplis, M. J., Dingwell, D. B. & Libourel, G. (1994). The effect of phosphorus on the iron redox ratio, viscosity, and density of an evolved ferro-basalt. Contributions to Mineralogy and Petrology 117, 293–304.
Vander Auwera, J. & Longhi, J. (1994). Experimental study of a jotunite (hypersthene monzodiorite): constraints on the parent magma composition and crystallization conditions (P, T, fO2) of the Bjerkreim–Sokndal layered intrusion (Norway). Contributions to Mineralogy and Petrology 118, 60–78.
Wiebe, R. A. (1990). Evidence for unusually feldspathic liquids in the Nain complex, Labrador. American Mineralogist 75, 1–12.[Abstract]
Wiebe, R. A. (1992) Proterozoic anorthosite complexes. In: Condie, K. C. (ed.) Proterozoic Crustal Evolution. Amsterdam: Elsevier, pp. 215–261.
Wilson, J. R., Robins, B., Nielsen, F. M., Duchesne, J. C. & Vander Auwera, J. (1996). The Bjerkrem–Sokndal layered intrusion, southwest Norway. In: Cawthorn, R. G. (ed.) Layered Intrusions. Amsterdam: Elsevier, pp. 231–255.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  B. R. Frost and C. D. Frost A Geochemical Classification for Feldspathic Igneous Rocks J. Petrology, November 20, 2008; (2008) egn054v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. D. Higgins and D. Chandrasekharam Nature of Sub-volcanic Magma Chambers, Deccan Province, India: Evidence from Quantitative Textural Analysis of Plagioclase Megacrysts in the Giant Plagioclase Basalts J. Petrology, May 1, 2007; 48(5): 885 - 900. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. G. Cawthorn Delayed accumulation of plagioclase in the Bushveld Complex Mineralogical Magazine, December 1, 2002; 66(6): 881 - 893. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. MUKHERJEE and S. DAS The Plagioclase-Magma Density Paradox Re-examined and the Crystallization of Proterozoic Anorthosites: a Discussion J. Petrology, October 1, 2002; 43(10): 1975 - 1978. [Full Text] [PDF]
|
|
|
|
|
|
J. S. SCOATES The Plagioclase-Magma Density Paradox Re-examined and the Crystallization of Proterozoic Anorthosites: a Reply J. Petrology, October 1, 2002; 43(10): 1979 - 1983. [Full Text] [PDF]
|
|
|
|
|
|
B. E. OWENS and R. F. DYMEK Petrogenesis of the Labrieville Alkalic Anorthosite Massif, Grenville Province, Quebec J. Petrology, August 1, 2001; 42(8): 1519 - 1546. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. O'HARA Flood Basalts, Basalt Floods or Topless Bushvelds? Lunar Petrogenesis Revisited J. Petrology, November 1, 2000; 41(11): 1545 - 1651. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||