| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Petrology Volume 42 Number 3 Pages 487-505 2001
© Oxford University Press 2001
Formation of Diatexite Migmatite and Granite Magma during Anatexis of Semi-pelitic Metasedimentary Rocks: an Example from St. Malo, France
1SCIENCES DE LA TERRE, UNIVERSITÉ DU QUÉBEC À CHICOUTIMI, CHICOUTIMI, QUÉBEC G7H 2B1, CANADA
2LABORATORY FOR CRUSTAL PETROLOGY, DEPARTMENT OF GEOLOGY, UNIVERSITY OF MARYLAND, COLLEGE PARK, COLLEGE PARK, MD 20742, USA
Received January 19, 2000; Revised typescript accepted June 2, 2000
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGEOLOGICAL SETTINGMETAMORPHISMMORPHOLOGY AND PETROGRAPHY OF...BIOTITE COMPOSITIONGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Petrological and geochemical variations are used to investigate the formation of granite magma from diatexite migmatites derived from metasedimentary rocks of pelitic to greywacke composition at St. Malo, France. Anatexis occurred at relatively low temperatures and pressures (<800°C, 4–7 kbar), principally through muscovite dehydration melting. Biotite remained stable and serves as a tracer for the solid fraction during melt segregation. The degree of partial melting, calculated from modal mineralogy and reaction stoichiometry, was <40 vol. %. There is a continuous variation in texture, mineralogy and chemical composition in the diatexite migmatites. Mesocratic diatexite formed when metasedimentary rocks melted sufficiently to undergo bulk flow or magma flow, but did not experience significant melt–residuum separation. Mesocratic diatexite that underwent melt segregation during flow generated (1) melanocratic diatexites at the places where the melt fraction was removed, leaving behind a biotite and plagioclase residuum (enriched in TiO2, FeOT, MgO, CaO, Sc, Ni, Cr, V, Zr, Hf, Th, U and REE), and (2) a complementary leucocratic diatexite (enriched in SiO2, K2O and Rb) where the melt fraction accumulated. Leucocratic diatexite still contained 5–15 vol. % residual biotite (mg-number 40–44) and 10–20 vol. % residual plagioclase (An22). Anatectic granite magma developed from the leucodiatexite, first by further melt–residuum separation, then through fractional crystallization. Most biotite in the anatectic granite is magmatic (mg-number 18–22).
KEY WORDS: anatexis; diatexite; granite magma; melt segregation; migmatite
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHISMMORPHOLOGY AND PETROGRAPHY OF...BIOTITE COMPOSITIONGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Partial melting experiments show that protoliths of pelite and greywacke composition yield up to
40 vol. % granite melt over the temperature range 650–900°C typical of anatexis of the continental crust (Conrad et al., 1988
; Vielzeuf & Holloway, 1988; Patiño Douce & Johnston, 1991; Vielzeuf & Montel, 1994; Patiño Douce & Harris, 1998). Therefore, to make a granite magma the melt fraction must be segregated from a solid fraction of much greater volume, and then collected together. The type of melt–residuum separation process depends on the melt fraction present.
Melt–residuum separation at low degrees of partial melting (F), and consequently low melt fraction (Mf), has been documented from metatexite migmatites. In these rocks, melt segregation produced centimetre-scale leucosomes distributed in a host composed of layers that have melted and others that have not (e.g. Henkes & Johannes, 1981; Gupta & Johannes, 1982; Dougan, 1983; Sawyer, 1987; Tait & Harley, 1988; Barbey et al., 1990; Brown, 1994). Leucosomes represent small volumes of felsic magma formed when melt was squeezed out of the deforming matrix (McLellan, 1988; Sawyer, 1991, 1994; Brown, 1994; Brown et al., 1995) and collected in low-pressure sinks, such as shear bands and inter-boudin partitions. This type of melt segregation process appears to be relatively simple; the melt fraction and solid fraction are almost perfectly separated into leucosome and melanosome, respectively. The melt composition reflects the particular melting processes that occurred; some leucosomes appear to have formed by equilibrium batch melting (Sawyer, 1987), others by disequilibrium melting (Barbey et al., 1989; Sawyer, 1991; Watt & Harley, 1993) and, no doubt compositions reflecting other melting schemes also exist.
At higher melt fractions the physical nature of the partially molten rock changes from solid dominated to melt dominated. The melt fraction may increase because (1) metamorphic temperature rises cause more melting (Brown, 1973), (2) melt from elsewhere is injected (Greenfield et al., 1996), or (3) melt is redistributed within the melting layer (Sawyer, 1998). As the melt fraction rises above the liquid percolation threshold, but remains below the melt-escape threshold (Vigneresse et al., 1996) bulk flow (Sawyer, 1996), or granular flow (Rutter, 1997), may occur. At this stage there is sufficient melt between the solid grains that rapid deformation can occur by processes such as grain boundary sliding accommodated by grain boundary diffusion through the melt phase. At still higher melt fractions, above the melt escape threshold, the partially melted rock becomes a dilute suspension of crystals in a melt and magma flow occurs. Once bulk flow or magma flow occurs pre-migmatization structures are destroyed and a diatexite migmatite is formed (Brown, 1973; Bea, 1991; Sawyer, 1999). The melt and residuum fraction in the magma can become separated if their flow velocities are different (Sawyer, 1996).
Sawyer (1998) has shown that high melt fraction diatexite migmatites formed during melting of tonalitic–trondhjemitic to granodioritic crust in the Opatica subprovince of Québec were the source (i.e. parental) to the granites found at higher crustal levels there. However, the details of the petrogenetic relationship between diatexite migmatite and derivative granite magmas have not yet been investigated for other common protoliths, such as metasedimentary rocks. This study examines the petrological and geochemical variations that occur during the relatively low-temperature (700–800°C) anatexis of siliclastic metasedimentary rocks, where a relatively large volume of granitic magma developed from diatexite migmatites. The St. Malo migmatite belt of Brittany was chosen because melting there occurred predominantly in the stability field of biotite (Brown, 1979; Weber et al., 1985). Hence, the petrological and compositional variations caused by the separation of the melt fraction from the solid fraction are not complicated by the development of new phases such as garnet or orthopyroxene, which might have markedly different responses to magma flow and hence produce a differentiation as a result of their individual hydraulic properties.
|
GEOLOGICAL SETTING |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHISMMORPHOLOGY AND PETROGRAPHY OF...BIOTITE COMPOSITIONGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The Armorican massif in northwestern France is divided into three parts by two major shear zone systems (Fig. 1). In the North Armorican Composite Terrane the last major orogenic activity to affect the late Proterozoic Brioverian rocks occurred in the late Precambrian (Cadomian, 590–540 Ma; Brown et al., 1990
). South of the North Armorican Shear Zone, the Brioverian and early Palaeozoic rocks of the Central Armorican Terrane and the South Armorican Terrane were subject to major orogenic deformation and exhumation during the Variscan orogeny (Gapais & Le Corre, 1980) between 400 and 300 Ma (Brown & Dallmeyer, 1996).
|
The Tregor–La Hague Terrane, in the northwest of the North Armorican Composite Terrane (Fig. 2a), is a 700–615 Ma continental arc sequence (Power et al., 1990) with local fragments of an older (Icartian, 2200–1800 Ma; Brown, 1995) basement. The arc-related St. Brieuc Terrane has no Icartian basement, and its oldest rocks belong to the Neoproterozoic Penthièvre complex, which is believed to be the root of an early Cadomian island arc and continental margin. The Penthièvre rocks form the basement to a Brioverian succession of tholeiitic volcanic rocks and clastic sedimentary rocks. The two arc terranes are separated from a predominantly siliclastic Brioverian succession in the St. Malo and Mancellian terranes, by the Fresnaye shear zone (Brown, 1995).
|
The St. Malo Terrane consists principally of greywacke metasedimentary rocks with minor calc-silicate and pelitic layers, but there are some volcanic rocks in the north. These sedimentary rocks were extensively migmatized (Brown, 1979
; Martin, 1980) during the 536 ± 14 Ma metamorphic event (Peucat, 1986). Similar looking, but compositionally distinct (Denis & Dabard, 1988; Dabard, 1990) sedimentary rocks crop out south of the sinistral, strike-slip Cancale Shear Zone; they constitute the Mancellian Terrane, which is of lower metamorphic grade and intruded by numerous granites (550–530 Ma; Brown, 1995).
Metasedimentary rocks along the Rance estuary south of St. Malo show an increase in metamorphic grade northwards from upper greenschist facies near Langrolay to amphibolite facies at La Landriais and Cancaval. Bedding is well preserved, and there is a general increase in the proportion of subsolidus quartzo-feldspathic veins northward. Rocks north of Cancaval and La Richardais have been partially melted and some are diatexite migmatites (Brun & Martin, 1978; Brown, 1979; Weber et al., 1985). Thus, the limit of melting lies farther south near La Landriais (Fig. 2b). Broadly, the distribution of migmatite types within the anatectic domain is zoned (Fig. 2). The outer parts consist mostly of metatexite migmatite, and the inner part principally of diatexite migmatite (Brown, 1979).
Magmatic foliations and lineations in diatexite migmatites and granites, respectively, dip and plunge to the north or northeast (Fig. 2) in the core of the St. Malo anatectic region, and are associated with top-to-the-south kinematic indicators. At the margins of the migmatite domain, anatectic rocks (diatexites and granites) have subvertical magmatic foliations and subhorizontal mineral elongation lineations; asymmetric structures indicate sinistral shear. Thus, Brown (1995) concluded that partial melting occurred during sinistral transpressive deformation, and argued that anatexis resulted from the thermal relaxation following contraction and inversion of the St. Malo Terrane basin during the oblique subduction of an aseismic ridge beneath the Armorican continental margin at 570 Ma.
|
METAMORPHISM |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHISMMORPHOLOGY AND PETROGRAPHY OF...BIOTITE COMPOSITIONGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The common mineral assemblage in non-migmatitic metasedimentary rocks along the Rance estuary is muscovite + biotite + plagioclase + quartz with accessory amounts of tourmaline, apatite, zircon and ilmenite; in addition, Weber et al. (1985)
have reported graphite. Alumina-rich bulk assemblages can contain >30 modal % muscovite. Some low-grade metasedimentary rocks contain grey poikiloblasts, which represent cordierite later altered to mica (Martin, 1977). All the metasedimentary rocks sampled come from the valley of the Rance.
The disappearance of muscovite and appearance of sillimanite along the Rance estuary are coincident with the appearance of migmatites. The principal mineral assemblage in the St. Malo migmatites (Fig. 3) generated by partial melting of the metasedimentary rocks is quartz + plagioclase + biotite ± K-feldspar + apatite + zircon + ilmenite + graphite. Fibrolitic sillimanite is rare, and occurs either as small clots of fibres, or as scattered needles in other phases, such as plagioclase and muscovite. Tourmaline is an accessory phase in some anatectic rocks, notably the leucocratic diatexite and anatectic granite. Most accessory phases (e.g. zircon and oxides), except apatite, are concentrated in biotite grains, hence their modal proportion tends to correlate with biotite. The mineralogical variation in the anatectic rocks is principally defined by variation in the modal proportion of the same basic mineral assemblage (Fig. 3). A few rocks contain large muscovite flakes, which appear to overgrow the syn-migmatization structures in the rocks; these muscovites are retrograde, possibly late magmatic (Brown, 1979). Chlorite after biotite, and white mica replacing feldspar are common low-temperature alteration products.
|
Garnet and orthopyroxene are absent from the samples examined in this study; however, a few samples of diatexite contain minor amounts (<5 vol. %) of cordierite, some of which is idioblastic and may have grown in a melt. Brown (1979) reported minor garnet at some localities, and Weber et al. (1985) reported minor amounts of cordierite in some rocks. Thus, minor incongruent breakdown of biotite may have occurred locally contributing <5 vol. % anatectic melt as suggested by Weber et al. (1985). Recently, however, Pickering & Johnston (1998) have shown that small amounts of garnet can form during muscovite dehydration melting reactions without involving the breakdown of biotite. Hence, melting of the St. Malo metasedimentary rocks occurred largely within the stability field of biotite, i.e. at temperature <800°C (Vielzeuf & Holloway, 1988; Patiño Douce & Harris, 1998), and involved principally the breakdown of muscovite (Brown, 1979). Relevant melting reactions involving muscovite, plagioclase and quartz (Thompson & Tracy, 1979) are
 |
 |
Whether reaction (1) or (2) was the principal melt-producing reaction will be discussed below, after the residual rocks have been identified. Rise in metamorphic temperature, and hence melting, appears to have stopped at about the time that muscovite disappeared, possibly because the widespread generation of anatectic melt buffered metamorphic temperatures; most of the biotite in the St. Malo migmatites is, therefore, relict.
|
MORPHOLOGY AND PETROGRAPHY OF THE PRINCIPAL ROCK TYPES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHISMMORPHOLOGY AND PETROGRAPHY OF...BIOTITE COMPOSITIONGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Four principal rock types can be distinguished within the St. Malo terrane: (1) metasedimentary rock; (2) metatexite migmatite; (3) diatexite migmatite; (4) anatectic granite. In many cases, the contact between the types is transitional. It is, therefore, necessary to clearly define the various rock types and to describe how they are related.
Metasedimentary rocks
Figure 2 shows a large region of metasedimentary rocks bordering the St. Malo migmatites to the south. This sequence of metasedimentary rocks consists predominantly of centimetre- to decimetre-scale layers (i.e. layers of contrasting composition inferred to be bedding) of fine- to medium-grained (>0·7 mm), quartzo-feldspathic metagreywacke (Fig. 4a). Thin, dark grey, fine-grained (<0·4 mm) pelitic layers and minor amounts of thin, pale green, calc-silicate rocks rich in plagioclase are interbedded with the greywackes. The metagreywackes and metapelites have a strong foliation defined by the shape-preferred orientation of biotite, and are locally folded.
|
Metatexite migmatite
Metatexite is defined as a migmatite containing evident pre-migmatization layering, foliation or banding, which survived partial melting (Brown, 1973
). Metatexites are migmatites that retained a low melt fraction, and in most only short-range transport of melt has occurred (Sawyer, 1996, 1999); their rheological properties are those of rock, rather than a magma. The St. Malo metatexite migmatites (Fig. 4b) consist of a dark grey, fine-grained (0·2–1 mm), compositionally layered host, which contains a small volume (<20 vol. %) of quartzo-feldspathic veins and pods (i.e. leucosomes). The dark grey host consists of two parts: (1) a lighter grey part that has the same texture, compositional layering and appearance as the metapelitic, metagreywacke and calc-silicate metasedimentary rocks; (2) a darker, well-foliated portion that is enriched in biotite and accessory phases, but depleted in leucocratic minerals compared with the metapelitic or metagreywacke protoliths. Typically, the biotite-rich portions are located at the edges of the leucosomes, and hence represent melanosomes or mafic selvedges. The leucosomes contain quartz, plagioclase, K-feldspar and minor biotite, and are texturally more homogeneous, and coarser grained than either the melanosomes or the metasedimentary rocks. The majority of leucosomes are parallel to the pre-migmatization structures (foliation and/or primary compositional layering) in their host, but some are cross-cutting. The metasedimentary rocks are considered to be the protolith, or palaeosome, to the metatexites, as the same type of compositional layering (bedding) and rock types can be recognized in them.
Diatexite migmatite
Diatexite has been defined as a migmatite in which the pre-migmatization structures are destroyed (Brown, 1973), and a homogenization and coarsening of texture occurs. A banding caused by flow in which the mafic minerals form schlieren is characteristically developed (Sawyer, 1996). Thus, diatexites represent rocks in which the melt fraction was large. As the melt fraction increases, the rheology of the whole becomes one of magma, and this allows large-scale material transport.
The diatexites at St. Malo show a considerable range in morphology. All have undergone a textural homogenization that has destroyed the primary centimetre-scale bedding typical of the metasedimentary rock and the metatexite. Pre-migmatization structures are preserved only in enclaves, which are locally abundant. Quartz-rich metagreywacke and calc-silicate enclaves predominate, but locally rounded quartz-rich enclaves are common, and are believed to represent the remains of vein quartz that existed in the palaeosome before migmatization. Generally, diatexite is characterized by increased grain size (0·5–5 mm), relative to either the metasedimentary palaeosome or the metatexite. There are systematic mineralogical (Fig. 3) and textural variations within the diatexite migmatites, and three subdivisions are made. Most diatexites have a strong preferred orientation of biotite and idioblastic–euhedral, elongate plagioclase and, as quartz domains remain equant, this suggests a magmatic or submagmatic foliation. However, many samples show some degree of deformation during cooling and small shear bands contain retrograde mineral assemblages.
Melanocratic diatexite
Melanocratic diatexite (Fig. 4c) has the highest biotite contents (>30 vol. %; Fig. 3a), and a significantly greater proportion of accessory phases (zircon, rutile, ilmenite, graphite, apatite and tourmaline) than the other diatexite types. They are generally well-foliated rocks (containing oriented biotite and plagioclase) and many are also flow banded on a centimetre scale; schlieren rich in biotite and accessory phases alternate with leucocratic bands (termed leucosomes in this study) rich in quartz, plagioclase, K-feldspar and apatite. Petrographically, melanocratic diatexite has a lepidoblastic, schistose texture in the mafic bands, and a granitic texture in the leucocratic bands commonly with flow-aligned plagioclase crystals. Some melanocratic diatexite contains biotite augen, which are either individual biotite porphyroblasts or aggregates of biotite grains; in many cases the biotite is crenulated or kinked, indicating that folding and disruption of mafic schlieren occurred during diatexite formation.
Mesocratic diatexite
Diatexite that contains between 10 and 30 vol. % biotite is mesocratic, and its composition overlaps that of the metasedimentary rocks. Biotite-rich schlieren may be present and define the foliation. Generally, biotite outside of the schlieren is not strongly oriented and, in some cases, these groundmass biotite grains are kinked or crenulated. Mesocratic diatexite generally has a medium to coarse grain size and a relatively uniform texture (Fig. 4d) with locally elongate euhedral plagioclase defining a magmatic foliation. Mesocratic diatexite typically occurs as concordant, or subconcordant, bodies ranging up to kilometre scale within the melanocratic diatexite.
Leucocratic diatexite
Leucocratic diatexite (Fig. 4e) contains <10 vol. % biotite, which is, in general, uniformly distributed, i.e. there are no biotite-rich schlieren, but some biotite aggregates do occur. These rocks are generally coarse grained and locally domains have a well-defined igneous texture (with crystal faces developed on feldspar against quartz) and, locally flow-oriented plagioclase and K-feldspar phenocrysts. Leucocratic diatexite contains more K-feldspar than the other types of diatexites (Fig. 3b), and forms small cross-cutting dykes, as well as much larger concordant and subconcordant sheets (commonly in reverse shear zones) and bodies within the other types of diatexite.
Besides the fractionation of biotite, the diatexite-forming process also fractionates the feldspar components (Fig. 3). Thus, in the sequence from mesocratic diatexite, the pale coloured, most uniform textured leucocratic diatexite has the highest K-feldspar/plagioclase ratio and the lowest biotite content, and represents the most evolved type of diatexite.
Anatectic granite
The leucocratic diatexite and the anatectic granite are generally found together, and their contacts are commonly gradational. Anatectic granite (Fig. 4f) is the most leucocratic (Fig. 3) and most homogeneous rock type in the St. Malo migmatite terrane, although pegmatitic layers and patches occur locally. They are predominantly medium- to coarse-grained (1–7 mm, locally 10 mm) monzogranites with minor biotite. The biotite crystals are texturally different from those in the diatexite; they are not crenulated. Biotite crystals are not generally well oriented, but in some rocks they define a magmatic foliation. Anatectic granite occurs as small veins located in late extensional shears and fractures within the diatexite migmatite, and as larger (up to kilometre-scale) bodies intruded late in the history of the major strike-slip shear zones. There are three groups of anatectic granite: (1) biotite granite, which comprises most of the larger bodies; (2) two-mica granite (biotite < muscovite) typical of the small veins in the diatexite, but also forming small patches in the biotite granite; (3) tourmaline–mica granite, which occurs as small patches in the biotite granite where skeletal tourmaline is intergrown with quartz.
|
BIOTITE COMPOSITION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHISMMORPHOLOGY AND PETROGRAPHY OF...BIOTITE COMPOSITIONGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Examination of the anatectic granites and leucocratic diatexites reveals two types of biotite: large equant or slightly elongate grains and a population of smaller, more bladed grains. They are also compositionally different. Biotite compositions have been determined (typically on 15–20 grains per thin section) by electron microprobe analyses (wavelength-dispersive mode) for two melanocratic diatexites and two anatectic granites (one biotite and one tourmaline granite). The mg-number [100Mg/(Mg + Fe)] for the large biotites in the melanocratic diatexite ranges between 41 and 47, and for the smaller bladed biotite in the anatectic granite between 12 and 18. The mg-number for small bladed biotite grains in the tourmaline-bearing anatectic granite ranges between 38 and 42. The titanium content of biotite, expressed as Ti atoms per formula unit (based on 22 oxygens), ranges from 0·19 to 0·29 a.p.f.u. for the small bladed biotite grains in anatectic granite, and from 0·20 to 0·37 for biotite in the other rock types.
|
GEOCHEMISTRY |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHISMMORPHOLOGY AND PETROGRAPHY OF...BIOTITE COMPOSITIONGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Major and trace element contents have been determined on five metasedimentary rocks, nine melanocratic diatexites, 13 mesocratic diatexites, 17 leucocratic diatexites, five biotite schlieren, four anatectic granites and three leucosome samples from St. Malo. Between 2 and 3 kg of each sample were crushed, and
200 g taken and pulverized in an aluminium oxide ceramic mill. Major oxides (except Na2O) and the trace elements Ni, V, Cu, Pb, Zn, Ba, Sr, Ga, Nb, Zr and Y were determined by standard X-ray fluorescence (XRF) methods at McGill University. The trace elements Cr, Co, Sc, Au, Rb, Cs, Ta, Hf, Th, U, La, Ce, Nd, Sm, Eu, Tb, Yb and Lu, plus Na2O were determined at Université du Québec à Chicoutimi using instrumental neutron activation analysis (INAA) after irradiation in a SLOWPOKE II reactor at the Université de Montréal. Representative results are presented in Table 1. The full geochemical dataset is available on the Journal of Petrology Web site at http://www.petrology.oupjournals.org. Geochemical variation diagrams also include XRF data for 25 St. Malo samples from Brown (1979) and five from Martin (1979).
|
Metasedimentary rocks
Field observations suggest that the metasedimentary rocks are the protolith to the migmatites. Therefore, the main geochemical features of the siliclastic metasedimentary rocks are described first to characterize the palaeosome, or protolith. Then, the composition of the migmatites will be described.
The SiO2 content for the metasedimentary rocks ranges between 63 and 73 wt % (Fig. 5). Most of the major oxides and trace elements, e.g. TiO2, Al2O3, FeOT, MgO, K2O, P2O5, Cr, Ni, Sc, V, Rb, Cs, Ba, Ta, Nb, U, Th and the heavy rare earth elements (HREE) show negative correlation with SiO2; only CaO, Na2O and Sr show a clear, positive correlation with SiO2 in the metasedimentary rocks. The light rare earth elements (LREE), Hf and Zr show no clear correlation with SiO2 content. The decrease in FeOT, MgO, TiO2 and Sc, but increasing CaO and Na2O contents of the metasedimentary rocks with increasing SiO2 content indicate a decrease in the ferromagnesian mineral to feldspar ratio as the SiO2 content rises, a pattern characteristic of siliclastic sediments in general (Shaw, 1956; Taylor & McLennan, 1985; Sawyer, 1986).
|
Multi-element variation diagrams (Fig. 6a) normalized to the primitive mantle of Taylor & McLennan (1985) show that the metasedimentary rocks all have similar-shaped patterns in which the portion from Ba to K is much flatter than the portion from La to Lu. All the samples have marked negative Ta, Nb, Sr, P and Ti anomalies. Negative Ta and Nb anomalies are characteristic of volcanic and plutonic rocks formed above subduction zones (Saunders et al., 1980), and are consistent with the St. Malo sedimentary rocks being derived through weathering of a volcanic arc (Dabard, 1990). Negative Sr, P and Ti anomalies (Fig. 6a) may reflect the crystallization of plagioclase, apatite and either ilmenite or rutile from the magma from which the source rocks to the sediments formed, rather than as a product of the sedimentary processes.
|
Diatexite migmatite and anatectic granite
About a third of the diatexite samples fall in the compositional field defined by the metasedimentary rocks (Fig. 5). These diatexites are predominantly those termed mesocratic on the basis of their biotite content. The diatexite samples outside the metasedimentary rock field define two groups, which also correspond to petrographic subdivisions. Melanocratic diatexite samples have lower SiO2 and Na2O, but higher TiO2, FeOT, MgO, Sc, Ni, Cr, V, Hf, Zr, U, Th and REE contents than the metasedimentary rocks, whereas the leucocratic diatexite samples display generally higher SiO2, but lower TiO2, FeOT, MgO, Sc, Ni, Cr, V, Hf, Zr, U, Th and REE contents.
The mesocratic diatexite samples have multi-element patterns (Fig. 6b) that closely resemble those of the St. Malo metasedimentary rocks. The samples of melanocratic diatexite (Fig. 6c) have slightly higher-level patterns than either the metasedimentary rock or the mesocratic diatexite samples, but they still have strong negative Ta, Nb, Sr, P and Ti anomalies. In contrast, the leucodiatexite samples show greater variability in both the level and shape of their patterns (Fig. 6d). Generally, they have lower patterns, which retain the negative Ta, Nb and Ti anomalies, but their negative Sr and P anomalies are greatly reduced or, in some cases, are reversed (are positive).
The anatectic granites lie close to the field of leucocratic diatexite, but extend to higher SiO2 and K2O, and lower TiO2, FeOT and MgO contents (Fig. 5). Their multi-element patterns are lower than for the diatexites (Fig. 6d), but still have large negative Nb, Ta and Ti anomalies, although some samples have positive P, Sr and K anomalies.
The metasedimentary rock, mesocratic and melanocratic diatexite samples have very similar REE patterns, with La/LuN ratios of between nine and 15, and negative Eu anomalies (Fig. 7). However, the leucocratic diatexite and anatectic granite show much more variation in the total REE content; decreasing LREE levels cause the La/LuN ratio to range between 23·4 and 2·9; moreover; the samples with the lowest total REE contents have positive Eu anomalies.
|
The mafic schlieren consist principally of biotite (65–80 vol. %), quartz (15–30 vol. %) and accessory phases; only the schlieren with the lowest biotite contents contain plagioclase (<10 vol. %). Consequently, the mafic schlieren are strongly enriched in TiO2, FeOT + MgO, K2O, Rb, Sc, V, Cr, Co, Zr, Hf, U, Th, Nb, Ta and REE, but depleted in SiO2, CaO, Na2O, Sr and P relative to the metasedimentary rock and diatexite samples (Figs 5 and 6c). Leucosome bands plot on the opposite side of the melanocratic diatexite field to the mafic schlieren, at high SiO2 contents and low K2O contents; however, some have relatively high FeOT + MgO, CaO and TiO2 contents. Mafic schlieren have the highest level multi-element and REE patterns (Figs 6 and 7), consistent with their high abundance of accessory phases. The leucosome band with least biotite has one of the lowest patterns in Figs 6 and 7, and a conspicious positive Eu anomaly.
The composition fields of the diatexite and metasedimentary rock samples overlap (Figs 5 and 6). This observation, together with field and petrographic evidence, suggests that migmatization of the metasedimentary palaeosome to mesocratic diatexite was an isochemical process—at least on the sample scale. However, about half of the diatexite samples lie outside the palaeosome field, and for these samples migmatization was a geochemically open process at the scale of the sample. Thus, the next section discusses the causes of the compositional diversity shown by the diatexite samples.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHISMMORPHOLOGY AND PETROGRAPHY OF...BIOTITE COMPOSITIONGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
As partial melting occurred in the St. Malo migmatites, some compositional variation is expected to result from (1) variable degrees of melt–residuum separation between samples and (2) fractional crystallization of the melt phase. The effect of these processes on the diatexite will be examined in terms of vectors for the principal minerals (biotite, plagioclase, K-feldspar and quartz) plus the melt, using diagrams that plot a component characteristic of the melt fraction against one representing the residuum.
Choice of a melt composition
If it is assumed that the trend of leucocratic diatexite samples in Fig. 5 results from mixing of anatectic melt with residual phases, then the average composition of the melt can be obtained from Fig. 5 by projecting the trend formed by the leucocratic diatexite and anatectic granite to the low TiO2 and FeOT + MgO contents of the anatectic melt. Assuming TiO2 0·05 wt %, this exercise yields a melt composition of 75 wt % SiO2, from which 15 wt % Al2O3, 0·6 wt % FeOT + MgO, 0·7 wt % CaO, 3·5 wt % Na2O and 3 wt % K2O can be derived from the other plots. Alternatively, a suitable experimental melt composition can be used. Patiño Douce & Harris (1998) partially melted a muscovite–biotite schist, which happens to have a composition close to the St. Malo sedimentary rocks, at P–T conditions similar to those inferred by Brown (1979) to have caused anatexis at St. Malo. In particular, the glass obtained from the muscovite–biotite schist at 6 kbar and 750°C generally matches the melt composition estimated from the diatexite data (Fig. 5); the main discrepancy is the higher K2O content of the experimental melt. Nevertheless, the Patiño Douce & Harris (1998) 6 kbar, 750°C melt of muscovite–biotite schist (hereafter called MBS melt) is used as the melt composition in modelling the St. Malo migmatites.
Partial melting
Figure 8 is a plot of a component (FeOT + MgO) representing the residual ferromagnesian phase biotite versus a component (K2O) representing the anatectic melt. The metasedimentary rock field is described by the quartz, plagioclase, biotite and muscovite vectors—in accordance with the petrographic observations. The diatexite field lies along the quartz, plagioclase and biotite trends, but not towards muscovite, indicating that muscovite is not a constituent in the diatexite.
|
Melanocratic diatexite defines a field on the opposite side of the metasedimentary rock field to the melt composition, and between the biotite and quartz–plagioclase vectors, consistent with the presence of a large residuum component. There is no trend towards cordierite in Fig. 8; it is either absent, or present in minor amounts. Thus, there is no geochemical signature that the incongruent breakdown of biotite is a major factor in generating the composition of the St. Malo migmatites, as was also concluded by Weber et al. (1985).
Patiño Douce & Harris (1998) determined the stoichiometry of reactions (1) and (2) to be
 |
 |
Assuming that the muscovite-consuming, melt-producing reaction at St. Malo had the same stoichiometry as reaction (1a) or (2a), then the maximum degree of partial melting can be calculated from the modal mineralogy of the metasedimentary rocks. Selecting metasedimentary rocks that represent the full range of composition shows (Table 2) that the most siliceous protolith compositions (HM3) are the least fertile, because of their low modal muscovite. Because of the high plagioclase/muscovite ratio in reaction (1a), melt yield is limited by the plagioclase content in the case of H2O-fluxed melting, and some residual compositions lacking plagioclase are generated. In contrast, dehydration melting by reaction (2a) terminates when muscovite is entirely consumed, and all residual rocks still contain plagioclase (Table 2). The ranges for F are 0–46 vol. % for H2O-fluxed partial melting [reaction (1a)], and 0–37 vol. % for muscovite dehydration melting [reaction (2a)].
|
Origin of the residual rocks
Subtracting the melt composition in proportion to F (Table 2) from the whole-rock metasediment compositions yields an estimate of the residuum composition. The residuum trends using MBS melt are shown in Figs 8 and 9 for muscovite dehydration melting. The melt depletion vectors are shortest for the muscovite-poor (highest SiO2, lowest K2O) samples and longest for the most pelitic, muscovite-rich (lowest SiO2, highest K2O) samples. Significantly, the residuum vector obtained for the most typical of metasediment compositions closely matches the position of the melanocratic diatexite samples. Repeating the procedure for other melt compositions gives similar results. The melanocratic diatexite can, therefore, be modelled (Figs 8 and 9) as the simple residuum left after the extraction of a moderate ((40 vol. %) melt fraction generated in the palaeosome by muscovite breakdown. The mesocratic diatexites represent partially melted metasedimentary rocks from which little or no melt extraction occurred.
|
The mafic schlieren also have residual compositions, but a more complex origin. The position of the mafic schlieren in Figs 8 and 9 could be modelled as a simple melt-extraction residuum; this would require a minimum of 60 vol. % and typically 70 vol. % melting. Such high degrees of partial melting are similar to those proposed by Weber et al. (1985) to form the melanosomes in the metatexites. For F to be so high the palaeosome bulk composition must closely match the stoichiometry of the melt-producing reaction, with additional biotite and quartz to be the excess phases. If the results in Table 2 are representative of the metasedimentary rocks, none has such a composition. A mass balance could be made if, in addition to muscovite-generated melt, a further melt fraction derived from biotite breakdown was lost, but the minor presence of cordierite, and the absence of garnet and orthopyroxene in the schlieren render this unlikely. Extreme contamination of a melt by up to 80 vol. % residual biotite could generate compositions similar to the biotite schlieren, but this mechanism is unlikely too, as the mafic schlieren lack K-feldspar and most lack plagioclase. The origin of the mafic schlieren is problematic, and may require a mechanism that collects together residual biotite flakes during the magma-like flow of the mesocratic diatexite (I. Milord, unpublished data, 1999).
Unmixing of the residuum and melt contamination
About one-third of the diatexite samples lie to the high K2O and SiO2, but low FeOT + MgO side of the mesocratic diatexite field (Fig. 5); these are the leucocratic diatexites. These samples also form separate trends in Fig. 5a and b (and also Sc vs SiO2), where they lie below the mesocratic diatexite field, whereas, in an Al2O3 vs SiO2 plot (Fig. 5e), they lie above it. This trend can be modelled (Fig. 9a and b) as the mixture of melt and residual biotite (mg-number 44), but not as a mixture of melt plus magmatic biotite (neither biotite with mg-number 12–18, from the anatectic granites, nor biotite with mg-number 22 obtained from MBS melt), which yields a trend (2 in Fig. 9) that lies above the array of leucocratic diatexites. The modelled trend for residual biotite confirms that melt MBS is too siliceous and K2O rich compared with the actual anatectic melt generated at St. Malo. The calculated mixing trends (Figs 8 and 9) indicate that the leucocratic diatexites represent a range of mixtures between melt and up to 15 vol. % residual biotite.
The model MBS melt + residual biotite trend does not match the sample trend well in Fig. 8, which is displaced to lower K2O. Thus, the St. Malo melt must have been less potassic than MBS melt. However, the sample trend can be satisfactorily modelled in (FeOT + MgO)–K2O composition space if the 6 kbar, 750°C melt of muscovite schist with 2 vol. % H2O added (MS melt) from Patiño Douce & Harris (1998) is used. Significantly, this melt composition matches that inferred for St. Malo from the Harker diagrams, although the composition of the muscovite schist from which MS melt was derived does not match the St. Malo metasedimentary rocks as closely as the muscovite–biotite schist does.
Residual plagioclase is another common contaminant of the melt in anatectic systems. The leucocratic diatexite samples follow the melt + residual plagioclase (An22) trend in Fig. 9, but are displaced slightly to lower SiO2, consistent with a reduced SiO2 content as a result of the presence of residual biotite. The addition of residual plagioclase should decrease the K2O content, particularly for samples with little residual biotite, and some leucocratic diatexite samples show this (Figs 8d and 9d). Thus, the mixing trends shown in Figs 8d and 9 show that most leucocratic diatexite contains between 10 and 20 vol. % residual plagioclase; however, some samples contain 30–40 vol. % residual plagioclase, in addition to the residual biotite they contain.
Melt crystallization
Figure 9 shows that the anatectic granite (except samples 740 and 742, see below) have much higher K2O, but lower CaO contents, whereas the leucosome bands have much lower K2O contents and higher CaO than either the MBS or MS melts. The K2O-rich samples are enriched in Ba relative to the low-K2O samples. In both groups there is an increase in Rb/Sr ratio with K2O content; from 0·11 to 0·59 for the low-K2O group and from 0·23 to 1·23 for the high-K2O group. Silica content ranges from 74 to 77 wt %, and is lowest in the high-K2O group. These compositional changes may be due to fractional crystallization of the anatectic melt yielding a (quartz + plagioclase)-dominated cumulate and a K-feldspar-rich, fractionated (residual) liquid. The low-K2O samples contain between 3 and 13 vol. % biotite, whereas the high-K2O samples contain only 0·2–4 vol. % biotite. The bulk mg-number for all the low-K2O samples is 41, but for the high-K2O samples it decreases to 20. This indicates that the biotite present in the plagioclase-rich, low-K2O, samples are from the residuum (mg-number 40–44), but that most of the biotite in the K-feldspar-rich samples crystallized from the anatectic melt (mg-number 18–22). Thus, much of any remaining residual material in the St. Malo anatectic magma was removed during the early stages of crystallization.
Because Eu is compatible with crystallizing feldspar, and the other REE are incompatible, early-formed feldspar cumulates are enriched in Eu and depleted in the other REE, resulting in positive Eu anomalies. Fractionated melts are depleted in Eu and enriched in the other REE, so they have negative Eu anomalies. In the St. Malo case, the plagioclase-rich sample with the lowest biotite content (IM30Hf) has a large positive Eu anomaly (Fig. 7d) and a low Rb/Sr ratio (0·11), consistent with being a cumulate of early-crystallized plagioclase. However, other plagioclase-rich rocks with higher (9–13 vol. %) residual biotite contents and similar Eu contents have much higher LREE and HREE contents, resulting in weak, negative Eu anomalies. This effect, caused by the abundance of accessory phase inclusions in the residual biotite, can be modelled by taking a plagioclase-free biotite schlieren and normalizing the REE content to a quartz-free basis (Table 3), and adding this residuum biotite, plus its included accessory phases, to leucosome IM30Hf (Fig. 10). Adding 12 vol. % residual biotite yields a REE pattern similar to the biotite-rich samples IM30Cf and IM35Hf. Adding >17 vol. % residual biotite, with its complement of LREE- and HREE-bearing accessory phases, completely masks any positive Eu anomally owing to plagioclase accumulation, and generates a negative anomaly. This effect also applies to the melanocratic diatexites (Fig. 7c) where their increased content of residual biotite plus accessory phases produces strong negative Eu anomalies.
|
|
All the K2O-rich samples have positive Eu anomalies (Fig. 7d); three samples (IM11A, IM30L and IM38B) have Eu contents similar to the plagioclase-rich samples. Hence, their positive Eu anomalies indicate that they were derived from melts depleted in the other REE, probably because the accessory phases containing them remained armoured in the biotite and were not available to the melt. Significantly, the most potassic samples have the highest Rb/Sr ratios (1·23 and 0·83, respectively), lowest biotite contents (<2 vol. %), and the lowest Eu contents. Thus, there is a strong geochemical signature of fractional crystallization in the melt-rich rocks at St. Malo.
Water-fluxed melting or dehydration melting?
Water-fluxed melting of muscovite schist at high pressure yields initial melts with higher Na2O/K2O ratios than melts formed at low pressure (Patiño Douce & Harris, 1998). Brown (1979) sampled two trondhjemite veins (740 and 742) from the Rance estuary, well south of, and structurally below, the St. Malo migmatites; these indicate that H2O-fluxed melting of the metasedimentary rocks occurred. Moreover, the trondhjemites have Na2O/K2O ratios (3·2 and 3·8, respectively) similar to the 10 kbar runs (Na2O/K2O ratio 3–4), but far higher than the 6 kbar runs (Na2O/K2O = 1·7) of Patiño Douce & Harris (1998). The trondhjemites from the Rance estuary could be initial melts tapped from a level well below the 4–7 kbar migmatites now exposed at St. Malo.
The residual mineral assemblage produced by H2O-fluxed melting should be plagioclase rich (Harris et al., 1995), but variations in the critical mineral proportions result in some quartz-depleted and muscovite-rich assemblages (Table 2). All the melanocratic diatexites contain significant amounts of quartz and plagioclase, which suggests they are the products of muscovite dehydration melting. However, the diatexites at some localities (e.g. Plage du Val) contain small, rounded enclaves (IM30D) consisting of sillimanite, muscovite, biotite and quartz, with little or no plagioclase. As the composition of the enclaves lies on the opposite side of the metasedimentary rock field to the melt (Fig. 5), and muscovite probably replaced K-feldspar, they could be the remains of the plagioclase-free residuum from H2O-fluxed melting. Furthermore, the mafic schlieren in some diatexites are plagioclase free and could also be the residuum of H2O-fluxed melting.
Significantly, it is within the metatexite migmatite envelope around the diatexite migmatites that plagioclase- and quartz-poor assemblages are most abundant. Weber et al. (1985) documented the widespread occurrence of biotite-rich melanosomes with 5–15 vol. % quartz + plagioclase, which they were able to model as the residuum left after 60–70 vol. % H2O-fluxed partial melting.
The progression from migmatite to granite at St. Malo
Most studies of diatexite migmatites (Brown, 1979; Wickham, 1987; Obata et al., 1994) could not identify a major residuum component. Thus, it could be argued, on mass balance constraints, that granites are injected into migmatite terranes (White & Chappell, 1990). Recently, an extensive volume of diatexite migmatite with a residual composition was described in the Opatica Subprovince (Sawyer, 1998). This study confirms the widespread occurrence of residual migmatites (the melanocratic diatexite) in the St. Malo migmatite terrane first suggested by Weber et al. (1985). The case can now be made at St. Malo that, as in the Opatica Subprovince, crustal anatexis generated granite magma from diatexite migmatite, in an essentially closed-system process. The steps in this process are summarized in Fig. 11.
|
Anatexis at St. Malo began with H2O-fluxed melting (Weber et al., 1985) that was of very limited extent, but produced high (up to 75 vol. %) degrees of partial melting locally. The extent of melting was limited by the rapid exhaustion of the available water, but generated a melt fraction in excess of the liquid percolation threshold (Vigneresse et al., 1996), which migrated short distances to collect into lensoid and vein-like leucosomes. These rocks are the metatexite migmatites best preserved in the southern part of the migmatite belt. Rocks in the centre of the St. Malo Terrane reached higher temperatures and underwent a further reaction in which muscovite dehydration melting generated a much larger and more pervasive melt fraction. This melt fraction was close to, or above, the melt escape threshold, and so allowed magma flow to occur, which formed the mesocratic diatexite. Remarkably little remains of the metasedimentary sequence in the diatexites, suggesting that virtually all the metasedimentary sequence was fertile and generated enough melt that flow occurred on a pervasive, regional scale, destroying pre-migmatization structures. The mafic schlieren may represent relics of the earlier H2O-fluxed melting reaction in the protolith to the diatexites. This is in marked contrast to melting of the Opatica plutonic arc, where most of the crust was infertile and retained its pre-migmatization structures (Sawyer, 1998).
The solid and melt fractions in the diatexite magma became progressively separated during flow (Sawyer, 1996) generating residuum-enriched and melt-enriched portions (2, Fig. 11). The melanocratic diatexites are residuum left after the extraction of up to 40 vol. % melt from partially molten mesocratic diatexite. Mafic schlieren represent a second, smaller-scale, more residuum-enriched part of the diatexite migmatites. The leucocratic diatexites are the melt-enriched part of migmatites, but contain between 5 and 50 vol. % residuum (25 vol. % is typical); 5–15 vol. % biotite, and the rest mostly plagioclase. The decreasing residual component is reflected in coarser grain size, presence of magmatic textures locally, and a less conspicious foliation in the leucocratic diatexite as it grades into anatectic granite.
When metamorphic temperature fell and the diatexite began to crystallize (3, Fig. 11) the geochemical signature of feldspar fractional crystallization replaces that of melt–residuum separation. Because regional deformation continued during crystallization, the early-crystallized plagioclase and the now fractionated melt were separated, giving rise to progressively more evolved, K2O-rich compositions of the anatectic granite (4, Fig. 11). The first accumulations of early-crystallized plagioclase contain the last residual biotite; later plagioclase accumulations and all the anatectic melts contain predominantly, or exclusively, magmatic biotite.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHISMMORPHOLOGY AND PETROGRAPHY OF...BIOTITE COMPOSITIONGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
There is a morphological, mineralogical and geochemical continuity from diatexite migmatite to anatectic granite at St. Malo. The identification of melanocratic diatexite as the in situ residuum after melt separation makes it possible that the formation of granitic magma during anatexis of the St. Malo metasedimentary rocks was an isochemical (closed-system) process at the regional or crustal scale. The various migmatites and the anatectic granite are related by the processes of melt–residuum separation and fractional crystallization. Melt–residuum separation dominates the early stages of anatexis, and fractional crystallization dominates as metamorphic temperatures begin to decline. Leucocratic diatexite and anatectic granite are closely related genetically and spatially; typically, they occur in the same type of structurally controlled low-pressure sites.
These results indicate that the most successful approach in determining the petrogenesis of anatectic rocks is the use of a petrographically based division from which the geochemical signatures of specific petrogenetic processes can be recognized. Field relations provide the temporal and spatial relationships between the rock types and processes that formed them. In contrast, a subdivision of rock types based on morphological, or structural, differences in the anatectic domain may facilitate mapping, but it restricts petrogenetic interpretation because it is difficult to identify individual petrogenetic processes within divisions established in such a manner.
Melting occurred principally through the dehydration melting of muscovite, and affected virtually the whole succession, as there is very little metasedimentary rock preserved; diatexite migmatite is pervasive. This suggests that nearly all the rocks generated enough melt to flow, which requires Mf to be above the melt-escape threshold of 20 vol. % melt. Given that melting was synchronous with non-coaxial shearing, this build-up of melt in the matrix, rather than the expected continuous and episodic draining of melt, indicates that the rate of melting during anatexis at St. Malo was rapid.
|
ACKNOWLEDGEMENTS |
|---|
We thank everybody at the Anthéus and Paramé who made our work and stay in the St. Malo region so pleasant. We would also like to thank journal editor George Bergantz and reviewers Alberto Patiño Douce, Pierre Barbey and an anonymous reviewer for their thorough, constructive and very supportive comments. In particular, we are grateful to Pierre Barbey for sharing his insight and understanding of anatexis in general and the St. Malo migmatites especially, and also for pointing out the connection between processes occurring in the metatexite migmatites and those in the diatexite migmatites. This research was funded by Natural Science and Engineering Research Council of Canada collaborative project grant 183274, and an individual operating grant A1927 to E.W.S.
|
FOOTNOTES |
|---|
*Corresponding author. Telephone: 418-545-5011-5636. Fax: 418-545-5012. E-mail: ewsawyer{at}uqac.uquebec.ca
Extended dataset can be found at: http://www.petrology.oupjournals.org
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHISMMORPHOLOGY AND PETROGRAPHY OF...BIOTITE COMPOSITIONGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Barbey, P., Bertrand, J. M., Angoua, S. & Dautel, D. (1989). Petrology and U/Pb geochronology of the Telohat migmatites, Aleksod, Central Hoggar, Algeria. Contributions to Mineralogy and Petrology 101, 207–219.[Web of Science]
Barbey, P., Macaudière, J. & Nzenti, J. P. (1990). High-pressure dehydration melting of metapelites: evidence from the migmatites of Yaoundé (Cameroon). Journal of Petrology 31, 401–427.
Bea, F. (1991). Geochemical modeling of low melt-fraction anatexis in peraluminous system: the Pena Negra Complex (central Spain). Geochimica et Cosmochimica Acta 55, 1859–1874.
Brown, M. (1973). The definition of metatexis, diatexis and migmatite. Proceedings of the Geologists’ Association 84, 371–382.
Brown, M. (1979). The petrogenesis of the St-Malo migmatite belt, Armorican Massif, France, with particular reference to the diatexites. Neues Jahrbuch für Mineralogie, Abhandlungen 135, 48–74.
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. (1995). Late-Precambrian geodynamic evolution of the Armorican Segment of the Cadomian Belt (France): distortion of the active continental margin during south-west directed convergence and subduction of a bathymetric high. Géologie de la France 3, 3–22.
Brown, M. & Dallmeyer, R. D. (1996). Rapid Variscan exhumation and role of magma in core complex formation: Southern Brittany metamorphic belt, France. Journal of Metamorphic Geology 14, 361–379.
Brown, M. & Solar, G. (1998). Shear zone systems and melts: feedback relations and self-organisation in orogenic belts. Journal of Structural Geology 20, 211–227.
Brown, M., Power, G. M., Topley, C. G. & D’Lemos, R. S. (1990). Cadomian magmatism in the North Armorican Massif. In: D’Lemos, R. S., Strachan, R. A. & Topley, C. G. (eds) The Cadomian Orogeny. Geological Society, London, Special Publication 51, 181–213.
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]
Brun, J. P. & Martin, H. (1978). Relation métamorphisme–déformation au cours de l’évolution dynamique d’un dôme migmatitique: la massif de St-Malo (France). Bulletin de la Société Géologique de France XX, 91–101.
Conrad, W. K., Nicholls, I. A. & Wall, V. J. (1988). Water-saturated and -undersaturated melting of metaluminous and peraluminous crustal compositions at 10 kb: evidence for the origin of rhyolites in the Taupo Volcanic Zone, New Zealand, and other granitoids. Journal of Petrology 29, 765–803.
Dabard, M.-P. (1990). Lower Brioverian formations (Upper Proterozoic) of the Armorican massif (France): geodynamic evolution of source area revealed by sandstone petrography and geochemistry. Sedimentary Geology 69, 45–68.
Denis, E. & Dabard, M.-P. (1988). Sandstone petrography and geochemistry of Late Proterozoic sediment of the Armorican massif: a key to basin development during the Cadomian Orogeny. Precambrian Research 42, 189–206.
Dougan, T. W. (1983). Textural relations in melanosomes of selected specimens of migmatitic pelitic schists: implications for leucosome-generating processes. Contributions to Mineralogy and Petrology 83, 82–98.
Gapais, D. & Le Corre, C. (1980). Is the Hercynian belt of Brittany a major shear zone? Nature 288, 574–576.
Greenfield, J. E., Clarke, G. L., Bland, M. & Clarke, D. J. (1996). In-situ migmatite and hybrid diatexite at Mt. Stafford, central Australia. Journal of Metamorphic Geology 14, 413–426.[Web of Science]
Gupta, L. N. & Johannes, W. (1982). Petrogenesis of a stromatic migmatite (Nelaug, southern Norway). Journal of Petrology 23, 548–567.
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.
Henkes, L. & Johannes, W. (1981). The petrology of a migmatite (Arvika, Vramland, western Sweden). Neues Jahrbuch für Mineralogie, Abhandlungen 141, 113–133.
Martin, H. (1977). Contribution à l’étude de l’anatexie: le Massif de St-Malo (Massif Armoricain). Mobilisation des éléments durant la fusion. Proposition d’un modèle. Ph.D. thesis, University of Rennes.
Martin, H. (1979). Comportement de quelques éléments en traces durant la fusion incongruente de la biotite. (Rapport d’ATP) Sciences géologiques (mémoire) 53, 125–130.
Martin, H. (1980). Comportement de quelques éléments en traces au cours de l’anatexie: exemple du Massif de St-Malo (Bretagne). Canadian Journal of Earth Sciences 17, 927–941.[Web of Science]
McLellan, E. L. (1988). Migmatite structures in the Central Gneiss Complex; Boca de Quadra, Alaska. Journal of Metamorphic Geology 6, 517–542.[Web of Science]
Obata, M., Yoshimura, Y., Nagakawa, K., Odawara, S. & Osanai, Y. (1994). Crustal anatexis and melt migrations in the Higo metamorphic terrane, west–central Kyushu, Kumamoto, Japan. Lithos 32, 135–147.
Patiño Douce, A. E. & Harris, N. (1998). Experimental constraints on Himalayan anatexis. Journal of Petrology 39, 689–710.
Patiño Douce, A. E. & Johnston, A. D. (1991). Phase equilibria and melt productivity in the pelitic system: implications for the origin of paraluminous granitoids and aluminous granulites. Contributions to Mineralogy and Petrology 107, 202–218.[Web of Science]
Peucat, J. J. (1986). Behavior of Rb–Sr whole rock and U–Pb zircon systems during partial melting as shown in migmatitic gneisses from the St. Malo Massif, N. E. Brittany, France. Journal of the Geological Society, London 143, 875–886.
Pickering, J. M. & Johnston, A. D. (1998). Fluid-absent melting behaviour of a two-mica metapelite: experimental constraints on the origin of Black Hills granite. Journal of Petrology 39, 1787–1804.
Power, G. M., Brewer, T., Brown, M. & Gibbons, W. (1990). Late Precambrian foliated plutonic complexes of the Channel Islands and La Hague: Early Cadomian plutonism. In: D’Lemos, R. S., Strachan, R. A. & Topley, C. G. (eds) The Cadomian Orogeny. Geological Society, London, Special Publication 51, 215–229.
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. London: Chapman and Hall, pp. 82–110.
Saunders, A. D., Tarney, J. & Weaver, S. D. (1980). Transverse geochemical variations across the Antarctic Peninsula; implication for the genesis of calc-alkaline magmas. Earth and Planetary Science Letters 46, 344–360.
Sawyer, E. W. (1986). The influence of rock source type, chemical weathering and sorting on the geochemistry of clastic sediments from Quetico Metasedimentary Belt, Superior Province, Canada. Chemical Geology 55, 77–95.
Sawyer, E. W. (1987). The role of partial melting and fractional crystallization in determining discordant migmatite leucosome compositions. Journal of Petrology 28, 445–473.
Sawyer, E. W. (1991). Disequilibrium melting and the rate of melt–residuum separation during migmatization of mafic rocks from the Grenville Front, Quebec. Journal of Petrology 32, 701–738.
Sawyer, E. W. (1994). Melt segregation in the continental crust. Geology 22, 1019–1022.
Sawyer, E. W. (1996). Melt-segregation and magma flow in migmatites: implications for the generation of granite magmas. Transactions of the Royal Society of Edinburgh: Earth Sciences 87, 85–94.[Web of Science]
Sawyer, E. W. (1998). Formation and evolution of granitic 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.
Shaw, D. M. (1956). Geochemistry of peltic rocks. Part III: Major elements and general geochemistry. Geological Society of America Bulletin 67, 919–934.
Tait, R. E. & Harley, S. L. (1988). Local processes in the generation of migmatites within mafic granulites. Transactions of the Royal Society of Edinburgh: Earth Sciences 79, 209–222.
Taylor, S. R. & Gorton, M. P. (1977). Geochemical application of spark source mass spectrography—III. Element sensitivity, precision and accuracy. Geochimica et Cosmochimica Acta 41, 1375–1380.
Taylor, S. R. & McLennan, S. M. (1985). The Continental Crust: its Composition and Evolution. Oxford: Blackwell.
Thompson, A. B. & Tracy, R. J. (1979). 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.
Vielzeuf, D. & Holloway, J. R. (1988). Experimental determination of the fluid-absent melting relations in the pelitic system: consequences for crustal differentiation. Contributions to Mineralogy and Petrology 98, 257–276.[Web of Science]
Vielzeuf, D. & Montel, J. M. (1994). Partial melting of metagreywackes. 1. Fluid-absent experiments and phase relationships. Contributions to Mineralogy and Petrology 117, 375–393.[Web of Science]
Vigneresse, J. L., Barbey, P. & Cuney, M. (1996). Rheological transitions during partial melting and crystallization with application to felsic magma segregation and transfer. Journal of Petrology 37, 1579–1600.
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.
Weber, C., Barbey, P., Cuney, M. & Martin, H. (1985). Trace element behaviour during migmatization. Evidence for a complex melt–residuum–fluid interaction in the St-Malo migmatitic dome (France). Contributions to Mineralogy and Petrology 90, 52–62.
White, A. J. R. & Chappell, B. W. (1990). Per migma ad magma down-under. Geological Journal 25, 221–225.[Web of Science]
Wickham, S. M. (1987). Crustal anatexis and granite petrogenesis during low-pressure regional metamorphism: the Trois Seigneurs Massif, Pyrenees, France. Journal of Petrology 28, 127–169.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  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]
|
|
|
|
 |
|
 R. F. Weinberg Melt segregation structures in granitic plutons Geology, April 1, 2006; 34(4): 305 - 308. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. T. R. DROOP, J. D. CLEMENS, and D. J. DALRYMPLE Processes and Conditions During Contact Anatexis, Melt Escape and Restite Formation: the Huntly Gabbro Complex, NE Scotland J. Petrology, June 1, 2003; 44(6): 995 - 1029. [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]
|
|
|
|
|
|
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]
|
|
|
|
|
|
G. S. SOLAR and M. BROWN Petrogenesis of Migmatites in Maine, USA: Possible Source of Peraluminous Leucogranite in Plutons? J. Petrology, April 1, 2001; 42(4): 789 - 823. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||