© F. Roger et al., Published by EDP Sciences 2025

1 Introduction

Since early U–Th–Pb laser ablation dating studies (e.g., Feng et al., 1993; Fryer et al., 1993), and recent major technological improvements to laser and ICPMS instrumentation, U-Th-Pb LA-ICPMS (Laser Ablation − Inductively Coupled Plasma Mass Spectrometry) has evolved significantly and established itself in the community because of its spatial resolution and speed of data acquisition. This technique has made it possible to obtain very valuable geochronological data, without which it would never have been possible to advance our understanding of geological phenomena.

However, the conversion of a date (i.e., a number calculated from measured isotopic ratios and the decay equation) into an age (i.e., a date of geological significance) may be agreed upon by the whole geochronological community, but this does not mean that the interpretation in terms of geological processes will be unanimous. The geological interpretation of ages is not always straightforward and can be influenced by our vision of the geological object under study, i.e., the proposed geodynamic model, the amount of data available, and/or additional data given by geochemistry, petrology or structural analysis.

The Montagne Noire Variscan gneiss dome, located in the southern part of the French Massif Central, is a perfect illustration of this issue (Fig. 1A). It is one of the most extensively studied metamorphic domes in the world since the pioneering work of Gèze in 1949. The acquisition of the first LA-ICPMS instrument by the Magmas and Volcans laboratory at the University of Clermont-Ferrand in 2010, followed by that of the University of Rennes, has sparked renewed interest in this geological object. In the space of a decade, we have gone from a single U-Pb age (TIMS) published by Matte et al. (1998) to over 68 today (Roger et al., 2020 and references therein; Pitra et al., 2022; Hamelin et al., 2022) (Tab. S1). The vast majority of these LA-ICPMS U-Th-Pb data from the Montagne Noire, primarily obtained to shed light on the P–T–t-D evolution of metamorphic rocks, were obtained at the Magmas and Volcans laboratory in collaboration with Jean-Louis Paquette (Fig. 1; Tab. S1), whose research work has made a significant contribution to our understanding of orogenic evolution, particularly the Variscan tectonic cycle.

Proposed mechanisms to explain the Montagne Noire Axial Zone dome formation within the timeframe of 420 to 290 Ma span a wide range of tectonic settings. Diapirism has been invoked, including diapiric rise during shortening (Beaud, 1985) or during gravitational collapse (Gèze, 1949; Schuiling, 1960; Faure and Cottereau, 1988; Soula et al., 2001; Faure et al., 2010). Nappe stacking and folding during contractional tectonics have also been suggested (Arthaud, 1970; Mattauer et al., 1996; Matte et al., 1998; Demange, 1999; Matte, 2007), potentially involving significant erosion during dome exhumation (Malavieille, 2010). Extensional settings are widely considered, including (i) dome formation within vertical strike-slip shear zones (Nicolas et al., 1977), (ii) oblique extension and development of metamorphic core complexes after crustal thickening (Van den Driessche and Brun, 1989, 1992; Echtler and Malavieille, 1990; Brun and Van den Driessche, 1994, 1996; Franke et al., 2011), (iii) syn-contractional gravitational collapse (Aerden, 1998; Aerden and Malavieille, 1999), and (iv) extension-driven convergent flow of deep crust, leading to the formation of “double domes” separated by high-strain zones (Rey et al., 2011, 2012, 2017), coeval with overall extensional to transtensional tectonics in the upper crust (Whitney et al., 2015, 2020). While the “double dome” structure has been observed in various settings, its interpretation remains debated in some cases (e.g., the eastern Montagne Noire, where alternative interpretations propose a high-strain augen gneiss slab instead; Van Den Driessche and Pitra, 2012; Rey et al., 2012).

Whitney et al. (2020) and Hamelin et al. (2022) suggested that the deep crust within the axial zone of the Montagne Noire underwent horizontal flow from north to south, followed by rapid exhumation through vertical upward flow between 315 and 305 Ma. This process is interpreted as a transition from compressive tectonics to post-thickening extensional regimes. Pitra et al. (2022) proposed a similar model, in which the deep crust flowed horizontally over a distance of approximately 50 km during compression, before being exhumed during post-thickening extension that began around 320 Ma and lasted until roughly 290 Ma (Poujol et al., 2017). These different studies and various interpretations demonstrate the challenge of understanding the Montagne Noire gneiss dome and deep-crustal flow. A clearer depiction of the region’s polyphase metamorphic and deformation history remains crucial for resolving these debates.

In this manuscript, as a tribute to the work of Jean-Louis Paquette, we first present the results of the last 10 yr of work on the Montagne Noire dome, highlighting in particular the various contributions of LA-ICPMS U-Th-Pb geochronology. Secondly, we present new structural, petrological, and U-Th-Pb data obtained from the southeastern part of the Montagne Noire Axial Zone (MNAZ), specifically within the narrow valley known as “Les Gorges d’Héric”. This valley represents a singular and famous geological site within the Montagne Noire gneiss dome. We provide new insight into the structure and strain partitioning in this region of the MNAZ, where we reevaluate the presence of a crustal-scale dextral shear zone, which we have named the “Gorges d’Héric” shear zone (GHSZ) (Fig. 1B). The study focuses on two key outcrops, each exhibiting distinct relationships with the GHSZ strain fabrics. The first outcrop shows boudins of amphibole-rich metamafic rocks, with their margins that were affected by metasomatism during dextral shearing along the GHSZ. The second outcrop displays late pegmatites that crosscut the GHSZ vertical structures. A combined analysis of structural, petrographical, and U-Th-Pb LA-ICP-MS dating reveals that dextral strike-slip shearing remained active until 280 Ma and was responsible for localized and significant fluid-rock interaction. Our results, combined with previously published ages, highlight the contribution of U-Th-Pb geochronology to understanding the history of metamorphism, magmatism, fluid flow and deformation in the Montagne Noire gneiss dome.

Fig. 1 (A) Simplified structural map of the southern French Massif Central (after Faure et al., 2009) and the simplified map of some of the Variscan massifs (in dark pink colour) in western Europe: Armorican (ARM), Iberian (IB), Massif Central (MC), and Pyrenees (PYR). The rectangle at the southern end of the Massif Central shows the location of the Montagne Noire. (B) Simplified geological sketch map of the Montagne Noire (after Rabin et al., 2015, and Roger et al., 2020) with sample locations dated by the U-Th-Pb (See Tab. S1 for detail). The foliation trajectories show the Gorges d’Hérie Shear Zone (GHSZ) where the trend of the foliations is mainly E-W. (1) : this study; (2) Roger et al. (2020); (3) Roger et al. (2015); (4) Whitney et al. (2015); (5) Hamelin et al. (2022); (6) Trap et al. (2017); (7) Poilvet et al. (2011); (8) Pitra et al. (2012); (9) Poujol et al. (2017); (10) Pitra et al. (2022); (11) Franke et al. (2011); (12) Faure et al. (2010); (13) Faure et al. (2014) and (14) Matte et al. (1998). (C) and (D): Histograms of the U-Th-Pb Variscan ages recorded on the metamorphic (C) and plutonic (D) rocks from the Axial Zone of the Montagne Noire (modified after Roger et al., 2020). The data are reported in Table S1. All monazite EPMA and two zircon U-Pb (#46, 55) dates obtained by Faure et al. (2010) and Matte et al. (1998) are not considered because they are significantly older (∼320‐335 Ma) than the other data set (∼290‐315 Ma) (see more detail in Roger et al., 2020, Roger et al., 2015; Trap et al., 2017; Poujol et al., 2017). The yellow circles below each diagram denote the individual data used to generate the diagrams. The purple dashed line corresponds to the age Sm/Nd (Grt-Cpx-WR) obtained by Faure et al. (2014) on an eclogite boulder sampled near the Peyrambert farm (Fig.1A). (A) Carte structurale simplifiée du sud du Massif Central français (d’après Faure et al., 2009) et la carte simplifiée des massifs varisques d’Europe occidentale : Armoricain (ARM), Ibérique (IB), Massif Central (MC) et les Pyrénées (PYR). Le rectangle à l’extrémité sud du Massif Central montre l’emplacement de la Montagne Noire. (B) Carte géologique simplifiée de la Montagne Noire (d’après Rabin et al., 2015, et Roger et al., 2020) avec les emplacements des échantillons datés par U-Th-Pb (voir le Tab. S1 pour plus de détails). Les trajectoires des foliations montrent la zone de cisaillement des Gorges d’Héric (GHSZ) où la tendance des foliations est principalement E-W. (1) : cette étude; (2) Roger et al. (2020); (3) Roger et al. (2015); (4) Whitney et al. (2015); (5) Hamelin et al. (2022); (6) Trap et al. (2017); (7) Poilvet et al. (2011); (8) Pitra et al. (2012); (9) Poujol et al. (2017); (10) Pitra et al. (2022); (11) Franke et al. (2011); (12) Faure et al. (2010); (13) Faure et al. (2014) and (14) Matte et al. (1998). (C) et (D) : Histogrammes des âges varisques U-Th-Pb enregistrés sur les roches métamorphiques (C) et plutoniques (D) de la Zone Axiale de la Montagne Noire (modifié d’après Roger et al., 2020). Les données sont dans le Tab. S1. Tous les âges EPMA monazite ainsi que deux âges U-Pb zircon (#46, 55) obtenues par Faure et al. (2010) et Matte et al. (1998) ne sont pas pris en compte car ils sont significativement plus vieux (∼320-335 Ma) que l’ensemble des données (∼290-315 Ma) (voir Roger et al., 2020, Roger et al., 2015; Trap et al., 2017; Poujol et al., 2017 pour plus de détail). Les cercles jaunes sous chaque diagramme indiquent les données individuelles utilisées pour générer les diagrammes. La ligne violette en pointillé correspond à l’âge Sm/Nd (Grt-Cpx-WR) obtenu par Faure et al. (2014) sur un bloc d’éclogite échantillonné près de la ferme de Peyrambert (Fig.1A).

2 Geological setting

2.1 Architecture and strain pattern of the Montagne Noire gneiss dome

The Montagne Noire gneiss dome, located at the southernmost edge of the Variscan French Massif Central, is a noteworthy geological structure in the orogen’s foreland (Franke et al., 2011) (Fig. 1A). It is bounded by a SW-directed thrust system to the north, involving low-grade Paleozoic metasedimentary units, and by foreland nappes to the south, where slightly metamorphosed Paleozoic sedimentary units form large, south-verging recumbent folds (Gèze, 1949; Arthaud, 1970; Demange, 1996, 1999; Aerden, 1998; Doublier et al., 2014). The emplacement age of these nappes is stratigraphically constrained to either between the late Visean and Serpukhovian (∼335–323 Ma) (Engel et al., 1980; Feist and Galtier, 1985) or slightly younger than 320 Ma (Franke et al., 2011).

The metamorphic core of the Montagne Noire, known as the Axial Zone, is primarily composed of migmatized orthogneisses and crustally-derived granitic intrusions (Géze, 1949; Schuiling, 1960; Demange et al., 1996; Aerden, 1998; Roger et al., 2015), with interlayers of highly strained fine-grained gneisses, amphibolites, micaschists, marbles/calc-silicates, and mafic to ultramafic pods hosted in migmatites (Demange, 1985; Faure et al., 2014; Whitney et al., 2015, 2020) (Fig. 1B). The Axial Zone is bordered to the north by the Mont de Lacaune Fault (MLF), a ductile to brittle-ductile fault with recent normal motion (Echtler and Malavieille, 1990; Pitra et al., 2012), and to the south by the Mons fault cluster. In its eastern sector, this cluster includes the Poujol Fault, with gently south-dipping structures dominating the west (Aerden, 1998) (Figs. 1A, 1B). A structurally-overlying Proterozoic to Ordovician metasedimentary unit (Schistes X) separates the gneissic core from the thrust belts and is affected by normal ductile shear zones at its eastern and western edges (Van Den Driessche and Brun, 1989; Brunel and Lansigu, 1997; Rabin et al., 2015).

Structural analyses reveal at least three planar fabrics that alternate between steep and flat-lying orientations (Aerden, 1998; Rabin et al., 2015). Two elongate subdomes trending ENE-WSW, the Espinouse-Laouzas subdome to the north and the Nore-Somail-Caroux subdome to the south, are defined by S1 foliation (Fig. 1B). These are separated by a median high-strain zone characterized by a NE-SW-trending zone of steeply dipping S2 foliation associated with folding and transposition of S1 (Rabin et al., 2015; Trap et al., 2017). Along the southern flank of the Espinouse-Laouzas subdome, S2 foliation shifts to E-W, and subvertical orientation (Rabin et al., 2015). Fold axes and stretching lineations trend ENE, subparallel to the dome’s elongation (Van Den Driessche and Brun, 1992; Rabin et al., 2015). In the core of the dome, orthogneisses and migmatites exhibit a prolate finite strain ellipsoid, indicating constriction with a subhorizontal long axis aligned with the dome’s elliptical shape (Echtler and Malavieille, 1990; Matte et al., 1998; Rabin et al., 2015).

Between the migmatitic core and micaschist envelope, S1 and S2 foliations are transposed into a flat-lying S3 foliation with top-to-NE and top-to-SW shearing in the NE and SW dome terminations, respectively (Aerden, 1998; Rabin et al., 2015). These structures define a D3 transition zone that acted as a detachment surface accommodating strain partitioning, with NE-SW to E-W stretching in the upper zone and NW-SE contraction with horizontal flow below (Rabin et al., 2015; Whitney et al., 2015). Metamorphism during D3 ranged from partial melting in the core to subsolidus above the transition zone. D2 and D3 likely operated simultaneously, with strain partitioning across the anatectic front, possibly marking a major rheological boundary (Rabin et al., 2015).

2.2 Metamorphism and P-T data in the Montagne Noire gneiss dome

The first petrological studies were mainly conducted from the 1960s to the early 1980s, focusing on both the Schist X carapace (Bard and Ramboloson 1973; Bouchardon et al., 1979; Thompson and Bard 1982) and the gneissic core (Schuiling, 1960, 1963; Schuiling and Widt, 1962; Den Tex, 1975; Demange, 1982; Soula et al., 2001). It soon became clear that deformation, especially the development of a strong linear fabric parallel to the long axis of the dome, occurred simultaneously with a HT/LP metamorphism and crustal anatexis (Brunel and Lansigu 1997, Soula et al., 2001). This is indicated by the presence of migmatite and some index minerals such as cordierite bearing rocks (e.g., the Vialais and Montalet granites) and the presence of elliptical sillimanite-quartz nodules in the Espinouse subdome. Most recent P-T estimations give conditions of P ∼ 0.4-0.5 GPa and T ∼ 620−730°C for quartzofeldspathic rocks within the migmatitic core (Fréville et al., 2016, Ourzik et al., 1991, Rabin et al., 2015, Trap et al., 2017). The schist-micaschist carapace also contains low-pressure (LP) index minerals such as andalusite and cordierite. The metamorphic grade decreases from the sillimanite zone near the gneissic core to slate-phyllite away from the dome (Thompson and Bard, 1982; Doublier et al., 2014), and a thermal gradient of 30 °C.km⁻¹ was estimated from Raman spectroscopy (Fréville et al., 2016).

In metapelitic rocks, relics of kyanite (Bouchardon et al., 1979) and cordierite coronas around garnet (Rabin et al., 2015) attest to higher pressure conditions of metamorphism (M1) that preceded LP/HT metamorphism (M2). P-T calculations through pseudosection modelling indicate a clockwise P-T path with high pressure conditions at ∼0.8 ± 0.15 GPa and 725 ± 25 °C followed by low pressure conditions at 0.4 ± 0.1 GPa and 690 ± 25 °C (Rabin et al., 2015; Fréville et al., 2016; Trap et al., 2017).

The highest-pressure records are given by rare eclogite facies assemblages preserved in metabasaltic fragments enclosed in the migmatitic gneisses (Demange, 1985; Soula et al., 2001; Alabouvette et al., 2003; Faure et al., 2014; Whitney et al., 2015, 2020; Pitra et al., 2022). Demange (1985) estimated a peak pressure of 0.9 ± 0.2 GPa from eclogite lenses. Recently, these mafic eclogites have been the subject of renewed interest, as evidenced by several publications (Faure et al., 2014; Whitney et al., 2015, 2020; Pitra et al., 2022; Hamelin et al., 2022). Peak-P conditions of ∼ 1.5 ± 0.2 GPa at T ∼ 700 ± 20°C have been determined from pseudosections and trace-element thermobarometry and interpreted by Whitney et al. (2015, 2020) as indicating orogenic eclogitization by crustal thickening at high temperature. Pitra et al. (2022) documented a prograde P-T path from ∼1.95 GPa, 700°C to a pressure peak at ∼2.1 GPa, 750°C, interpreted as a subduction-related metamorphic evolution before lower-crustal flow. These authors also provide P-T conditions of exhumation (∼6 kbar, 730°C) recorded within sillimanite-bearing migmatites enclosing the eclogites (Pitra et al., 2022).

2.3 Timing of deformation and metamorphism

In the Montagne Noire massif, most authors now agree that the U-Th-Pb geochronological record gives ages in the range of 320-290 Ma and that this age range corresponds to the period of LP-HT metamorphic conditions and crystallization of anatectic granites at shallow crustal levels, which post-dated regional contraction and nappe emplacement (> 320 Ma) (Fig. 1C) (e.g., Poujol et al., 2017; Roger et al., 2015, 2020, and references therein). This LP–HT metamorphism was accompanied by granitic magmatism and pervasive migmatization of the gneiss dome (Soula et al., 2001). During this interval, the orogenic crust experienced widespread ENE stretching and strike-slip deformation (Nicolas et al., 1977; Beaud, 1985; Faure and Cottereau, 1988; Echtler and Malavieille, 1990; Aerden, 1998; Matte et al., 1998; Rabin et al., 2015).

However, there is no consensus on whether there was a single continuous event over about 30 Ma, or whether there were several discrete episodes. In the Espinouse subdome, using the U-Th-Pb monazite LA-ICPMS and muscovite ⁴⁰Ar-³⁹Ar methods, two age groups at ∼319‐318 Ma and 298‐295 Ma were identified by Poujol et al. (2017). The ∼319 Ma event is interpreted as a first stage of migmatization and as the emplacement age of the anatectic granite, whereas the second event at ∼298–295 Ma is given as a fluid-induced event possibly related to a second partial melting event, also identified through the syn-extensional emplacement of the Montalet and Vialais leucogranites (Poilvet et al., 2011; Roger et al., 2015). Poujol et al. (2017) also highlighted the probable occurrence of a third event, dated at around 285 Ma, the significance of which is still uncertain. It could correspond either to an artifact due to fluid-induced modification of some monazite crystals or the evidence of Permian fluid circulations. Based on the compilation of U-Th-Pb Variscan ages recorded by 55 plutonic and metamorphic rocks from the Axial Zone of the Montagne Noire (Figs. 1C and 1D and Tab. S1), we propose that the age range between ca. 320 and ca. 290 Ma represents a single event corresponding to the maximum temperature under suprasolidus conditions.

Within retrogressed eclogites, the different U-Th-Pb dating of zircons carried out in these recent studies (LA-ICPMS, SHRIMPS, SIMS) indicate the presence of three age clusters at 470-450 Ma (inherited zircon cores), ∼360 Ma (a few zircon cores) and 315-310 Ma (zircon rims) (Faure et al., 2014; Whitney et al., 2015, 2020; Pitra et al., 2022; Hamelin et al., 2022) (Fig. 1C and Tab. S1). According to Whitney et al., 2020 and Hamelin et al., 2022, the emplacement of the mafic to ultra-mafic magmatic protoliths occurred during the Ordovician (500-450 Ma). This therefore overlaps with the Late Ordovician emplacement ages (470-450 Ma) of the Nore-Somail-Caroux orthogneiss felsic protoliths (e.g., Roger et al., 2004, 2015, 2020; Cocherie et al., 2005; Faure et al., 2010; Pitra et al., 2012; Trap et al., 2017).

Based on a Sm-Nd isochron age (grt-cpx-whole rock) partially derived from resorbed garnet, Faure et al. (2014) interpreted that the eclogite-facies stage occurred at ca. 360 Ma. They also identified a hydrothermal overprint dated at ca. 315-310 Ma is recorded in zircon and rutile grains (SHRIMP and SIMS). They attributed the eclogitization of ultramafic to mafic protoliths to intra-continental subduction during the Devonian-Lower Carboniferous. Zircon grains dated at 315 Ma show typical eclogitic zircon REE pattern (flat HREE, no Eu anomaly) (Whitney et al., 2015, 2020; Pitra et al., 2022; Hamelin et al., 2022). Building on the age obtained by the Sm-Nd isochron of Faure et al. (2014), and assuming decoupling of the U-Pb and REE systems during zircon recrystallization, Pitra et al. (2022) proposed a geodynamic evolution involving subduction of oceanic crust at ca. 360 Ma, followed by its partial incorporation into LP-HT crust and recrystallized at low P during doming (315-310 Ma). The eclogitization of ultramafic to mafic protoliths, thought to have occurred further north in the French Massif Central, is attributed to pre-Carboniferous subduction followed by southward migration of allochthonous eclogite-bearing lower crust during syn-orogenic lateral flow (Pitra et al., 2022). Complementary geochemical data including zircon trace element signatures (U/Yb-Hf and U/Yb-Y), REE and Th/U in zircon cores (ca. 450 Ma), as well as rutile trace-element composition and oxygen isotope signature of zircon and garnet, have led Hamelin et al. (2022) to interpret the Montagne Noire eclogites as continental eclogites in which Cambro-Ordovician protoliths (∼520–400 Ma) are derived from different mafic (gabbroic) sources. The authors also suggested that the eclogites were translated and buried during crustal flow towards the foreland, thereby contributing to crustal thickening, followed by rapid exhumation in Late Carboniferous. The studied eclogites may have experienced varying degrees or magnitudes of lateral flow in the deep crust, depending on their point of origin relative to their final position (Hamelin et al., 2022). The c. 360 Ma Sm-Nd date of Faure et al. (2014) has been interpreted as that of prograde metamorphism recorded in the preserved and dated low-Mg, pre-eclogite-facies garnet cores, rather than that of eclogite-facies metamorphism (Whitney et al., 2015; Hamelin et al., 2022). Based on zircon and rutile U-Pb ages (LASS-ICPMS), REE in zircon (flat HREE, no Eu anomaly), and P-T estimates, these authors suggest that peak HP metamorphism occurred at 315-310 Ma (zircon rim ages), immediately followed by cooling/decompression at 308-304 Ma (rutile ages) (Whitney et al., 2015, 2020; Hamelin et al. (2022).

The timing of deformation events is given by U-Th-Pb ages obtained on zircon and monazite grains in syn-kinematic plutons (Matte et al., 1998; Poilvet et al., 2011; Roger et al., 2015; Trap et al., 2017), in pegmatites and dykes (Franke et al. 2011), or in association with metamorphic assemblages (Trap et al., 2017). Micas ⁴⁰Ar/³⁹Ar ages have been obtained at three structural levels (Maluski et al., 1991) which are: i) Migmatites and gneisses of the axial zone record high-temperature deformation and metamorphism at ∼315 Ma, clearly post-dating regional contraction and nappe emplacement; ii) Mylonites from the northern and southern shear zones formed during major syn-metamorphic deformation related to the dextral wrench-fault zone on the flank of the Axial Zone at ca. 310 Ma and iii) Metasedimentary rocks along the basal sole thrust of the Paleozoic nappes and the leucogranitic mylonites of the Espinouse detachment shear zone have ages of 297 ± 3 Ma, interpreted as the age of dynamic recrystallization at the base of the southern nappes during dextral shearing along the southern margin (Maluski et al., 1991). The fine-grained, folded and intensely sheared gneiss layers show late, high-temperature deformation localized at the core of the dome, which has been dated at 295 Ma by monazite U-Th-Pb LA-ICPMS (Roger et al., 2020).

3 Results

3.1 Regional-scale strain analysis: eastern half of the dome

In this section we present the strain pattern of the eastern half of the Montagne Noire gneiss dome, based on our structural dataset obtained from our previous works (Rabin et al., 2015; Fréville et al., 2016; Trap et al., 2017) and complemented by new field observations and measurements (Figs. 2 and 3). As in the rest of the dome three deformations (D1, D2, and D3) are recognised. An early flat-lying to moderately dipping S1 foliation was folded during D2, producing ENE-WSW oriented F2 upright folds. Intense D2 shortening resulted in transposition of S1 into a steeply dipping, NE-SW trending S2 foliation. Along S2, a shallowly plunging mineral and stretching lineation is consistently observed along which dextral shear criteria predominate. On either side of the central part of the studied area (Fig. 2), the S2 generally strikes N40-60E with dips of 60−70° toward the SE. Toward the center, S2 becomes progressively steeper and rotates toward a N80-90E orientation. The orientation changes of S2 makes a curved or sigmoidal pattern of anastomosing S2 foliations, observable at all scales, and reflects the dextral sense of shear, as indicated by asymmetric shear criteria. The N80-90E S2 corresponds to the highest strain plane where strong boudinage and stretching is observed. This N80-90 fabrics was considered as shear plane, named C2, by Rabin et al. (2015). However actual C2-surfaces are less penetrative than S2 foliation and the strain gradient does not lose its regular curvature so that N80-N90E directed planar fabrics are S-surfaces deflected into near parallelism to the flow plane of the shear deformation (Fig. 2). At the outcrop scale, C2 shear planes typically exhibit N100 ± 5E orientations with a steep dip (>65) toward either the north or the south. At a larger scale, N100E trending planar fabric trajectories form up to 1-2 km wide and 2-4 km long shear planes as observed in the deepest structural level, south-east of La Salvetat village. The N70-80E preferential directions of strike define a N100E directed domain the boundaries of which are two parallel and imaginary lines, i.e., the shear zone walls outside of which the S2 foliation is N60-70E trending. We refer to this shear zone as the Gorges d’Héric shear Zone (GHSZ). The strain gradient along the GHSZ boundaries is evidenced by the geometry of lithological units in map view. For example, the Larn gneisses form an east-west-oriented belt that cartographically coincides with the zone where S2 foliation trends are predominantly east-west, aligning with the core of the shear zone.

Another example is given by the Caroux augen orthogneiss (the equivalent of the Héric Gneiss aforementioned) that is reduced by a half, from 5 km north of Brassac to 2.5 km at St-Vincent-d’Olargues showing the strain gradient toward the north (Fig. 2). Similarly, at a shallower structural position, examination of the geological map of the entire axial zone reveals that the thickness of the micaschist envelope is extremely reduced by more than a half, from 600 meters along the NE-SW section (Brassac) to 200 meters along the E-W section (west of St-Vincent d’Olargues). The 10 km long and N80 striking corridor, from St-Vincent-d’Olargues to St-Martin de l’Arçon represents the zone of higher strain, i.e., the GHSZ. Here, the E-W oriented Pujol Fault that bounds the MN Axial Zone to the south (Figs. 1A and B, 2, and 3) may represent the brittle upward continuation of the ductile, mylonitic GHSZ.

Within the gneissic core of the axial zone and along the anatectic front, S1 and S2 foliations were transposed into flat-lying S3 foliation, with top-to-NE shearing (Aerden 1998, Rabin et al., 2015). Figure 2 also shows vertical large quartz veins, whose preferential orientations are illustrated by the stereographic plot. Four main orientations are identified: N10–15E, N35–40E, N80–85E, and N115–120E, with the last set being well-represented along the GHSZ. These sets of veins, which correspond to kilometer-scale fractures, might be interpreted as Riedel fractures, particularly the N115–120E set, which are likely Riedel shears forming an “en echelon” pattern at an acute angle to the GHSZ direction.

In the following, we focus on the D2 deformation and the GHSZ, we present petrological and geochronological data obtained on rocks sampled in the “Gorges d’Héric”, a N-S oriented enclosed valley (Fig. 3). Due to the quality and accessibility of outcrops, the “Gorges d’Héric” valley provides a continuous cross-section through the migmatitic gneisses of the Caroux subdome. The mean measured foliation in the Gorges d’Héric valley is oriented N70-90, with a stretching lineation showing the same direction with a shallow plunge (Figs. 2 and 3). This change in orientation of planar and linear features is due to the GHSZ along which the shear foliation (XY plane) tends to parallelise the shear plane (shear zone boundary) (Fig. 2).

Fig. 2 Structural map on the eastern half of the Montagne Noire gneiss dome showing the orientation of planar fabrics and interpreted foliation trajectories (modified from Rabin et al., 2015). The stereogram shows the preferential orientation of the quartz-bearing dykes that cut across the migmatitic rocks of the axial zone. Carte structurale de la partie orientale du dôme gneissique de la Montagne Noire montrant l’orientation et les trajectoires de foliation interprétées (modifiée d’après Rabin et al., 2015). La projection stéréographique montre l’orientation préférentielle des filons de quartz qui recoupent les migmatites de la zone axiale.

Fig. 3 Simplified geological map of the Gorges d’Héric valley, modified from Raynal (2004) and Roger et al. (2020). Letters within black circle correspond to monazite U-Th-Pb ages (Tab. S1; Roger et al., 2020): A: Vialais granite (MN 17/303.6 ± 2 Ma); B: F.G.G. (MN11-59/294.3 ± 2.2 Ma); C: Fine grained gneiss (MN11-57/294 ± 2 Ma); D: Augen Gneiss (MN14-03/314.1 ± 1.3 Ma); E: Fine grained gneiss (MN11-117/445.8 ± 3.8 Ma; MN11- 118/441 ± 2.5 Ma; MN11- 119A/316.9 ± 5.5 Ma-297.3 ± 1.6 Ma); F: Fine grained gneiss (MN11-47/450.6 ± 5.6 Ma-301 ± 4 Ma; MN11-121/294.9 ± 2.9 Ma); G: Fine grained gneiss (MN11-27/294.6 ± 1.6 Ma). Carte géologique simplifiée de la vallée des Gorges d’Héric (d’après Raynal, 2004 et Roger et al., 2020). Les lettres inscrites dans les ronds noirs correspondent aux âges U-Th-Pb sur monazites (Tab. S1; Roger et al., 2020): A: granite du Vialais (MN 17/303.6 ± 2 Ma); B: F.G.G. (MN11-59/294.3 ± 2.2 Ma); C: Gneiss fin (MN11-57/294 ± 2 Ma); D: Augen Gneiss (MN14-03/314.1 ± 1.3 Ma); E: Gneiss fin (MN11-117/445.8 ± 3.8 Ma; MN11- 118/441 ± 2.5 Ma; MN11- 119A/316.9 ± 5.5 Ma-297.3 ± 1.6 Ma); F: Gneiss fin (MN11-47/450.6 ± 5.6 Ma-301 ± 4 Ma; MN11-121/294.9 ± 2.9 Ma); G: Gneiss fin (MN11-27/294.6 ± 1.6 Ma).

3.2 Petro-structural analysis of the Gouffre du Cerisier outcrop

The ∼100 m long “Gouffre du Cerisier“ outcrop is located in a deep structural position in the Gorges d’Héric (from N43°35’08”/E2°57’34” to N43°35’10”/E2°57’30”; Fig. 3). Heterogeneous deformation at meter-scale form sigmoidal curvature of the shear foliation and a lense-shaped strain pattern (Fig. 4). This section shows migmatites derived from strongly sheared orthogneiss, some of them ultramylonitic (fined grained gneiss in Roger et al., 2020) and sheared lenses of metamafic rocks. Foliation is mainly E-W trending and subvertical with a subhorizontal stretching lineation (Fig. 3). Intense shearing, boudinage, and transposition of S1 and S2 along the GHSZ resulted in lense-shaped trajectories of planar fabrics.

Metamafic rocks appear as meter-long lenses aligned in strings of 3 to 6 bodies. Lenses are more or less widely spaced, depending on the intensity of deformation (Figs. 4A and C). The core of the lenses shows an isotropic granoblastic texture or a subtle planar fabric. In contrast, the edges of the lenses show very pronounced foliation due to the preferential alignment of amphibole, biotite and chlorite (Fig. 4C). The foliated appearance of the lens edges may result more from both mineral transformation and strain. The adjacent gneissic surroundings exhibit intense shearing. In addition, microstructural observation clearly shows strong stretching in the MN13-02 sample in comparison to the MN13-01 sample (see next Section 3.2.2).

At the outcrop scale a change of mineral assemblages is observed from the core to the edge of the lenticular bodies (Fig. 4) but is only well identifiable at the scale of the thin-section. Two representative samples of the metamafic lense were studied, one from the undeformed core (MN13-01) and the other from the deformed rim (MN13-02).

Fig. 4 Lithology and structure of the Gouffre du Cerisier outcrop. (A) Strain pattern showing the sigmoidal and lense shaped arrangement of the steeply dipping S2 foliation. The green coloured boudins represent the metamafic rocks (MN13-01/02) that are embedded within the sheared orthogneiss (white coloured). (B) Sketch showing the sheared metamafic lenses with the garnet − pyroxe rich core (sample MN13-01) mantled by amphibole + chlorite rich domains (sample MN13-02). The orientation of the drawing is NW (left)- SE (right) making a different orientation of the mafic pods between A & B (C) Zoom picture in the metamafic lenses showing the strain strain gradient and textural changes from the core to the rim of MN13 metamafic. Lithologie et structure de l’affleurement du Gouffre du Cerisier. (A) Schéma de déformation montrant la géométrie sigmoïde et en lentilles de la foliation S2 à fort pendage. Les boudins de couleur verte représentent les roches métamafiques (MN13-01/02) au sein de l’orthogneiss cisaillé (de couleur blanche). (B) Schéma illustrant les lentilles métamafiques cisaillées, avec un cœur à grenat −pyroxène (échantillon MN13-01) entouré de domaines riches en amphibole + chlorite (échantillon MN13-02). L’orientation du dessin est NW (gauche)- SE (droite), ce qui donne une orientation différente des boules mafiques entre A et B. (C) Vue rapprochée des lentilles métamafiques montrant le gradient de déformation du cœur jusqu’à la bordure des boudins.

Fig. 5 (A) Microphotographs of sample MN13-01 showing the microstructural arrangement of pyroxene, garnet and amphibole. (B) Close-up image showing the relationships between orthopyroxene (Opx), clinopyroxene (Cpx), garnet, and amphiboles. (C–F) X-ray maps highlighting the elemental distribution of magnesium (C), calcium (D), aluminium (E), and sodium (F), providing both textural and chemical information (see text for explanations). (G) Microphotograph showing a thin olivine corona around ilmenite. (A) Photographies au microscope de l’échantillon MN13-01 montrant la disposition microstructurale des grains de pyroxène, grenat et amphibole. (B) Vue rapprochée montrant les relations entre l’orthopyroxène (Opx), le clinopyroxène (Cpx), le grenat et les amphiboles. (C–F) Cartes en rayons X illustrant la distribution des éléments magnésium (C), calcium (D), aluminium (E) et sodium (F), apportant des informations texturales et chimiques (voir texte pour les explications). (G) Microphotographie montrant une fine couronne d’olivine autour de l’ilménite.

Fig. 6 Microphotographs showing the mineralogical assemblages and microstructure of sample MN13-02. (A) and (B) Coarse grained facies showing the amphibole + biotite + olivine stbale assemblage. Chlorite appears as a secondary phase. (C) Chlorite-rich fine-grained facies showing the layering defined by amphibole rich and chlorite rich layers. (D) Zoom at the amphibole-olivine grain boundary showing destabilization of amphibole into olivine. (E). Strongly foliated fine −grained facies with boudinage of olivine porphyroblasts with syn-kinematic chlorite within boudin neck. Remnants of large biotite grains are observed. (F) Serpentinite + magnetite corona around olivine. The asymmetric appearance of the corona with almost no serpentine along the left side of the olivine grain suggests a syn-kinematic reaction. Photographies au microscope illustrant les assemblages minéralogiques et la microstructure de l’échantillon MN13-02. (A) et (B) Faciès à grains grossiers montrant l’assemblage stable amphibole + biotite + olivine. La chlorite apparaît comme une phase secondaire. (C) Faciès à grains fins riche en chlorite présentant une alternance de couches riches en amphibole et en chlorite. (D) Zoom sur la limite de grain amphibole-olivine montrant la déstabilisation de l’amphibole en olivine. (E) Faciès à grains fins fortement folié, avec boudinage de porphyroblastes d’olivine et chlorite syn-cinématique dans les queues des boudins. Des vestiges de gros grains de biotite sont observés. (F) Couronne de serpentinite + magnétite autour des grains d’olivine. L’aspect asymétrique de la couronne, avec très peu de serpentinite du côté gauche du grain d’olivine, suggère une réaction syn-cinématique.

3.2.1 Garnet-bearing assemblage (samples MN13-01)

The rock sample MN13-01 is a garnet-bearing mafic rock with an amphibole + garnet + orthopyroxenes ± clinopyroxene ± ilmenite main assemblage (Fig. 5A). Garnet appears as 5-10 mm large aggregates made of several 1-5 mm euhedral to subhedral grains. Each garnet grain (Alm_41-43Prp_42-45 Grs_9-11Sps_02-03) contains inclusions of amphiboles, ilmenite and spinelle (Figs. 5B-C). Orthopyroxene, preponderant over clinopyroxene, forms 0.2-5 mm euhdral to sub-euhedral grains (Figs. 5B-C). Orthopyroxene is an enstatite with a high Mg content (Mg# = 0.73; Tab. S2). Clinopyroxene, a diopside (X_Wo = 0.47-0.48, Mg# = 0.91-0.92; Tab. S2), is observed as relictual mineral with an irregular cuspate shape within the amphibole matrix (Figs. 5B-E). Compare to orthopyroxene, clinopyroxene is never in contact with garnet. Two amphiboles can be distinguished based on differences in their chemical composition and textural positions. Amphibole 1, as inclusions in garnet core and as matrix mineral around diopside (Figs. 5B and 5C) is a pale green magnesio-hornblende (Al(iv) = 0.1–0.8 p.f.u, Mg# =0.79–0.88) (Tab. S2). Amphibole 2 is a green to dark green tschermakite (Al(iv) = 0.5–2.1 p.f.u, Mg# =0.54–0.87) that forms symplectitic coronas (amphibole + spinel) around garnet and late aggregates of subhedral prismatic grains in the matrix (Figs. 5D-E). The kelyphytic amph 2 + spinel corona separates orthopyroxene from garnet. Amphibole compositions evolve from a low-Al type (Amph1) to a more aluminous amphibole (Amph2), indicating a compositional trend within a single mineral generation. The X-ray maps reveal that some amphiboles (Amph2) display well-defined concentric zoning patterns, with chemical boundaries that are partly independent of grain boundaries. In several cases, the chemical zoning appears cross-cutting relative to grain interfaces, suggesting a growth pattern that is not controlled by the surrounding crystal framework. Spinel grains are <100 μm size and their chemical composition change depending on their textural location (Fig. 5E). Spinel in inclusion in garnet or in equilibrium with amph 2 (kelyphytic corona) are Mg-rich and Cr-poor (Mg#=0.41-0.46 and Cr#=0.02-0.07) compare to spinel in the matrix associated with ilmenite (Mg#0.28-0.31 and Cr#0.25-0.26). Within the amphibole-rich matrix, thin coronas of forsterite (Mg# = 75; Tab. S2) develop around ilmenite (Fig. 5G). Apatite is present as few grains of <100μm in size.

3.2.2 Chlorite-bearing assemblage (MN13-02)

In sample MN13-02, the mineral assemblage is made of amphibole + biotite + olivine + chlorite assemblage (Fig. 6A). Amphibole is a pale green magnesio-hornblende (Al(iv) = 0.9–1.0 p.f.u, Mg# =0.63–0.65). Biotite is a phlogopite with Al(iv) = 1.1–1.2 p.f.u and Mg# =0.6-0.7. Olivine and chlorite are Mg-rich, with Mg#=0.73 and Mg# =0.8-0.9 (Tab. S2). Two types of facies and assemblages could be distinguished. A first, coarse grained facies shows mm-scale amphibole, biotite and olivine (Fig. 6A). Some chlorite appears as a replacement of biotite. Grain boundaries between amphibole and olivine suggest the two minerals to be at equilibrium (Fig. 6B). In a second, fine grained facies, chlorite and amphibole are the most abundant and their preferential orientation clearly defines the planar fabric (Fig. 6C). Forsterite-rich olivine is also finer grained and some irregular cuspate grain boundary microstructures may suggest it crystallized after amphibole consumption (Fig. 6D). Boudinage of olivine with chlorite-rich inter-boudins attest to syn-kinematic retrogression (Fig. 6E). Serpentine occurs in fracture infilling or as a corona around olivine. The serpentine corona is separated from the chlorite-rich matrix by a fringe of magnetite (Fig. 6F). U-Th-bearing accessory minerals, mainly zircon grains, are evenly distributed throughout the rock with the exception of some zircon-rich clusters (Fig. S1A).

3.3 U-Th-Pb geochronology

3.3.1 Instrumentation and analytical methods

U-Th-Pb geochronology of zircon and monazite was conducted by LA-ICPMS at the University of Clermont-Auvergne (Laboratoire Magmas et Volcans). The analyses involved ablation of grains with a Resonetics Resolution M-50 powered by an ultra-short pulse ATL Atlex Excimer laser system operating at a wavelength of 193 nm. The detailed analytical procedures are described in Paquette et al. (2014) and detailed in Hurai et al. (2010) and in the Supplementary material (Tab. S3). Data reduction was carried out with the GLITTER® software package from Macquarie Research Ltd (van Achterbergh et al., 2001). Dates and diagrams were generated using the Isoplot/Ex v. 2.49 software package (Ludwig, 2001). In the text and figures, all uncertainties in dates are given at ±2σ. The decay constants used for the U-Pb system are those determined by Jaffey et al. (1971) and recommended by the IUGS (Steiger and Jäger, 1977).

3.3.2 U-Th-Pb results

Zircon grains in sample MN13-02 are small (<100 mm), commonly rounded and transparent and colorless, to slightly opaque and white color. They are typically fractured and show no sign of internal zoning in BSE images (Fig. S1B; Zr17) but some crystals clearly show the presence of inherited cores (Fig. S1B; Zr15 and 16). In addition, most zircon grains are characterized by CL-dark zones, in many cases with a slightly brighter rim (Fig. S1C; Zr34). However, some grains show partial replacement textures like those observed in monazite, which are interpreted as the result of fluid-induced coupled dissolution-reprecipitation processes (Harlov et al., 2023). A few grains also showed magmatic oscillatory zoning in CL images (Fig. S1C; Zr33). Twenty-four analyses were carried out on the cores and rims of 16 crystals; results are plotted in a Tera-Wasserburg diagram in which three concordant to subconcordant populations can be identified (Fig. 7A; Tab. S4):

• The first consists of two zircon grains (white ellipses) whose cores and rims are dated at ca. 550-520 Ma and 480-460 Ma, respectively. These four analyses have high Pb and U contents of 35-86 ppm and 372-753 ppm, respectively, and Th/U ratios of 0.97-1.5, typical of magmatic zircon from mafic rocks (Tiepel et al., 2004)

• The second group (grey ellipses) comprises three analyses (2 cores and one rim) obtained on 2 grains and yields a concordia age of 314. 8 ± 4.6 Ma (MSWD_(C+E) = 0.7). The U and Pb contents vary between 225 −266 ppm and 11-12 ppm, respectively, with Th/U ratios ranging from 0.01 to 0.27.

• The last and largest group includes 17 analyses (11 cores and 6 rims). U and Pb content range from 71 ppm to 803 ppm and from 3 ppm to 36 ppm, respectively, with Th/U ratios ranging from 0.2 to 0.42. The data yield a concordia age of 280.6 ± 2.5 Ma (MSWD_(C+E) = 1.9, N=17) (Fig. 7A).

The second area studied, called the “Pont des Soupirs” outcrop, is located near the southern egde of the Caroux subdome, 200 m to the north of the micaschist envelope (Fig. 2). There, we sampled an undeformed tourmaline-bearing pegmatite several meters thick (sample MN 16), cutting the N76/64-trending gneissic foliation (Fig. S2A and B). Monazite grains in the Pont des Soupirs pegmatite (MN 16) are subhedral transparent and green and show no sign of internal zoning in BSE images (Fig. S2C). Twenty analyses were carried out on 17 crystals. The very high Pb, Th and U contents ranging from 1055-2017 ppm, 64114-113948 ppm and 15477-51651 ppm, respectively, as well as the low Th/U ratios from 1.28 to 8.08 are typical of magmatic monazite (Schandl and Gorton, 2004) (Tab. S4). Plotted in a concordia diagram, only two data are discordant (dotted ellipses), probably due to the presence of common Pb (Fig. 7B). The 18 other data give a U-Pb concordia age of 296.8 ± 2.3 Ma (MSWD(C+E) = 1.2).

Fig. 7 U-Th-Pb results plotted in Tera-Wasserburg diagram (A) for zircon from the Gouffre du Cerisier sample (MN13-02) and (B) for monazite from the Pont des Soupirs pegmatite (MN 16), respectively. Error ellipses and age uncertainties are 2σ. The dotted ellipses are not taken into account for age calculation. MSWD(C+E) = mean square of weighted deviations for concordance and equivalence. Les résultats U-Th-Pb reportés dans un diagramme Tera-Wasserburg (A) pour les zircons de l’échantillon du Gouffre du Cerisier (MN13-02) et (B) pour les monazites de la pegmatite du Pont des Soupirs (MN 16), respectivement. Les ellipses d’erreur et les incertitudes sur les âges sont à ± 2σ. Les ellipses en pointillés ne sont pas prises en compte pour le calcul de l’âge. MSWD(C+E) = déviation pondérée de la régression par moindre carrés pour la concordance et l’équivalence.

4 Discussion

4.1 Origin and metamorphic history of the metamafic lenses (MN13 sample)

The MN13 studied samples, shows a garnet + amphibole + orthopyroxene + clinopyroxene + ilmenite main mineralogical assemblage (MN13-01) that evolved into a chlorite–amphibole-rich rock (MN13-02). Such evolution is strongly similar to that of the Airette meta-ultramafic rocks, located 1.5 km westward along the same structural trend (Fig. 2). Based on major and REE compositions, Demange (1985) suggests that the Airette meta-ultramafic rocks (also called "garnet lherzolites" by Cohen, 1975) represent a metamorphosed ferromagnesian cumulate derived from a calc-alkaline basaltic magma. Although no complete relics of the magmatic assemblage are preserved in the Airette rocks, relictual spinel and sulfides are locally observed, while plagioclase has been entirely destabilized during high-pressure metamorphism (Demange, 1985). In our sample, spinel appears as a metamorphic phase in equilibrium with Na-Al-rich amphibole. The chemical zoning observed in amphibole (from Amph 1 to Amph 2 end-members), with zoning boundaries that cross-cut grain boundaries, strongly suggests mineral growth from a reactive fluid, where diffusion and advection govern chemical gradients rather than crystal boundaries. Compared to clinopyroxene, which is relictual, orthopyroxene shows euhedral grain shapes and amphibole–orthopyroxene grain boundaries with no reaction textures, suggesting that both phases crystallized synchronously. These observations provide compelling evidence for metasomatic amphibole and orthopyroxene formation under fluid-assisted conditions. Considering the textures of garnet aggregates and the fact that they contain amphibole inclusions, it is suggested that garnet grew together with amphibole 1, which is also in equilibrium with orthopyroxene. Two reactions may be proposed to account for the observed assemblage: (1) cpx + pl + H₂O → garnet + amphibole and (2) cpx + H₂O → opx + amphibole. The first reaction may generate SiO₂(aq), which could contribute to the formation of orthopyroxene. Demange (1985) described early orthopyroxene over which clinopyroxene developed during eclogite-facies metamorphism. We do not observe such textural evidence in the MN13-01 sample, where diopside appears to be the earliest and relictual phase.

The MN13-01 sample mostly corresponds to the first retrograde stage described by Demange (1985), resulting from decompression and increasing fluid activity, with formation of amphibole, olivine, and secondary spinels through hydration of the eclogitic assemblage. Demange (1985) proposes that amphibole grows on clinopyroxene, and the appearance of small olivine grains at orthopyroxene rims is interpreted as driven by water supply through the reaction (3) enstatite + diopside + H₂O → forsterite + tremolite. Consistent with our observations, Demange (1985) also described kelyphite made of amphibole and spinel droplets, attributed to this retrogression stage. In our samples, olivine is observed as thin coronas in MN13-01 and as large porphyroblasts in MN13-02. In MN13-02, olivine grains are arranged in elongated aggregates that mantle amphibole-rich domains, where pseudomorphic textures after pyroxene are observed, consistent with reaction (3) above.

Additionally, Demange (1985) argues for localized potassic metasomatism evidenced by the crystallization of phlogopite. Sample MN13-02 shows large synkinematic biotite, which we also interpret as the result of potassium metasomatism. In the Airette rock, further retrogression at lower temperatures led to extensive replacement of pyroxenes and olivine by amphibole and chlorite (Demange, 1985). In the final retrogression stage, characterized by the dominance of amphibole and chlorite, Demange describes the rock as belonging to the "chlorite–amphibole peridotite facies". A similar amphibole + chlorite assemblage is observed in MN13-02, although several generations of chlorite are likely. Indeed, in MN13-02, the last synkinematic reactions involve the late replacement of biotite by chlorite and the growth of serpentine. Serpentine rims around olivine are separated from the chlorite + amphibole-rich matrix by a magnetite fringe, likely resulting from the reaction (4) olivine + H₂O → serpentine + magnetite + H₂.

We propose that samples MN13 may derived from a protolith similar to the meta-ultramafic Airette rock that shared a comparable origin and metamorphic/metasomatic evolution. The P–T estimates proposed for the Airette rock indicate a nearly isothermal decompression at 700–800°C from peak pressure conditions of 15–20 kbar down to ∼5 kbar, which is consistent with the P–T range obtained for the late metamorphic evolution in the Gorges-d’Héric area (Rabin et al., 2015; Fréville et al., 2016) and summarized in Trap et al. (2017). However, more quantitative thermodynamic modelling is needed to obtain precise P–T estimates for the “Gouffre du Cerisier“ metamafic rocks; a task beyond the scope of this contribution.

4.2 Zircon inheritance in sample MN13-02

Field and petrographic observations suggest that the chlorite-bearing MN13-02 rock represents the more deformed and highly metasomatized equivalent of MN13-01, affected by syn-kinematic metasomatim. The presence of two zircon crystals, with Cambrian and Ordovician dates for their cores and rims respectively, might be related to the early Paleozoic magmatic event(s) that formed the protolith of the Montagne Noire dome rocks, as evidenced by their Th/U ratios (Tiepel et al., 2004). These igneous bodies of Cambrian or Ordovician age are a common protolith for the Montagne Noire migmatites (Roger et al., 2004, 2015, 2020; Faure et al., 2010; Pitra et al., 2012) and more widely for the French Massif Central (e.g., Vanderhaeghe et al., 2020 and references therein). Nonetheless, the hypothesis of contamination from the surrounding rocks cannot be entirely discounted. Two preserved zircon grains, one fully neoformed (both core and rim) and one inherited (core only), yielded a concordia age of ∼315 Ma. Despite limited data (n = 3), this date appears reliable (MSWD_(C+E) = 0.7) and is consistent with the known history of the Montagne Noire. This 315 Ma age might reflect high-pressure metamorphic conditions (e.g., Whitney et al., 2020) or the initial stage of dome emplacement, marked by LP-HT metamorphism, granitic magmatism, and widespread migmatization of the gneiss dome (e.g., Pitra et al., 2022). To date, it remains unclear whether the ca. 315 Ma high-temperature metamorphism recorded by these zircons occurred under high- or low-pressure conditions (Roger et al., 2015, 2020; Trap et al., 2017; Poujol et al., 2017; Whitney et al., 2015, 2020; Hamelin et al., 2022; Pitra et al., 2022).

4.3 Record of an early Permian event

Most of the analyses performed on zircon grains from sample MN13-02 give a concordia age at 280.6 ± 2.5 Ma, which we interpret as the zircon (re)crystallization age during strong deformation and fluid circulation along E-W trending localized shearing within the GHSZ. The mineralogical changes recorded from the core to the edge of the metamafic lenses, with the crystallization of amphibole, biotite and then chlorite attest to an important water supply. In addition, the numerous quartz veins observed around the sheared lenses also testify to fairly significant hydrothermal activity in the Gouffre du Cerisier area. As suggested by Pidgeon (1992), fluids are able to alter the composition and internal zonation of zircon, and also to dissolve and reprecipitate it. While this idea has long been debated in the literature, it is now accepted that the U-Th-Pb geochronometer of zircon can therefore be totally reset by metasomatism. The presence of alkali-bearing or F-rich fluids favours the alteration and dissolution-precipitation of zircon (e.g., Harlov et al., 2023 and references therein). In comparison to the garnet-clinopyroxene assemblage of the lens core (MN13-01), the chlorite and biotite-bearing metasomatized sample MN13-02 contains a very large proportion of zircon grains that show cathodoluminescence images typical of hydrothermal zircons (Fig. S2).

So far, only three rock samples dated by LA-ICPMS U-Th-Pb on monazite yielded a similar ²⁰⁸Pb/²³²Th Permian age in the Montagne Noire dome (Fig. 1B; Tab. S1). In the eastern Caroux subdome termination, a La Fage fine-grained gneiss yielded two ²⁰⁸Pb/²³²Th ages: one at 305.7 ± 3.9 Ma, interpreted as the age of peak temperature metamorphism synchronous with D3 shearing, and a younger, poorly-defined event, obtained on only 2 monazites at 284 ± 13 Ma (N=4), interpreted as the age of late deformation and fluid circulation during retrograde evolution by Trap et al. (2017). Further to the west, the Laouzas anatectic granite sampled at the Laouzas dam and a migmatite sampled 2 km further south, gave U-Pb and U-Th-Pb ages at ca. 319 Ma and 298-295 Ma. In addition, the U-Th-Pb system also yielded younger ²⁰⁸Pb/²³²Th dates at 285.2 ± 2.2 Ma and 284.7 ± 2.1 Ma, respectively (Poujol et al., 2017). Poujol et al. (2017) advanced two interpretations for the younger data at ca 285 Ma. On the one hand, they propose an incomplete reset of the monazite Th-Pb system during later fluid circulation events. On the other hand, they suggest that these data could reflect the age of an as yet unknown Permian event in the region. Nevertheless, they express a slight preference for the first hypothesis despite the lack of evidence. Whereas, our results suggest that the imprint of an intense fluid circulation was indeed recorded at 285-280 Ma throughout the GHSZ.

The source of the fluid responsible for retrograde metasomatism remains uncertain, requiring further petrological and geochemical investigation beyond the scope of this study. Some hypotheses can be proposed. One possibility is the circulation of hydro-magmatic fluids expelled during the crystallization of migmatites and pegmatites. However, in the Gorges d’Héric, late pegmatite crystallization ages (∼290 Ma; K-Ar muscovite, Franke et al., 2011 and 298 Ma U-Pb monazite in this study) predate the fluid-rock interaction documented in this study by at least around 10 Ma, making this an unlikely direct source. A more plausible origin for the metasomatism could be Permian magmatism. This period is characterized by widespread magmatic activity across the Variscan basement, with mafic to acidic intrusions and effusive volcanism documented in the Briançonnais, Austroalpine, and South-Alpine regions (e.g., Schaltegger and Brack, 2007; Ballèvre et al., 2018). While Permian magmatism is more limited in the external crystalline massifs and the French Massif Central, volcaniclastic layers in intracontinental Stephano-Permian basins indicate early Permian activity in these areas. Around the Montagne Noire gneiss dome, the late to post-Variscan transition is marked by volcanic episodes recorded in nearby sedimentary basins such as Graissessac, Lodève, and Saint-Affrique (Fig. 1A). Recent U-Pb zircon dating of volcanic ash and tuff interbeds (∼280–285 Ma; Michel et al., 2015; Pfeifer et al., 2016; Poujol et al., 2023) confirms the presence of early Permian volcanism in the region. Thus, we propose that early Permian magmatism may have sourced the late-stage hydrothermal fluids responsible for metasomatism in the Montagne Noire Axial Zone, a hypothesis also supported for the central Southern Alps (Zanchetta et al., 2022).

4.4 Shear localization in the Montagne Noire gneiss dome

The exhumation of the migmatitic core of the Montagne Noire is widely accepted to have occurred in a strike-slip or transtensional tectonic setting (Echtler and Malavieille, 1990; Franke et al., 2011; Rabin et al., 2015; Rey et al., 2017; Chardon et al., 2020), involving a NE-SW-oriented high-strain zone. Recent U–Th-Pb LA-ICPMS dating of Axial Zone syn-kinematic granites indicates that high-temperature deformation within this zone occurred between ∼315 and 295 Ma (Roger et al., 2020 and references therein). Monazite grains from the Pont des Soupirs pegmatite (MN 16) have been dated by LA-ICPMS at 296.8 ± 2.3 Ma. Monazite is known to preserve its crystallisation age, as it is not sensitive to Pb diffusion (e.g., Seydoux-Guillaume et al., 2002; Gardés et al., 2007). Also, it recrystallises easily by dissolution/precipitation processes when fluids or magmas are percolating, especially during deformation (Williams et al., 2011; Tartèse et al., 2012; Didier et al., 2013). Syn-kinematic fluid-assisted dissolution-precipitation mechanisms are effective at modifying the chemical and isotopic composition of monazite, even at low temperature (down to ∼300°C) (Hawkins and Bowring, 1997; Townsend et al., 2000). The Pont des Soupirs pegmatite (MN 16) does not show any plastic deformation. We interpret the date of 296.8 ± 2.3 Ma as the crystallization age of monazite during the emplacement of the pegmatite. It is consistent with the monazite U-Th-Pb ages of ∼295 Ma obtained in fine-grained gneisses from the Montagne Noire dome, notably those from the Gouffre du Cerisier in the Gorges d’Héric, which corresponds to high-temperature mylonites (Roger et al., 2020). These results argue for strain localization at metric to hectometric scale across the GHSZ.

In this study, we further detail the median high-strain zone by identifying the GHSZ as the late-stage expression of strain localization during D2 strike-slip deformation. We propose that the development of a ∼N100°-trending shear zone caused the reorientation of S2 foliations into an E-W strike, likely due to strain softening. Magontier (2023) suggested that a kilometre-scale shear zone could be present in the median section of the dome, based on mylonitic orthogneiss facies, although its boundaries are still unknow. The main facies of the orthogneisses of the Somail-Nore form a continuous spectrum, ranging from augen facies known as the Héric gneisses to predominantly banded, non-augen facies referred to as the Larn gneisses. There are also intermediate facies between the two with sporadic augen features (Demange, 1975, 1982; Alabouvette et al., 2003). In an exhaustive literature review, Magontier (2023) highlights that the Larn gneisses were previously identified, in whole or in part, as highly deformed equivalents of the Héric augen gneisses (e.g., Demange, 1975). We delineate the GHSZ boundaries, oriented ∼N100°, which emerge at the south-eastern dome margin where the mica-schist envelope is notably thinned. At higher structural levels, the GHSZ is expressed by E-W-trending brittle faults of the Mons corridor. Thinning associated with the GHSZ led to the tightening of HT/LP metamorphic isograds in the mica-schist envelope (Thompson and Bard 1982), suggesting that part of this deformation post-dates HT/LP metamorphism (∼305 Ma; Trap et al., 2017). Permian ages (285–280 Ma) obtained from rocks within or near the GHSZ (Fig. 1B) further constrain its activity. For example, LA-ICPMS U-Th-Pb dating of monazite from a pegmatite cross-cutting structures on the southern shear zone margin indicates ductile deformation ceased at ∼295 Ma. Further north, at the Gouffre du Cerisier, a deformation corridor accompanied by fluid circulation is dated to ∼280 Ma (MN13-02). These data argue for heterogeneous deformation a shear localization though time.

The crustal-scale, steeply dipping dextral GHSZ likely acted as a pathway for deep-seated fluid migration during late-orogenic unroofing and cooling. Between 320 and 300 Ma, strike-slip shearing affected a ∼10 km-wide segment of the orogenic crust. Between 300 and 290 Ma, deformation localized to a narrower GHSZ (a few kilometers wide), and from 290 to 280 Ma, strain was further restricted to zones 1–100 m wide, such as those at the Gouffre du Cerisier. The GHSZ represents a major structure accommodating the final stages of Variscan tectonogenesis and its inheritance may have partly contributed to the dislocation of Pangea.

5 Conclusion

In this contribution, we provide details regarding the structure and localization of deformation within the eastern part of the Montagne Noire gneiss dome and in the renowned Gorges d’Héric valley. We describe the GHSZ as the late expression of the strike-slip deformation that structured the core of the Montagne Noire gneiss dome. The LA-ICPMS U-Th-Pb monazite age obtained on a pegmatite cross-cutting structures on the southern margin of this shear zone shows that ductile deformation at the Pont des Soupirs outcrop ceased ca. 295 million years ago. Further north, at the outcrop of the Gouffre du Cerisier, a localized deformation corridor is accompanied by significant metasomatism recorded in retrogressed mafic lenses. LA-ICPMS U-Th-Pb dating of zircon from this metasomatic zone results helped us to specify the age of deformation along this crustal scale shear zone GHSZ and to argue for a syn-kinematic aqueous fluid circulation at ca. 285-280 Ma.

Acknowledgements

This research was supported by grant EAR-1050020 and 1946911 from the National Science Foundation (US), by research support from the College of Science and Engineering at the University of Minnesota (USA) and by an INSU/SYSTER project from the French CNRS (France). We sincerely thank Valérie Bosse for her assistance with instrument setup and support during the LA-ICPMS monazite analysis session at the University of Clermont-Ferrand (France). We also deeply appreciate her contributions to the editorial work on this manuscript. We thank Y. Branquet and the anonymous reviewer for their valuable comments and insightful suggestions, which significantly contributed to improving the quality of this manuscript.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Supplementary Material

Fig. S1. (A) Microphotographs of sample MN13-02 showing the structural relationships between the zircon grains and the Amph + Srp + Chl + opaque assemblage. (B) BSE and (C) CL images of zircon from sample MN13-02. Zircon grains that suffered hydrothermal alteration often possess abundant fluid/mineral inclusions, and show dark characteristics in CL images. The red circles indicate spot analyses with a diameter of 21 µm, and the dates correspond to ²⁰⁶Pb/²³⁸U dates (± 2σ) reported in Table S4.

Fig. S1. (A) Photographie prise au microscope de l’échantillon MN13-02 montrant les relations structurales entre les zircons et l’assemblage minéralogique Amph + Srp + Chl + opaques. Images en électrons rétrodiffusés (B) et en cathodoluminescence (C) des zircons de l’échantillon MN13-02. Les zircons ayant subi une altération hydrothermale possèdent souvent d’abondantes inclusions fluides/minérales et présentent des caractéristiques sombres sur les images CL. Les cercles rouges indiquent les analyses ponctuelles d’un diamètre de 21 µm, et les âges correspondent aux âges ²⁰⁶Pb/²³⁸U (± 2σ) indiqués dans le Tableau S4.

Fig. S2. Pictures of The Pont des Soupirs outcrop (A), an undeformed tourmaline-bearing pegmatite several meters thick (sample MN 16; (B)), cutting the N76/64-trending gneissic foliation. (C) BSE image of monazite from pegmatite MN 16. The red circles indicate spot analyses with a diameter of 12 µm, and the dates correspond to ²⁰⁶Pb/²³⁸U dates (± 2σ) reported in Table S4.

Fig. S2. Photographies de l’affleurement du Pont des Soupirs (A), une pegmatite à tourmaline non déformée de plusieurs mètres d’épaisseur (échantillon MN 16; B), recoupant la foliation gneissique en N76/64. (C) Image MEB en électrons rétrodiffusés des monazites de la pegmatite MN 16. Les cercles rouges indiquent les analyses ponctuelles d’un diamètre de 12 µm, et les âges correspondent aux âges ²⁰⁶Pb/²³⁸U (± 2σ) indiqués dans le Tableau S4.

Table S1 Compilation of U-Th-Pb ages obtained in this study and of previously published Variscan ages (modified from Roger et al., 2020). Mz: monazite; Z: zircon; X: xenotime and R: rutile. *: ²⁰⁸Pb/²³²Th Permian age and red diamond: inherited age.

Table S1 Compilation des âges U-Th-Pb obtenus dans le cadre de cette étude et des âges varisques publiés antérieurement (modifiée d’après Roger et al., 2020). Mz : monazite; Z : zircon; X : xénotime et R : rutile. *: ²⁰⁸Pb/²³²Th age permien et losange rouge: âge hérité.

Table S2 Representative mineral compositions of samples MN13-01 and MN13-02.

Table S2. Compositions minérales représentatives des échantillons MN13-01 et MN13-02.

Table S3 Table of instrumentation and analytical method.

Table S3. Table des instruments et méthodes d’analyse.

Table S4 U-Th-Pb LA-ICPMS data for the zircon grains from the MN13-02 sample of Gouffre du Cerisier and for the monazite grains of the Pont des Soupirs pegmatite (MN 16 sample).

Table S4 Données U-Th-Pb LA-ICPMS obtenues sur les zircon de l’échantillon MN13-02 du Gouffre du Cerisier et sur les monazites de la pegmatite du Pont des Soupirs (échantillon MN 16).

References

All Figures

Fig. 1 (A) Simplified structural map of the southern French Massif Central (after Faure et al., 2009) and the simplified map of some of the Variscan massifs (in dark pink colour) in western Europe: Armorican (ARM), Iberian (IB), Massif Central (MC), and Pyrenees (PYR). The rectangle at the southern end of the Massif Central shows the location of the Montagne Noire. (B) Simplified geological sketch map of the Montagne Noire (after Rabin et al., 2015, and Roger et al., 2020) with sample locations dated by the U-Th-Pb (See Tab. S1 for detail). The foliation trajectories show the Gorges d’Hérie Shear Zone (GHSZ) where the trend of the foliations is mainly E-W. (1) : this study; (2) Roger et al. (2020); (3) Roger et al. (2015); (4) Whitney et al. (2015); (5) Hamelin et al. (2022); (6) Trap et al. (2017); (7) Poilvet et al. (2011); (8) Pitra et al. (2012); (9) Poujol et al. (2017); (10) Pitra et al. (2022); (11) Franke et al. (2011); (12) Faure et al. (2010); (13) Faure et al. (2014) and (14) Matte et al. (1998). (C) and (D): Histograms of the U-Th-Pb Variscan ages recorded on the metamorphic (C) and plutonic (D) rocks from the Axial Zone of the Montagne Noire (modified after Roger et al., 2020). The data are reported in Table S1. All monazite EPMA and two zircon U-Pb (#46, 55) dates obtained by Faure et al. (2010) and Matte et al. (1998) are not considered because they are significantly older (∼320‐335 Ma) than the other data set (∼290‐315 Ma) (see more detail in Roger et al., 2020, Roger et al., 2015; Trap et al., 2017; Poujol et al., 2017). The yellow circles below each diagram denote the individual data used to generate the diagrams. The purple dashed line corresponds to the age Sm/Nd (Grt-Cpx-WR) obtained by Faure et al. (2014) on an eclogite boulder sampled near the Peyrambert farm (Fig.1A). (A) Carte structurale simplifiée du sud du Massif Central français (d’après Faure et al., 2009) et la carte simplifiée des massifs varisques d’Europe occidentale : Armoricain (ARM), Ibérique (IB), Massif Central (MC) et les Pyrénées (PYR). Le rectangle à l’extrémité sud du Massif Central montre l’emplacement de la Montagne Noire. (B) Carte géologique simplifiée de la Montagne Noire (d’après Rabin et al., 2015, et Roger et al., 2020) avec les emplacements des échantillons datés par U-Th-Pb (voir le Tab. S1 pour plus de détails). Les trajectoires des foliations montrent la zone de cisaillement des Gorges d’Héric (GHSZ) où la tendance des foliations est principalement E-W. (1) : cette étude; (2) Roger et al. (2020); (3) Roger et al. (2015); (4) Whitney et al. (2015); (5) Hamelin et al. (2022); (6) Trap et al. (2017); (7) Poilvet et al. (2011); (8) Pitra et al. (2012); (9) Poujol et al. (2017); (10) Pitra et al. (2022); (11) Franke et al. (2011); (12) Faure et al. (2010); (13) Faure et al. (2014) and (14) Matte et al. (1998). (C) et (D) : Histogrammes des âges varisques U-Th-Pb enregistrés sur les roches métamorphiques (C) et plutoniques (D) de la Zone Axiale de la Montagne Noire (modifié d’après Roger et al., 2020). Les données sont dans le Tab. S1. Tous les âges EPMA monazite ainsi que deux âges U-Pb zircon (#46, 55) obtenues par Faure et al. (2010) et Matte et al. (1998) ne sont pas pris en compte car ils sont significativement plus vieux (∼320-335 Ma) que l’ensemble des données (∼290-315 Ma) (voir Roger et al., 2020, Roger et al., 2015; Trap et al., 2017; Poujol et al., 2017 pour plus de détail). Les cercles jaunes sous chaque diagramme indiquent les données individuelles utilisées pour générer les diagrammes. La ligne violette en pointillé correspond à l’âge Sm/Nd (Grt-Cpx-WR) obtenu par Faure et al. (2014) sur un bloc d’éclogite échantillonné près de la ferme de Peyrambert (Fig.1A).

In the text

Fig. 2 Structural map on the eastern half of the Montagne Noire gneiss dome showing the orientation of planar fabrics and interpreted foliation trajectories (modified from Rabin et al., 2015). The stereogram shows the preferential orientation of the quartz-bearing dykes that cut across the migmatitic rocks of the axial zone. Carte structurale de la partie orientale du dôme gneissique de la Montagne Noire montrant l’orientation et les trajectoires de foliation interprétées (modifiée d’après Rabin et al., 2015). La projection stéréographique montre l’orientation préférentielle des filons de quartz qui recoupent les migmatites de la zone axiale.

In the text

Fig. 3 Simplified geological map of the Gorges d’Héric valley, modified from Raynal (2004) and Roger et al. (2020). Letters within black circle correspond to monazite U-Th-Pb ages (Tab. S1; Roger et al., 2020): A: Vialais granite (MN 17/303.6 ± 2 Ma); B: F.G.G. (MN11-59/294.3 ± 2.2 Ma); C: Fine grained gneiss (MN11-57/294 ± 2 Ma); D: Augen Gneiss (MN14-03/314.1 ± 1.3 Ma); E: Fine grained gneiss (MN11-117/445.8 ± 3.8 Ma; MN11- 118/441 ± 2.5 Ma; MN11- 119A/316.9 ± 5.5 Ma-297.3 ± 1.6 Ma); F: Fine grained gneiss (MN11-47/450.6 ± 5.6 Ma-301 ± 4 Ma; MN11-121/294.9 ± 2.9 Ma); G: Fine grained gneiss (MN11-27/294.6 ± 1.6 Ma). Carte géologique simplifiée de la vallée des Gorges d’Héric (d’après Raynal, 2004 et Roger et al., 2020). Les lettres inscrites dans les ronds noirs correspondent aux âges U-Th-Pb sur monazites (Tab. S1; Roger et al., 2020): A: granite du Vialais (MN 17/303.6 ± 2 Ma); B: F.G.G. (MN11-59/294.3 ± 2.2 Ma); C: Gneiss fin (MN11-57/294 ± 2 Ma); D: Augen Gneiss (MN14-03/314.1 ± 1.3 Ma); E: Gneiss fin (MN11-117/445.8 ± 3.8 Ma; MN11- 118/441 ± 2.5 Ma; MN11- 119A/316.9 ± 5.5 Ma-297.3 ± 1.6 Ma); F: Gneiss fin (MN11-47/450.6 ± 5.6 Ma-301 ± 4 Ma; MN11-121/294.9 ± 2.9 Ma); G: Gneiss fin (MN11-27/294.6 ± 1.6 Ma).

In the text

Fig. 4 Lithology and structure of the Gouffre du Cerisier outcrop. (A) Strain pattern showing the sigmoidal and lense shaped arrangement of the steeply dipping S2 foliation. The green coloured boudins represent the metamafic rocks (MN13-01/02) that are embedded within the sheared orthogneiss (white coloured). (B) Sketch showing the sheared metamafic lenses with the garnet − pyroxe rich core (sample MN13-01) mantled by amphibole + chlorite rich domains (sample MN13-02). The orientation of the drawing is NW (left)- SE (right) making a different orientation of the mafic pods between A & B (C) Zoom picture in the metamafic lenses showing the strain strain gradient and textural changes from the core to the rim of MN13 metamafic. Lithologie et structure de l’affleurement du Gouffre du Cerisier. (A) Schéma de déformation montrant la géométrie sigmoïde et en lentilles de la foliation S2 à fort pendage. Les boudins de couleur verte représentent les roches métamafiques (MN13-01/02) au sein de l’orthogneiss cisaillé (de couleur blanche). (B) Schéma illustrant les lentilles métamafiques cisaillées, avec un cœur à grenat −pyroxène (échantillon MN13-01) entouré de domaines riches en amphibole + chlorite (échantillon MN13-02). L’orientation du dessin est NW (gauche)- SE (droite), ce qui donne une orientation différente des boules mafiques entre A et B. (C) Vue rapprochée des lentilles métamafiques montrant le gradient de déformation du cœur jusqu’à la bordure des boudins.

In the text

Fig. 5 (A) Microphotographs of sample MN13-01 showing the microstructural arrangement of pyroxene, garnet and amphibole. (B) Close-up image showing the relationships between orthopyroxene (Opx), clinopyroxene (Cpx), garnet, and amphiboles. (C–F) X-ray maps highlighting the elemental distribution of magnesium (C), calcium (D), aluminium (E), and sodium (F), providing both textural and chemical information (see text for explanations). (G) Microphotograph showing a thin olivine corona around ilmenite. (A) Photographies au microscope de l’échantillon MN13-01 montrant la disposition microstructurale des grains de pyroxène, grenat et amphibole. (B) Vue rapprochée montrant les relations entre l’orthopyroxène (Opx), le clinopyroxène (Cpx), le grenat et les amphiboles. (C–F) Cartes en rayons X illustrant la distribution des éléments magnésium (C), calcium (D), aluminium (E) et sodium (F), apportant des informations texturales et chimiques (voir texte pour les explications). (G) Microphotographie montrant une fine couronne d’olivine autour de l’ilménite.

In the text

Fig. 6 Microphotographs showing the mineralogical assemblages and microstructure of sample MN13-02. (A) and (B) Coarse grained facies showing the amphibole + biotite + olivine stbale assemblage. Chlorite appears as a secondary phase. (C) Chlorite-rich fine-grained facies showing the layering defined by amphibole rich and chlorite rich layers. (D) Zoom at the amphibole-olivine grain boundary showing destabilization of amphibole into olivine. (E). Strongly foliated fine −grained facies with boudinage of olivine porphyroblasts with syn-kinematic chlorite within boudin neck. Remnants of large biotite grains are observed. (F) Serpentinite + magnetite corona around olivine. The asymmetric appearance of the corona with almost no serpentine along the left side of the olivine grain suggests a syn-kinematic reaction. Photographies au microscope illustrant les assemblages minéralogiques et la microstructure de l’échantillon MN13-02. (A) et (B) Faciès à grains grossiers montrant l’assemblage stable amphibole + biotite + olivine. La chlorite apparaît comme une phase secondaire. (C) Faciès à grains fins riche en chlorite présentant une alternance de couches riches en amphibole et en chlorite. (D) Zoom sur la limite de grain amphibole-olivine montrant la déstabilisation de l’amphibole en olivine. (E) Faciès à grains fins fortement folié, avec boudinage de porphyroblastes d’olivine et chlorite syn-cinématique dans les queues des boudins. Des vestiges de gros grains de biotite sont observés. (F) Couronne de serpentinite + magnétite autour des grains d’olivine. L’aspect asymétrique de la couronne, avec très peu de serpentinite du côté gauche du grain d’olivine, suggère une réaction syn-cinématique.

In the text

Fig. 7 U-Th-Pb results plotted in Tera-Wasserburg diagram (A) for zircon from the Gouffre du Cerisier sample (MN13-02) and (B) for monazite from the Pont des Soupirs pegmatite (MN 16), respectively. Error ellipses and age uncertainties are 2σ. The dotted ellipses are not taken into account for age calculation. MSWD(C+E) = mean square of weighted deviations for concordance and equivalence. Les résultats U-Th-Pb reportés dans un diagramme Tera-Wasserburg (A) pour les zircons de l’échantillon du Gouffre du Cerisier (MN13-02) et (B) pour les monazites de la pegmatite du Pont des Soupirs (MN 16), respectivement. Les ellipses d’erreur et les incertitudes sur les âges sont à ± 2σ. Les ellipses en pointillés ne sont pas prises en compte pour le calcul de l’âge. MSWD(C+E) = déviation pondérée de la régression par moindre carrés pour la concordance et l’équivalence.

In the text