Alpine Burial and Heterogeneous Exhumation of Variscan Crust in the West Carpathians: Insight From Thermodynamic and Argon Diffusion Modelling

By Jerabek, P Faryad, W S; Schulmann, K; Lexa, O; Tajcmanova, L

Abstract: Phase equilibrium modelling of metagranitoids and metapelites in the MnO-Na^sub 2^O-CaO-K^sub 2^O-FeO-MgO-Al^sub 2^O^sub 3^-SiO^sub 2^-H2O system was used to characterize Variscan and Alpine metamorphism in the crystalline basement of the Vepor Unit, West Carpathians. The calculated P-T conditions range between 570-670 [degrees]C and 6-8.5 kbar for the Variscan and 430-600 [degrees]C and 5-11 kbar for the Alpine event. These two events show contrasting metamorphic field gradients and P-T evolutions, indicating 22-27 [degrees]C km^sup -1^ during the retrograde Variscan metamorphism and 15-18 [degrees]C km^sup -1^ during the prograde Alpine metamorphism. The prograde Alpine metamorphism is associated with the Early Cretaceous overthrusting of the southern Gemer Unit, which resulted in apparently contemporaneous burial and horizontal ductile spreading of the underlying Vepor basement. The argon diffusion modelling was used to interpret the existing Variscan, Alpine and mixed ^sup 40^Ar/^sup 39^Ar cooling ages, and to constrain the T-t evolution and Alpine thermal overprint. Constructed Alpine metamorphic isograds and isotherms show horizontal P-T gradients resulting from heterogeneous exhumation of the deeper parts of the Vepor basement along two narrow belts during later folding. Here the structurally deeper metapelites form large- scale anticlinal cusp-like structures that separate structurally higher metagranitoids. The distribution of metamorphic isograds is commonly used to identify metamorphic field gradients associated with large-scale tectonic processes, such as collision, subduction or extension of the continental crust. Because of the strong influence of the temperature on the metamorphic reaction kinetics (Spear 1993), metamorphic isograds are interpreted as the traces of isograd surfaces related to certain thermal conditions. Although early thermal numerical models suggested that the peak pressure and temperature conditions are not attained synchronously (England & Thompson 1984; Thompson & England 1984), the distribution of metamorphic isograds, showing spatial variations in the peak temperature conditions, remains an important observation that can help to identify dominant tectonothermal regimes (Todd & Engi 1997). Until recently, only fragments of P-T paths could be obtained from conventional thermobarometry; however, recent advances in thermodynamic modelling (Powell et al. 1998; Connolly 2005) allow reconstruction of more complete parts of P-T evolutions. Mapping of metamorphic field gradients and the deciphering of P-T evolutions can contribute to our understanding of burial and exhumation of either coherent crustal blocks or their constituent parts (Stipska et al. 2006).

In this study, we use P-T sections, calculated with the Perple_X thermodynamic software (Connolly 2005), to distinguish prograde Alpine and retrograde Variscan P- T evolutions and metamorphic field gradients within the polymetamorphic Vepor basement in the Central West Carpathians. The phase-equilibrium modelling of Alpine metamorphic conditions, combined with field structural analysis and reassessment of an existing ^sup 40^Ar/^sup 39^Ar dataset, is used to discuss the mechanisms of Cretaceous burial and subsequent heterogeneous exhumation of the Vepor basement.

Geological setting

The Central West Carpathians consist of three major crustal units, from south to north, the Gemer, Vepor and Tatra (Fig. 1). The differences and similarities in lithological content and tectonometamorphic history of the three units suggest an affinity to two major crustal blocks, the northern and the southern block, separated by a crustal-scale thrust zone. The southern block, represented by the Gemer Unit, comprises volcano-sedimentary sequences of Early Palaeozoic age, exhibiting Variscan low- to medium-grade metamorphism (Faryad 1991), and Carboniferous-Triassic cover. The Gemer Unit is overlain by a Jurassic subduction- accretionary complex of the Meliata Ocean (Faryad 1995; Faryad & Henjes-Kunst 1997). The northern block, represented by the Tatra and Vepor units, comprises Variscan high-grade metamorphic rocks, Carboniferous granitoids and Permian-Early Cretaceous cover. Both the blocks are tectonically overlain by the Turna, Silica, Choc and Krizna nappes (Fig. 1), which are composed of the Late Palaeozoic- Cretaceous platform and rift-related sedimentary sequences (Plasienka et al. 1997, and references therein). During the Cretaceous, the overall convergence of the region led to juxtaposition of the three crustal units, emplacement of the above- mentioned nappes and, consequently, to tectonometamorphic reworking of both blocks. However, the grade of the Alpine metamorphic overprint differs significantly in the two blocks. The southern block exhibits very low-grade to greenschist Alpine metamorphic conditions (Faryad 1997), whereas the northern block exhibits a northward decrease in metamorphic conditions from upper greenschist- to amphibolite-facies conditions in the Vepor Unit (Vrana 1966, 1980; Plasienka et al. 1999; Janak et al. 2001; Koroknai et al. 2001) to prehnite-pumpellyite-facies conditions in the Tatra Unit (Krist et al. 1992).

The Vepor Unit

The overall structure of the Vepor Unit is characterized by alternation of the NE-SW-trending belts with different lithologies (Fig. 1). The basement consists of two complexes exhibiting different lithology and structural position (Klinec 1966). The structurally lower Hron Complex (Fig. 2) is formed of paragneisses, micaschiste and amphibolites, whereas the Kral’ova Hol’a Complex hanging wall is formed of Upper Devonian migmatites and anatectic orthogneisses, Lower Carboniferous S-type granitoids and Upper Carboniferous tonalites (Bibikova et al. 1988, 1990; Michalko et al. 1998). The division of the Vepor basement into two lithotectonic complexes (Klinec 1966) was later questioned, mostly because of the differences in the lithology and degree of Alpine tectonometamorphic reworking of the southern and northern parts of the Hron Complex (Fig. 1). Nevertheless, the mutual structural relationship of the two dominant lithologies within the Vepor Unit is identical to that of the Tatra Unit and to that of the eastern segment of the Vepor Unit (Fig. 1), exhibiting lower degree of Alpine reworking (Janak 1994; Jacko et al. 1996). Consequently, the inverted crustal structure (granitoids overlying metasediments) within both units argues for an inherited Variscan crustal stratification (Bezak et al. 1997).

The ages of the Variscan and Alpine tectonometamorphic events were established by the U-Pb and ^sup 40^Ar/^sup 39^Ar ages. The U- Pb magmatic zircon ages from the Kral’ova Hol’a Complex migmatites and granites as well as hornblende 40^Ar/^sup 39^Ar cooling ages from the Hron Complex amphibolites indicate that Variscan metamorphism occurred between 370 and 340 Ma (Kral’ et al. 1996; Michalko et al. 1998; Putis et al. 2001). The intrusions of Carboniferous granitoids (Bibikova et al. 1988, 1990) were followed by minor post-orogenic Late Permian-Early Triassic A-type granites and subalkaline volcanites (Petrik et al. 1995; Kotov et al. 1996; Putis et al. 2000). The Alpine metamorphic event is defined by numerous hornblende, muscovite and biotite 40^Ar/^sup 39^Ar cooling ages, which show two distinct age populations ranging between 115- 105Ma and 87-80Ma (see Table 1 for 40^Ar/^sup 39^Ar data and references).

Variscan metamorphism in metapelites of the Hron Complex reached kyanite grade (Meres & Hovorka 1991; Kovacik 1993; Table 2), although the northern part of this complex also contains remnants of Variscan eclogites (Janak et al. 2007). Janak et al. (2001) established three Alpine metamorphic zones within the southern part of the Hron Complex: (1) chloritoid + chlorite + garnet; (2) garnet + staurolite + chlorite; (3) staurolite + biotite + kyanite. Corresponding P-T estimates range from 500 [degrees]C and 7 kbar to 620 [degrees]C and 10 kbar. Variscan metamorphism of the Kral’ova Hol’a Complex reached conditions of partial melting indicating 680- 730 [degrees]C and 4-6 kbar (Siman et al. 1996). Subsequent Alpine metamorphism is associated with greenschist-facies reworking at 400- 500 [degrees]C and 5 kbar (Vrana 1980).

The southern part of the Vepor Unit, the Gemer-Vepor Contact Zone, is formed by schists and meta-arkoses or metasandstones. These rocks have traditionally been related to the Variscan basement (Klinec 1966; Vrana 1966); however, pollen analysis suggests both Early and Late Palaeozoic depositional ages (Planderova & Vozarova 1978; Klinec & Planderova 1981). Alpine metamorphism of the Gemer- Vepor Contact Zone is characterized by the presence of chloritoid- kyanite schists (Vrana 1964) with estimated P-T conditions of 500- 590 [degrees]C and 4.5-7.7 kbar (Luptak et al. 2000; see also Table 2). The area locally exhibits subsequent HT-LP metamorphic overprint (Vozarova 1990), related to intrusion of the Late Cretaceous (82-75 Ma) Rochovce I-type granite (Hrasko et al. 1998, 1999; Poller et al. 2001).

Two metamorphosed cover successions, each with a distinct lithostratigraphy, have been recognized in the Vepor Unit (Biely 1964). The Triassic para-autochthonuous metasediments of the Foederata Cover (Rozlozsnik 1935) overlie the Gemer-Vepor Contact Zone and the Kral’ova Hol’a Complex. Alpine metamorphic conditions of the Foederata Cover did not exceed 380 [degrees]C and 4.5 kbar (Luptak et al. 2003; see also Table 2). Permian to Early Cretaceous allochthonous metasediments of the Vel’ky Bok Cover (Biely 1964), exhibiting low-grade Alpine metamorphic overprint (Plasienka et al. 1989; Luptak et al. 2003), occur at the northern margin of the Vepor Unit. Structural record within the Vepor Unit

Three main deformational events were identified in the basement of the Vepor Unit and two in the Permian to Early Cretaceous cover. The earliest deformation D^sub V^, recorded exclusively within the Vepor basement, is interpreted as pre-Alpine, whereas the later two deformations D^sub A1^ and D^sub A2^, also affecting the cover, are interpreted as the first and second Alpine phases (Fig. 2).

The earliest deformation D^sub V^ is associated with the development of high-grade metamorphic schistosity S^sub V^ in the Hron Complex and Gemer-Vepor Contact Zone, and high-grade orthogneiss fabric and migmatite layering in the Kral’ova Hol’a Complex. Where unaffected by later deformation, the S^sub V^ fabric generally shows east-west trends.

In the Hron Complex and Gemer-Vepor Contact Zone, the D^sub A1^ deformation event is characterized by transposition of the previous fabric into new penetrative metamorphic foliation S^sub A1^, which is axial planar to the locally preserved isoclinal folds. In the Kral’ova Hol’a Complex, the D^sub A1^ event is represented by a heterogeneously developed deformation fabric defined by shape- preferred orientation of micas and elongated quartz aggregates. In the Foederata and Vei’ky Bok Cover, the S^sub A1^ low-grade metamorphic fabric is mostly parallel to the bedding, being axial planar to centimetre- to kilometre-scale isoclinal folds (Plasienka 2003). With the exception of the Hron Complex, where the S^sub A1^ fabric dips towards the NW and SE at shallow to steep angles, the S^sub A1^ fabric is subhorizontal and/or gently dipping to the east or south (Fig. 2). The S^sub A1^ fabric bears ENE-WSW-trending L^sub A1^ stretching Hneation defined by alignment of feldspar, quartz, and/ or micas. L^sub A1^ is locally subparallel to the axis of the isoclinal F^sub A1^ folds.

During the D^sub A2^ deformation event, large-scale folds developed in the Hron Complex and Gemer-Vepor Contact Zone, and crenulation cleavage affected the D^sub A1^ high-strain domains in the Kral’ova Hol’a Complex. Generally, the F^sub A2^ folds have subvertical NE-SW- to ENE-WSW-trending axial planes and subhorizontal axis. In the southern Hron Complex, the S^sub A2^ axial planar cleavage exhibits large-scale inverse fan structure (Fig. 2). In the Foederata and Vei’ky Bok Cover, the D^sub A2^ deformation resulted in the development of metre-scale upright folds.

Petrography

The rocks studied here are: (1) paragneisses and micaschists (with semipelitic to pelitic compositions) of the Hron Complex, (2) metagranitoids (variably deformed granodiorite-tonalite) of the Kral’ova Hol’a Complex and (3) schistose metasediments (originally sandstones, shales and volcaniclastic rocks) of the Gemer-Vepor Contact Zone (see Fig. 1 for locations). Two metamorphic mineral assemblages (M^sub 1^ and M^sub 2^), as well as a primary igneous (Ign) mineral assemblage (Table 3), were identified on the basis of the textural relations and chemical compositions (mineral abbreviations after Kretz 1983).

Paragneisses and micaschists of the Hron Complex

The Hron Complex paragneisses-micaschists consist of garnet, biotite, chlorite, muscovite, plagioclase, quartz and accessory ilmenite, titanite, epidote, allanite, zircon, monazite and tourmaline. Two varieties of garnet, muscovite, biotite and plagioclase are related to metamorphic stages M^sub 1^ and M^sub 2^ (Table 3). Garnet (I) represents cores of large garnet porphyroblasts, whereas garnet (II) either envelops garnet (I) or occurs in the matrix as small single grains. Garnet (I) and (II) can be further distinguished on the basis of different amounts of opaque mineral inclusions or compositional differences, visible in the backscattered electron (BSE) images (Fig. 3). Muscovite (I) is coarse-grained (0.2-1 mm) and differs from the fine-grained muscovite (II) in the matrix in composition (see below). Biotite (I) is coarse-grained and dark brown and is characterized by the presence of exsolution rutile needles. It is commonly replaced by chlorite. In contrast, biotite (II) is fine-grained and lighter brown. Large plagioclase (I) porphyroblasts are overgrown by fine- grained muscovite (II), whereas small plagioclase (II) grains are muscovite-free.

Two samples of paragneiss (PP207, PP473) and one of micaschist (PP482) from different locations in the Hron Complex (Fig. 1) were selected for detailed petrological study because of the clear textural relations between the M^sub 1^ and M^sub 2^ metamorphic mineral assemblages. Paragneiss sample PP207 is characterized by the presence of two varieties of garnet (Fig. 3a) and by biotite (II) overgrowing chlorite (Fig. 4a). The core garnet (I) is rich in biotite, ilmenite, plagioclase and quartz inclusions, whereas the inclusion-poor rim garnet (II) contains only ilmenite and quartz. Sample PP473 is rich in plagioclase and contains only one variety of garnet (c. 0.3 mm), related to the first metamorphic stage M^sub 1^, which is intensely fractured and partly replaced by both biotite (II) and chlorite (Fig. 4b). Sample PP482 is characterized by high muscovite and chlorite contents and by relics of garnet (I), enclosed by garnet (II) (Fig. 3c). Only one variety of biotite, almost completely replaced by chlorite, is related to the first metamorphic stage M^sub 1^.

Metagranitoids of the Kral’ova Hol’a Complex

Five Kral’ova Hol’a Complex metagranitoid samples (PP150, PP28, PP225, PP229 and MM136) from the central part of the Vepor Unit (Fig. 1) were selected for detailed petrological study, because they contain garnet and provide clear textural relations between Ign and M^sub 2^ mineral assemblages. Common minerals in metagranitoids are biotite, muscovite, chlorite, clinozoisite, plagioclase, quartz and accessory zircon, apatite, monazite, alanite, titanite, rutile and ilmenite. Some samples also contain garnet and relics of K- feldspar. The K-feldspar phenocrysts, plagioclase (I), tabular muscovite (I) and dark brown tabular biotite (I) are interpreted as primary igneous phases (Table 3). Large biotite (I) grains are characterized by the presence of exsolution rutile needles (Fig. 4c). Plagioclase (I) porphyroblasts are replaced by fine-grained muscovite (II) and clinozoisite. Samples PP150 and PP28 contain porphyroblastic garnet, which is formed by partly corroded garnet (I) cores, probably of igneous origin, and idioblastic rims, formed by metamorphic garnet (II) (Fig. 3e). The metamorphic mineral assemblage, related to the M^sub 2^ metamorphic stage (Table 3), comprises small plagioclase (II) grains, chlorite intergrowing with biotite (II), clinozoisite and small garnet (II) grains, occurring within or adjacent to plagioclase porphyroblasts (Fig. 4c).

Schists of the Gemer-Vepor Contact Zone

The Gemer-Vepor Contact Zone schists consist of biotite, muscovite, chlorite, plagioclase, quartz and accessory magnetite and ilmenite. Some samples also contain garnet, clinozoisite, tourmaline and cordierite, the last resulting from a local HT-LP metamorphic overprint (Vozarova 1990). Two samples of Gemer-Vepor Contact Zone schists (BL1 and PP304) were selected for detailed petrological study because of the clear textural relations between the M^sub 1^ and M^sub 2^ metamorphic mineral assemblages. Two varieties of biotite and one variety of plagioclase can be distinguished optically. Biotite (I) is idiomorphic and dark brown, whereas biotite (II) is anhedral and light brown (Fig. 4d). Two textural varieties of garnet, described in these rocks by Korikovsky et al. (1990), were detected only in sample BL1. Idioblastic garnet (I) in the core is enveloped by garnet (II), which exhibits abundant quartz and ilmenite inclusions (Figs 3g and 4d). Sample PP304, containing only the garnet (II) variety, exhibits relics of chlorite, which are overgrown by biotite (II).

Chemical composition of the minerals

The minerals were analysed using the CAMSCAN 54 SEM with energy- dispersive spectrometers and microanalytical system LINK ISIS 300 at the Institute of Petrology and Structural Geology, Charles University, Prague, and the CAMECA SX 50 electron microprobe with four wavelength-dispersive spectrometers at the Institute of Mineralogy, Technische Universitat Stuttgart. Mineral formulae and ferric/ferrous iron ratios were calculated on the basis of 12 oxygens and eight cations for garnet and 11 oxygens for micas. Representative analyses of minerals, exhibiting two compositional varieties and/or used in thermodynamic or conventional thermometry calculations, are presented in Table 4 (garnet) and Table 5 (muscovite, biotite).

The two varieties of garnet (Fig. 3, Table 4) usually exhibit sharp compositional transition (Fig. 5). Apart from garnet (I) in the Kral’ova Hol’a Complex metagranitoids (sample PP150 and PP28 in Table 4), interpreted as igneous on the basis of textural relations and the high Mg and Mn contents, garnets (I) in the Hron Complex paragneisses-micaschists and the Gemer-Vepor Contact Zone schists (samples PP207, PP473, PP482 and BL1 in Table 4 and Fig. 3) exhibit similar compositions with high Fe and Mg contents. In addition, these garnets exhibit similar compositional trends, manifested by a minor decrease in Mg and X^sub Mg^ (X^sub Mg^ = Mg/(Mg + Fe)), and a minor increase in Mn towards the rim. Gamet (II) is enriched in Ca at the expense of the Fe content and exhibits pronounced compositional zoning, manifested by a rimward increase in Fe, Mg and X^sub Mg^, and a decrease in Ca and Mn (Fig. 3, Table 4). The two textural varieties of biotite mostly reveal identical composition with Ti contents ranging between 0.08 and 0.15 atoms per formula unit (a.p.f.u.) and X^sub Mg^ between 0.3 and 0.5 (Table 5). The biotite (I) composition, established in biotite inclusion enclosed in plagioclase (I) within garnet (I) (sample PP207; Fig. 3a), exhibits a higher Ti content (0.29 a.p.f.u.) and lower X^sub Mg^ (0.22).

Chlorite is present in most samples and its X^sub Mg^ value varies between 0.33 and 0.48.

In the Hron Complex paragneisses-micaschists and Kral’ova Hol’a Complex metagranitoids, the two textural varieties of muscovite differ in their Si and Na contents. The Si and Na contents in muscovite (I) are in the range of 3.00-3.11 a.p.f.u. and 0.04-0.21 a.p.f.u., respectively, whereas muscovite (II) with higher proportion of phengite has Si contents of 3.11-3.32 a.p.f.u. and Na contents of 0.02-0.14 a.p.f.u. (Table 5). Only one textural variety of muscovite in the Gemer-Vepor Contact Zone schists contains 3.06- 3.33 a.p.f.u. of Si and 0.04-0.15 a.p.f.u. of Na.

The two textural varieties of plagioclase, present in most samples from the Hron Complex and Kral’ova Hol’a Complex, exhibit identical albite composition with less than 5 mol% of anorthite component. The original plagioclase (I) composition (An 23-24 mol%) was established in porphyroblasts of the Hron Complex paragneiss (sample PP473) or in inclusions within garnet (I) (sample PP207) (Fig. 3a). In the Gemer-Vepor Contact Zone schists, the plagioclase (I) with 13-28 mol% An is locally replaced by plagioclase (II) of albite composition. It is not clear whether the newly formed albite and clinozoisite in the Gemer-Vepor Contact Zone schists are part of the M^sub 2^ metamorphic mineral assemblage, or if these are late- stage alteration products.

P- T conditions

The textural relations within the studied rocks demonstrate the polymetamorphic nature of the Vepor basement complex. The first metamorphic event (M^sub 1^) was recognized in the Hron Complex paragneisses-micaschists and in the Gemer-Vepor Contact Zone schists. It is characterized by the presence of garnet (I), coarse- grained muscovite (I) and biotite (I), and plagioclase (I) porphyroblasts (Table 3). The second metamorphic event (M^sub 2^) identified in all the investigated rock complexes is represented by relatively Ca-rich garnet, phengite muscovite (II), albite plagioclase (II), chlorite, +- biotite (II), and +- clinozoisite. Considering the existing geochronological data for the Vepor Unit, especially the ^sup 40^Ar/^sup 39^Ar ages (Table 1), the recognized M^sub 1^ and M^sub 2^ events relate to Variscan and Alpine regional metamorphic events, respectively.

To interpret the phase relations and metamorphic evolution of the studied rocks, the P-T conditions of the two metamorphic events were estimated by means of phase equilibrium modelling and the P-T section approach (Powell et al. 1998; Connolly & Petrini 2002). The P-T sections were calculated using the thermodynamic software Perple_X (Connolly 2005), with the internally consistent thermodynamic dataset of Holland & Powell (1998; 2002 upgrade). The bulk-rock compositions used in the modelling were determined by three methods. (1) The wholerock composition obtained by X-ray fluorescence (XRF) analysis was used to model P- T conditions for the older (Mi) metamorphic event, as well as for the younger (M^sub 2^) metamorphic event in those samples exhibiting complete M^sub 2^ overprint. (2) The modified whole-rock composition was used for the M^sub 2^ event in samples preserving a significant modal proportion of the mi phases. Modification of the whole-rock composition involved exclusive removal of garnet (I), as the other mineral phases of the M^sub 1^ assemblage re-equilibrated their compositions (plagioclase (I) and biotite (I)) or represent a minor modal portion of the rock (muscovite (I) in the sample PP207). (3) The estimated equilibration volume (Stiiwe & Powell 1995) was used for the M^sub 2^ event in samples with abundant pre-M^sub 2^ minerals, which inhibited modification of the whole-rock composition. The equilibration volume, covering a representative modal portion of the M^sub 2^ mineral assemblage, was derived from the BSE images using image analysis. For conversion to the proportions of oxides, the chemically zoned garnets were approximated by their mean composition; this did not affect our modelling results because of the low modal content of garnet (<1 modal%). The resulting bulk- rock compositions are indicated in each P-T section (Figs 6 and 7).

As for the whole-rock composition, mineral chemistry (particularly the high Mn content in garnets) and modal proportion of the mineral phases, all P-T sections were calculated in the MnO- Na^sub 2^O-CaO-K^sub 2^O-FeO-MgO-Al^sub 2^O^sub 3^-SiO^sub 2^-H2O (MnNCKFMASH) system with excess water. Titanium was not included in our calculations because of its low content (<0.5 wt%) and the minor role of Ti-bearing phases, such as rutile (within biotite (I)) and ilmenite, in the metamorphic reactions in question. Although the manganese content in the analysed rocks is also low, it was included in modelling because of its striking effect on the stability of garnet (Tinkham et al. 2001). In addition to the plagioclase solution model by Newton et al. (1980), we used the solution models of White et al. (2001) for NCKFMASH end-member phases expanded by the Mn endmembers for biotite, chlorite, garnet, cordierite and staurolite (Tinkham et al. 2001).

As an independent method (different solution models and thermodynamic data from those used in P-T section calculations), the conventional garnet-biotite thermometer of Kleemann & Reinhard! (1994) was employed to evaluate our phase equilibrium modelling results (Table 6). Garnet rim-biotite analyses used for the calculations are listed in Tables 4 and 5.

Paragneisses and micaschiste of the Hron Complex

Two P- T sections, constructed to estimate the P- T conditions of sample PP207, were modelled using the whole-rock composition for the m^sub 1^ event and modified whole-rock composition for the M^sub 2^ event (see above). For the m^sub 1^ event P-T section, the observed equilibrium mineral assemblage (Table 3) corresponds to the five- variant Gn-Bt-Ms-Pl-QtZ-H2O field (Fig. 6a). Peak P- T conditions for the M^sub 1^ event were established using the garnet (I) core composition and isopleths of X^sub Mg^ (0.11), grossular (5 mol%), and spessartine (5 mol%) component (Table 4). The estimated P-T conditions of c. 680 [degrees]C and c. 8.5 kbar are in good agreement with the results of garnet-biotite thermometry, which yielded 660[degrees]C (Table 6). This temperature was calculated using garnet (I) enclosing biotite (I) (Fig. 3a, Tables 4 and 5). The M^sub 2^ assemblage in sample PP207 (Table 3), exhibiting biotite overgrowths of chlorite, indicates a prograde transition from the four-variant Grt-Bt-Ms-Chl-Pl-Qtz-H2O field into the five- variant Gn-Bt-Ms-Pl-QtZ-H2O field (Fig. 6b). The P- T increase from 540 [degrees]C and 9 kbar to 600 [degrees]C and 11 kbar, is suggested by a rimward increase of X^sub Mg^ (from 0.055 to 0.085) and a decrease of the grossular (from 24 to 17mol%) and spessartine (from 3 to 1 mol%) contents within garnet (II) (Fig. 3b, Table 4). Garnet-biotite thermometry, using the garnet (II) rim and matrix biotite (II), yields 550 [degrees]C for the M^sub 2^ event (Table 6).

Based on the textural observations in sample PP473 (Fig. 4b), the peak metamorphic mineral assemblage (Grt-Bt-Ms-PlQtz) relates to the M^sub 1^ metamorphic event (Table 3). The corresponding P-T section (Fig. 6c) was calculated using the whole-rock composition. For the isopleths of X^sub Mg^, grossular and spessartine component, the garnet (I) core composition (X^sub Mg^ = 0.165, Grs 4mol%, Sps 8mol% in Table 4) plots into the four-variant Gn-Bt-Ms-ChI-Pl-QtZ-H2O field and records peak m^sub 1^ P-T conditions of c. 580 [degrees]C and c. 6.5 kbar.

As minerals formed during the mi event (Table 3) are rarely preserved in sample PP482 (see garnet (I) in Fig. 3c), the wholerock composition was used to determine the P- T conditions for both metamorphic events (Fig. 6d). The stability of biotite, together with the chemical composition of garnet (I), restricts the P-T range for the M^sub 1^ event to two possible fields: the fourvariant Gn- Bt-Ms-ChI-Pl-QtZ-H2O and the three-variant Gn-Bt-Ms-ChI-St-Pl-QtZ- H2O fields. The actual lack of chlorite (I) and staurolite in this sample can be explained by an intense M^sub 2^ overprint. Considering both possible fields and the garnet (I) composition (Grs 10mol%, Sps 6mol%), the P- T estimate for the mi event is in the range of 6.5-7.5 kbar and 580-600 [degrees]C (dashed ellipse in Fig. 6d). Although this estimate remains problematic because of the lack of chlorite and staurolite and the lower X^sub Mg^ in garnet (I), it is similar to the M^sub 1^ estimate from nearby sample PP473 (Fig. 6c; see Fig. 1 for locations). For the M^sub 2^ event in sample PP482, the observed mineral assemblage (Table 3) corresponds to the five-variant Grt-Chl-Ms-Pl-QtzH2O field (Fig. 6d). An increase in X^sub Mg^ (from 0.061 to 0.083), together with a decrease in spessartine (from 12 to 8 mol%) and grossular (from 19 to 18 mol%) contents towards the garnet (II) rim, suggests a prograde metamorphic path and an increase in P-T conditions from 535 [degrees]C and 6.5 kbar to 555 [degrees]C and 7 kbar. As garnet (II) represents c. 6 modal% of the rock, considerable changes in the effective bulk-rock composition must have occurred during the garnet growth. Therefore, the P-T section, calculated from the whole-rock composition, characterizes only the initial stages of the garnet (II) growth and, consequently, the outermost rim-composition of garnet (II) (Table 4) was not taken into consideration (see Gaidies et al. 2006).

Metagranitoids of the Krai Ova Hol’a Complex Because of the large proportion of relict igneous phases (Fig. 4c, Ign in Table 3) in metagranitoids, the P-T sections for the M^sub 2^ event were calculated using the estimated equilibration volume. As the studied metagranitoid samples (PP28, PPl50, PP225, PP229, and MM 136) exhibit nearly identical bulk-rock composition, their garnet (II) compositions (Table 4) were plotted into a single P- T section (Fig. 7a). The observed M^sub 2^ mineral assemblage (Table 3) corresponds to the three-variant Grt-Bt-Ms-Chl-CzoAb-QtZ-H2O field. Using the garnet (II) compositions, the estimated P-T conditions are in the range of 430-450 [degrees]C and 5.5-6.5 kbar (sample PP225) to 460- 480 [degrees]C and 7.5-8.5 kbar (sample PP150). A prograde metamorphic P- T path is suggested by the garnet (H) compositional zoning (Fig. 7a). The garnetbiotite thermometry, used for the garnet (II) rim and matrix biotite (H) (Tables 4 and 5), yields temperatures ranging between 430 [degrees]C (sample PP225) and 530 [degrees]C (sample PP150) (Table 6).

Schists of the Gemer-Vepor Contact Zone

P-T conditions of sample BLl (Table 3) were calculated using the whole-rock composition for the M^sub 1^ event and modified whole- rock composition for the M^sub 2^ event. Although biotite (I) and garnet (I) are the only preserved M^sub 1^ phases in this sample, the presence of former muscovite, chlorite and plagioclase is suggested by the composition of garnet (I), which plots into the four-variant Gn-Bt-Ms-ChI-Pl-QtZ-H2O field (Fig. 7b). Peak P-T conditions established using the garnet (I) core composition (X^sub Mg^ = 0.134, Grs 5 mol%, Sps 10 mol% in Table 4) indicate c. 580 [degrees]C and c. 6 kbar. The observed M^sub 2^ mineral assemblage in sample BLl corresponds to the four-variant Grt-Bt-Ms-ChlPl-QtZ- H2O field (Fig. 7c). An increase in X^sub Mg^ (from 0.086 to 0.093), together with a decrease in the spessartine (from 9 to 3 mol%) and grossular (from 29 to 26 mol%) contents, towards the garnet (II) rim indicates a prograde metamorphic path and an increase in P-T conditions from 520 [degrees]C and 8 kbar to 540 [degrees]C and 9 kbar. Garnet-biotite thermometry, applied for the garnet (H) rim and matrix biotite (H), indicates a temperature of 560 [degrees]C for the M^sub 2^ event (Table 6).

As a result of the strong M^sub 2^ overprint, relics of biotite (I), revealing identical composition to biotite (H), are the only sign of the M^sub 1^ event in sample PP304 (Table 3). Therefore, the whole-rock composition was used to calculate the P-T conditions for the M^sub 2^ event. Considering the biotite (II) overgrowths of chlorite and using the garnet compositional isopleths, the P-T conditions of the M^sub 2^ event indicate transition from the four- variant Gn-Bt-Ms-ChI-Pl-QtZ-H2O field to the five-variant Gn-Bt-Ms- Pl-QtZ-H2O field (Fig. 7d). A prograde P- T path, indicating an increase in the P-T conditions from 510[degrees]C and 8 kbar to 530 [degrees]C and 9 kbar, is suggested by the rimward increase of X^sub Mg^ (from 0.073 to 0.087), the decrease in spessartine (from 21 to 13 mol%) and the constant grossular (26 mol%) content. This P- T estimate is consistent with the results of garnet-biotite thermometry, indicating a temperature of 525 [degrees]C for the M^sub 2^ event (Table 6).

Re-evaluation of the existing ^sup 40^Ar/^sup 39^Ar dataset

Heterogeneity of the ^sup 40^Ar/^sup 39^Ar ages obtained from the Vepor basement is commonly explained by the complex P-T-1 history during the Variscan and Alpine tectonometamorphic events (Maluski et al. 1993; Kovacik et al. 1996; Kral’ et al. 1996). Nevertheless, there appear to be systematic variations in the age spectra across the Vepor basement, which, using numerical modelling of argon diffusion (Lister & Baldwin 1996), can be directly related to a particular T-1 history. The aim of this section is to demonstrate regional differences in the thermal history of the Vepor basement rocks, pointing to Alpine heterogeneous exhumation of the original Variscan crustal stratification.

There are three types of apparent age spectra within the existing ^sup 40^Ar/^sup 39^Ar dataset (see Table 1 for references), with the following features: (1) well-developed plateaux (Cretaceous ages between 80 and 87 Ma); (2) either well-developed plateaux or low- temperature release of excess argon, followed by broad pseudo- plateaux encompassing the rest of ^sup 39^Ar (Palaeozoic ages older than 340 Ma); (3) a stepwise increase in the apparent ages (reaching a maximum between 85 and 180Ma) (Fig. 8a and b). The integrated ages, obtained in the last case, clearly represent mixtures between the Cretaceous and Palaeozoic age populations.

The spatial distribution of the existing hornblende, muscovite and biotite ^sup 40^Ar/^sup 39^Ar ages reveals a systematic pattern across the Vepor basement. The centre of the southern Hron Complex exhibits a concentration of Cretaceous plateau ages, which, for hornblende, changes to mixed ages along strike to the NE and SW (Fig. 9). In the northern Hron Complex, Palaeozoic plateau ages prevail over the mixed age spectra for hornblende, whereas muscovite and biotite exhibit Cretaceous plateau ages. The ^sup 40^Ar/ ^sup 39^Ar spectra from the Kral’ova Hol’a Complex exhibit Palaeozoic plateau ages for hornblende and Cretaceous or mixed ages for micas.

Applying the concept of closure and blocking temperatures, the existing ^sup 40^Ar/^sup 39^Ar spectra can be re-evaluated in terms of Alpine thermal overprint. Argon closure temperatures (i.e. temperatures below which the mineral grain is effectively closed to argon diffusion; McDougal & Harrison 1988), define minimum Alpine temperatures for rocks exhibiting Cretaceous plateau ages. On the other hand, argon blocking temperatures (i.e. temperatures above which significant argon loss will occur in the time frame of ongoing diffusion; Lister & Baldwin 1996), must be taken into consideration when estimating maximum Alpine temperatures for rocks exhibiting Palaeozoic plateau and mixed ages.

To explain the development of the stepwise increase in the apparent age spectra for mixed ages, we have simulated the effects of possible P-T-t histories on the argon solid-state diffusion and ^sup 40^Ar/^sup 39^Ar apparent age spectra, using the MacArgon software and associated MacSpectrometer (Lister & Baldwin 1996). The diffusion domain parameters and the P-T-t history must be specified for the purposes of modelling. Because our argon diffusion modelling is based on the published dataset of argon age spectra and detailed sample description (grain size, mineral defects and compositional variations) is usually not available, the experimentally derived diffusion parameters of McDougal & Harrison (1988) had to be complemented by some simplifying assumptions, as follows: (1) the development of the stepped age spectra is related to partial argon loss from a single mineral phase of homogeneous chemical composition; (2) the diffusion is homogeneous, intra-granular and restricted to a 40 [mu]m wide cylindrical domain. The possible P-T- t history for the Vepor basement is determined by the following assumptions: (1) the Variscan crystallization of hornblende and micas (onset of their argon isotopie system) was followed by cooling; (2) despite the fact that the region exhibits a record of Permian postorogenic and Jurassic rifting (Plasienka 1995; Plasienka et al. 1997), the Permian-Early Cretaceous sedimentation in the area suggests c. 200 Ma of relative tectonic quiescence between the end of Variscan magmatism (c. 300Ma, Bibikova et al. 1990) and the onset of Cretaceous convergence (c. 100 Ma, Table 1); (3) the whole crustal column was uplifted during the declining stages of Alpine orogeny, as indicated by the Late Cretaceous apatite and zircon fission-track ages (Kral’ 1982; Koroknai et al. 2001).

Our diffusion modelling suggests a critical dependence of the ^sup 40^Ar/^sup 39^Ar spectral pattern on the ambient temperatures (and hence depths) attained after the Variscan event. For mixed ages, the best match between simulated and observed ^sup 40^Ar/^sup 39^Ar apparent age spectra occurs for P-T-t histories that involve high ambient temperatures during the 200 Ma of tectonic quiescence (compare the real and modelled hornblende spectra and the corresponding T-1 paths in Fig. 8).

Although stepped age spectra can also develop during a shorter period of residence in the argon partial retention zone (that portion of the stable crust where the temperatures are insufficient to completely reset the argon isotopie system; Baldwin & Lister 1998) (e.g. Permian and Jurassic thermal spikes related to rifting or the Cretaceous tectonometamorphic event), the temperature interval for which the stepped age spectra develop is then very small (c. 10 [degrees]C). In general, the greater the period of residence in the argon partial retention zone, the greater is the temperature interval (difference between closure and blocking temperature) for which the stepped age spectra develop. Consequently, the widespread occurrence of stepped age spectra across the Vepor basement argues for a long period of residence in the argon partial retention zone.

Blocking temperatures, calculated for the 200 Ma of quiescence, constrain the ambient temperatures above which the stepped age spectra develop and below which the Palaeozoic ages are preserved. Our phase equilibrium modelling suggests a temperature increase during the Alpine event. However, unless the Alpine overprint reached the argon closure temperatures, the original ^sup 40^Ar/ ^sup 39^Ar apparent age spectra did not experience significant modification. Thus, the occurrences of stepped age spectra suggest that the magnitude of the Alpine temperature increase did not exceed the difference between closure and blocking temperatures and the rocks remained at elevated temperatures for a relatively short period of time before the final Late Cretaceous cooling.

For a cooling rate of 5-50 [degrees]C Ma^sup -1^ across the closure temperature and 200 Ma of residence time in the argon partial retention zone, the argon closure and blocking temperatures are 490-531 and 415 [degrees]C for hornblende, 337-378 and 267 [degrees]C for muscovite, and 301-333 and 240[degrees]C for biotite (see Lister & Baldwin 1996). These closure and blocking temperatures have been used to approximate the ambient temperatures, attained by samples from the Vepor basement, during the Alpine event. The spatial distribution of the estimated ambient temperatures, together with our P-T estimates and microstructural variations across the Kral’ova Hol’a Complex (Jerabek et al. 2007), has been used to construct idealized Alpine isotherms across the Vepor basement (Fig. 9). Discussion

Polymetamorphic record within the Vepor basement

The distinct chemical composition and zoning of Variscan (M^sub 1^) and Alpine (M^sub 2^) garnets were used to interpret the P- revolution during these two metamorphic events. The question arises as to the extent to which the composition of garnet (I) was affected by Alpine metamorphism. The compositional profile across the garnet (I)-garnet (II) boundary in the Hron Complex paragneiss sample PP207 (Figs 3a and 5) exhibits a sharp compositional transition, without diffusion-related smoothing for all the elements except manganese, suggesting very limited internal diffusion at medium temperatures (Alpine temperatures of 540-600 [degrees]C). Nevertheless, the preservation of such a sharp compositional transition may also indicate relatively short exposure to the Alpine peak P-T conditions.

The Variscan garnet (I) core-to-rim compositional changes plotted on the calculated P-T sections (Figs 6 and 7) follow a garnet consumption trend (decrease in modal proportion) and indicate retrograde P- T evolution. The chemical composition of the garnet (I) core plots on the calculated P- T section far above the predicted garnet-in line (Figs 6 and 7). Such gaps in the P-T path can be explained by diffusional homogenization of the garnet (I) during Variscan peak metamorphism, which, however, has been subsequently affected by Variscan retrograde diffusional modification. The absence of more complete P- T paths for the Alpine garnets, plotting above the garnet-in line and exhibiting prograde growth zoning, can be explained using the results of our argon diffusion modelling. The modelling suggests that certain domains of the Vepor basement had already experienced temperatures near the closure and blocking temperatures for the argon isotopie system in hornblende (c. 490-531 and 415 [degrees]C) prior the onset of Cretaceous metamorphism.

P- T estimates derived from the phase equilibrium modelling range between 570-670 [degrees]C and 6-8.5 kbar for the Variscan and 430- 600 [degrees]C and 5-11 kbar for the Alpine metamorphism and show good correlation with the results of garnet-biotite thermometry. Our P- T estimates for Alpine metamorphism are consistent with the majority of published P-T data. However, for the Variscan metamorphism in the Hron Complex, they exhibit considerably higher values (see Table 2). Because our Variscan P- T estimates are similar to those from the neighbouring Tatra Unit (Krist et al. 1992; Janak et al. 1996), where only a lowgrade Alpine overprint occurs, this discrepancy can be explained by problems related to application of conventional thermobarometry in polymetamorphic domains.

The phase equilibrium modelling of the M^sub 1^ and M^sub 2^ mineral assemblages indicates systematic differences between the Variscan and Alpine metamorphism. The estimated P- T ranges for the retrograde Variscan and the prograde Alpine events exhibit contrasting metamorphic field gradients of 22-27 [degrees]C km^sup – 1^ and 15-18 [degrees]C km^sup -1^, respectively (Fig. 10). The relatively hot Variscan metamorphic field gradient is a transient feature typical of the late exhumation stages of the hot orogenic lower crust, commonly reported for Variscan rocks of the adjacent Bohemian Massif (e.g. O’Brien 2000; Tajcmanova et al. 2006; Kosulicova & Stipska 2007). In contrast, the Alpine metamorphic field gradient is considerably cooler and corresponds to geothermal gradients, commonly reported for the Alpine collisional belts (Hoinkes et al. 1999, and references therein).

The starting points of the Alpine P- T path vectors depicted in Figure 10 suggest that the studied samples were located at different crustal depths prior to burial, but shared a common field geotherm. The preserved P-T path segments exhibit nearcontemporaneous increases in pressure and temperature, indicating that the peak P- T conditions correspond to metamorphic conditions at maximum pressures (P^sub max^). Consequently, the metamorphic field gradient of 15-18 [degrees]C km^sup -1^ at peak-burial corresponds to an instantaneous field geotherm, which is slightly lower than a standard geothermal gradient (Ranalli 1995), implying minor crustal thickening and stretching of the isograds. The northernmost sample PP482 deviates from the field geotherm of 15-18 [degrees]C km^sup -1^ (Fig. 10) and exhibits a significantly higher initial temperature. This difference can be explained by a thermal anomaly related to the Jurassic rifting documented in the vicinity of the northern edge of the Vepor Unit (Plasienka 1995).

Interpretation of the burial P-T history

Phase equilibrium modelling allowed us to reconstruct segments of the Alpine prograde P- T path by using the garnet core-to-rim compositional changes. The fractionation of the bulk-rock composition, which could affect garnet growth zoning by depletion of some elements (Mn, Fe) from the matrix (Gaidies et al. 2006), can be neglected as the modal proportion of garnet in the studied samples is low (c. 1.5modal%). The garnet compositional changes indicate an increase in temperature and pressure of 20-60[degrees]C and 1-1.5 kbar, respectively (Fig. 10).

The prograde compositional zoning of the Alpine garnets is traditionally attributed to burial of the Vepor Unit, related to its southward underthrusting below the supracrustal Gemer Unit (e.g. Kovacik et al. 1996; Plasienka et al. 1999; Janak et al. 2001) during the north-south shortening of the region (Lexa et al. 2003). Nevertheless, the thickness of the overriding rock pile is unclear and is disputed, and depends on the interpretation of the status of the Gemer-Vepor Contact Zone. If this contact zone is considered to be part of the Vepor cover, then the Alpine P- T conditions of c. 550 [degrees]C and 8 kbar (Luptak et al. 2000; and this work) suggest a c. 28 km thick overburden (e.g. Plasienka et al. 1999; Luptak et al. 2000). However, if the Gemer-Vepor Contact Zone is considered to be part of the Vepor basement, which we prefer on the basis of its polymetamorphic record, then the overburden is much thinner (c. 15 km), being constrained by the metamorphic conditions of the Foederata Cover, indicating 380 [degrees]C and 4.5 kbar (Luptak et al. 2003).

The magnitudes and slopes of the P-T path vectors in Figure 10 should be interpreted with caution, as they were obtained from different lithologies. Nevertheless, real magnitudes of Alpine burial can be assessed for the pre-Alpine surface sediments (c. 350 [degrees]C and 4 kbar in the Foederata Cover; e.g. Luptak et al. 2003) and for the deepest parts of the Vepor basement. In the deep southern Hron Complex, the magnitude of Alpine burial displacement probably corresponds to the modelled P- T path vector, indicating a 60 [degrees]C and 1.5 kbar P- T increase (sample PP207). This is suggested by the presence of numerous hornblende stepped age spectra in this belt, which, according to our argon diffusion modelling, implies that the total temperature increase in the Hron Complex during the Alpine event must have been less than the difference between the argon closure and blocking temperature for hornblende (c. 75 [degrees]C). A downward decrease in the Alpine burial displacement (Fig. 11a) is also documented in the Kral’ova Hol’a Complex metagranitoids (Figs 7a and 10); however, the magnitude of the P-T path vectors in the Kral’ova Hol’a Complex metagranitoids is lower than that of the deeper Hron Complex (Fig. 11a).

The hypothesis of downward decrease in magnitude of Alpine burial implies that the internal part of the Vepor basement had to accommodate significant vertical shortening and horizontal stretching. This observation is consistent with the synmetamorphic (synburial) development of the subhorizontal mylonitic fabric S^sub A1^ in the Vepor basement. Consequently, the vertical shortening and horizontal east-west (orogen-parallel) ductile stretching of the Vepor basement may have been coeval with northward overthrusting of the Gemer Unit (Fig. 11b).

Distribution of the Alpine isograds and isotherms

Our interpretation of the Vepor basement burial history suggests that the Alpine metamorphic isograds developed subhorizontally and subparallel to the synmetamorphic deformation fabric S^sub A1^. This is supported by the structurally downwards increase in the P-T conditions and associated decrease in ^sup 40^Ar/^sup 39^Ar ages. If the Alpine metamorphic isograds originally developed subhorizontally, then the post peak-metamorphic tectonic events must have deformed the isograds to explain their current partially steep disposition.

Our structural field observations indicate that the Alpine synburial fabric S^sub A1^ is folded and heterogeneously reworked by a steep axial cleavage S^sub A2^, which chiefly developed within the NE-SW-trending belts of the Hron Complex. Moreover, the southern belt of the Hron Complex exposes rocks recording the highest Alpine metamorphic conditions (see also Vrana 1964; Plasienka et al. 1999; Janak et al. 2001). Janak et al. (2001) documented an internal metamorphic zonation within this belt, marked by east-west-trending metamorphic isograds, indicating a southward decrease in the metamorphic grade (see Fig. 9 for the location of the zones). Our P- T data suggest that a similar decrease in the metamorphic grade also occurs towards the north, where the P-T estimates from the southern Hron Complex and neighbouring central part of the Kral’ova Hol’a Complex exhibit a difference of up to 150[degrees]C and 5 kbar (Fig. 9). The spatial distribution of our Alpine P-T data is in good agreement with the regional variations of the existing ^sup 40^Ar/ ^sup 39^Ar ages (Fig. 9), pointing to the folding-related redistribution of the Alpine isograds and isotherms (Fig. 12). Significance of the orogen-parallel extension in the Vepor Unit for the Cretaceous evolution of the Central West Carpathians: alternative models

Based on the existing deep seismic transect 2T and available structural data, Tomek (1993) and Plasienka et al. (1997) proposed that burial of the Vepor Unit is associated with the crustal-scale thrust sheet stacking of the Central West Carpathians. According to this geometry, the Veporic sheet was surrounded by two strongly non- coaxial crustal-scale thrust zones (Plasienka 1993, 2003) and, as described here, experienced synconvergent orogen-parallel stretching. A similar tectonic scenario has been recently proposed by Indares et al. (2000), who described orogen-parallel stretching, perpendicular to thrusting, within the Theologically weaker units bounded by strong thrust sheets. This model, however, fails to explain the deformation compatibility problem, which can be eliminated only if the synconvergent orogen-parallel stretching is compensated by extrusion of weakly constrained lateral foreland (Seyferth & Henk 2004). Nevertheless, the evidence for laterally unconfined boundary conditions of the Cretaceous Central West Carpathians convergent system (see Tapponnier et al. 1982; Ratschbacher et al. 1991a) remains poorly defined (Frank & Schlager 2006). Therefore, the synburial lateral escape model for the Vepor basement must be considered with caution.

An alternative explanation of orogen-parallel stretching is associated with a model of synconvergent exhumation of deeply buried crustal rocks (e.g. Selverstone 1988; Ratschbacher et al. 1991a, b; Mancktelow & Pavlis 1994; Fugenschuh & Schmid 2005). This concept was favoured by Plasienka et al. (1999) and Janak et al. (2001), who suggested that the Vepor Unit evolved as a metamorphic core complex during Cretaceous growth of the Central West Carpathians orogenic wedge.

In contrast, we are of the opinion that the spatial variations in the Alpine metamorphic grade across the Vepor basement resulted from D^sub A2^ folding of the synburial deformation fabric S^sub A1^ (Fig. 12). Differential exhumation occurred mainly along two belts, where structurally deeper paragneisses-micaschists of the Hron Complex separate higher metagranitoids of the Kral’ova Hol’a Complex by forming large-scale F^sub A2^ anticlines (Fig. 2). A similar model of differential exhumation of lower crustal rocks within the hinge zones of crustal-scale anticlines has been proposed for the Namche Barwa region on the basis of field documentation and numerical modelling (Burg et al. 1997; Burg & Podladchikov 1999).

The previously described, low-angle extensional shearing, related to exhumation and tectonic unroofing of the Vepor Unit (Hok et al. 1993; Plasienka 1993; Plasienka et al. 1999; Janak et al. 2001), is restricted in our model to the contact between the Vepor and Gemer units. Similarly to the exhumational model for the Tauern Window of Mancktelow & Pavlis (1994), this low-angle detachment fault zone could operate synchronously with synconvergent large-scale folding.

On the other hand, because of the low resolution of existing geochronological data, the incomplete pattern of Alpine metamorphic isograds and the lack of knowledge of Cretaceous plate configuration in the Central West Carpathians, we cannot rule out the hypothesis of rapid switches between contraction and extension. This model may involve (1) north-south contraction during the Early Cretaceous Central West Carpathians thrust sheet stacking, followed by (2) east- west orogen-parallel crustal extension localized within the ductile Vepor Unit and resulting in the development of a detachment zone between the Vepor and overlying Gemer units and (3) Late Cretaceous north-south contraction resulting in the formation of upright F^sub A2^ folds. A tectonic switching model, thermally and Theologically justified by the calculations of Thompson et al. (2001), is currently being proposed in many regions (e.g. Collins 2002a, b; Schulmann et al. 2002). Based on modern and ancient examples, tectonic switching occurs when a slab retreat induces an upper plate extension, transiently followed by a flat subduction (or slab flip) and crustal thickening (Collins 2002a). If the tectonic switching model is applied to the Central West Carpathians region, then the driving force may be represented by the south-dipping lithospheric subduction operating during Cretaceous time (Plasienka 2003).

Further structural work and tectonic modelling may be required to distinguish between these models.

Conclusions

The polymetamorphic record in the Vepor basement related to the Variscan and Alpine regional metamorphic events has been characterized using phase equilibrium modelling. The Variscan metamorphism indicates P-T conditions of 570-670 [degrees]C and 6- 8.5 kbar, retrograde P-T evolution and a metamorphic field gradient of 22-27 [degrees]C km^sup -1^, corresponding to the late exhumation stages of the hot Variscan orogenic lower crust. In contrast, the Alpine metamorphism indicates P-T conditions of 430-600 [degrees]C and 5-11 kbar, prograde P-T evolution and a metamorphic field gradient of 15-18 [degrees]C km^sup -1^, characteristic of Alpine collisional belts.

The existing ^sup 40^Ar/^sup 39^Ar cooling ages from the Vepor basement have been re-evaluated in terms of thermal history and Cretaceous temperature overprint. The examination of existing argon age spectra, by means of the argon diffusion modelling, suggests that certain parts of the Vepor basement had already been located deep in the crust before the onset of the Cretaceous tectonometamorphic event.

During the Cretaceous, the Vepor Unit was thrust under the Palaeozoic metasediments of the Gemer Unit, resulting in apparently contemporaneous burial and orogen-parallel ductile spreading of the Vepor Unit.

Spatial variations in P-T estimates and Palaeozoic to Cretaceous ^sup 40^Ar/^sup 39^Ar ages indicate that the post-burial isotherms were deformed during subsequent large-scale folding. The folding is responsible for differential exhumation of the Vepor basement, where the structurally deeper metapelites separate structurally higher metagranitoids by forming large-scale anticlines.

This work was supported by research grants from the Czech Science Foundation (GACR 205/03/1490), the Charles University Science Foundation (GAUK 373/2004) and the MSM project No. 0021620855. Internal funding of CNRS UMR 7517 to O.L. is gratefully acknowledged. Two anonymous reviewers are thanked for their critical comments. M. Krabbendam is thanked for his careful editorial work. The valuable comments of J. Konopasek and V Janousek are appreciated.

References

BALDWIN, S.L. & LISTER, G.S. 1998. Thermochronology of the South Cyclades Shear Zone, los, Greece: Effects of ductile shear in the argon partial retention zone. Journal of Geophysical Research: Solid Earth, 103, 7315-7336.

BEZAK, V. 1991. Metamorphic conditions of the Veporic unit in the Western Carpathians. Geologica Carpathica, 42, 219-222.

BEZAK, V., JACKO, S., JANAK, M., LEDRU, P., PETRIK, I. & VOZAROVA, A. 1997. Main Hercynian litiiotectonic units of the Western Carpathians. In: GRECULA, P., HOVORKA, D. & PUTIS, M. (eds) Geological Evolution of the Western Carpathians. Mineralia Slovaca, Bratislava, 261-268.

BIBIKOVA, E.V., CAMBEL, B., KORIKOVSKY, S.P., BROSKA, I., GRACHEVA, T.V., MAKAROV, V.A. & ARAKEUANTS, M.M. 1988. U-Pb and K- Ar isotopie dating of Sinec (Rimavica) granites (Kohut zone of Veporides). Geologicky Zbornik Geologica Carpathica, 39, 147-157.

BIBIKOVA, E.V., KORIKOVSKY, S.P., PUNA, M., BROSKA, I., GOLTZMAN, Y.V. & ARAKELIANTS, M. 1990. U-Pb, Rb-Sr and K-Ar dating of Sihla tonalites of Vepor Pluton (Western Carpathian Mts.). Geologicky Zbornik Geologica Carpathica, 41, 427-436.

BIELY, A. 1964. Ueber die ‘Veporiden’. Geologicky Zbornik Geologica Carpathica, 15, 263-266.

BURG, J.P. & PODLADCHIKOV, Y. 1999. Lithospheric scale folding: numerical modelling and application to the Himalayan syntaxes. International Journal of Earth Sciences, 88, 190-200.

BURG, J.P., DAVY, P., NIEVERGELT, P., OBERLI, F., SEWARD, D., DIAO, Z.Z. & MEIER, M. 1997. Exhumation during crustal folding in the Namche-Barwa syntaxis. Terra Nova, 9, 53-56.

COLLINS, W.J. 2002a. hot orogens, tectonic switching, and creation of continental crust. Geology, 30, 535-538.

COLLINS, WJ. 2002b. Nature of extensional accretionary orogens. Tectonics, 21, 1024.

CONNOLLY, J.A.D. 2005. Computation of phase equilibria by linear programming: A tool for geodynamic modeling and its application to subduction zone decarbonation. Earth and Planetary Science Letters, 236, 524-541.

CONNOLLY, J.A.D. & PETRINI, K 2002. An automated strategy for calculation of phase diagram sections and retrieval of rock properties as a function of physical conditions (0.4 Mb). Journal of Metamorphic Geology, 20, 697-708.

DALLMEYER, R.D., NEUBAUER, F., HANDLER, R., FRITZ, H., MUELLER, W., PANA, D. & PUTIS, M. 1996. Tectonothermal evolution of the internal Alps and Carpathians; evidence from ^sup 40^Ar/^sup 39^Ar mineral and whole-rock data. In: SCHMID, S.M., FREY, M., FROITZHEIM, N., HEILBRONNER, R. & STUENITZ, H. (eds) Alpine Geology; Proceedings of the second Workshop. Eclogue Geologicae Helvetiae. Birkhaeuser, Basel, 203-227.

ENGLAND, P.C. & THOMPSON, A.B. 1984. Pressure temperature time paths of regional metamorphism 1. Heat-transfer during the evolution of regions of thickened continental crust. Journal of Petrology, 25, 894-928. FARYAD, S.W. 1991. Pre-Alpine metamorphic events in Gemericum. Mineralia Slovaca, 23, 395-402.

FARYAD, S.W. 1995. Phase petrology and P-T conditions of mafic blueschists from the Meliate unit, West Carpathians, Slovakia. Journal of Metamorphic Geology, 13, 701-714.

FARYAD, S.W. 1997. Metamorphic petrology of the Early Paleozoic low-grade rocks in the Gemericum. In: GRECULA, P., HOVORKA, D. & PUTIS, M. (eds) Geological Evolution of the Western Carpathians. Mineralia Slovaca, Bratislava, 309-314.

FARYAD, S.W. & HENJES-KUNST, F. 1997. Petrological and K-Ar and Ar-40-Ar-39 age constraints for the tectonothermal evolution of the high-pressure Meliate unit, Western Carpathians (Slovakia). Tectonophysics, 280, 141-156.

FRANK, W. & SCHLAGER, W. 2006. Jurassic strike slip versus subduction in the Eastern Alps. International Journal of Earth Sciences, 95, 431-450.

FUGENSCHUH, B. & SCHMID, S.M. 2005. Age and significance of core complex formation in a very curved orogen: Evidence from fission track studies in the South Carpathians (Romania). Tectonophysics, 404, 33-53.

GAIDIES, F., ABART, R., DE CAPITANI, C, SCHUSTER, R., CONNOLLY, J.A.D. & REUSSER, E. 2006. Characterization of polymetamorpmsm in the Austroalpine basement east of the Tauern Window using garnet isopleth thermobarometry. Journal of Metamorphic Geology, 24, 451- 475.

HOINKES, G., KOLLER, F., RANTITSCH, G., DACHS, E., HOCK, V., NEUBAUER, F. & SCHUSTER, R. 1999. Alpine metamorphism of the Eastern Alps. Schweizerische Mineralogische und Petrographische Mitteilungen, 79, 155-181.

HOLLAND, T. & POWELL, R. 1998. An internally consistent thermodynamic data set for phases of petrological interest. Journal of Metamorphic Geology, 16, 309-343.

HOK, J., KOVAC, P. & MADARAS, J. 1993. Extensional tectonics of the western part of the contact area between Veporicum and Gemericum. Mineralia Slovaca, 25, 172-176 [in Slovak].

HRASKO, L., KOTOV, A.B., SALNIKOVA, E.B. & KOVACH, V.P. 1998. Enclaves in the Rochovce granite intrusion as indicators of the temperature and origin of the magma. Geologica Carpathica, 49, 125- 138.

HRASKO, L., HATAR, J., HUHMA, H., MANTARI, I. & MICHALKO, J. 1999. U/Pb zircon dating of the Upper Cretaceous granite (Rochovce type) in the Western Carpamians. Krystalinikum, 25, 163-171.

JACKO, S., SASVARI, T., ZACHAROV, M., SCHMIDT, R. & VOZAR, J. 1996. Contrasting styles of Alpine deformations at the eastern part of the Veporicum and Gemericum units, Western Carpathians. Slovak Geological Magazine, 2, 151-164.

JANAK, M. 1994. Variscan uplift of the crystalline basement, Tatra Mts., Central Western Carpathians: evidence from ^sup 40^Ar/ ^sup 39^Ar laser probe dating of biotite and P-T-t paths. Geologica Carpathica, 45, 293-300.

JANAK, M., O’BRIEN, PJ., HURAI, V. & REUTEL, C. 1996. Metamorphic evolution and fluid composition of garnet-clinopyroxene amphibolites from the Tatra Mountains, Western Carpathians. Lithos, 39, 57-79.

JANAK, M., PLASIENKA, D., FREY, M., COSCA, M., SCHMIDT, SX, LUPTAK, B. & MERES, S. 2001. Cretaceous evolution of a metamorphic core complex, the Veporic unit, Western Carpathians (Slovakia): P-T conditions and in situ Ar-40/Ar-39 UV laser probe dating of metapelites. Journal of Metamorphic Geology, 19, 197-216.

JANAK, M., MERES, S. & IVAN, P. 2007. Petrology and metamorphic P- T conditions of eclogites from the northern Veporic Unit (Western Carpathians, Slovakia). Geologica Carpathica, 58, 121-131.

JERABEK, P., STUNITZ, H., HEILBRUNNER, R., LEXA, O. & SCHULMANN, K. 2007. Microstructural -deformation record of an orogen-parallel extension in the Vepor Unit, West Carpathians. Journal of Structural Geology, 29, 1722-1743.

KLEEMANN, U. & REINHARDT, J. 1994. Garnet-biotite thermometry revisited; the effect of Al (super VI) and Ti in biotite. European Journal of Mineralogy, 6, 925-941.

KLINEC, A. 1966. On the structure and evolution of the Veporic crystalline unit. Zbornik Geologickych Vied, 6, 7-28 [in Slovak].

KLINEC, A. & PLANDEROVA, E. 1981. A question of stratigraphic homogeneity of the Hladomorna Valley formation. Geologicke Prace, Spravy, 75, 13-18 [in Slovak].

KORIKOVSKY, S.P., DUPEJ, J., BORONIKHIN, V.A. & ZINOVIEVA, N.G. 1990. Zoned garnets and their equilibria in mica schists and gneisses of Kohut crystalline complex, Hnust’a region, Western Carpathians. Geologica Carpathica, 41, 99-124.

KOROKNAI, B., HORVATH, P., BALOGH, K. & DUNKL, I. 2001. Alpine metamorphic evolution and cooling history of the Veporic basement in northern Hungary: new petrological and geochronological constraints. International Journal of Earth Sciences, 90, 740-751.

KOSULICOVA, M. & STIPSKA, P. 2007. Variations in the transient prograde geothermal gradient from chluritoid -staurolite equilibria: a case study from the Barrovian and Buchan-type domains in the Bohemian Massif. Journal of Metamorphic Geology, 25, 19-35.

KOTOV, A.B., MIKO, O. & PUTIS, M. ET AL. 1996. U/Pb dating of zircons of postorogenic acid metavolcanics and metasubvolcanics: A record of Permian-Triassic taphrogeny of the West-Carpathian basement. Geologica Carpathica, 47, 73-79.

KOVACIK, M. 1993. Polyphase evolution of Lower Paleozoic metamorphites in the middle part of die Kohut zone,Veporic unit, West Carpathians. Mineralia Slovaca, 25, 379-385.

KOVACIK, M. 1996. Kyanite-magnesian chlorite schist and its petrogenetic significance (the Sinec massif, southern Veporic unit, Western Carpathians). Geologica Carpathica, 47, 245-255.

KOVACIK, M., KRAL’, J. & MALUSKI, H. 1996. Metamorphic rocks in the Southern Veporicum basement: their Alpine metamorphism and thermochronologic evolution. Mi