Journal of Petrology Advance Access originally published online on November 30, 2006
Journal of Petrology 2007 48(3):435-457; doi:10.1093/petrology/egl066
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Shear Zone-hosted Migmatites (Eastern India): the Role of Dynamic Melting in the Generation of REE-depleted Felsic Melts, and Implications for Disequilibrium Melting
1Department of Geology and Geophysics, Indian Institute of Technology, Kharagpur 721302, India
2Department of Earth Sciences, Pondicherry University, Kalapet, Pondicherry 605014, India
RECEIVED JULY 11, 2005; ACCEPTED OCTOBER 24, 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGEOLOGICAL SETTINGGEOCHEMISTRYPETROGENETIC CONSIDERATIONSCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
In the Ranmal migmatite complex, non-anatectic foliated granite protoliths can be traced to polyphase migmatites. Structural–microtextural relations and thermobarometry indicate that syn-deformational segregation–crystallization of in situ stromatic and diatexite leucosomes occurred at 800°C and 8 kbar. The protolith, the neosome, and the mesosome comprise quartz, K-feldspar, plagioclase, hornblende, biotite, sphene, apatite, zircon, and ilmenite, but the modal mineralogy differs widely. The protolith composition is straddled by element abundances in the leucosome and the mesosome. The leucosomes are characterized by lower CaO, FeO+MgO, mg-number, TiO2 , P2O5 , Rb, Zr and total rare earth elements (REE), and higher SiO2 , K2O, Ba and Sr than the protolith and the mesosome, whereas Na2O and Al2O3 abundances are similar. The protolith and the mesosome have negative Eu anomalies, but protolith-normalized abundances of REE-depleted leucosomes show positive Eu anomalies. The congruent melting reaction for leucosome production is inferred to be 0·325 quartz+0·288 K-feldspar+0·32 plagioclase+0·05 biotite+0·014 hornblende+0·001 apatite+0·001 zircon+0·002 sphene=melt. Based on the reaction, large ion lithophile element, REE and Zr abundances in model melts computed using dynamic melting approached the measured element abundances in leucosomes for >0·5 mass fraction of unsegregated melts within the mesosome. Disequilibrium-accommodated dynamic melting and equilibrium crystallization of melts led to uniform plagioclase composition in migmatites and REE depletion in leucosome.
KEY WORDS: migmatite; REE; trace element; partial melting; P–T conditions
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGGEOCHEMISTRYPETROGENETIC CONSIDERATIONSCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Barring a few granulite terrains (Clemens, 1990
; Vielzeuf et al., 1990), the expected levels of incompatible element depletion in the lower crust (McCarthy, 1976; Tarney & Windley, 1977; Nesbit, 1980; Taylor & McLennan, 1985) corresponding to large-scale melt extraction cannot be explained by batch or fractional melting models (Rudnick, 1990, 1992; Watt & Harley, 1993). Experimental results, field studies and theoretical arguments (Vigneresse et al., 1996; Brown & Solar, 1998; Sawyer, 2001; Vanderhaeghe, 2001; Mecklenburgh & Rutter, 2003; Barraud et al., 2004) emphasize that melt generation, segregation and dispersal in mineral-supported matrix depend critically on strain imposed by the far-field stress. For small-volume fraction of melting (Vigneresse et al., 1996), melts in a strained partially molten crust are unable to migrate from the source region and therefore crystallize in situ. With increase in melt volume, melts segregate along a network of dislocations and grain boundaries, and eventually the segregated melts escape when the melt volume exceeds the melt escape threshold (Vigneresse et al., 1996). Melts segregated by stress-induced continuous extraction are rapidly separated from the residual solids experiencing melting. Therefore, a single melt phase in complete or partial equilibrium with the solid undergoing anatexis, as in classical models of batch and fractional melting (Schilling & Winchester, 1967; Gast, 1968; Shaw, 1970), may not be an adequate analogue for melt generation and migration in a deforming crust experiencing melting. Instead, dynamic melting (Langmuir et al., 1977; McKenzie, 1985) approximates syn-deformation melting more closely because dynamic melting involves two melt components; for example, an unsegregated melt that remains in equilibrium with the residual solid, and a segregated melt that is continuously extracted from the residual melt (see Vigneresse & Burg, 2000), and is isolated from the residual solid (Zou, 1998; Zou & Reid, 2001). The consequences of dynamic melting on the petrogenesis of mantle-derived basaltic rocks are well known (Langmuir et al., 1977; McKenzie, 1985; Williams & Gill, 1989; Sobolev & Shimizu, 1992), but the application of dynamic melting in explaining element abundances in crustally derived melts has not been attempted. Melt segregation, migration and accumulation are independent processes that may operate prior to, synchronous with, and after deformation. Clearly, the temporal relations between melt emplacement and crystallization processes and the far-field stress are crucial in determining melt extraction processes in a deforming partially molten crust. The present study examines the effects of syn-deformational dynamic melting on trace and rare earth element (REE) abundances in leucosome based on structural, microtextural and geochemical characterization of a ductile shear zone hosted migmatite complex.
|
GEOLOGICAL SETTING |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGGEOCHEMISTRYPETROGENETIC CONSIDERATIONSCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
In the northwestern sector of the Eastern Ghats Mobile Belt (Fig. 1), granulites are thrust against Archaean to Palaeoproterozoic cratonic lithologies to the west along a north–south-trending east-dipping ductile shear zone (Gupta et al., 2000
; Bhadra et al., 2003, 2004). The granulites are represented by Grt±Opx-bearing quartzofeldspathic and Bt+Sil-bearing metapelite gneisses with distended bands of mafic granulites, calcsilicate granulites and Mg, Al-metapelites. Hornblende–biotite granite with amphibolite-facies enclaves of metamafic and calc-silicate gneisses constitute the cratonic foreland. Cratonic granites are also exposed in north–south elongated domains within the main mass of granulites (Gupta et al., 2000; Bhadra et al., 2003, 2004). In one such domain near the village of Ranmal (Fig. 1), foliated granite protoliths grade from non-migmatitic varieties to polyphase migmatites (Fig. 1).
|
, n=140); (3) mean orientation of S2 (32° to 076°N; dashed line); (4) F2 fold axes (
; n=34). The F2 fold axes lie on the average S2 plane, indicating the non-cylindrical nature of F2 folds.In the Ranmal migmatite complex, non-migmatitic granites are characterized by a monophase fabric (S1, Fig. 2a) defined by biotite±hornblende-rich discontinuous layers of a few millimetres width and ribbon quartz (with sweeping deformation bands, chessboard twinning and sub-grains; Fig. 3a) that wrap around partly to wholly recrystallized K-feldspar augen. The foliated granite protolith grades from unmelted yet metamorphosed rocks into migmatitic varieties (Fig. 4). In the sparingly migmatitic varieties, the leucocratic segregations occur as unconnected pods and lenses, with zonal development of S1 and S2 (see below) fabrics (Fig. 4). With increasing melt fraction, leucosomes mantled by biotite+hornblende-rich selvedges (melanosome) segregated parallel to the augen-defined S1 fabric (Figs 2b, c and 4).
|
|
|
Mesoscopic folds on S1 are rare in non-migmatitic foliated granites, except in zones proximal to partially melted granites (Figs 2a and 4). The S1 fabric is folded into foreland-vergent, asymmetric, non-cylindrical F2 folds, with development of a set of NNE–SSW-trending, east-dipping shear fabric S2 (Figs 1 and 4). The intensity of S1 and S2 fabrics and the tightness of the F2 folds intensify with increasing volume fraction of melts. Diatexite leucosomes parallel to S2 shears (Figs 2c, d and 4) are coarser grained, show pinch-and-swell structures, invariably disrupt the stromatic layering (Fig. 2c), and contain coarse euhedral K-feldspar grains sub-parallel to the leucosome walls (Fig. 2d). Mafic selvedges (melanosome) rimming diatexite leucosomes are absent.
The stromatic leucosomes (quartz>K-feldspar>plagioclase, biotite and hornblende, with accessory amounts of apatite, ilmenite, zircon, titanite) are characterized by coarse, sub-equant and weakly strained grains that share randomly dispersed interphase boundaries (Fig. 3b and c). However, trains of S1 biotite grains and subhedral to anhedral plagioclase grains show weak shape-preferred orientation. Rare plagioclase grains with well-developed rational faces grow into neighboring quartz and K-feldspar films (Fig. 3b). The quartz–microcline boundary is plumose, with regularly spaced, optically homogeneous quartz lobes protruding into the feldspar grains (Fig. 3c). Quartz grains have amoeboid shapes and share bulbous margins against feldspar grains (Fig. 3c). Larger quartz grains show a chessboard extinction pattern (Fig. 3d). Minerals in the melanocratic layers are identical to those in leucosomes, but are dominated by biotite (Fig. 3e) and hornblende, with volumetrically subordinate amounts of quartz, plagioclase>K-feldspar and ilmenite, and accessory phases apatite, zircon and apatite.
In diatexites, the grain and phase boundaries are highly non-equilibrated and interdigitating as in igneous rocks. Nearly all plagioclase grains are optically homogeneous and euhedral to subhedral in shape. Tiled grains of plagioclase (Fig. 3f) enclosed in quartz are common.
Temporal relation between migmatization and deformation
The correlated increase in the intensity of S1 or S2 fabrics and the tightening of folds on S1 with increased melt fraction suggests that deformation and melting were synchronous processes (Hand & Dirks, 1992; Brown & Solar, 1998). Trains of tiled grains of euhedral plagioclase in diatextite leucosomes parallel to leucosome walls are instantly recognizable evidence of deformation-aided melt-supported crystal flow (Fig. 3f). Syn-deformational emplacement of diatexite leucosomes is supported by pinch-and-swell structure in igneous-textured diatexite leucosomes. In stromatic leucosomes, subhedral plagioclase grains with rational faces growing into neighboring quartz films (Fig. 3b) are indicative of the presence of melts during deformation (see Garlick & Gromet, 2004). Plumose quartz grains against K-feldspars in anatectic migmatites and the absence of such textures in neighboring melt-free foliated granite protolith suggest the presence of melts during high-temperature deformation (see Berger & Rosenberg, 2003). The apophytic nature of stromatic (Figs 2a and 4) and diatexite leucosomes (Figs. 2c and d) and the textural relations (Figs. 3b and f) do not support a cumulus origin for the leucosomes. The outcrop-scale variation between non-migmatitic foliated granite and its migmatitic counterpart (Figs 2a and 4) cannot be explained by variation in temperature or in composition, except for the higher proportion of biotite in the migmatitic part. Melting and deformation were promoted by H2O released by prograde dehydration reaction. Focused (Johnson et al., 2003) fluxing of the released water lowered the solidus for melting higher in the column, leading to strain softening in the axial zones of F2 folds (Figs 2a and 4; see Hand & Dirks, 1992).
Mesocratic domain vis-à-vis foliated granite
The mesocratic domains adjacent to the neosomes (=leucosome+melanosome) show tonal variations manifested by the alternation of thin (< 1 cm wide) biotite + hornblende-rich layers and wider (up to 10 cm) quartz+feldspar-rich layers parallel to S1 (Figs 2b, c and 4). Because the tonal variation in the mesosome is not a pre-melting feature of the melt-free foliated granite protolith, the feature was acquired synchronous with melting and deformation. The mesosomes have textural features similar to those of leucosomes; for example: (1) the preponderance of weakly strained to unstrained grains having high-energy boundaries (see Garlick & Gromet, 2004); (2) weak shape-preferred alignment among strain-free subhedral feldspar grains with their rational faces parallel to and limited by aggregates of biotite showing prominent lattice preferred alignment (parallelism of 001 cleavage; Milord & Sawyer, 2003); (3) plumose margins of quartz against subhedral feldspars; (4) quartz films along feldspar grain boundaries. The similar-to-leucosome microtextures in mesosomes, the absence of deformation textures akin to non-migmatitic granite protoliths (Fig. 3a), and the tonal variations taken together suggest that melts were present during high-temperature deformation of mesosomes.
|
GEOCHEMISTRY |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGGEOCHEMISTRYPETROGENETIC CONSIDERATIONSCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Analytical techniques
Analyses of minerals were carried out using a Cameca four-channel wavelength-dispersive spectrometry microprobe analyzer [analytical details have been provided by Mukhopadhyay & Bhattacharya (1997
)]. Bulk chemical analyses were made of samples weighing at least 5 kg. As a result of the laminated nature of the migmatites, thin (
3 mm wide) melanosome layers uncontaminated by the leucosome and the mesosome proved difficult to obtain. The melanosome layers in the samples were carefully ‘shaved off’ by diamond cutter, and discarded. The remaining portions of leucosome and mesosome were disintegrated into smaller fragments that were crushed using a Fritsch tungsten carbide jaw crusher to chips smaller than 2 mm. In the case of diatexites (not rimmed by melanosome), 2 cm wide portions along the mesosome–leucosome interface were sliced off so that the leucosome and the mesosome were not contaminated by each other. After repeated coning and quartering, the samples were ground to finer that 800 mesh in a tungsten-carbide ball mill. X-ray fluorescence (XRF) analyses on the fused rock pellets for major and trace elements (Sr, Ba, Rb, Zr and Hf) were made using a Philips-PW 1480 instrument at the Mineralogisch-Petrologisches Institute (Bonn). The OXIQUANT software package (Vogel & Kuipers, 1987) was used for correcting peak overlap and matrix effects. The rock powders were analyzed by inductively coupled plasma mass spectrometry (ICP-MS) for whole-rock REE and other trace elements at the National Geophysical Research Laboratory, Hyderabad, India using an ELAN DRC II system. Instrumental parameters were as follows: RF power 1100 W; argon gas-flow—nebulizer 0·86 l/min, auxiliary 1·20 l/min and plasma 15 l/min; lens voltage 5·00 V; sample uptake rate 0·80 ml/min. GSJ Standard (Japan), and JG2 were used as standards for analysis (Govindaraju, 1994).
Mineral chemistry
For the samples analyzed, non-migmatitic granite protoliths, leucosomes and mesosomes are characterized by overlapping compositions of plagioclase (An22–29), amphibole (ferropargasite; XMg0·25–0·46) and biotite (XMg0·32–0·50; TiO22·53–4·67 p.f.u.) (Table 1). Ti content in biotite in leucosomes is lower compared with those in the mesosome (Fig. 5a). For leucosome–mesosome pairs, the albite content in liquidus plagioclase is similar or slightly higher compared with plagioclase in the mesosome and the protolith, although bulk Na2O/CaO ratio varies considerably (Fig. 5b). In diatexite leucosomes, XAb in coarse liquidus plagioclase (long axis >1000 µm) lies in the narrow range of ±1 mol%.
|
Major and trace elements
The stromatic and diatexite leucosomes (Table 2) are metaluminous to weakly peraluminous (‘Shand's index’; Maniar & Piccoli, 1989
), whereas the non-migmatitic granite protoliths are metaluminous. The leucosomes plot close to, but do not correspond to, the experimental minimum melt compositions for moderate to low a(H2O) (< 0·5) and pressure 5–10 kbar (Fig. 6; Johannes & Holtz, 1996). The compositional disparity between leucosomes and ‘minimum’ melt suggests that some components of the leucosome may not represent melts. The leucosomes are enriched in SiO2 and K2O, and depleted in CaO, FeO+MgO, mg-number and TiO2 (Fig. 7a–h) relative to the protolith and the mesosome. Barring sample RW21B, the leucosomes are characterized by higher concentration of Ba (Fig. 8a). Sr is marginally (< 75 ppm) enriched in the leucosome (Fig. 8b), whereas Rb in the leucosome is marginally (< 50 ppm) depleted (Fig. 8c), although sample RW21B in Fig. 8b and c shows an oppositely correlation. P2O5, Zr, REE and Hf are depleted in the leucosome relative to the mesosome (Fig. 8d–g). Ba and Rb in leucosomes are inversely correlated (Fig. 8h).
|
|
and bulk-rock (Na2O/CaO) between protolith, leucosome and mesosome. Bold lines join leucosome–mesosome pairs. Stromatic migmatite: •, leucosome; 
, leucosome; 
, mesosome. Star, non-migmatitic foliated granite (protolith). Vertical bars in (b) indicate the range of XAb in each sample.
|
|
|
The Zr abundance in leucosomes varies from 120 to 350 ppm (Fig. 9a). Zr concentrations in the leucosomes from the interior of the migmatite complex are either comparable with or higher than the equilibrium Zr abundance predicted from experimentally determined zircon solubility in felsic melts (Watson & Harrison, 1983
). The higher-than-equilibrium Zr values in leucosomes imply entrainment of zircon from the protolith. This is supported by the existence of complexly zoned zircon in leucosomes (Fig. 9b) similar to those hosted within biotite in the mesosome (Fig. 9c). Smaller unzoned zircon grains in the leucosomes may have crystallized from melts (Fig. 9d). On the other hand, the measured Zr abundance in the highly strained margin of the migmatitic complex is lower than the predicted Zr abundance. The lower-than-equilibrium Zr concentrations in these leucosomes suggest that melt extraction rates were higher than rates of zircon dissolution, or zircons may have been effectively filtered by the solid residua and trapped in the mesosome without sufficient time in contact with the melt to attain chemical equilibrium. Apparently, deformation-aided disequilibrium melting (see Sawyer, 1991) was responsible for leucosome generation along the strongly deformed marginal zones of the migmatite complex.
|
Rare earth elements
The protolith and the ‘mesosome’ are characterized by negative Eu anomalies and enrichment of light REE (LREE) over heavy REE (HREE) (Fig. 10a). In comparison, stromatic and diatexite leucosomes are depleted in REE (Fig. 10a and b). Whereas stromatic leucosomes do not show a Eu anomaly (Fig. 10a), diatexite leucosomes are characterized by varying magnitudes of negative Eu anomalies (Fig. 10b). Typically, leucosomes with lower
REE are those with a smaller chondrite-normalized negative Eu anomaly (Fig. 10a and b) or protolith-normalized larger positive Eu anomaly (Fig. 10c). Chondrite-normalized REE-depleted peraluminous leucogranites produced by partial melting of metasedimentary protoliths are known (Sevigny et al., 1989; Wark & Miller, 1993; Watt & Harley, 1993; Nabelek & Glascock, 1995). This study is possibly the first report of REE-depleted leucosomes produced by partial melting of metaluminous granite.
|
The
REE–Eu variation in the leucosomes may reflect the effects of fractional crystallization; for example, diatexites may correspond to melts derived by fractionation of plagioclase, and the stromatic leucosomes represent the complementary plagioclase-rich cumulates. This possibility does not explain the following features: (1) the overlapping abundances of plagioclase-compatible and -incompatible elements (Figs 7 and 8) in diatexites and stromatic leucosomes; (2) the similar or higher-than-protolith REE abundances with negative Eu anomalies in the nearest neighbor mesosomes; (3) the intrusive nature of stromatic leucosomes (Figs 2a and 4); (4) the high melt/crystal ratio in stromatic leucosomes suggested by trains of euhedral plagioclase and grain tiling (see Den Tex, 1969; Nicolas, 1992; Nicolas & Ildefonse, 1996; Figs 2d and 3f); (5) the absence of chemical zoning in magmatic plagioclase, and the lack of compositional differences between plagioclase in stromatic and diatexite leucosomes; (6) the discordant relationship between stromatic leucosomes and diatexites (Fig. 2c). Alternatively, the leucosomes may represent weakly or unfractionated melts containing varying proportion of restitic and liquidus phases. |
PETROGENETIC CONSIDERATIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGGEOCHEMISTRYPETROGENETIC CONSIDERATIONSCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
If the segregated melts (leucosomes) and adjacent melanosome (=neosome) constitute a closed system, then the mesocratic domain beyond the neosome must approximate the protolith composition. In the case of open-system melting, if all melts were expelled from the mesosome, the mesosome and the leucosome should plot on opposite sides of the non-migmatitic granites in Harker-type variation diagrams. If partial melt loss occurred in mesosomes, the collinearity between plotted compositions of the mesosome, the leucosome and the protolith will still be preserved, but the composition of the mesosome will be smeared along the line bracketed by the hypothetical melanosome and the segregated leucosome.
In Figs 7 and 8, the leucosomes and the corresponding mesosome straddle the protolith composition. Most mesosomes (Figs 7 and 8) plot in and around the protolith composition, but a few samples (RW28, RW13 and RW14; all located near the margin of the migmatite complex) are far removed from the protolith–mesosome cluster. Because the protolith represents the pre-melting composition of the migmatites, the tie-line for the first group of ‘mesosomes’ implies that either (1) a small fraction of melt was expelled from the protolith, if the volume fraction of total melt generated was small, or (2) the mesosome contains a substantial volume of residual melt, if the volume fraction of melt generated was large. For the set of samples from the marginal parts of the Ranmal complex (Fig. 2), the relative positions of the plotted mesosome composition suggest that a significantly larger volume fraction of melt was extracted from the marginal zone of the complex, compared with the more centrally located parts. The larger fraction of extracted melt is to be expected because of the prevalent higher strain (see Barraud et al., 2004) along the margin.
P–T conditions of melting
The P–T conditions of melting were retrieved from a garnet–hornblende–plagioclase–orthopyroxene–quartz assemblage (Fig. 11a) in a mafic granulite (RW10; location shown in Fig. 1) mylonite located 2 m from the margin of the migmatite complex and hornblende–plagioclase–quartz assemblages in mesosomes of migmatites. For the mafic granulite (analytical data in Table 3), temperatures were estimated from garnet–orthopyroxene pairs using the thermometric formulations of Sen & Bhattacharya (1984) and Lee & Ganguly (1988), and from hornblende–plagioclase pairs based on the formulation of Holland & Blundy (1994). Pressures were calculated using the barometric formulations of Bohlen et al. (1983), Perkins & Chipera (1985) and Moecher et al. (1988) for the Fe end-member reaction, Opx+Pl=Grt+Qtz. P–T intersections of the thermobarometers yield 800±50°C and 8·0±1 kbar (Fig. 11b). The P–T values are similar to the thermobarometric estimates obtained for migmatites using the formulations of Holland & Blundy (1994) and the recently proposed barometric formulation of the reaction tremolite+tschermakite+2 albite=2 pargasite+8 quartz (Bhadra & Bhattacharya, 2006). The estimated temperature is consistent with chessboard extinction (T700°C; Kruhl, 1996; Rosenberg & Berger, 2001; Garlick & Gromet, 2004) in deformed quartz grain in the leucosome (Fig. 3d). Apparently, melting and deformation in the Ranmal migmatite complex occurred at comparable temperature.
|
|
The model melting reaction
In the absence of incongruently produced minerals, congruent melting of protolith phases was assumed for leucosome generation. The modal proportion (and composition) of mineral phases in the leucosomes therefore corresponds to the melting proportion of the phases in the protolith. The modal abundances of minerals were determined by the matrix solution method using the equation (Nabelek, 1999
) |
|
|
| (1) |
|
| (2) |
0·2 wt% water. At the P–T conditions of melting (800°C and 8 kbar), a ‘dry’ melt could not have existed, and therefore, melting was in all likelihood fluid-present. As the proposed P–T conditions are above the experimentally determined water-present incongruent melting reaction of biotite, melting may have been controlled by focused fluid flux. An externally derived fluid source, in addition to internally generated fluid by biotite or hornblende breakdown, appears to be a realistic scenario for melt generation, especially as foliated granite protoliths are observed to physically grade into migmatites over short distances (Fig. 4).
|
|
Trace element modeling
During melting or crystallization, fractionation of large ion lithophile elements (LILE; Rb, Sr, and Ba) is determined by silicate phases, but the behavior of accessory phases strongly influences the fractionation of high field strength elements (HFSE) and REE. Among the accessory phases, zircon, sphene and apatite do not discriminate between Eu and the middle REE (MREE) (Table 5; Henderson, 1984
, fig. 1.14). Therefore, the protolith-normalized positive Eu anomaly in the REE-depleted leucosomes cannot be explained by equilibrium (batch) or fractional melting, although the overall decrease in REE may be explained. REE bulk partition coefficients in Table 5 computed from reactions (1) and (2) indicate that, for conditions of batch and fractional melting, the mesosome is expected to retain Eu in preference to other REE. This would cause the melt to be depleted in Eu, but other REE should be enriched for reaction (1) and depleted for (2). The abundance of Eu in the leucosomes is opposite to this trend (Fig. 10c). Similarly, for the modal mineralogy (Table 4), Sr and Ba are predicted to be depleted in the melt, whereas Rb and Zr should be enriched in the melt. However, the measured abundances of the elements show opposite relationships (Figs 7 and 8). Clearly, equilibrium (batch) or fractional melting processes cannot account for the measured abundances of LILE, HFSE and REE in the leucosomes. REE depletion in leucosomes has been explained by preferential retention or incomplete dissolution of accessory phases (Sawyer, 1991; Watt & Harley, 1993; Johannes et al., 1995; Nabelek & Glascock, 1995; Rapp & Watson, 1995; Bea, 1996; Solar & Brown, 2001). None of these disequilibrium melting processes involving the accessory phases will lead to Eu enrichment in melts relative to the protolith in the Ranmal migmatites, nor will these options account for the measured abundances of the LILE.
|
Alternatively, mesosome microtextural (Fig. 3) relations and petrogenetic considerations (Figs 7 and 8) indicate that variable proportions of melt may have been locked up within mesosomes and were not segregated to form leucosomes. Trace element abundances in leucosomes were computed using dynamic melting based on the formulation of Zou (1998
, 2000) and Zou & Reid (2001). Calculations were carried out using modal dynamic melting (MCM; constant fraction of each solid phase and constant bulk partition coefficient) and non-modal dynamic melting (NCM; bulk partition coefficient varying as a function of changing solid fraction). The concentration of an element (CL) in the extracted melt (leucosome) subject to modal and non-modal melting, respectively, can be expressed following Zou (1998) and Zou & Reid (2001) as
|
| (3) |
|
| (4) |
are mass fraction of the extracted melt and residual melt, respectively. These two parameters are related to the degree of partial melting (F) by F =+(1–)·X (Zou, 1998; Zou & Reid, 2001). D0 is the bulk distribution coefficient of the mineral phases in the protolith and is expressed as 
, where 
is the mass fraction of a particular mineral phase (i) in the protolith. P0 is defined as 
, where 
is the fractional contribution of mineral i to the melting reaction and 
is the distribution coefficient of an element between mineral i and the liquid. However, as residua from their anatectic source were probably entrained in melts, bulk partition coefficients computed using leucosome mineralogy and mineral–melt partition coefficients are not without their inaccuracies. is obtained from the equation 
, 
being the mass of a particular mineral phase (i) involved in the congruent melting reaction. Model concentrations of Rb, Sr, Ba, Zr, Ce, Eu, and Yb in the extracted melt relative to the protolith (CL/C0) were computed from equation (3) (MCM-1 and MCM-2; Fig. 13a and b) and equation (4) (NCM-1 and NCM-2; Fig. 13c and d) for different values (0·5, 0·8 and 0·95) at X=0·2 and X=0·5.
|
). The vertical bars correspond to the range of measured abundances for each element; the median values of elements in leucosomes are joined by dashed lines. The values of NCM and MCM computed Ba abundances in model melts merely approach the lowest values of measured Ba abundances in the leucosomes (Fig. 13). For Ba, the measured abundances (CL) need to be scaled down by
50% to compare favorably with the computed abundances at high values. On other hand, the measured Rb and Sr abundances in the leucosomes are adequately explained by both modal and non-modal congruent melting, but for values >0·5 (Fig. 13). The agreement between the computed and measured abundances of Ba and Rb may be improved if unmelted biotite and K-feldspar from the protolith are assumed to become entrained in melts during melt migration and segregation. Subtracting the entrained biotite and K-feldspar ‘xenocrysts’ from the leucosome will have a twofold effect. First, the ‘true melt’ abundances of Ba and Rb in leucosomes will be scaled downward. Second, because the two elements are compatible with biotite and K-feldspar, Ba and Rb bulk partition coefficients will be reduced. The reduction in bulk partition coefficients will cause Ba to become less compatible, and Rb will become incompatible. This in turn will lead to upward revision of iso- CL/C0 values of Ba and Rb in melt (Fig. 13). Although the nets changes cannot be quantified because the proportion of xenocryst minerals in melts is unknown, it appears that the measured and computed abundances are likely to correspond better, especially for values >0·5.
Among the REE, Eu abundances (Fig. 13) in leucosomes (especially diatexites), are best explained at high values (>0·5) by modal congruent melting with or without accessory phases, and by non-modal congruent melting reaction (1); that is, without accessory phases. For similar range of values, the measured abundances of Ce and Yb in diatexites are best explained by a non-modal congruent melting model with and without involving accessory phases [reaction (1)]. The measured Zr abundances in diatexites are explained only by non-modal congruent melting with or without accessory phases, and again only for high values. Summarizing, abundances of trace elements in diatexites, except Ba, are adequately explained by non-modal congruent melting based on reaction (1), as opposed to a modal congruent melting model. In contrast, the measured abundances of REE and HFSE such as Zr in stromatic leucosomes are at best approached by the model melt calculations. The disparity may be traced to errors in computing bulk partition coefficients from leucosomes that contain varying proportion of entrained protolith phases.
Mesosome vis-à-vis anatexis
Bulk partition coefficients computed based on the mineralogy of leucosome containing an unknown proportion of unmelted protolith minerals are at best semi-quantitative. In spite of this limitation, the inferred close correspondences between measured abundances of elements in leucosomes and element abundances computed in model melts, albeit for high mass fraction (>50%) of residual melts, are encouraging. By implication, substantial fractions of melt generated during anatexis were not segregated as leucosomes and instead remained locked up elsewhere in the partially melted rock. Ever since Mehnert (1968) published his classic work on migmatites, the debate on the origin of migmatites has centered on field observations and petrological–geochemical characterization of neosomes (=leucosome+melanosome). The relevance of mesosome to the melting process, however, has received inadequate attention (see Johannes & Gupta, 1982; Johannes, 1988). The enigmatic nature of the mesosome was succinctly presented by Johannes & Gupta (1982; p. 114): ‘If mesosomes (paleosomes) do not represent the parent rock, then what are they?’
Mesosomes depicted in numerous publications show considerable mineralogical heterogeneity. The following is a list of some of these depictions: Brown (1994, fig. 1; p. 86); Johannes (1988, fig. 3; p. 456); Johannes & Gupta (1982, figs 6 and 7; p. 117); Vanderhaeghe (2001, fig. 2d; p. 215); Olsen (1982, plate 1; p. 1602); Watt et al. (1996, fig. 2a; p. 103). The tonal variations within the mesosomes in the illustrations are sharp, repetitive, and not systematic or gradational with respect to the neosome as is to be expected if the mesosome formed as a result of systematic layer-perpendicular extraction or addition of material related to neosome formation. One suggestion could be that the heterogeneities may have been acquired prior to melting. This is difficult to assess in rocks characterized by pre-melting mineralogical anisotropy or that were multiply deformed following melting. However, in the Ranmal migmatite complex, the foliated granite protoliths are mineralogically and structurally isotropic, and therefore the tonal variations in the mesosome must reflect changes imposed by melting and melt extraction processes. The complementary element abundances in the mesosome and leucosome suggest a communion between the segregated melts in the neosome and the unsegregated melts. The results of dynamic modeling demonstrate that a considerable portion of the melt generated during crustal anatexis was actually locked up as residual melt dispersed within the mesosome matrix as films, layers and discontinuous lenses. It is tempting to suggest that melt extraction even in zones of intense far-field stress (as in this study) may not be an efficient mechanism for melt extraction in partially molten rocks. Taking the argument a step further, the chemical disparity between the leucosome and the mesosome is one of varying fractions of melt localization; that is, leucosomes constitute higher volume fraction of segregated melts.
Uniformity of plagioclase composition
The uniformity of plagioclase composition in the leucosome and residual rocks (Misch, 1968; Johannes, 1988; Fitzsimons, 1996; Jung et al., 1998) is a paradox because experimentally determined melts in equilibrium with plagioclase are known to have a lower Ca/Na ratio. We may envisage a scenario in which melting may proceed at rates faster than time scales of diffusive element redistribution in plagioclase; that is, the consuming margin of plagioclase retreats inwards into the grain at a velocity faster than the rate of Na+Si
Ca+Al coupled exchange (see Prasad et al., 2005). Disequilibrium melting of plagioclase into its own composition for conditions of rapid melting is similar to impact-induced flash melting reported in meteorites (Scott et al., 1997; James et al., 2003). In such a case, the surface of the plagioclase grain will melt into its own composition. Equilibrium crystallization of such disequilibrium melts will produce plagioclase having the same composition as the partially melted plagioclase. In other words, liquidus plagioclase in the extracted melt and the restite plagioclase will have the same composition. Because of the lack of textural evidence favoring hydration induced by fluid released from solidifying melts, the extent and effects of fluid- or melt-driven chemical rehomogenization (see Fourcade et al., 1992) of melt and solid residua, although unassessed, may have been small.
Eu anomaly in REE-depleted leucosome
Disequilibrium melting leads to decreasing incompatibility of incompatible elements, the effect being higher for the more incompatible elements (see Bedard, 1989; Qin, 1992). In the case of rapid melting of plagioclase, the equilibrium mineral–melt partitioning of REE will cease if the surface of feldspar grains is consumed at rates faster than the rate of NaSi=CaAl exchange (see Prasad et al., 2005). For the extreme condition, disequilibrium melting ensures that REE in feldspars will be wholly partitioned into the melt phase (e.g. 
), and the REE pattern of the resultant melt will be essentially parallel to that of the protolith. The depleted nature of the leucosomes relative to protolith should then be explained by the entrainment of variable proportions of plagioclase xenocrysts in melts. It stands to reason that higher proportions of plagioclase entrainment will lead to lower REE and reduced chondrite-normalized negative Eu anomaly in melts (see Fig. 10a and b), or protolith-normalized larger positive Eu anomaly (Fig. 10c). Taking the argument a step forward, stromatic leucosomes with lower REE and lacking chondrite-normalized negative Eu anomalies must contain higher proportions of entrained plagioclase xenocrysts, compared with the diatexite leucosomes.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGGEOCHEMISTRYPETROGENETIC CONSIDERATIONSCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
This study demonstrates that dynamic non-modal congruent melting—as opposed to batch and fractional melting—provides improved convergence of measured and computed abundances of LILE, REE and HFSE in syn-deformational in situ melts, albeit for a substantial weight fraction of the residual melt locked up in the partially melted rock. Mesosomes characterized by element abundances complementary to those in the segregated leucosomes are argued to have retained the residual fraction of the total melt produced during anatexis.
Disequilibrium melting among accessory phases has been invoked to explain the depletion of HFSE or REE in crustally derived melts (Barbey et al., 1990; Sawyer, 1991; Watt & Harley, 1993; Carrington & Watt, 1994; Bea, 1996). In this study, disequilibrium melting is extended to major rock-forming silicate phases (namely plagioclase). Disequilibrium-accommodated dynamic melting is argued to provide a possible explanation for the compositional similarity between liquidus plagioclases in leucosomes and plagioclase in the foliated granite protolith, and the REE–Eu variations in leucosomes that are REE depleted relative to the protolith.
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGGEOCHEMISTRYPETROGENETIC CONSIDERATIONSCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Supplementary data for this paper are available at Journal of Petrology online.
|
ACKNOWLEDGEMENTS |
|---|
The XRF analysis and electron probe microanalysis were carried out in the Mineralogisch-Petrologisches Institut, Universität Bonn. We express our gratitude to Professor Michael Raith (Universität Bonn) for providing access to the EPMA and XRF facilities at the University of Bonn. The DST–DAAD–PPP exchange program provided financial support to S.B. for the analytical work. The ICP-MS work was carried out at NGRI, Hyderabad. Dr V. P. Dimri, Director, is thanked for extending the analytical facility. The CSIR is thanked for financial support (24/263/03 EMR II) for the work. The authors are grateful to the journal reviewers P. O'Brien, T. Slagstad and G. Stevens for their in-depth analysis of the manuscript, and for educative and constructive suggestions that improved the content, style and overall presentation of the manuscript. We are also indebted to Nick Arndt for helping us to tie up the loose ends. The manuscript greatly benefited from these reviews.
*Corresponding author. Present address: Department of Geology and Geophysics, Indian Institute of Technology, Kharagpur 721302, India. Telephone: (+91)-3222-283350. Fax: (+91)-3222-255303/282268. E-mail: abbhat55{at}yahoo.com
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGGEOCHEMISTRYPETROGENETIC CONSIDERATIONSCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Barbey P, Macaudiere J, Nzenti JP. (1990) High pressure dehydration melting of metapelites: evidence from the migmatites of Yaounde, Cameroon. Journal of Petrology 31:401–427.
Barraud J, Gardien V, Allemand P, Grandjean P. (2004) Analogue models of melt flow networks in folding migmatites. Journal of Structural Geology 26:307–324.[CrossRef][Web of Science]
Bea F. (1996) Residence of REE, Y, Th and U in granites and crustal protoliths; implications for the chemistry of crustal melts. Journal of Petrology 37:521–552.
Bedard JH. (1989) Disequilibrium mantle melting. Earth and Planetary Science Letters 91:359–366.[CrossRef][Web of Science]
Berger A and Rosenberg CL. (2003) Preservation of chemical residue–melt equilibria in natural anatexite: the effects of deformation and rapid cooling. Contributions to Mineralogy and Petrology 144:416–427.[Web of Science]
Bhadra S and Bhattacharya A. (2006) The barometer tremolite + tschermakite+2albite=2pargasite+8quartz: constraints from experimental data at unit silica activity, with application to garnet-free natural assemblages. American Mineralogist (in press).
Bhadra S, Banerjee M, Bhattacharya A. (2003) Tectonic restoration of a polychronous mobile belt–craton assembly: constraints from corridor study across the western margin of the Eastern Ghats Belt, India. Memoir, Geological Society of India (Milestone in Petrology) 52:109–130.
Bhadra S, Gupta S, Banerjee M. (2004) Structural evolution across the Eastern Ghats Mobile Belt–Bastar craton boundary, India: hot over cold thrusting in an ancient collision zone. Journal of Structural Geology 26:233–245.[CrossRef][Web of Science]
Bohlen SR, Wall VJ, Boetcher AL. (1983) Geobarometry in granulites. In Saxena SK (Ed.). Kinetics and Equilibrium in Mineral Reactions. Advances in Physical Geochemistry 3: pp. 141–171.
Brown M. (1994) The generation, segregation, ascent and emplacement of granite magma: the migmatite-to-crustally-derived granite connection in thickened orogens. Earth-Science Reviews 36:83–130.
Brown M and Solar GS. (1998) Granite ascent and emplacement during contractional deformation in convergent orogens. Journal of Structural Geology 20:1365–1393.[CrossRef][Web of Science]
Carrington DP and Watt GR. (1994) Contrasting geochemistry of granulite-facies leucosomes and S-type granites: K-feldspar a culprit. Mineralogical Magazine 58a:152–153.
Clemens JD. (1990) The granite–granulite connexion. In Vielzeuf D and Vidal Ph (Eds.). Granulites and Crustal Evolution(Kluwer, Dordrecht) pp. 25–36.
Deer WA, Howie RA, Zussman J. (1985) An Introduction to the Rock-forming Minerals(English Language Book Society, Longman, Hong Kong) pp. 528.
Den Tex E. (1969) Origin of ultramafic rocks, their tectonic settings and history: a contribution to the discussion of the paper: ‘The origin of ultramafic and ultrabasic rocks’ by P. J. Wyllie. Tectonophysics 7:457–488.[CrossRef][Web of Science]
Fitzsimons ICW. (1996) Metapelite migmatites from Battstrand Bluff, East Antarctica—metamorphism, melting and exhumation of the mid crust. Journal of Petrology 37:395–414.
Fourcade S, Martin H, de Brémond d' Ard J. (1992) Chemical exchange in migmatites during cooling. Lithos 28:43–53.[CrossRef][Web of Science]
Garlick SR and Gromet LP. (2004) Diffusion creep and partial melting in high temperature mylonitic gneisses, Hope Valley Shear Zone, New England Appalachians, USA. Journal of Metamorphic Geology 22:45–62.[CrossRef][Web of Science]
Gast PW. (1968) Trace element fractionations and the origin of tholeiitic and alkaline magma type. Geochimica et Cosmochimica Acta 32:1057–1086.[CrossRef][Web of Science]
Govindaraju K. (1994) . Geostandards Newsletter 18 (Special Issue): pp. 158.
Gupta S, Bhattacharya A, Raith M, Nanda JK. (2000) Contrasting pressure–temperature–deformation history across a vestigial craton–mobile belt boundary: the western margin of Eastern Ghats belt at Deobhog, India. Journal of Metamorphic Geology 18:683–697.[CrossRef][Web of Science]
Hand M and Dirks PHGM. (1992) The influence of deformation on the formation of axial planar leucosomes and the segregation of small melt bodies within the migmatitic Napperby Gneiss, central Australia. Journal of Structural Geology 14:591–604.[CrossRef][Web of Science]
Henderson P. (1984) Rare Earth Element Geochemistry(Elsevier, Amsterdam) pp. 510.
Holland T and Blundy J. (1994) Non ideal interactions in calcic amphiboles and their bearing on amphibole–plagioclase thermometry. Contributions to Mineralogy and Petrology 116:433–447.[CrossRef][Web of Science]
James OB, Cohen BA, Taylor A. (2003) Lunar meteorite DHOFAR 026: a shocked granulite breccia, not an impact melt. Lunar and Planetary Sciences 34:1149–1150.
Johannes W. (1988) What controls partial melting in migmatites? Journal of Metamorphic Geology 6:451–465.[Web of Science]
Johannes W and Gupta LN. (1982) Origin and evolution of migmatite. Contributions to Mineralogy and Petrology 79:114–123.[CrossRef][Web of Science]
Johannes W and Holtz F. (1996) Petrogenesis and Experimental Petrology of Granitic Rocks(Springer, Berlin) pp. 335.
Johannes W, Holtz F, Muller P. (1995) REE distribution in some layered migmatites; constraints on their petrogenesis. Lithos 35:135–152.
Johnson TE, Gibson RL, Brown M, Buick IS, Cartwright I. (2003) Partial melting of metapelitic rocks beneath the Bushveld Complex, South Africa. Journal of Petrology 44:789–813.
Jung S, Mezger K, Masberg P, Hoffer P, Hoerness S. (1998) Petrology of an intrusion related high grade migmatite: implication for partial melting of metasedimentary rocks and leucosome forming processes. Journal of Metamorphic Geology 16:425–445.[CrossRef][Web of Science]
Kruhl JH. (1996) Prism- and basal-plane parallel subgrain boundaries in quartz: a micro-structural geothermometer. Journal of Metamorphic Geology 14:581–589.[CrossRef][Web of Science]
Langmuir CH, Bender JF, Bence AE, Hanson GN, Taylor SR. (1977) Petrogenesis of basalts from the FAMOUS-area, Mid-Atlantic ridge. Earth and Planetary Science Letters 36:133–156.[CrossRef][Web of Science]
Lee HY and Ganguly J. (1988) Equilibrium compositions of coexisting garnet and orthopyroxene: experimental determination in the system FeO–MgO–Al2O3–SiO2 and applications. Journal of Petrology 29:93–113.
Maniar PD and Piccoli PM. (1989) Tectonic discrimination of granitoids. Geological Society of America Bulletin 101:635–643.
McCarthy TS. (1976) Chemical interrelationships in a low-pressure granulite terrain in Namaqualand, South Africa, and their bearing on the granite genesis and the compositions of the lower crust. Geochimica et Cosmochimica Acta 40:1057–1068.[CrossRef][Web of Science]
McKenzie D. (1985) The extraction of magma from the crust and mantle. Earth and Planetory Science Letters 74:81–91.
Mecklenburgh J and Rutter RH. (2003) On the rheology of partially molten synthetic granite. Journal of Structural Geology 25:1575–1585.[CrossRef][Web of Science]
Mehnert KR. (1968) Migmatites and the Origin of Granitic Rocks(Elsevier, Amsterdam) pp. 393.
Milord I and Sawyer EW. (2003) Schlieren formation in diatexite migmatite: examples from the St. Malo migmatite terrane, France. Journal of Metamorphic Geology 21:347–362.[Web of Science]
Misch P. (1968) Plagioclase compositions and non-anatectic origin of migmatite gneisses in Northern Cascade Mountains of Wahington State. Contributions to Mineralogy and Petrology 17:1–70.[CrossRef]
Moecher DP, Essene J, Anovitz LM. (1988) Calculation and application of clinopyroxene–garnet–plagioclase–quartz barometers. Contributions to Mineralogy and Petrology 100:92–106.[CrossRef][Web of Science]
Mukhopadhyay AK and Bhattacharya A. (1997) Tectonothermal evolution of the gneiss complex at Salur in the Eastern Ghats Granulite Belt of India. Journal of Metamorphic Geology 15:719–734.[CrossRef][Web of Science]
Nabelek PI. (1999) Trace element distribution among rock-forming minerals in Black Hills migmatites, South Dakota: a case for solid-state equilibrium. American Mineralogist 84:1256–1269.[Abstract]
Nabelek PI and Glascock MD. (1995) REE-depleted leucogranites, Black Hills, South Dakota: a consequence of disequilibrium melting of monazite-bearing schist. Journal of Petrology 36:1055–1071.
Nesbit HW. (1980) Genesis of the new Quebec and Adirondack granulites: evidence for their production by partial melting. Contributions to Mineralogy and Petrology 72:303–310.[CrossRef][Web of Science]
Nicolas A. (1992) Kinematics in magmatic rocks with special reference to gabbros. Journal of Petrology 33:891–915.
Nicolas A and Ildefonse B. (1996) Flow mechanism and viscosity in basaltic magma chambers. Geophysical Research Letters 23:2013–2016.[CrossRef][Web of Science]
Olsen SN. (1982) Open- and closed-system migmatites in the Front Range, Colorado. American Journal of Science 282:1596–1622.
Perkins D and Chipera SJ. (1985) Garnet–orthopyroxene–plagioclase–quartz barometry: refinement and application to the English River subprovince and the Minnesota River Valley. Contributions to Mineralogy and Petrology 89:69–80.[CrossRef][Web of Science]
Prasad S, Bhattacharya A, Raith MM, Bhadra S. (2005) The origin of orthopyroxene/biotite+plagioclase coronas from the Bolangir anorthosite complex (India), and implications for reconstructing P–T paths. American Mineralogist 90:291–303.
Qin Z. (1992) Disequilibrium partial melting model and its implications for trace element fractionations during mantle melting. Earth and Planetary Science Letters 112:75–90.[CrossRef][Web of Science]
Rapp RP and Watson EB. (1995) Dehydration melting of metabasalt at 8–32 kbar: implication for continental growth and crust–mantle recycling. Journal of Petrology 36:891–931.
Rollinson H. (1993) Using Geochemical Data: Evaluation, Presentation, Interpretation(Longman, Singapore) pp. 352.
Rosenberg CL and Berger A. (2001) Syntectonic melt pathways in granite, and melt induced transition in deformation mechanisms. Physics and Chemistry of the Earth (A) 26:287–293.[CrossRef]
Rudnick RL. (1990) Continental crust—growing from below. Nature 347:711–712.[CrossRef]
Rudnick RL. (1992) Restites, Eu anomalies, and the lower continental crust. Geochimica et Cosmochimica Acta 56:963–970.[CrossRef][Web of Science]
Sawyer EW. (1991) Disequilibrium melting and the rate of melt residuum separation during migmatization of mafic rocks from the Grenville Front, Quebec. Journal of Petrology 32:701–738.
Sawyer EW. (2001) Melt segregation in the continental crust: distribution and movement of melt in anatectic rocks. Journal of Metamorphic Geology 19:291–310.[CrossRef][Web of Science]
Schilling JG and Winchester JW. (1967) Rare-earth element fractionation and magmatic processes. In Runcorn SK (Ed.). Mantle of the Earth and Terrestrial Planets(Interscience, New York) pp. 267–283.
Scott ER, Yamaguchi A, Krot AN. (1997) Petrological evidence for shock melting of carbonates in the martian meteorite ALH84001. Nature 387:377–379.[CrossRef]
Sen SK and Bhattacharya A. (1984) An orthopyroxene–garnet thermometer and its application to the Madras charnockites. Contributions to Mineralogy and Petrology 88:64–71.[CrossRef][Web of Science]
Sevigny JH, Parrish RR, Ghent ED. (1989) Petrogenesis of peraluminous granites, Monashee Mountains, Southeastern Canadian Cordilera. Journal of Petrology 30:557–581.
Shaw DM. (1970) Trace element fractionation during anatexis. Geochimica et Cosmochimica Acta 34:237–243.[CrossRef][Web of Science]
Sobolev AV and Shimizu N. (1992) Superdepleted melts and ocean mantle permeability. Doklady Rossiyskoy Akademii Nauk 326:354–360.
Solar GS and Brown M. (2001) Petrogenesis of migmatites in Maine, USA: possible source of peraluminous leucogranite in plutons. Journal of Petrology 42:789–823.
Tarney J and Windley BF. (1977) Chemistry, thermal gradients and evolution of the lower continental crust. Journal of the Geological Society, London 134:153–172.
Taylor SR and McLennan SM. (1985) The Continental Crust: its Composition and Evolution(Blackwell, Oxford) pp. 312.
Vanderhaeghe O. (2001) Melt segregation, pervasive melt migration and magma mobility in the continental crust: the structural record from pores to orogens. Physics and Chemistry of the Earth 26:213–223.[CrossRef]
Vielzeuf D, Clemens JD, Pin C, Moinet E. (1990) Granites, granulites and crustal differentiation. In Vielzeuf D and Vidal Ph (Eds.). Granulites and Crustal Evolution(Kluwer, Dodrecht) pp. 59–85.
Vigneresse JL and Burg JP. (2000) Continuous vs. discontinuous melt segregation in migmatites: insights from a cellular automation model. Terra Nova 12:188.[CrossRef][Web of Science]
Vigneresse JL, Barbey P, Cuney M. (1996) Rheological transitions during partial melting and crystallization with application to felsic magma segregation and transfer. Journal of Petrology 37:1579–1600.
Vogel W and Kuipers G. (1987) A pre-calibrated program for geological applications. Phillips New Developments, X-Ray Spectrometry TA 11:2–8.
Wark DA and Miller CF. (1993) Accessory mineral behavior during evolution of a peraluminous granite suite, Sweetwater Wash Pluton, California. Chemical Geology 110:49–67.[CrossRef][Web of Science]
Watson EB and Harrison TM. (1983) Zircon saturation revisited: temperature and composition effects in a variety of crustal magma types. Earth and Planetary Science Letters 64:295–304.[CrossRef][Web of Science]
Watt GR and Harley SL. (1993) Accessory phase controls on the geochemistry of the crustal melts and restites produced during during water-undersaturated partial melting. Contributions to Mineralogy and Petrology 114:550–566.[CrossRef][Web of Science]
Watt GR, Burns IM, Graham GA. (1996) Chemical characteristics of migmatites: accessory phase distribution and evidence for fast melt segregation rates. Contributions to Mineralogy and Petrology 125:100–111.[CrossRef][Web of Science]
Williams RW and Gill JB. (1989) Effect of partial melting on the uranium decay series. Geochimica et Cosmochimica Acta 53:1607–1619.[CrossRef][Web of Science]
Zou H. (1998) Trace element fractionation during modal and nonmodal dynamic melting and open-system melting: a mathematical treatment. Geochimica et Cosmochimica Acta 62:1937–1945.[CrossRef][Web of Science]
Zou H. (2000) Modeling of trace element fractionation during non-modal melting with linear variation in mineral/melt distribution coefficients. Geochimica et Cosmochimica Acta 64:1095–1102.[CrossRef][Web of Science]
Zou H and Reid MR. (2001) Quantitative modeling of trace element fractionation during incongruent dynamic melting. Geochimica et Cosmochimica Acta 65:153–162.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||