| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Petrology Volume 42 Number 4 Pages 789-823 2001
© Oxford University Press 2001
Petrogenesis of Migmatites in Maine, USA: Possible Source of Peraluminous Leucogranite in Plutons?
LABORATORY FOR CRUSTAL PETROLOGY, DEPARTMENT OF GEOLOGY, UNIVERSITY OF MARYLAND, COLLEGE PARK, MD 20742, USA
Received June 24, 1999; Revised typescript accepted July 4, 2000
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGEOLOGY OF WESTERN MAINEPETROLOGY AND FIELD RELATIONS...GEOCHEMISTRYBIOTITE COMPOSITIONDISCUSSIONCONCLUSIONSAPPENDIXREFERENCES |
|---|
In Maine, Siluro-Devonian turbidites were metamorphosed under high-T–low-P facies series conditions during deformation within a Devonian crustal-scale shear zone system, defined by kilometer-scale straight belts of apparent flattening strain that anastomose around lozenges of apparent constrictional strain. At upper amphibolite facies grade, metapelites are partially melted, the onset of which is recorded by a migmatite front. The resulting migmatites are stromatic or heterogeneous, and smaller-volume granites form sheets or cylinders according to the structural zone in which they occur, suggesting that migmatites and granites record syntectonic melt flow through the deforming crust. Common leucogranite of the nearby coeval Phillips pluton, which was emplaced syntectonically, was sourced from crustal rocks with geochemical characteristics similar to those of the host Siluro-Devonian succession. Migmatites have melt-depleted compositions relative to metapelites. Leucosomes are peraluminous and represent the cumulate products of fractional crystallization and variable loss of evolved fractionated liquid. Among the heterogeneous migmatites are schlieric granites, the geochemistry of which suggests melt accumulation before fractional crystallization and loss of the evolved liquid. Smaller-volume granites are peraluminous with a range of chemistries that reflect variable entrainment of residual plagioclase and biotite, accumulation of products of fractional crystallization and loss of most of the evolved liquid. Common leucogranite of the Phillips pluton and larger granites in the migmatites have compositions that suggest crystallization of evolved liquids derived by fractional crystallization of primary muscovite dehydration melts. We infer that the leucogranite represents the crystallized fugitive liquid from a migmatite source similar to that exposed nearby. Water transported through the shear zone system dissolved in melt was exsolved at the wet solidus to cause retrogression in sub-solidus rocks and retrograde muscovite growth in migmatites.
KEY WORDS: anatexis of pelite; Maine; migmatite; peraluminous granite; plutons
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF WESTERN MAINEPETROLOGY AND FIELD RELATIONS...GEOCHEMISTRYBIOTITE COMPOSITIONDISCUSSIONCONCLUSIONSAPPENDIXREFERENCES |
|---|
The process of generation, segregation, ascent and emplacement of granite magma during orogeny has important implications because melt transfer affects the thermal and rheological behavior of the crust during orogenesis (e.g. Collins & Vernon, 1991
; Stüwe et al., 1993; Brown & Solar, 1999). Our conception of how melt is generated and segregated is well developed (e.g. Wickham, 1987; Johannes, 1988; Allibone & Norris, 1992; Sawyer, 1994, 1998; Brown et al., 1995; Rushmer, 1996; Milord et al., 2001). We also understand well how granite magma is emplaced in both extensional and contractional tectonic settings (e.g. Hutton & Reavy, 1992; Grocott et al., 1994; Paterson et al., 1996; Benn et al., 1998; Brown & Solar, 1998b; Cruden, 1998; Paterson & Miller, 1998). However, the mechanism by which melt is transferred from source to sink during orogeny remains a matter of debate (e.g. Clemens & Mawer, 1992; D’Lemos et al., 1992; Brown, 1994; Rutter, 1997; Sawyer, 1998; Brown & Solar, 1999; Miller & Paterson, 1999; Weinberg, 1999).
Within migmatites, the geometry of leucosomes and smaller-volume granites may record the melt flow network through the crust (e.g. Brown & Rushmer, 1997; Sawyer, 1998; Brown & Solar, 1999; Brown et al., 1999), particularly so if the leucosomes do not record solid-state strain. At the outcrop scale, the presence of granite located in structurally controlled sites within migmatite, such as interboudin partitions and strain shadows (e.g. Stromgard, 1973), fractures and fold hinge zones (e.g. Collins & Sawyer, 1996), and dilatant shear surfaces (e.g. Brown, 1994) suggests melt flow through the migmatite during deformation (e.g. Brown & Rushmer, 1997). In a partially molten rock, melt is segregated during deformation by moving down gradients in melt pressure to create leucosomes (Brown, 1994; Brown et al., 1995; Rutter, 1997; Marchildon & Brown, 2001). If deformation and melting are coeval, cyclic inflation and collapse of melt flow conduits caused by build-up of melt pressure and periodic draining of the source moves melt through the crust (e.g. Brown & Solar, 1998a, 1999; Weinberg, 1999).
This paper reports variations in the structure, mineral assemblage and geochemistry of migmatites, and the form, petrography and geochemistry of granites from an area in western Maine, USA. The area includes the northern limit of migmatite in the Central Maine belt, which marks the termination of a diachronous ‘metamorphic high’ that extends from eastern Connecticut, through central Massachusetts and New Hampshire, into Maine [formally called the ‘Acadian metamorphic high’ by, for example, Schumacher et al. (1990)]. In an earlier study we have shown that regional metamorphism and crystallization of migmatite leucosomes and granites were coeval (Solar et al., 1998). In this study we compare the geochemistry of metapelite source rocks, migmatites and leucogranites to evaluate the hypothesis that migmatites and leucogranites in western Maine are cogenetic. Our conclusions refute the common belief that there is no genetic relationship between migmatites and leucogranites, and have implications for the petrogenesis of migmatites and leucogranites in other orogens.
|
GEOLOGY OF WESTERN MAINE |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF WESTERN MAINEPETROLOGY AND FIELD RELATIONS...GEOCHEMISTRYBIOTITE COMPOSITIONDISCUSSIONCONCLUSIONSAPPENDIXREFERENCES |
|---|
Tectonic setting
The Central Maine belt (CMB) of the northern Appalachians is the principal tectonostratigraphic unit of the eastern part of New England and New Brunswick (Fig. 1). The CMB lies between Ordovician rocks of the Bronson Hill belt (BHB) to the WNW (Ratcliffe et al., 1998
, and references therein), and Neoproterozoic-to-Silurian rocks of the Avalon Composite Terrane (ACT) to the SSE (e.g. West et al., 1995; Cocks et al., 1997). The CMB is composed of a succession of Siluro-Devonian turbidites (interlayered pelite and psammite) that was deformed and metamorphosed at greenschist to upper amphibolite facies conditions during the Early Devonian Acadian orogeny (e.g. Bradley, 1983; Smith & Barreiro, 1990; Robinson et al., 1998). These metasedimentary rocks were intruded syntectonically by Early to Middle Devonian plutons (e.g. Bradley et al., 1998; Solar et al., 1998).
|
|
Strain was partitioned during Devonian dextral transpression of the northern Appalachians (Fig. 1) so that inboard oblique contraction was accommodated within the CMB shear zone system (Solar & Brown, 1999
, 2001) whereas concomitant dextral-transcurrent displacement was accommodated within the Norumbega shear zone system (West & Hubbard, 1997; West, 1999). Deformation had become localized within the Norumbega shear zone system by Carboniferous time (Hubbard et al., 1995; West & Hubbard, 1997; Ludman, 1998; West, 1999), which juxtaposed the ACT against the CMB by orogen-parallel translation.
Regional structural geology and metamorphism
The CMB shear zone system is the principal Devonian (Acadian) structure of western Maine (Fg. 1). It consists of two types of kilometer-scale structural zones (Solar & Brown, 2001). These are NE-striking straight belts of steeply to vertically dipping planar and moderately to steeply plunging linear structures associated with tight folds that alternate with and anastomose around zones in which the orientations and development of planar structures are variable and folds are gentle to open. The preferred orientation of bladed muscovite and ribbons or rods of polycrystalline quartz aggregates define the penetrative metamorphic mineral fabrics that are distinct in each structural zone (S > L fabrics in straight belts; L >> S fabrics in intervening zones). We interpret these fabrics to record contrasting styles of finite strain, and we refer to these as zones of apparent flattening-to-plane strain (AFZs; Fig. 1) and zones of apparent constrictional strain (ACZs; Fig. 1). Solar & Brown (2001) have postulated that this pattern indicates deformation localization, with greater strain accommodation, and therefore larger tectonic displacements within the AFZs. The pelite layers exhibit porphyroblast–matrix microstructures that suggest syntectonic growth of porphyroblast minerals during progressive tightening of regional-scale folds (Solar & Brown, 1999, 2000, 2001). Because a well-developed NE-plunging mineral elongation lineation, defined by the same metamorphic minerals at each grade, penetrates all zones, Solar & Brown interpreted matrix mineral growth as syntectonic. Thus, the fabrics of the prograde metamorphic minerals recorded the regional tectonic strain ellipsoid.
The high-T–low-P metamorphic field gradient reflects syntectonic polymetamorphism (e.g. Guidotti, 1989; Solar & Brown, 1999) related to pluton-driven thermal pulses (De Yoreo et al., 1989) that overprint a regionally elevated thermal gradient (Brown & Solar, 1999). Separate periods of metamorphism (e.g. Guidotti, 1963, 1989, 1993; Holdaway et al., 1982; Chamberlain & England, 1985; Eusden & Barreiro, 1988; Smith & Barreiro, 1990) are interspersed with plutonism (e.g. Tomascak et al., 1996; Bradley et al., 1998; Solar et al., 1998). Metamorphic grade in the area of Fig. 1 varies from garnet zone through staurolite zone and lower sillimanite zone to upper sillimanite zone, achieving anatexis in metapelites along the high-T prograde path at depths equivalent to 
15 km (Holdaway et al., 1997; Brown & Solar, 1998b). The age of regional metamorphism is 405 to 399 ± 2 Ma, based upon U–Pb data from metamorphic monazite grains from upper amphibolite-facies mica schist (Smith & Barreiro, 1990). In comparison, concordant U–Pb zircon and monazite ages from granite bodies in migmatite, as well as plutons, range from c. 408 to c. 404 Ma (Solar et al., 1998); this suggests that granite emplacement was contemporaneous with metamorphism (Brown & Solar, 1998a, 1998b, 1999). Because metamorphic mineral fabrics developed syntectonically (Solar & Brown, 1999, 2000), it follows that both metamorphism and plutonism were synchronous with deformation.
|
PETROLOGY AND FIELD RELATIONS OF MIGMATITE |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF WESTERN MAINEPETROLOGY AND FIELD RELATIONS...GEOCHEMISTRYBIOTITE COMPOSITIONDISCUSSIONCONCLUSIONSAPPENDIXREFERENCES |
|---|
The Tumbledown and Weld anatectic domains (TAD and WAD, respectively; Fig. 2) are separated based upon map pattern and regional structure (Solar & Brown, 2001
). Migmatite in these domains varies from strongly foliated metasedimentary rock with a few millimeter-scale leucosomes per square meter (Fig. 3), in which relict primary structures are preserved, to rocks structurally disrupted by the migmatization process (diatexis; see Brown, 1973; Fig. 4) and schlieric granite (Fig. 5a). Leucosome density and disruption of relict primary structures both increase across strike from the migmatite front (Solar, 1999). Similarly, sheets and cylinders of granite progressively dominate outcrops of migmatite both along and across strike of migmatite layers (Figs 3d and 5b). Many outcrops of migmatite in both the TAD and WAD consist of domains and/or blocks in which some relict primary pelite–psammite interlayers are preserved. Layers in such blocks resemble those common in the stratigraphic succession outside the TAD and WAD, down-temperature from the migmatite front, which suggests the protolith of the migmatites was the CMB metasedimentary succession. This is supported by the continuation of the regional structure across the migmatite front (Fig. 1).
|
|
The TAD and WAD consist of two subdomains where the structure of the migmatite and the distribution of leucosome reflect the pattern of strain. We map these subdomains as stromatic and heterogeneous migmatite that correspond typically, but not precisely, to AFZs and ACZs, respectively. This structural relation led Brown & Solar (1999) to interpret heterogeneous migmatite to be within the cores of regional thermal antiforms (in ACZs), flanked by stromatic migmatites (in AFZs). Transition zones are present between the migmatite subdomains, and between vein migmatite and diatexite within the heterogeneous migmatite (Fig. 2). Commonly, bodies of schlieric granite are found within transition zones (see Fig. 5a) that occur at AFZ–ACZ boundaries. In the northern part of the TAD and WAD, layers are progressively eliminated across strike from the migmatite front by increasing volume of leucosome and disruption by apparent flow, as stromatic migmatite grades into diatexite.
Migmatite domain 1: stromatic migmatite
Approximately half of the exposed migmatite in western Maine is stromatic (Fig. 3; Menhert, 1968), characterized by a planar structure in which each layer is mineralogically and texturally distinct. This type of migmatite is found mostly in AFZs (Fig. 2), and corresponds to metatexite (Brown, 1973; Ashworth, 1985), being composed of discrete millimeter- or centimeter-thick sheet-like but discontinuous bodies of granite (leucosome) separated from medium-colored high-grade metamorphic host rock by dark-colored selvedges (melanosome). The orientations of migmatite layers and mineral fabrics are concordant (Fig. 3), as reflected by the consistent steeply dipping orientation of these structures at all scales within each structural zone (Fig. 2). At the regional scale, the layers are parallel to those of metasedimentary rocks in the same structural zone (Fig. 2, see stereograms).
Leucosomes are trondhjemitic, being composed of plagioclase, quartz, biotite and muscovite (Fig. 3); they make up 3 vol. % of stromatic migmatite at outcrop. Millimeter-scale leucosomes range from 1 to 25 cm in length, whereas centimeter-scale leucosomes are typically 1–2 m long; both have low width-to-length ratios of 0·05 (Fig. 3a and b). Melanosomes range from 0·1 to 0·6 mm, rarely up to 1 mm, in thickness and are invariably in contact with leucosomes. They are composed of >80 vol. % biotite, accompanied by fibrolite, minor quartz, plagioclase and retrograde chlorite. Biotite grains, generally 1 mm long, are clustered, and show a strongly preferred orientation that defines a lepidoblastic texture and a foliation parallel to leucosome edges. The intervening host rock layers are 2–10 cm thick. They contain biotite, quartz, sillimanite (mostly fibrolite), garnet, pyrrhotite and/or ilmenite, muscovite (skeletal), and locally plagioclase, tourmaline and clinozoisite. Typically, fibrolite has grown at the expense of primary muscovite in the foliation; the fibrolite forms a fabric in addition to the penetrative biotite foliation, and these sillimanite–biotite folia alternate with quartz–feldspar folia (Fig. 3c). Fabrics are oriented sub-parallel to fabrics in the metasedimentary rocks outside the migmatite domains down-temperature from the migmatite front (Fig. 2, see stereograms). Elongate fibrolite aggregates define a steeply plunging lineation visible in the field, and elongate quartz aggregates define a weak sub-horizontal lineation seen only in cut hand specimens and suitably oriented thin sections.
In leucosomes, grains are equant and anhedral with average sizes ranging from 0·1 mm in the thinner leucosomes to 4 mm in the thicker leucosomes. Quartz is present in approximately equal proportion to plagioclase. An exception is rare leucosomes in which quartz is >70 vol. %, where the quartz grains are <0·1 to 3 mm in diameter and plagioclase is 1 mm in length. Quartz shows undulatory extinction, except in grains <0·1 mm in diameter. Quartz grains in the quartz-rich leucosomes display serrated edges to suggest grain-size reduction. In leucosomes <1 mm thick, a magmatic foliation is defined by biotite and opaque mineral grains (10 µm). Leucosomes >5 mm thick consist of plagioclase, with normal zoning in several irregular or patchy concentric shells, and interstitial quartz. Some leucosomes may contain up to 5 vol. % anhedral K-feldspar. Coarser irregular biotite is inferred to be residual, whereas finer-grained euhedral biotite grains are inferred to be neocrystalline. Although textures generally appear to be the result of crystallization in the presence of a melt phase (see Vernon & Collins, 1988) and, therefore, undeformed, many leucosomes of <1 cm thickness are locally overprinted by solid-state strain that has produced common sub-grain boundaries in quartz or serrated edges of quartz accompanied by coronas of grains (<0·1 mm in diameter) that lack sub-grain boundaries.
Leucosomes show ‘pinch-and-swell’ structure in three dimensions (Fig. 3b). A longer wavelength in the sub-horizontal dimension suggests that the maximum apparent ‘pinch’ is sub-vertical and down-dip, consistent with kinematic indicators in the metasedimentary rocks (Solar & Brown, 2001). This triaxial, oblate shape is similar to that defined by the mineral grains in the host rock layers. Centimeter- to meter-scale sub-vertical tabular bodies of granite cut outcrops of stromatic migmatite at concordant to weakly discordant angles to the planar structures. Many of these granite sheets are composite (Fig. 3d; see also Brown & Solar, 1999), and most have a ‘pinch-and-swell’ structure with a longer wavelength in the sub-horizontal direction. In one area, within 1 km along strike a progressive increase occurs in the proportion of meter-scale composite granite sheet to host stromatic migmatite such that the migmatite becomes disrupted ultimately to occur only as isolated schollen in granites (Fig. 5b) that make up a sheeted granite complex.
|
Migmatite domain 2: heterogeneous migmatite
A regular planar structure is absent in the remainder of the exposed migmatite in western Maine; we refer to this type of migmatite as heterogeneous. This type of migmatite is found exclusively in ACZs. The orientation of grain-shape fabrics and geometry of leucosomes vary more in these rocks than in the stromatic migmatite (Fig. 2, see stereograms). Weak foliations and lineations in heterogeneous migmatite are defined by sillimanite (mostly fibrolite) and biotite (Fig. 4b).
There are two types of heterogeneous migmatite, vein migmatite in the south and SW and diatexite in the north and NE [Fig. 2; terminology after Menhert (1968) and Brown (1973), as modified by Ashworth (1985)]. Contacts between the two types are gradational over tens of meters in transition zones. Vein migmatite shows phlebitic structure (Menhert, 1968), and meter-scale compositional layers interpreted to be relict from the protolith (Fig. 4e). Diatexite, in contrast, is a rock in which the protolith structures are not observed, suggesting destruction by diatexis (e.g. Fig. 4a and b; see Brown, 1973; Sawyer, 1998). Vein migmatite shows sharp leucosome contacts (Fig. 4e and f). In contrast, contacts between leucocratic and melanocratic domains in diatexite are diffuse and gradational at the centimeter scale (Fig. 4c).
Vein migmatite
Vein migmatite occurs as meter-scale units alternating with weakly migmatized meter-scale semi-pelitic units that together reflect original metasedimentary layers. Centimeter-scale pod- or lens-shaped trondhjemitic leucosomes are separated by centimeter-scale anastomosing darker host layers similar to the melanosomes of stromatic migmatite (Fig. 4e). Leucosomes make up 15 vol. % on outcrop surfaces, and display ‘pinch-and-swell’ structure (Fig. 4e and f), in which thickness varies from 3 to 10 cm. Individual pods are up to 20 cm long, yielding width-to-length ratios up to 0·5.
Leucosomes consist of subequant and anhedral quartz, plagioclase and K-feldspar with no discernible grain-shape fabric. Larger leucosomes may contain anhedral garnet grains of 1–2 mm diameter. Quartz grains have subgrain boundaries and undulatory extinction, but to a lesser degree than counterparts in stromatic migmatite. Quartz generally displays the largest grains ($$$5 mm in diameter), whereas plagioclase is usually 1 mm in length, and grains commonly show inhomogeneous normally compositionally zoned concentric shells of 50 µm width.
The melanosomic host rock contains sillimanite (mostly fibrolite), biotite, muscovite, plagioclase, quartz, pyrrhotite and/or ilmenite, garnet and locally K-feldspar. Sillimanite is found as clots within both muscovite and plagioclase (Fig. 4f). Some prismatic sillimanite grains are up to 0·1 mm across, but sillimanite more commonly occurs in clots up to 5 mm in diameter. Biotite (1–3 mm in length), muscovite, and elongate untwinned plagioclase (up to 6 mm in length) all show a distinct grain-shape fabric, defining a strong moderately dipping foliation (variable dip direction) and weak, moderately plunging, down-dip lineation.
Diatexite
Diatexite varies at outcrop from ‘patchy’ leucosome-dominated to biotite–sillimanite-dominated rock, and outcrop to outcrop from schlieren-rich migmatite to schlieric granite with schollen of vein migmatite and unmigmatized calc-silicate-rich psammite (Fig. 5a). Most types are characterized by a discontinuous, weakly defined foliation of variable attitude (Fig. 4a). Leucosomes and leucocratic domains make up 9–15 vol. % of diatexite outcrop surfaces, and are generally uniformly distributed.
Discrete leucosomes appear as centimeter-scale quartzo-feldspathic mineral segregations that vary from diffuse to sharp at their margins (Fig. 4c). Shapes of leucosomes are subequant in pavement outcrops, 1–2 cm in diameter, and elongate down-dip of the sillimanite fabric in the host rock, with lengths up to 20 cm and width-to-length ratios of up to 0·1. Some leucosomes anastomose along their lengths; these show a fabric defined by biotite and opaque minerals that we infer to be of magmatic origin, and formed where melt and entrained crystals flowed around residue and/or unmelted protolith. Thus leucosomes tend to be rod-shaped, with a long dimension that plunges moderately to steeply ENE. This linear structure is sub-parallel to both the weakly defined mineral elongation lineation in the host rock and the strongly defined mineral elongation lineation in the metasedimentary rocks outside the migmatite domains down-temperature from the migmatite front (Fig. 2). These fabrics define a triaxial, strongly prolate shape similar to that of the metasedimentary rocks in the same structural zones (ACZs; Fig. 2).
Leucosomes contain quartz, plagioclase, biotite, pyrrhotite and/or ilmenite, and locally K-feldspar and muscovite. Grains are 0·5 mm in size with subequant and anhedral shapes. Some quartz grains have moderate undulatory extinction with a few sub-grain boundaries. Plagioclase commonly shows several irregular or patchy normally compositionally zoned concentric shells of 50 µm width. Larger biotite with concentrations of opaque mineral grains at the edges is inferred to be residual, whereas smaller euhedral grains without the opaque grains are inferred to be neocrystalline.
Although the dark host rock is unlike the unmigmatized protolith it is not truly melanosomic, making the usual migmatite nomenclature unsuitable in this case. Diffuse domains of the host rock interfinger with the leucosome at the millimeter-scale, and consist of biotite, sillimanite (mostly fibrolite) and garnet, with accessory muscovite, plagioclase, quartz, pyrrhotite and/or ilmenite, and K-feldspar. Fibrolite has grown at the expense of muscovite and untwinned plagioclase to form a younger generation of foliation-forming minerals. Different proportions of these minerals account for the gradual variation of diatexite from more quartzo-feldspathic (leucocratic) to more ferro-magnesian (melanocratic) types. In the extreme case, either schlieric granite is formed (leucocratic diatexite; Fig. 5a), or the mineral assemblage is dominated by biotite, sillimanite and garnet with <10 vol. % plagioclase + quartz, to give the rock a melanocratic appearance (Fig. 5c and d). In all varieties of diatexite, sillimanite is found as clots within both plagioclase and retrograde muscovite. Although grain size varies, it is generally 3 mm, with sillimanite clots up to 5 mm. Biotite (1–3 mm in length), muscovite and elongate untwinned plagioclase (up to 6 mm in length) show a preferred grain-shape fabric that defines a weak, moderately to steeply dipping foliation and strong, moderately ENE-plunging, down-dip lineation.
Most outcrops of diatexite are cut by meter-scale, cylindrical granite bodies (Fig. 4d). These granite bodies are elongate subparallel to the mineral lineation in the diatexite, and to the rod-shaped leucosomes [see Brown & Solar (1999) for a discussion]. Entrained blocks of strongly foliated biotite–garnet schist are found in the interior of the granite cylinders. The granite cylinders lack a fabric, except proximal to the margin of these blocks (Fig. 5c).
Interpretation
In both subdomains, the common matrix hosting the leucosome is sillimanite (mostly fibrolite), biotite, garnet, quartz, plagioclase, opaque phases (usually ilmenite), and coarse, skeletal muscovite books that cut the fabric and are interpreted to be a retrograde feature. Fibrolite and biotite are the main fabric-forming phases (Fig. 3c), and fibrolite has apparently grown at the expense of primary fabric-forming muscovite to suggest it was produced by the breakdown of muscovite. Migmatite leucosomes are discrete to diffuse with a common mineralogy of plagioclase, quartz, muscovite, and locally K-feldspar and biotite. The microstructure of leucosomes shows crystal faces and mineral films along grain boundaries that suggest some crystallization from melt (Sawyer, 1999), and melt-present formation. Locally, leucosomes exhibit biotite foliation that is concordant with fabrics in the adjacent host rock.
On the basis of the petrography, the melt-producing reaction at the migmatite front probably was
 |
 |
). At a depth of 15 km these reactions indicate T of >700°C. Given the limited amount of water-rich metamorphic volatile phase that can be stored in rocks at upper amphibolite facies conditions, we expect that melting will be dominated by the muscovite dehydration reaction. Patiño Douce & Harris (1998) investigated experimentally melting of metapelites similar to those in the CMB, using two schists with different modes from the hanging wall of the Main Central Thrust in the Himalayas.
In Maine, the absence of primary muscovite in migmatites, in comparison with the metasedimentary rocks outside the migmatite front where muscovite averages 25 vol. % in the mode, and the universal occurrence of sillimanite as a fabric-forming phase with biotite in the migmatites suggest that the material that hosts the leucosome is depleted of melt. Further, the generally K-feldspar-poor nature of the leucosomes suggests melt has been lost from the migmatites as a whole. These observations are consistent with syntectonic migmatization, and consequent syntectonic melt extraction from the migmatite domains. We will evaluate this postulate in the light of the contemporaneous deformation after describing the geochemistry of all rock types, but for the remainder of this paper we will refer to darker host rock that does not form a distinct melanosome like those in the stromatic migmatite as melt-depleted host rock.
The Phillips pluton
Immediately NE of the WAD (Fig. 2) is the coeval Phillips pluton, which is interpreted to be hemi-ellipsoidal with long dimension parallel to the regional moderately NE-plunging lineation (Brown & Solar, 1998b, 1999; Pressley & Brown, 1999). It is located in an ACZ, similar to the diatexites, and it has a similar geometry to the smaller-volume cylinders of granite found in the heterogeneous migmatite domains. These observations have been used to suggest a relationship between structure, granite ascent and emplacement (Brown & Solar, 1999). The geochemistry of common leucogranite (95%) from the Phillips pluton has been interpreted to reflect an origin by muscovite dehydration melting of a source with geochemical characteristics similar to the metasedimentary rocks of the CMB (Pressley & Brown, 1999). The remaining 5% of the Phillips pluton is granodiorite interpreted to reflect an origin by biotite dehydration melting of a source geochemically similar to ‘Avalon-like’ rocks (see Fig. 1). The granites in the migmatites do not possess this latter component. For these reasons, we evaluate what relation exists between the migmatites, the smaller-volume granites in the migmatites and the common leucogranite of the Phillips pluton.
|
GEOCHEMISTRY |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF WESTERN MAINEPETROLOGY AND FIELD RELATIONS...GEOCHEMISTRYBIOTITE COMPOSITIONDISCUSSIONCONCLUSIONSAPPENDIXREFERENCES |
|---|
Major and trace element compositions were determined for representative whole-rock specimens of (1) pelite layers of the CMB metasedimentary rocks, (2) stromatic migmatite, (3) heterogeneous migmatite, and (4) granites from within the migmatite domains. For the Phillips pluton we use the published analyses of 10 common leucogranite specimens from Pressley & Brown (1999)
. All specimens were selected for analysis as described by Solar (1999). The metasedimentary rocks were collected from lower to upper amphibolite facies outcrops and from both AFZs and ACZs. This was done to account for any variation in composition resulting from progressive metamorphism and/or contrasting strain accommodation. The suite of metasedimentary rocks is, therefore, taken to be representative of the range of metapelite compositions of the Siluro-Devonian stratigraphic succession. Portions of the same specimens used for whole-rock analyses of two of the stromatic migmatites (95-103, Fig. 3a; and 96-286), two of the heterogeneous migmatites (vein migmatite: 96-65, Fig. 4e; diatexite: 95-35, Fig. 4c) and one of the schlieric granites (from the northern TAD transition zone: 95-215, Fig. 5a) were separated into the different structural components, and analyzed separately. Major, minor and trace element data for all rocks were analyzed by X-ray fluorescence (XRF) (Tables 1–4), and trace elements for 32 of the rocks, including the rare earth elements (REE), were analyzed by inductively coupled plasma mass spectrometry (ICP-MS) (Tables 5 and 6). Methods and analytical uncertainties concerning these data are given in the Appendix.
|
|
|
|
|
|
Metasedimentary rocks
Metasedimentary rocks range from
53 to 71 wt % SiO2 (Table 1). Al2O3 and (FeO* + MgO + TiO2) decrease, (CaO + Na2O) shows no systematic change, and K2O decreases slightly with increasing SiO2 (Fig. 6). V decreases systematically, Rb/Sr increases slightly, and Rb, Sr, Zr and Ba show no systematic change with increasing SiO2 (Fig. 7). In appropriate plots of major oxides (Fig. 6), the field defined by the sample of metasedimentary rocks lies between projected compositions of biotite, garnet and muscovite, and albite-rich plagioclase and quartz.
|
|
|
Two metasedimentary rocks from this study (95-97 and 95-173) and those reported by Cullers et al. (1974, 1997) have similar REE compositions and concentrations, and have similarly shaped chondrite-normalized REE patterns (Fig. 8). These patterns have slight negative Nd anomalies, otherwise they are smooth and straight to concave-upward in the heavy REE (HREE), with similar overall steepness over a large range of LaN/LuN (5·5–38). Although the two specimens of this study lie at the lower end of this range (Table 5), they have similar LaN/SmN and GdN/LuN ratios to those of the larger sample of Cullers et al. (1974, 1997). All patterns have negative Eu anomalies (Eu/Eu* = 0·65 and 0·88 for the two specimens of this study). Overall, the shape and steepness of the patterns are similar to those of the North American shale composite (Taylor & McLennan, 1985).
|
Stromatic migmatite
Relative to the metasedimentary rocks, the stromatic migmatite sample has less SiO2 and CaO, more Al2O3, (FeO* + MgO + TiO2) and K2O, and similar Na2O (Table 2; Fig. 6). With increasing SiO2, stromatic migmatite decreases in (FeO* + MgO + TiO2), Al2O3 and K2O (Fig. 6). With increasing SiO2, (Na2O + CaO) does not change systematically. The average Rb and V contents are greater than those of metasedimentary rocks, and both decrease with increasing SiO2 (Fig. 7). Sr, Zr and Ba contents are lower, whereas Rb/Sr ratios are higher than those of metasedimentary rocks; these data do not vary systematically with increasing SiO2 (Fig. 7). Stromatic migmatite compositions lie between projected compositions of biotite, garnet and sillimanite, and albite-rich plagioclase and quartz.
Chondrite-normalized REE patterns of stromatic migmatite have slight negative Nd (Fig. 9), otherwise the patterns are smooth and straight to concave-upward, with REE compositions and overall patterns that are similar to those of the metasedimentary rocks (see Figs. 8 and 9; Table 5). All patterns have negative Eu anomalies (Table 5) similar to metasedimentary rocks; the overall shape and steepness of these patterns are similar to those of the North American shale composite (Taylor & McLennan, 1985).
|
|
Diatexite
Compared with metasedimentary rocks and stromatic migmatite, the diatexite sample has a smaller range of SiO2, Al2O3 and (FeO* + MgO + TiO2) (Table 3; Fig. 6), it is poorer in CaO, and it displays similar Na2O, which increases with increasing SiO2. K2O contents tend to be greater than in metasedimentary rocks, especially at low SiO2 contents; K2O decreases systematically with increasing SiO2, and more steeply than for the stromatic migmatite sample (Fig. 6). K2O, Rb and Rb/Sr ratios are higher than those of stromatic migmatite, and Sr contents are lower than those of both metasedimentary rocks and stromatic migmatite (Fig. 7).
Chondrite-normalized REE patterns of diatexite (Fig. 9) are smooth, straight to concave-upward with negative Nd anomalies, and are similar in slope to those of the metasedimentary rocks and fall within the range of metasedimentary rocks (see Figs. 8 and 9; Table 5). All diatexite patterns have negative Eu anomalies (Table 5) similar to the metasedimentary rocks; the overall shape and steepness of these patterns are similar to those of the North American shale composite (Taylor & McLennan, 1985).
Granite
All leucogranites from within the migmatites are peraluminous (A/CNK = 1·1–1·5; Table 4), but in detail granite in sheets in the stromatic migmatite (within AFZs) is more peraluminous (A/CNK = 1·2–1·5) than granite in cylinders in diatexite (within ACZs; A/CNK = 1·1–1·2). All granites display a limited variation in SiO2 and Al2O3, and low (FeO* + MgO + TiO2) that decreases with increasing SiO2 (Fig. 6). With increasing SiO2, (Na2O + CaO) decreases whereas K2O increases (Fig. 6). There is a systematic difference between granites found in stromatic migmatite (in AFZs) and those in heterogeneous migmatite (in ACZs) with respect to their trace element compositions (Tables 4 and 6; Figs 7 and 9). Relative to granite in the diatexite domain, granite in the stromatic migmatite commonly has lower Sr, Zr, Ba, Th, Hf and light REE (LREE), and higher Rb. For granite 95-121 (from the west margin of the TAD, Fig. 2), Zr, Sr, Sr/Ba, LREE, middle REE (MREE), Th and Hf are lower, and Rb, V, Ta, Y, Nb and Rb/Sr are higher than the other three granite specimens from the stromatic migmatite, and are similar to the average of the common leucogranite from the Phillips pluton (Tables 4 and 6; Figs 7 and 9).
The two types of granite, sheets in stromatic migmatite (in AFZs) and cylinders in heterogeneous migmatite (in ACZs), show differences in LREE concentrations and chondrite-normalized REE patterns (Fig. 9). Granite from sheets has lower REE concentrations relative to metasedimentary rocks, and has steep LaN/LuN ratios (Table 6), but differently shaped REE patterns. Like metasedimentary rocks, REE patterns of granite in the sheets have negative Nd anomalies; however, the patterns are not as smooth, and have inflection in the HREE part. Otherwise, these patterns are straight to concave-upward in the HREE. The LREE patterns are similar, with the exception of 95-121, which is lower in concentration and has shallower slope (LaN/SmN = 3·0–5·2; 2·0 for 95-121). The HREE patterns of the granite in the sheets are similar in concentration and slope (Table 6; Fig. 9), and this slope is similar relative to HREE of the metasedimentary rocks. Two of these granite REE patterns have negative Eu anomalies [Eu/Eu* = 0·37 (95-121) and 0·81] and two have positive Eu anomalies (Eu/Eu* = 1·4 and 1·7). The overall shape and steepness of these patterns are similar to those of the North American shale composite (Taylor & McLennan, 1985).
Granite from cylinders has similar REE concentrations and steep chondrite-normalized REE patterns (Table 6; Fig. 9). Like metasedimentary rocks, REE patterns of the granite in the cylinders have negative Nd anomalies; however, the patterns are not as smooth, but have no inflection in the HREE part in the manner of granite in the sheets. The patterns are straight to concave-upward in the HREE. In comparison with patterns of the CMB metasedimentary rocks and granite in sheets (in AFZs), the LREE patterns of the granite in the cylinders (in ACZs) have steeper slopes, but the HREE patterns are similar in concentration and slope (Table 6; Fig. 9). Although Eu anomalies are small and unpronounced, unlike the granite in sheets, two of the granite patterns have the same positive Eu anomalies, whereas the other two have similar negative Eu anomalies [Eu/Eu* = 1·1 (two specimens), 0·80 and 0·83]. The overall shape and steepness of these patterns are similar to those of the North American shale composite (Taylor & McLennan, 1985).
Other rocks and migmatite components
The schlieric granite specimen (95-215, Fig. 5a) consistently plots between the migmatites and the granites (Figs 6 and 7). Three specimens from a block of biotite–garnet schist from one cylindrical body of granite (outcrop 95-49; Figs 4d and 5c) are poor in SiO2, and rich in Al2O3, (FeO* + MgO + TiO2) and K2O (Table 3), and plot between the field of CMB metasedimentary rocks and these two minerals (Figs 6 and 7). The two specimens of melt-depleted host from stromatic migmatite plot within the range of SiO2 for the whole-rock stromatic migmatite with similar K2O, Rb and V, but at higher (FeO* + MgO + TiO2) and Zr, and lower Al2O3, (Na2O + CaO), Na2O, Sr and Ba (Figs 6 and 7). These rocks have the highest Rb/Sr ratios of the suite (Fig. 7). The melt-depleted host from vein migmatite 96-65 commonly plots near the specimens from the block of biotite–garnet schist, with the exception of Al2O3 and Ba (higher), and V (lower).
Leucosomes from both types of migmatite are more peraluminous than the granites (A/CNK = 1·5–1·9). Leucosomes of the stromatic migmatite and diatexite 95-35 resemble the granite sheets of the stromatic migmatite in respect of Zr, Sr, Th, Hf and REE, but resemble the granite cylinders of the diatexite with respect to Rb. Most leucosomes do not plot near the field of common leucogranite from the Phillips pluton, except in SiO2 vs Al2O3, (Na2O + CaO), Zr and Ba (Figs 6 and 7). The leucosome of schlieric granite 95-215 is significantly different from the other leucosome separates, with the highest Na2O/K2O ratio and Sr contents, and the lowest Ba contents of the suite (Table 4). This leucosome also has a different composition from the average of the common leucogranite of the Phillips pluton. These observations suggest that the different leucosomes and granites are not related by simple melt segregation and accumulation. Assuming all of these rocks were derived from a similar source, other processes must account for these differences.
Relative to metasedimentary rocks, melt-depleted host and two of the three specimens of the biotite–garnet schist (a and c) have higher REE concentrations, but similarly shaped chondrite-normalized REE patterns (Figs 8 and 9). Specimen b is low in LREE and highest in HREE concentrations consistent with the higher garnet content in that part of the block (e.g. Hanson, 1980; see Fig. 5c for the centimeter-scale variation in the block). Like metasedimentary rocks, REE patterns for the biotite–garnet schist have negative Nd anomalies; otherwise the patterns are smooth, and are straight to concave-upward (Fig. 9). All LREE patterns are similar (Figs 8 and 9), but although most HREE patterns of melt-depleted host and the schist are similar in concentration and slope, schist specimen b is different (GdN/LuN = 0·9–2·2; 0·3 for b), and is shallower relative to HREE of the metasedimentary rocks. All of these patterns have negative Eu anomalies, similar to metasedimentary rocks and migmatites. The overall shape and steepness of these patterns are similar to those of the North American shale composite (Taylor & McLennan, 1985).
Relative to metasedimentary rocks, leucosomes are lower in REE concentrations, and have a larger range of slopes in their chondrite-normalized REE patterns (Tables 5 and 6; Figs 8 and 9), owing to fanning of the LREE patterns (Fig. 9). The HREE patterns of leucosomes are similar in slope, except leucosome 95-103 (GdN/LuN = 12–2·4; 1·1 for 95-103), but shallower relative to HREE of the metasedimentary rocks. Also, with the exception of 95-103, which has a negative Eu anomaly (Table 5; Fig. 9), all of these leucosome REE patterns have variously positive Eu anomalies).
|
BIOTITE COMPOSITION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF WESTERN MAINEPETROLOGY AND FIELD RELATIONS...GEOCHEMISTRYBIOTITE COMPOSITIONDISCUSSIONCONCLUSIONSAPPENDIXREFERENCES |
|---|
Biotite in smaller-volume granites and migmatite leucosomes varies from aggregates of grains forming schlieren and individual or small clusters of irregularly shaped grains, interpreted possibly to be residual, to smaller more bladed euhedral grains, interpreted to be magmatic. The composition of each type of biotite has been determined at multiple sites in several grains in each of one stromatic migmatite leucosome (95-103), one diatexite leucosome (95-100), two specimens of granite from cylinders in diatexite (95-49 and 96-11) and two specimens of granite from sheets in stromatic migmatite (95-72 and 95-103) using a JEOL 8900R electron microprobe analyzer in wavelength-dispersive mode at the University of Maryland. Although there is variation in XMg [where XMg = Mg/(Mg + Fe) on a molar basis] and Ti in atoms per formula unit (a.p.f.u.; based on 22 oxygens) between specimens, within each specimen compositional variation is insignificant (at 2
based on counting statistics). Granite in sheets has biotite with XMg of 0·36–0·37 and 0·52, and Ti a.p.f.u. of 0·22–0·26 and 0·19–0·20, respectively, in two specimens, whereas granite in cylinders has biotite with XMg of 0·48–0·49 and 0·40–0·39, and Ti a.p.f.u. of 0·26–0·32 and 0·33–0·35, respectively, in two specimens, regardless of size, shape or position in the rock texture. Leucosome in stromatic migmatite and diatexite has biotite with XMg of 0·49 and 0·44, and Ti a.p.f.u. of 0·23–0·22 and 0·15, respectively, regardless of size, shape or position in the rock texture. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF WESTERN MAINEPETROLOGY AND FIELD RELATIONS...GEOCHEMISTRYBIOTITE COMPOSITIONDISCUSSIONCONCLUSIONSAPPENDIXREFERENCES |
|---|
The central question is whether leucosomes and granites in migmatite are related to granite in the nearby coeval plutons. Melt loss from migmatites is implied by the K-feldspar-poor nature of the leucosomes. This is supported by the lack of mass balance of trace elements between migmatite components (Tables 2 and 4, stromatic migmatite and diatexite, respectively), suggesting open-system behavior at the scale of hand specimens. The ‘pinch-and-swell’ structure of granite sheets as described above is consistent with melt flow during deformation and weak strain during or after emplacement of these sheets. In addition, the strong correlation of regional fabrics across the migmatite front in the same structural zone supports the interpretation that migmatization occurred while the rocks were accommodating strain. Thus, we postulate that leucosomes and smaller-volume granites record evidence of syntectonic melt flow within and through the migmatite, and that granite in plutons apparently outside the migmatites at the level exposed represents evolved melt that escaped syntectonically from a similar source to the migmatites exposed in the TAD and WAD.
Origin of leucosomes and granites
We can constrain the processes involved using the experimental results of Patiño Douce & Harris (1998). The composition of the muscovite–biotite schist used in that study (MBS), lies within the field of CMB metasedimentary rocks (Figs 6 and 10), and melt compositions produced from MBS plot among the data for Maine granites and leucosomes (Figs 6 and 10). Most of the experiments of Patiño Douce & Harris (1998) involved the metamorphic volatile phase absent reaction
 |
This reaction is widely considered appropriate for producing leucogranitic melt from metasedimentary protoliths of metapelitic composition in collisional origins (e.g. Patiño Douce, 1999). Some of the experiments by Patiño Douce & Harris (1998) were run with added H2O, resulting in the reaction
 |
The amount of melt is limited by protolith muscovite content in the first case, and plagioclase content and the amount of added H2O in the second case. The compositions of these melts vary from granite (muscovite dehydration and lower-pressure water-fluxed melting) to trondhjemite (higher-pressure water-fluxed melting). Migmatite textures show that biotite was apparently stable on a regional basis (as described above), indicating that the biotite dehydration melting reaction was not crossed during regional metamorphism.
There are three options for the origin of the granites and leucosomes in the Maine migmatites:
- these rocks could represent equilibrium melts (residue equilibrates with liquid before melt segregation), in which case the range in composition observed is controlled by P, T, aH2O, protolith composition and the degree of melting, the amount of entrained residual material and any fractionation during crystallization.
- The rocks could represent disequilibrium melts (residue does not equilibrate with liquid before melt segregation) if melting is rapid and deformation allows fast segregation to preserve disequilibrium compositions (Ayers & Harris, 1997). Rapid melting and segregation have been argued for muscovite dehydration melting (Rubie & Brearley, 1987; Brearley & Rubie, 1990), and rapid melting may occur in response to fluid ingress. Although compositions may have been modified by fractionation during crystallization, fast segregation as expected during deformation may limit the potential for entrainment of residual material.
- These rocks could be the result of crystallization of mixtures of liquid and crystals, where the crystals may be residual or cumulate, or both, and where some or most of the evolved melt may have escaped from the crystallization site, aided by and/or enhancing the deformation.
In reality, equilibrium is an unlikely expectation during syntectonic melting in nature, and some degree of disequilibrium is to be expected. Thus, under conditions of limited disequilibrium we may anticipate equilibrium between melt and newly crystallized peritectic minerals, but not necessarily between melt and residual minerals. In the next section we discuss the petrogenesis of the migmatite leucosomes and smaller-volume granites, and we investigate any possible relationship with the common leucogranite of the Phillips pluton.
Petrogenesis
Protolith of the migmatites and chemical differentiation during migmatization
On the basis of the structural evolution of western Maine (Solar & Brown, 2001), and the field relations and geochemistry of the migmatites and granites described herein, we propose a model of progressive separation of melt and residue during deformation of the CMB metasedimentary succession. Below the solidus, these rocks were undergoing syntectonic prograde metamorphism that is assumed to have been effectively isochemical. Although centimeter-scale quartzo-feldspathic- and mica-rich domain segregations occur in the metapelite layers (Solar & Brown, 2001) to suggest redistribution of silica (Ague, 1994), we have collected specimens of protolith rocks that have undergone this process, and a reduction in silica by open-system behavior does not account for the lower SiO2 in migmatites relative to metasedimentary rocks sample. Therefore, we interpret differences in geochemistry as due to syntectonic melting and various secondary processes during melt–residuum separation, such as melt loss or gain, residue entrainment, fractional crystallization, and accumulation of cumulate phases and loss of evolved melt.
The three samples of the populations of metasedimentary rocks, stromatic migmatite and diatexite analyzed in this study are not statistically different in respect of their geochemical compositions (Solar, 1999). Thus, on the basis of comparative geochemical arguments we cannot reject the null hypothesis that partial melting of the metasedimentary rocks was the process that formed the migmatites. On the basis of the field relations and geochemistry discussed above, metasedimentary rocks similar in composition to those of the CMB are inferred to be the protolith for the TAD and WAD migmatites.
For (FeO* + MgO + TiO2), Al2O3, (Na2O + CaO) and V it is apparent that the depleted nature of stromatic migmatite and diatexite is not balanced by the smaller-volume granites alone (Figs 6 and 7). As linear trends on Harker plots are a characteristic feature of restite unmixing (e.g. Chappell et al., 1987), these plots indicate that differential separation of melt from residual solid material was not the sole petrogenetic process involved in producing the variation observed.
If mass balance is preserved at all scales during melting, melt segregation and transfer, and crystallization of the melt, then the processes involved may be tracked in the ternary plot K–(Fe* + Mg + Ti)–(Na + Ca) (Fig. 10). In such a plot, biotite lies along the edge (Fe* + Mg + Ti)–K; it represents the major residual phase. The feldspar join is represented by the edge (Na + Ca)–K, close to which lie melts produced from crustal protoliths. Residual compositions will be displaced from the field of metasedimentary rocks toward (Fe* + Mg + Ti)–K, whereas leucosome and granite compositions will trend toward the feldspar join. In Fig. 10, the field of CMB metasedimentary rocks lies between muscovite and garnet, and close to biotite. Migmatite compositions are displaced toward the (Fe* + Mg + Ti)–K edge in comparison with the CMB metasedimentary rock field, whereas granites and leucosomes are weakly clustered toward the feldspar join, closer to the (Na + Ca) apex than the K apex. The three specimens from the block of biotite–garnet schist plot between biotite and plagioclase, but much closer to biotite, and are displaced from the CMB metasedimentary rock field toward garnet, reflecting the dominance of these phases in the rock (Fig. 5d). The whole-rock chemistry of the schlieric granite specimen plots between the field of CMB metasedimentary rocks and the feldspar join, suggesting that the specimen is enriched in the feldspathic components compared with the protolith composition. Granite specimen 95-121 plots within the field of common leucogranite from the Phillips pluton.
In Fig. 10, the migmatite leucosomes are seen to define an array of compositions from melt dominated, plotting close to the MBS melt compositions in the experiments of Patiño Douce & Harris (1998), to cumulate dominated, plotting close to a cumulate composed of 80% plagioclase and 20% biotite. Variable loss of a K-rich liquid is implied. In contrast, the smaller-volume granites define a triangular field between the MBS melts and the cumulate join between plagioclase and biotite, with the leucosome array as the bottom edge and extending along the cumulate join from 20 to 35% biotite. The common leucogranite of the Phillips pluton and granite 95-121 crystallized from a K-enriched (evolved) liquid in comparison with the MBS melts.
From data in Fig. 10, we infer that none of the smaller-volume granites has a likely melt composition, although 95-103 appears to have a melt-dominated composition. Most of the smaller-volume granites have a variable cumulate composition that we presume incorporates some residual biotite; a K-rich liquid has been partially lost from these rocks. Indeed, all analyzed granites from within diatexite appear to have largely cumulate compositions probably with variable amounts of included residual biotite and/or plagioclase, and only a minor amount of retained K-rich liquid. However, in contrast to some studies (Friend et al., 1985; Sawyer, 1998; Milord et al., 2001), texturally distinct biotite in the Maine migmatites and smaller-volume granites we analyzed does not have significantly different composition, and we are unable to identify residual biotite on the basis of chemical composition.
Most oxide and trace element concentrations of migmatite components, except for the schlieric granite (95-215), have straight connecting lines between the components and their respective whole-rock compositions (Figs 6 and 7). The leucosome of schlieric granite 95-215 is lower in K2O and Rb, and higher in (Na2O + CaO), Na2O and Sr than other leucosomes. The difference is particularly conspicuous in Fig. 10, where all connecting lines between leucosomes and whole-rock compositions are subparallel except for the leucosome from the schlieric granite (95-215). We interpret these relations to indicate that migmatites do indeed represent in situ segregation, generally with variable loss of a K-rich liquid from partly cumulate leucosomes, except for the schlieric granite, which we suggest was formed by melt accumulation before fractional crystallization and loss of most of the K-rich liquid imposed the final composition on the largely cumulate leucosome.
Temperature regime of melting
To investigate further the temperature regime during melting, we estimate the saturation temperature of these rocks from the solubility of zircon and monazite in melt, obtained from Zr and LREE concentrations relative to the major element compositions [Tables 2–4 and 6; from expressions of Watson & Harrison (1983) and Montel (1993), respectively]. Temperatures in the range 750–800°C would be consistent with muscovite dehydration melting. Spurious high temperatures might indicate entrainment of residual accessory phases, whereas low calculated temperatures argue against significant entrainment of accessory phases. If the solubilities of accessory phases are kinetically controlled, the temperatures inferred from zircon solubilities should be systematically higher than inferred from monazite solubilities, as a result of the more rapid dissolution of zircon in a granite melt (e.g. Rapp & Watson, 1986). Monazite saturation temperatures (902–1090°C, with the exception of the leucosome from the vein migmatite) are much higher than zircon saturation temperatures (700–830°C). The zircon saturation temperatures are consistent with muscovite dehydration melting, whereas the spurious high monazite saturation temperatures indicate entrainment of residual monazite.
Rb/Sr ratios
Rb/Sr ratios have been used as a discriminator of water-fluxed vs dehydration melting (Harris & Inger, 1992; Harris et al., 1995), although care must be exercised as some degree of disequilibrium may be involved in the melting process, which may reduce the Rb/Sr ratio of the melt regardless of the aH2O during melting (Harris et al., 1993). Rb/Sr ratios of granite (excepting 95-121) and leucosome specimens are in the range 0·2–1, which taken at face value are consistent with a low Rb–Sr fractionation during water-fluxed melting of the CMB metasedimentary rocks (Rb/Sr = 0·8–2·5; Harris & Inger, 1992; Harris et al., 1995). However, the postulated loss of a K-rich liquid from some granite and leucosomes is likely to have lowered Rb/Sr ratios in those rocks by preferential fractionation of Rb from Sr (e.g. Harris & Inger, 1992). Further, the incorporation of residual biotite in most of these granites will have raised Rb relative to Sr (e.g. Hanson, 1980). Thus, any conclusions about type of melting based on Rb and Sr distributions in these smaller-volume granites are likely to be flawed.
In contrast, common leucogranite from the Phillips pluton has Rb, Sr and Ba covariations and K2O contents that are interpreted to reflect control by peritectic K-feldspar in the source (Pressley & Brown, 1999). Rb/Sr ratios for the Phillips pluton of 1·5–8·5 (of which most are 3–5), suggest a high Rb–Sr fractionation during derivation by moderate aH2O muscovite dehydration melting of the CMB metasedimentary rocks. Granite specimen 95-121, which plots commonly with the Phillips pluton leucogranite, also has a large Rb/Sr ratio (4·5; Fig. 7), which suggests that it too may have crystallized from a liquid composition evolved from melt that was produced by muscovite dehydration. We suggest that conclusions about type of melting based on Rb–Sr distributions in these relatively large volume granites may be more robust than for the smaller-volume granites.
Modeling major element compositions
From arguments presented above, the apparent residual composition of the migmatites may be the result of open-system behavior whereby melt was removed. If the CMB metasedimentary rocks are a reasonable analogue for the protolith for the migmatites and granites, then the petrogenetic relations can be investigated by modeling subtraction of MBS melt from the representative metasedimentary rock 95-97 (Fig. 10). The residuum trend shown for 20% melt extraction from that specimen (thick line beginning at the rock composition and projected away from MBS melt) extends to the limit of data for migmatites, closer to the (Fe* + Mg + Ti)–K edge than the field of CMB metasedimentary rocks. This is consistent with a calculated average stoichiometric melt production of 28 vol. %, based on the average modal muscovite in CMB metasedimentary rocks, assuming some of the melt is retained in the migmatites as leucosome, consistent with the observations described above. Therefore, migmatite specimens can be modeled as residual after melt removal from rocks similar in composition to the CMB metasedimentary rocks. In this light, the SiO2-poor, [Al2O3–(FeO* + MgO + TiO2)–K2O]-rich composition of the block of biotite–garnet schist in the cylindrical body of granite 95-49 (Fig. 5) suggests it is an extreme example of residue from partial melting of CMB metasedimentary rock.
The dashed lines in Fig. 10 show the trends produced by fractional crystallization of a cumulate composed of 20% biotite + 80% plagioclase (a1) and 35% biotite + 65% plagioclase (a2), respectively, from a melt composition similar to the 6 kbar MBS melt of Patiño Douce & Harris (1998). Three important conclusions follow from this exercise:
- the migmatite leucosomes could be produced by fractional crystallization of assemblage a1 and variable melt loss, except for leucosome 95-103, which must include entrained residual biotite;
- the smaller-volume granites could be produced by fractional crystallization of an assemblage similar to a2, with variable melt loss and entrained residual plagioclase and/or biotite;
- the common leucogranite of the Phillips pluton can be modeled by subtraction of 0–20% cumulate biotite and plagioclase (with biotite << plagioclase) from a melt similar to the MBS melt composition and crystallization of the resultant evolved K-rich liquid. Because the migmatite leucosomes and smaller-volume granites individually are very much smaller in volume than the Phillips pluton, this suggests accumulation at the site of the Phillips pluton of many small batches of evolved melt, consistent with the interpretation of Rb–Sr–Ba and Nd isotope data by Pressley & Brown (1999).
Modeling Rb–Sr–Ba relations
Leucosomes and smaller-volume granites are enriched in Sr and depleted in Rb relative to the metasedimentary rocks, the migmatites and the common leucogranite of the Phillips pluton, although Ba contents are similar, except for the common leucogranite of the Phillips pluton, which has Ba contents at the low end of the range for leucosomes and smaller-volume granites. In principle, it should be possible to test our modeling of the major element compositions by using the results to model the Rb–Sr–Ba relations of each of the different groups of rocks. There are, however, significant problems inherent in modeling the large ion lithophile elements (LILE).
The first issue concerns the assumption of equilibrium melting, with no entrainment of residual minerals in the melt, and with the LILE partitioned between crystals and melt in accordance with experimentally determined partition coefficients. Equilibrium melting may be unlikely during crustal anatexis, particularly during deformation, and there are few LILE mineral–melt partition data for conditions of crustal melting and crystallization of granitic systems. Further, the existing data exhibit sufficient variation that modeled trends at best will be qualitative. This variation reflects a number of important controls on the partition coefficients, such as the structure and composition of the melt (e.g. Lagache & Carron, 1982), mineral composition (e.g. Blundy & Wood, 1991; Icenhower & London, 1996), temperature (Icenhower & London, 1995; Chappell, 1996) and water content (Mahood & Hildreth, 1983). The second issue is the tendency for leucosome to maintain mineral–mineral equilibrium with melanosome or the melt-depleted host during crystallization (Nabelek, 1999), which makes the use of mineral–melt partition coefficients inappropriate for some of the processes involved.
The mineral–melt partition coefficients we chose to use are given in Table 7, together with the modes for the initial mineral assemblage in the metasedimentary protolith and the residue from partial melting. Although our modeling produced realistic Rb, Sr and Ba concentrations in residue and melt, modeling fractional crystallization of this melt composition according to the results obtained from our modeling of the major elements does not reproduce the observed data for leucosomes and smaller-volume granites very well (Fig. 11). We conclude that modeling the LILE distributions during complex, multistage processes of crustal melting and fractional crystallization in granitic systems is premature.
|
|
The REE
Chondrite-normalized REE plots of these rocks are consistent with a petrogenetic relation between the metasedimentary rocks, the migmatites and the granites. Migmatite whole-rock REE patterns are similar to those of the CMB metasedimentary rocks. REE patterns in the granites are also similar in slope, but are systematically lower in abundance. Typical upper-crustal granites that are derived from metasedimentary protoliths have enriched LREE contents relative to chondritic values (e.g. Taylor & McLennan, 1985). This is true for the suite of specimens from Maine, where granites from the diatexite domains are the most enriched (La 100 times chondrites) and common leucogranite from the Phillips pluton is the least enriched (La 20 times chondrites). Chondrite-normalized REE patterns show that most of the granites and all of the leucosomes are LREE depleted relative to the metasedimentary rocks, consistent with the postulate that these are the putative protolith for the melts (Miller & Mittlefehldt, 1982; Mittlefehldt & Miller, 1983; Le Fort et al., 1987; Sevigny et al., 1989; Breaks & Moore, 1992; Wark & Miller, 1993; Nabelek & Glascock, 1995; Tomascak et al., 1996; Pressley & Brown, 1999; Milord et al., 2001). Taking the CMB metasedimentary rocks as the protolith for the TAD and WAD migmatites, and the source for the granites, we recast the REE data by normalizing to one representative metasedimentary rock (specimen 95-97). We choose this metasedimentary rock because it was collected from close proximity to the migmatite front. Further, the chondrite-normalized REE pattern for this rock plots at about the middle of the sample range (Fig. 8).
Examination of the protolith normalized REE patterns in Fig. 12 shows that whole-rock compositions of the migmatites resemble the protolith closely, but are generally enriched in the total REE, probably reflecting melt loss at the scale of the hand specimen relative to the protolith. All granites have REE patterns that lie well below the protolith composition, and all but granite 95-121 have positive Eu anomalies. One granite specimen from the diatexite shows LREE enrichment relative to the protolith, and two have positive Ce anomalies that may reflect incorporation of monazite from the source. The melt-depleted host material in the migmatites is REE enriched, with the exception of one specimen that shows MREE depletion. Leucosomes show variable total REE depletion, shallow LREE slopes, inflected and variable HREE patterns and distinct positive Eu anomalies. Schlieric granites are slightly depleted in the REE, although 95-215 shows slight HREE enrichment, and both specimens show positive Eu anomalies. The common leucogranite from the Phillips pluton is REE depleted in a similar fashion to the smaller-volume granites, and is closely similar to the pattern of granite 95-121. Complementing the suite of granites, the three analyses from the block of biotite–garnet schist (95-49) show strong total REE enrichment, with the exception of the LREE of one specimen (b) that may reflect the higher proportion of garnet in that specimen (e.g. Hanson, 1980). Each, however, has a negative Eu anomaly similar to the patterns from the migmatite whole rock and melt-depleted host.
|
We have not tried to model REE patterns for the multistage model we postulate based on the major element compositions because assumptions necessary for such modeling are not met by the Maine migmatites and granites. REE modeling requires that (1) leucosomes and granites do not contain significant amounts of xenocrystic material, (2) the protolith composition can be unambiguously identified, (3) leucosome and granite REE concentrations have not been significantly disturbed either by subsolidus processes or by fluid infiltration, and (4) leucosomes and granites have melt compositions and that have undergone fractional crystallization. None of these criteria can be satisfied given the dynamic nature of the syntectonic migmatization and melt flow in our Maine example. In addition, mineral–melt partition coefficients for trace elements in silicic melts are dependent on composition (Mahood & Hildreth, 1983; Blundy & Wood, 1991), and insufficiently well known to justify the complexity of modeling required in this case.
The role of biotite is likely to be the key to understanding the REE variations. As widespread biotite breakdown did not occur, biotite in the protolith probably did not contribute significantly to the REE budget of the melt except as an entrained phase. Because accessory minerals in migmatites commonly are included in the major phases (e.g. Bea, 1996), minerals such as biotite control the behavior of these phases. As common accessory minerals in granites have distinctively different REE compositions (Fig. 13c), exactly which phases are included in biotite will control in detail the effect of biotite addition to or subtraction from the melt. Thus, REE-bearing accessory minerals such as apatite, monazite and zircon may not contribute significantly to the melt if sequestered in residual biotite, leaving the melt depleted in the REE (e.g. Rapp & Watson, 1986; Sevigny, 1993; Watt & Harley, 1993; Nabelek & Glascock, 1995). Correspondingly, the REE enrichment of the migmatites is probably due to concentration of biotite and its included accessory minerals in the residue (Bea, 1996; Watt et al., 1996).
|
The LREE concentrations in the granites and the leucosomes are inconsistent with the calculated temperatures for monazite saturation (902–1090°C), because monazite would be totally dissolved under those conditions. As a test of residual biotite incorporation into the granites to elevate LREE contents, we model two-component mixing of the REE composition of biotite (and its included accessory minerals) from the migmatites with the migmatite leucosomes by incremental addition (e.g. Milord et al., 2001). Two of the results are listed in Table 8, and plotted as protolith-normalized patterns in Fig. 13a and b. Mixing residual biotite with leucosome from stromatic migmatite 95-103 does not reproduce the REE pattern of the granite sheet that cuts it (Fig. 13a). Addition of >2·5 vol. % residual biotite to the leucosome composition results in higher LREE and HREE concentrations than in the granite. In contrast, the leucosome from diatexite 95-35 is depleted in LREE relative to the granite cylinder that cuts it (Fig. 13b), and although addition of 25 vol. % residual biotite causes the LREE pattern of the mixture to resemble that of the granite, the HREE content is enriched relative to the granite composition. If biotite were removed from the magma during ascent, the REE composition would decrease proportionally, but more strongly in the LREE, and REE patterns similar to the common leucogranite of the Phillips pluton may be the result (Figs 9 and 12). The LREE-enriched granite in cylinders in diatexite (Fig. 12) may be a consequence of incorporation of monazite into the melt (Fig. 13) by the physical destruction of entrained biotite (e.g. Sawyer, 1998).
H2O recycling
Poikilitic muscovite crystals and minor chlorite occur parallel to foliation and lineation throughout migmatites of the TAD and WAD. Although we postulate loss of a K-rich melt from the migmatites, K-feldspar does occur locally in some leucosomes and its absence from other leucosomes may be due to replacement by muscovite. Given the steep orientation of these fabrics, we interpret the retrograde growth of muscovite and chlorite to record buoyancy-aided fluid flow parallel to the fabrics in the rocks. This fluid is likely to have been derived from crystallizing melts within the migmatites (Fig. 14). As melt crystallizes in the deforming rocks, liberated H2O may promote melting in adjacent units at suprasolidus conditions or retrogression at subsolidus conditions. Such a mechanism has been postulated by Holk & Taylor (1997) to explain homogenization of oxygen isotopic compositions in mid-crustal rocks of the Thor–Odin metamorphic core complex in British Columbia. No influx of water-rich volatile phase is necessarily implied or required, and we postulate that this is the principal cause of the regional syntectonic retrogression of staurolite and andalusite in rocks outside the migmatite front (Solar & Brown, 1999).
|
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF WESTERN MAINEPETROLOGY AND FIELD RELATIONS...GEOCHEMISTRYBIOTITE COMPOSITIONDISCUSSIONCONCLUSIONSAPPENDIXREFERENCES |
|---|
In Maine, metamorphism and granite crystallization are contemporaneous; and syntectonic magma ascent was controlled by deformation and development of strain fabrics. Mineral assemblages and geochemical data are consistent with muscovite dehydration melting, and suggest that migmatite leucosomes and smaller-volume granites represent cumulate rocks (± residual material and some retained fractionated melt) that complement common leucogranite in the Phillips pluton. The residual geochemistry of stromatic migmatite and diatexite relative to the metasedimentary rocks, and the ‘pinch-and-swell’ structure of granite sheets, are consistent with melt loss from those rocks, perhaps driven by the accommodation of deformation. Rare schlieric granites suggest some melt redistribution before melt loss, in this case leaving a more felsic cumulate than the more residual migmatites.
Exposed migmatites preserve evidence in cumulate leucosomes of the flow network that drained these rocks of an evolved melt. We postulate that migmatites similar to those exposed represent the source of common leucogranite in the Phillips pluton. Thus, the Phillips pluton may be connected at depth to granites similar to those found in the migmatites of the TAD and WAD in the manner described by Brown & Solar (1999) where the heterogeneous migmatites and granites in ACZs formed in the cores of thermal antiforms developed during regional contraction. However, the relation between migmatite leucosomes, smaller-volume granites and leucogranite in plutons is not straightforward.
In Maine, we have demonstrated that migmatite leucosomes and smaller-volume granites are cumulate rocks, whereas the leucogranite of the adjacent Phillips pluton is consistent with the putative liquid lost from these rocks. Any expectation that migmatite leucosomes and leucogranite in plutons should show simple melt compositions without entrained residue or modification by fractional crystallization is naive. The popular notion that migmatites represent ‘failed’ granites should be reconsidered in the light of multiple syntectonic processes, and there is no a priori reason to suppose that the process of crustal melting and the evolution of leucogranites will be any less complex than for other magma suites.
|
APPENDIX |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF WESTERN MAINEPETROLOGY AND FIELD RELATIONS...GEOCHEMISTRYBIOTITE COMPOSITIONDISCUSSIONCONCLUSIONSAPPENDIXREFERENCES |
|---|
Analytical methods
After removal of weathered material,
0·5 kg of each specimen (depending upon homogeneity and grain size) was reduced to powder at the University of Maryland using a SPEX Industries shatterbox with an alumina grinding chamber and puck [see Tomascak (1995) for details]. Approximately 200 mg of each powdered specimen was fused at Washington University in St Louis, Missouri, and analyzed by XRF (Couture et al., 1993; R. Couture, analyst; data in Tables 1–4). Thirty-two of the powders were processed at Washington State University, for REE composition primarily, using a fusion–dissolution method adapted from Crock & Lichte (1982), and analyzed by ICP-MS using a Sciex Elan model 250 instrument (C. Knaack, analyst; data in Tables 5 and 6).
Analytical uncertainty and correlation
XRF
Precision of XRF data is estimated from results for standard GRD8083B, analyzed 36 times (18 runs) with each set of analyses of unknown specimens of this study. The means of standard values compared favorably with reference values (Solar, 1999). The 2 uncertainty is 1% of the value for SiO2, Al2O3 and FeO*, 1·5–6% of the value for MgO, CaO, Na2O and K2O, 2·5% of the value for TiO2, and 8 and 31% of the value for MnO and P2O5, respectively. The high uncertainty for P2O5 is due to drift in concentration of the standard disk over time (R. Couture, personal communication, 1999); however, given the low concentration of P2O5 in samples (<0·30 wt %; Tables 1–4), this uncertainty in the standard translates to <0·1 wt % in samples. The 2 uncertainties for trace elements are in the range 1–5 ppm for Nb, Zr, Y, Sr, Rb, Pb, Ga, Zn, Ni and Co, 7 ppm for V, 13 ppm for Cr and 17 ppm for Ba. Sn was below detection limit in the standard.
Univariate and multivariate tests of the significance of XRF whole-rock data to represent the geochemical variation of the various rock groups, and a discussion of these tests based upon arguments of Power (1993) have been given by Solar (1999). Solar (1999) determined that distributions of major element oxide data, and of trace element data, are normal and asymmetric for the chosen metasedimentary rocks, stromatic migmatite and diatexite samples of these populations.
ICP-MS
Precision of ICP-MS data is estimated from results for standard BCR-P, analyzed 24 times (12 runs) over a 4 month period of 1994. The means of standard values compare well with reference values of Govindaraju (1994). The 2 uncertainty is <5% with the exception of Th (19%), U (19%), Pb (6·5%), Cs (6%), Ta (5·4%) and Eu (5%). The larger uncertainties for Th and U translate to 
5 and 2 ppm, respectively, in samples of this study (Tables 5 and 6). The 2 uncertainties for the REE are smaller than the plotted symbol size in the normalized plots. Three specimens were re-fused as a check on reproducibility [schlieric granite 95-215 (whole rock), 95-215 leucosome, stromatic migmatite 96-286 (whole rock)], and these data are listed in Tables 5 and 6. As a result of difficulties in determining the concentration of thulium that stem from its low abundance and mono-isotopic nature, complicated by its low abundance in chondrites (S. M. McLennan & S. R. Hemming, personal communication, 2000), the measured Tm concentrations of all 32 of our specimens produced uniform slight positive anomalies on chondrite-normalized plots (Tm/Tm* = 1·06 ± 0·02). Because of this uniform anomaly, and because we cannot find a mineralogical explanation to account for such a uniform anomaly across different rock types, we have left Tm off our REE plots.
Correlation between XRF and ICP-MS data
Six trace elements were analyzed by both XRF and ICP-MS (Nb, Y, Sr, Rb, Pb and Ba). Thirty-one of these 192 data (six elements by 32 specimens) are in >10% disagreement, but only eight translate to a difference of >10 ppm. Twenty-two of these 31 data show a lower concentration in the XRF analyses (Tables 5 and 6). The most common disagreement is found in Y (17 of the 32 specimens), but only five of those are different by >10 ppm, and none is different by >20 ppm. The next most common disagreement is found in Pb (six specimens), none different by >10 ppm. We find from the comparison that the correlation between the XRF and ICP-MS data is within reasonable uncertainty.
|
ACKNOWLEDGEMENTS |
|---|
We wish to thank those who have helped us in the field and in the interpretation of the chemical data, in particular E. Sawyer, T. Rushmer, A. B. Thompson, G. Hanson, S. McLennan, J. Hurowitz, T. Rasbury, S. Hemming and P. Tomascak. G.S.S. is grateful to the ‘Dirt Group’ of the Department of Geosciences at SUNY Stony Brook for guidance concerning the quality and interpretation of geochemical data. Of course, any misconceptions and errors are our own. We thank T. Rushmer, C. Barnes and an anonymous referee for thorough reviews and critical comments, and Sorena Sorensen for her extensive and helpful editorial comments. We acknowledge partial support of this work from the Department of Geology, University of Maryland, the Department of Geological Sciences, Virginia Tech, the Department of Geosciences, SUNY Stony Brook, the Geological Society of America Research Grants Program and NSF Grant EAR-9705858 (to M.B.).
|
FOOTNOTES |
|---|
*Corresponding author. Present address: Department of Earth Sciences, SUNY College at Buffalo, 1300 Elmwood Avenue, Buffalo, NY 14222, USA. Telephone: 716-878-6731. Fax: 716-878-4009. E-mail: solargs{at}bscmail.buffalostate.edu
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGY OF WESTERN MAINEPETROLOGY AND FIELD RELATIONS...GEOCHEMISTRYBIOTITE COMPOSITIONDISCUSSIONCONCLUSIONSAPPENDIXREFERENCES |
|---|
Ague, J. J. (1994). Mass transfer during barrovian metamorphism of pelites, south–central Connecticut, I: evidence for composition and volume change. American Journal of Science 294, 989–1057.
Allibone, A. H. & Norris, R. J. (1992). Segregation of leucogranite microplutons during synanatectic deformation: an example from the Taylor Valley, Antarctica. Journal of Metamorphic Geology 10, 589–600.[Web of Science]
Ashworth, J. R. (1985). Introduction. In: Ashworth, J. R. (ed.) Migmatites. Glasgow: Blackie, pp. 1–35.
Ayers, M. & Harris, N. (1997). REE fractionation and Nd-isotope disequilibrium during crustal anatexis: constraints from Himalayan leucogranites. Chemical Geology 139, 247–267.
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 39, 521–552.
Benn, K., Cruden, A. R. & Sawyer, E. W. (1998). Preface. Journal of Structural Geology 20, Special Issue: Extraction, Transport and Emplacement of Granitic Magmas.
Blundy, J. D. & Wood, B. J. (1991). Crystal-chemical controls on the partitioning of Sr and Ba between plagioclase feldspar, silicate melts, and hydrothermal solutions. Geochimica et Cosmochimica Acta 55, 193–209.
Bradley, D. C. (1983). Tectonics of the Acadian orogeny in New England and adjacent Canada. Journal of Geology 91, 381–400.
Bradley, D. C., Tucker, R. D., Lux, D. R., Harris, A. G. & McGregor, D. C. (1998). Migration of the Acadian orogen and foreland basin across the northern Appalachians. US Geological Survey, Open-File Report 98-770, 79 pp.
Breaks, F. W. & Moore, J. M. (1992). The Ghost Lake batholith, Superior Province of northwestern Ontario—a fertile, S-type, peraluminous granite–rare-element pegmatite system. Canadian Mineralogist 30, 835–875.[Web of Science]
Brearley, A. J. & Rubie, D. C. (1990). Effects of H2O on the disequilibrium breakdown of muscovite + quartz. Journal of Petrology 31, 925–956.
Brown, M. (1973). The definition of metatexis, diatexis and migmatite. Procedings of the Geological Association 84, 371–382.
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. & Rushmer, T. (1997). The role of deformation in the movement of granite melt: views from the laboratory and the field. In: Holness, M. B. (ed.) Deformation-enhanced Fluid Transport in the Earth’s Crust and Mantle. London: Chapman and Hall, pp. 111–144.
Brown, M. & Solar, G. S. (1998a). Shear zones and melts: positive feedback in orogenic belts. Journal of Structural Geology 20, 211–227.
Brown, M. & Solar, G. S. (1998b). Granite ascent and emplacement during contractional deformation of convergent orogens. Journal of Structural Geology 20, 1365–1393.[Web of Science]
Brown, M. & Solar, G. S. (1999). The mechanism of ascent and emplacement of granite magma during transpression: a syntectonic granite paradigm. Tectonophysics 312, 1–33.
Brown, M., Averkin, Y. A., McLellan, E. L. & Sawyer, E. W. (1995). Melt segregation in migmatites. Journal of Geophysical Research 100, 15655–15679.[Web of Science]
Brown, M. A., Brown, M., Carlson, W. D. & Denison, C. (1999). Topology of syntectonic melt flow networks in the deep crust: inferences from three-dimensional images of leucosome geometry in migmatites. American Mineralogist 84, 1793–1818.[Abstract]
Chamberlain, C. P. & England, P. C. (1985). The Acadian thermal history of the Merrimack Synclinorium in New Hampshire. Journal of Geology 93, 593–602.
Chappell, B. W. (1996). Compositional variation within granite suites of the Lachlan Fold Belt: its causes and implications for the physical state of granite magma. Transactions of the Royal Society of Edinburgh: Earth Sciences 88, 159–170.
Chappell, B. W., White, A. J. R. & Wyborn, D. (1987). The importance of residual source material (restite) in granite petrogenesis. Journal of Petrology 28, 1111–1138.
Clemens, J. D. & Mawer, C. K. (1992). Granitic magma transport by fracture propagation. Tectonophysics 204, 339–360.[Web of Science]
Cocks, L. R. M., McKerrow, W. S. & van Staal, C. R. (1997). The margins of Avalonia. Geological Magazine 134, 627–636.[Abstract]
Collins, W. J. & Sawyer, E. W. (1996). Pervasive granitoid magma transfer through the lower–middle crust during non-coaxial compressional deformation. Journal of Metamorphic Geology 14, 565–579.[Web of Science]
Collins, W. J. & Vernon, R. H. (1991). Orogeny associated with anticlockwise P–T–t paths: evidence from low-P, high-T metamorphic terranes in the Arunta Inlier, central Australia. Geology 19, 835–838.
Couture, R. A., Smith, M. S. & Dymek, R. F. (1993). X-ray fluorescence analysis of silicate rocks using fused glass discs and a side-window Rh source tube: accuracy, precision and reproducibility. Chemical Geology 110, 315–328.[Web of Science]
Crock, J. G. & Lichte, F. E. (1982). Determination of rare earth elements in geologic materials by inductively couple argon plasma/atomic emission spectrometry. Analytical Chemistry 54, 1329–1332.
Cruden, A. R. (1998). On the emplacement of tabular granites. Journal of the Geological Society, London 154, 853–862.
Cullers, R. L., Long-Tsu, Y., Sambhudas, C. & Guidotti, C. V. (1974). Rare earth elements in Silurian pelitic schists from N.W. Maine. Geochimica et Cosmochimica Acta 38, 389–400.
Cullers, R. L., Bock, B. & Guidotti, C. (1997). Elemental distributions and neodymium isotopic composition of Silurian metasediments, western Maine, USA: redistribution of rare earth elements. Geochimica et Cosmochimica Acta 61, 1847–1861.
Deer, W. A., Howie, R. A. & Zussman, J. (1997). Rock Forming Minerals, Volumes 1, 3 and 4. Harlow, UK: Longman.
De Yoreo, J. J., Lux, D. R., Guidotti, C. V., Decker, E. R. & Osberg, P. H. (1989). The Acadian thermal history of western Maine. Journal of Metamorphic Geology 7, 169–190.[Web of Science]
D’Lemos, R. S., Brown, M. & Strachan, R. A. (1992). Granite magma generation, ascent and emplacement within a transpressional orogen. Journal of the Geological Society, London 149, 487–490.
Eusden, J. D., Jr & Barreiro, B. (1988). The timing of peak high-grade metamorphism in central–eastern New England. Maritime Sedimentation and Atlantic Geology 24, 241–255.
Evansen, M. N., Hamilton, P. J. & O’Nions, R. K. (1978). Rare-earth abundances in chondritic meteorites. Geochimica et Cosmochimica Acta 42, 1199–1212.[Web of Science]
Friend, C. R. L., Brown, M., Perkins, W. T. * Burwell, A. D. M. (1985). The geology of the Qôrqut granite complex north of Qôrqut, Godthåbsfjord, southern West Greenland. Bulletin Grønlands Geologiske Undersøgelse 151, 43 pp.
Govindaraju, K. (1994). Compilation of working values and sample descriptions for 383 geostandards. Geostandards Newsletter, Special Issue 18, 15–53.
Grocott, J., Brown, M., Dallmeyer, R. D., Taylor, G. K. & Treloar, P. J. (1994). Mechanisms of continental growth in extensional arcs: an example from the Andean Plate Boundary Zone. Geology 22, 391–394.
Guidotti, C. V. (1963). Metamorphism of the pelitic schists in the Bryant Pond quadrangle, Maine. American Mineralogist 48, 772–791.
Guidotti, C. V. (1989). Metamorphism in Maine: an overview. In: Tucker, R. D & Marvinney, R. G. (eds) Studies in Maine Geology, Volume 3: Igneous and Metamorphic Geology. Maine Geological Survey, pp. 1–19.
Guidotti, C. V. (1993). Textural aspects of high T–low P polymetamorphism in the Rangeley Area, western Maine: general implications for studies of Acadian metamorphic rocks in New England. Geological Society of America, Northeastern Section Meeting, Abstracts with Programs 25, A21.
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]
Harris, N. B. W. & Inger, S. (1992). Trace element modelling of pelite-derived granites. Contributions to Mineralogy and Petrology 110, 46–56.
Harris, N. B. W., Inger, S. & Massey, J. (1993). The role of fluids in the formation of High Himalayan leucogranites. In: Treloar, P. J. & Searle, M. P. (eds) Himalayan Tectonics. Geological Society, London, Special Publications 74, 391–400.
Harris, N., Ayres, M. & Massey, J. (1995). Geochemistry of granitic melts produced during the incongruent melting of muscovite: implications for the extraction of Himalayan leucogranite magmas. Journal of Geophysical Research 100, 15767–15777.
Holdaway, M. J., Guidotti, C. V., Novak, J. M. & Henry, W. E. (1982). Polymetamorphism in medium- to high-grade pelitic metamorphic rocks, west–central Maine. Geological Society of America Bulletin 93, 572–584.
Holdaway, M. J., Mukhapodhyay, B., Dyar, M. D., Guidotti, C. V. & Dutrow, B. L. (1997). Garnet–biotite geothermometry revised: new Margules parameters and a natural specimen data set from Maine. American Mineralogist 82, 582–595.[Abstract]
Holk, G. J. & Taylor, H. P., Jr (1997). 18O/16O homogenization of the middle crust during anatexis: the Thor–Odin metamorphic core complex, British Columbia. Geology 25, 31–34.
Hubbard, M. S., West, D. P., Jr, Ludman, A., Guidotti, C. V. & Lux, D. R. (1995). The Norumbega fault zone, Maine: a mid- to shallow-level crustal section within a transcurrent shear zone. Atlantic Geology 31, 109–116.
Hutton, D. H. W. & Reavy, R. J. (1992). Strike-slip tectonics and granite petrogenesis. Tectonics 11, 960–967.[Web of Science]
Icenhower, J. & London, D. (1995). An experimental study of element partitioning among biotite, muscovite and coexisting peraluminous silicic melt at 200 MPa (H2O). American Mineralogist 80, 1229–1251.[Abstract]
Icenhower, J. & London, D. (1996). Experimental partitioning of Rb, Cs, Sr and Ba between alkali feldspar and peraluminous melt. American Mineralogist 81, 719–734.[Abstract]
Johannes, W. (1988). What controls partial melting in migmatites? Journal of Metamorphic Geology 6, 451–465.[Web of Science]
Kretz, R. (1983). Symbols for rock-forming minerals. American Mineralogist 34, 345–368.
Lagache, M. & Carron, J. P. (1982). Zonation des éléments en traces au cours de la croissance des cristaux dans les bains silicatés: l’exemple de Rb, Cs, Sr and Ba dans le système Qz–Ab–Or–H2O. Geochimica et Cosmochimica Acta 46, 2151–2158.
LeFort, P., Cuney, M. Deniel, C., France-Lanord, C., Sheppard, S., Upreti, B. N. & Vidal, P. (1987). Crustal generation of Himalayan leucogranite. Tectonophysics 134, 39–57.
Ludman, A. (1998). Evolution of a transcurrent fault system in shallow crustal metasedimentary rocks: the Norumbega fault zone, eastern Maine. Journal of Structural Geology 20, 93–107.
Mahood, G. & Hildreth, W. (1983). Large partition coefficients for trace elements in high-silica rhyolites. Geochimica et Cosmochimica Acta 47, 11–30.[Web of Science]
Marchildon, N. & Brown, M. (2001). Melt segregation in late-tectonic migmatites: an example from the Onawa contract aureole, Maine, U.S.A. Physics and Chemistry of the Earth (in press).
Menhert, W. D. (1968). Migmatites and Origin of Granitic Rocks. Amsterdam: Elsevier.
Miller, C. F. & Mittlefehldt, D. W. (1982). Depletion of light rare-earth elements in felsic magmas. Geology 10, 129–133.
Miller, R. B. & Paterson, S. R. (1999). In defense of magmatic diapirs. Journal of Structural Geology 21, 1161–1173.
Milord, I., Sawyer, E. W. & Brown, M. (2001). Formation of diatexite migmatite and granite magma during anatexis of semi-pelitic metasedimentary rocks: an example from St. Malo, France. Journal of Petrology 42, 487–505.
Mittlefehldt, D. W. & Miller, C. F. (1983). Geochemistry of the Sweetwater Wash pluton, California: implications for ‘anomalous’ trace element behavior during differentiation of felsic magmas. Geochimica et Cosmochimica Acta 47, 109–124.
Montel, J. M. (1993). A model for monazite/melt equilibrium and application to the generation of granitic magmas. Chemical Geology 110, 127–146.
Nabelek, P. I. (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, P. I. & Glascock, M. D. (1995). REE-depleted leucogranites, Black Hills, South Dakota: a consequence of disequilibrium melting of monazite-bearing schists. Journal of Petrology 36, 1055–1071.
Nash, W. P. & Crecraft, H. R. (1985). Partition coefficients for trace elements in silicic magmas. Geochimica et Cosmochimica Acta 49, 2309–2322.[Web of Science]
Patiño Douce, A. E. (1999). What experiments tell us about the relative contributions of crust and mantle to the origin of granitic magmas. In: Castro, A., Fernandez, C. & Vigneresse, J.-L. (eds) Understanding Granites. Integrating New and Classical Techniques. Geological Society, London, Special Publications 168.
Patiño Douce, A. E. & Harris, N. (1998). Experimental constraints on Himalayan anatexis. Journal of Petrology 39, 689–710.
Paterson, S. R. & Miller, R. B. (1998). Magma emplacement during arc-perpendicular contraction. Tectonics 17, 571–856.
Paterson, S. R., Fowler, T. K., Jr & Miller, R. B. (1996). Pluton emplacement in arcs: a crustal-scale exchange process. Transactions of the Royal Society of Edinburgh 87, 115–123.
Power, G. M. (1993). Geochemical differences between the Cadomian granites of Mancellia and the St. Malo migmatites, Armorican Massif, France. Journal of the Geological Society, London 150, 465–468.
Pressley, R. A. & Brown, M. (1999). The Phillips Pluton, Maine, USA: evidence of heterogeneous crustal sources and implications for granite ascent and emplacement mechanisms in convergent orogens. Lithos 49, 335–366.
Rapp, R. P. & Watson, E. B. (1986). Monazite solubility and dissolution kinetics: implications for the Th and LREE chemistry of felsic magmas. Contributions to Mineralogy and Petrology 94, 304–316.
Ratcliffe, N. M., Hames, W. E. & Stanley, R. S. (1998). Interpretation of ages of arc magmatism, metamorphism, and collisional tectonics in the Taconian orogen of western New England. American Journal of Science 298, 791–797.[Web of Science]
Robinson, P., Tucker, R. D., Bradley, D., Berry, H. N., IV & Osberg, P. H. (1998). Paleozoic orogens in New England, USA. GFF 120, 119–148.[Web of Science]
Rubie, D. C. & Brearley, A. J. (1987). Metastable melting during the breakdown of muscovite + quartz at 1-kbar. Bulletin de Minéralogie 110, 533–549.
Rushmer, T. (1996). Melt segregation in the lower crust: how have experiments helped us? Transactions of the Royal Society of Edinburgh: Earth Sciences 87, 73–83.[Web of Science]
Rutter, E. H. (1997). The influence of deformation on the extraction of crustal melts: a consideration of the role of melt-assisted granular flow. In: Holness, M. (ed.) Deformation-enhanced Melt Segregation and Metamorphic Fluid Transport. Mineralogical Society Series 8, 82–110.
Sawyer, E. W. (1994). Melt segregation in the continental crust. Geology 22, 1019–1022.
Sawyer, E. W. (1998). Formation and evolution of granite magmas during crustal reworking, the significance of diatexites. Journal of Petrology 39, 1147–1167.
Sawyer, E. W. (1999). Criteria for the recognition of partial melting. Physics and Chemistry of the Earth (A) 24, 269–279.
Schumacher, J. C., Hollicher, K. T., Robinson, P. & Tracy, R. J. (1990). Progressive reactions and melting in the Acadian metamorphic high of central Massachusetts in southwestern New Hampshire, USA. In: Ashworth, J. R. & Brown, M. (eds) High-temperature Metamorphism and Crustal Anatexis. New York: Unwin Hyman.
Sevigny, J. H. (1993). Monazite controlled Sm/Nd fractionation in leucogranite: an ion microprobe study of garnet phenocrysts. Geochimica et Cosmochimica Acta 57, 4095–4105.
Sevigny, J. H. Parrish, R. R. & Ghent, E. D. (1989). Petrogenesis of peraluminous granites, Monashee Mountains, southeastern Canadian Cordillera. Journal of Petrology 30, 557–581.
Smith, H. A. & Barreiro, B. (1990). Monazite U–Pb dating of staurolite grade metamorphism in pelitic schists. Contributions to Mineralogy and Petrology 105, 602–615.[Web of Science]
Solar, G. S. (1999). Structural and petrological investigations in the Central Maine belt, western Maine, with special reference to the migmatites. Ph.D. thesis, University of Maryland, College Park.
Solar, G. S. & Brown, M. (1999). The classic high-T–low-P metamorphism of west–central Maine, USA: is it post-tectonic or syn-tectonic? Evidence from porphyroblast–matrix relations. Canadian Mineralogist 37, 289–311.
Solar, G. S. & Brown, M. (2000). The classic high-T–low-P metamorphism of west–central Maine, USA: is it post-tectonic or syn-tectonic? Evidence from porphyroblast–matrix relations.—Reply. Canadian Mineralogist 38, 10071026.
Solar, G. S. & Brown, M. (2001). Contrasting patterns of strain and kinematic partitioning of flow during oblique contraction in the Central Maine belt, northern Appalachian orogen, USA. Journal of Structural Geology 22 (in press).
Solar, G. S., Pressley, R. A., Brown, M. & Tucker, R. D. (1998). Granite ascent in convergent orogenic belts: testing a model. Geology 26, 711–714.
Stromgard, K. E. (1973). Stress distribution during formation of boudinage and pressure shadows. Tectonophysics 16, 215–248.
Stüwe, K., Sandford, M. & Powell, R. (1993). Episodic metamorphism and deformation in low-pressure, high-temperature terranes. Geology 21, 829–832.
Taylor, S. R. & McLennan, S. M. (1985). The Continental Crust: its Composition and Evolution. An Examination of the Geochemical Record Preserved in Sedimentary Rocks. Oxford: Blackwell Science.
Thompson, A. B. & Tracy, R. J. (1976). Model systems for anatexis of pelitic rocks. II, facies series melting and reaction in the system CaO–KAlO2–NaAlO2–Al2O3–SiO2–H2O. Contributions to Mineralogy and Petrology 70, 429–438.
Tomascak, P. B. (1995). The petrogenesis of granitic rocks in southwestern Maine. Ph.D. thesis, University of Maryland, College Park.
Tomascak, P. B., Krogstad, E. J. & Walker, R.J. (1996). Nature of the crust in Maine, USA: evidence from the Sebago batholith. Contributions to Mineralogy and Petrology 125, 45–59.[Web of Science]
Vernon, R. H. & Collins, W. J. (1988). Igneous microstructures in migmatites. Geology 16, 1126–1129.
Wark, D. A. & Miller, C. F. (1993). Accessory mineral behavior during differentiation of a granite suite: monazite, xenotime and zircon in the Sweetwater Wash pluton, southeastern California, U.S.A. Chemical Geology 10, 49–67.
Watson, E. B. & Harrison, T. M. (1983). Zircon saturation revisited: temperature and compositional effects in a variety of crustal magma types. Earth and Planetary Science Letters 64, 295–304.[Web of Science]
Watt, G. R. & Harley, S. L. (1993). Accessory phase controls on the geochemistry of crustal melts and restites produced during water-undersaturated partial melting. Contributions to Mineralogy and Petrology 114, 550–566.
Watt, G. R., Burns, I. M. & Graham, G. A. (1996). Chemical characteristics of migmatites: accessory phase distribution and evidence for fast melt segregation rates. Contributions to Mineralogy and Petrology 125, 100–111.
Weinberg, R. F. (1999). Mesoscale pervasive felsic magma migration: alternatives to dyking. Lithos 46, 393–410.
West, D. P., Jr (1999). Timing of displacements along the Norumbega fault system, south–central and south–coastal Maine. In: West, D. P., Jr & Ludman, A. (eds) Norumbega Fault System of the Northern Appalachians. Geological Society of America, Special Paper 331, 167–178.
West, D. P., Jr & Hubbard, M. S. (1997). Progressive localization of deformation during exhumation of a major strike-slip shear zone: Norumbega fault zone, south–central Maine. Tectonophysics 273, 185–202.[Web of Science]
West, D. P., Jr, Guidotti, C. V. & Lux, D. R. (1995). Silurian orogenesis in the western Penobscot Bay region, Maine. Canadian Journal of Earth Sciences 32, 1845–1858.[Web of Science]
Wickham, S. M. (1987). The segregation and emplacement of granite magmas. Journal of the Geological Society, London 144, 281–297.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  F. Nehring, S. F. Foley, P. Holtta, and A. M. Van Den Kerkhof Internal Differentiation of the Archean Continental Crust: Fluid-Controlled Partial Melting of Granulites and TTG-Amphibolite Associations in Central Finland J. Petrology, January 7, 2009; (2009) egn070v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Brown Metamorphic patterns in orogenic systems and the geological record Geological Society, London, Special Publications, January 1, 2009; 318(1): 37 - 74. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. F. Weinberg and G. Mark Magma migration, folding, and disaggregation of migmatites in the Karakoram Shear Zone, Ladakh, NW India Geological Society of America Bulletin, July 1, 2008; 120(7-8): 994 - 1009. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Brown Crustal melting and melt extraction, ascent and emplacement in orogens: mechanisms and consequences Journal of the Geological Society, July 1, 2007; 164(4): 709 - 730. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Bhadra, S. Das, and A. Bhattacharya Shear Zone-hosted Migmatites (Eastern India): the Role of Dynamic Melting in the Generation of REE-depleted Felsic Melts, and Implications for Disequilibrium Melting J. Petrology, March 1, 2007; 48(3): 435 - 457. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. SLAGSTAD, R. A. JAMIESON, and N. G. CULSHAW Formation, Crystallization, and Migration of Melt in the Mid-orogenic Crust: Muskoka Domain Migmatites, Grenville Province, Ontario J. Petrology, May 1, 2005; 46(5): 893 - 919. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Brown Synergistic effects of melting and deformation: an example from the Variscan belt, western France Geological Society, London, Special Publications, January 1, 2005; 243(1): 205 - 226. [Abstract] [PDF]
|
|
|
|
|
|
T. JOHNSON and M. BROWN Quantitative Constraints on Metamorphism in the Variscides of Southern Brittany--a Complementary Pseudosection Approach J. Petrology, June 1, 2004; 45(6): 1237 - 1259. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. E. Johnson, T.E. Johnson, N.F.C. Hudson, and G.T.R. Droop Evidence for a genetic granite-migmatite link in the Dalradian of NE Scotland Journal of the Geological Society, May 1, 2003; 160(3): 447 - 457. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. E. Johnson, T. E. Johnson, M. Brown, and G. S. Solar Low-pressure subsolidus and suprasolidus phase equilibria in the MnNCKFMASH system: Constraints on conditions of regional metamorphism in western Maine, northern Appalachians American Mineralogist, April 1, 2003; 88(4): 624 - 638. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. G. BARNES, A. S. YOSHINOBU, T. PRESTVIK, O. NORDGULEN, H. R. KARLSSON, and B. SUNDVOLL Mafic Magma Intraplating: Anatexis and Hybridization in Arc Crust, Bindal Batholith, Norway J. Petrology, December 1, 2002; 43(12): 2171 - 2190. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. FORNELLI, G. PICCARRETA, A. DEL MORO, and P. ACQUAFREDDA Multi-stage Melting in the Lower Crust of the Serre (Southern Italy) J. Petrology, December 1, 2002; 43(12): 2191 - 2217. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. E. Johnson, T. E. Johnson, N. F. C. Hudson, and G. T. R. Droop Melt segregation structures within the Inzie Head gneisses of the northeastern Dalradian Scottish Journal of Geology, November 1, 2001; 37(2): 59 - 72. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||