Permian A-type granites of the Western Carpathians and Transdanubian regions: products of the Pangea supercontinent breakup

Permian biotite leucogranites to granite porphyries and rhyolites form small intrusions in several Alpine tectonic units in the Western Carpathians and the Pannonian region (Slovakia and Hungary). Their A-type signature is inferred from main- and trace-element geochemistry, with high K, Rb, Y, REE, Zr, Th, Nb, Fe/Mg and Ga/Al, low Al, Mg, Ca, P, Sr, V and strong negative Eu-anomaly. This geochemical signature is further supported by the mineralogy comprising local hypersolvus alkali feldspars, annitic biotite and the presence and composition of HFSE accessory minerals. The δ18O values measured for zircon (mean value 8.3 ‰ ± 0.36) may be explained by the melting of igneous material of crustal origin and/or mantle basalts which interacted with low-temperature fluids. The in-situ SHRIMP U–Pb isotope dating of zircon from the granites highlights two different periods of magmatic crystallisation and pluton emplacement: the older 281 ± 3 Ma Cisuralian age in the southern part, Velence Hills in the Pannonian region (Transdanubian Unit) and younger Guadalupian ages in the northern part, the West-Carpathian area: 262 ± 4 Ma (Turčok, Gemeric Unit), 267 ± 2 Ma (Hrončok, Veporic Unit) and 264 ± 3 Ma (Upohlav, granitic pebbles in Cretaceous conglomerates of the Pieniny Klippen Belt). The ~ 280 to 260-Ma interval is simultaneous with post-orogenic or anorogenic, rift-related and mainly alkaline (A-type) magmatism on the broader European scale. Our study documents a close relationship between the Permian continental rifting and the Neotethyan Meliatic oceanic basin opening in the Middle Triassic. The A-type granites originated from the partial melting of the ancient lower crustal quartzo-feldspatic rocks with the possible contribution of meta-basic material from the mantle in an extensional tectonic regime consistent with disintegration of the Pangea supercontinent during the Permian–Triassic period.


Introduction
The A-type granites were originally distinguished by Loiselle and Wones (1979) as a specific group of granitic rocks with peculiar chemistry and geotectonic setting (e.g., Bonin 2007Bonin , 2008. The A-type granites' origin is generally connected with an extensional regime in the lithosphere (Collins et al. 1982;Whalen et al. 1987;Bonin 2007) and it is related to the geodynamic settings, which are consistent with both crustal and mantle sources (Bonin 2004;Shellnutt and Zhou 2007;Grebennikov 2014;Lu et al. 2020, and references therein). The A-type granites are subdivided into two groups on the basis of trace element abundances, particularly Y/Nb ratio (Eby 1992). The A 1 group with Y/Nb < 1.2 includes felsic rocks chemically similar to those observed in oceanic islands and continental rifts (ocean island basalts, OIB source). The second A 2 group with Y/Nb > 1.2 is proposed to form by several different mechanisms from an island arc or continental margin basalt to the partially melted continental crust sources (Eby 1992). The occurrences of A-type granites can also indicate collided plate suture zones (Balen et al. 2020).
The A-type granites are distinguishable from the S-, Iand M-type genetic groups by major and trace element data. These data include elevated high field strength elements (HFSE: especially Zr, Nb, Ta), REE (except Eu) and F contents, and high FeO Tot /MgO and Ga/Al ratios and low CaO and trace elements compatible with mafic silicate minerals (Co, Sc, Cr, Ni) or feldspars-compatible large-ion lithophile elements (LILE: Ba, Sr) + Eu 2+ (Loiselle and Wones 1979;Collins et al. 1982;Whalen et al. 1987;Eby 1990;Bonin 2007;Whalen and Hildebrand 2019;Bonin et al. 2020).
The A-type granites can also be identified by the specific textural and compositional features of rock-forming and accessory minerals, including hypersolvus alkali feldspars, Fe-rich mafic silicates (annite-dominant biotite, locally also alkali amphiboles, pyroxenes and fayalite), crystal morphology and high Zr/Hf ratio of zircon and in some cases also by fluorite, topaz, gadolinite, REE-Nb-Ta oxides and other exotic Zr, Ti, Nb and REE minerals (Pupin 1992;Uher and Broska 1996;Bonin et al. 1998Bonin et al. , 2020Bonin 2007;Uher et al. 2009;Breiter et al. 2014 among others).
Small intrusions of post-Variscan granites and rhyolites with documented A-type affinity have been described within various tectonic units incorporated in Alpine edifice of the Western Carpathian mountain belt in Slovakia and the adjacent Transdanubian Central Range in Hungary. Moreover, analogous granite to rhyolite pebbles to boulders commonly occur as clasts in Cretaceous to Palaeogene flysch conglomerate beds of the Pieniny Klippen Belt. Their petrography, rock-forming and accessory minerals as well as basic geochemical characteristics were investigated by many authors (Buda 1985;Uher and Gregor 1992;Broska et al. 1993;Uher and Marschalko 1993;Uher et al. , 2002aUher et al. , 2009Uher et al. , 2015Uher et al. , 2018aBroska 1994, 1996;Petrík et al. 1994Petrík et al. , 1995Buda and Nagy 1995;Broska and Uher 2001;Gyalog and Horváth 2004;Demko and Hraško 2013;Ondrejka et al. 2015, b;Sobocký et al. 2020). Monazite chemical and multi-zircon isotope dating revealed their Permian to Triassic age (Cambel et al. 1977;Uher and Pushkarev 1994;Kotov et al. 1996;Putiš et al. 2000Putiš et al. , 2016Putiš et al. , 2019aFinger et al. 2003;Lelkes-Felvári and Klötzli 2004;Radvanec et al. 2009;Vozárová et al. 2009Vozárová et al. , 2012Vozárová et al. , 2016Demko and Hraško 2013;Sobocký et al. 2020;Szemerédi et al. 2020a). In contrast, Neogene mantle-derived A 1 -type granite and syenite xenoliths containing high concentrations of HFSE occur in alkali basalts of northern Pannonian Basin (Slovakia) erupted during Pliocene-Pleistocene extension in Carpathian back-arc (Huraiová et al. 2017(Huraiová et al. , 2019. However, a complex characterization of this magmatic event using recent petro-chronological data (geochemistry, isotopes and dating) as a base for understanding of their geotectonic role in a broader post-Variscan Alpine-Carpathian domain has been still missing. Consequently, herein we determine major and trace element geochemistry including isotopic data, zircon SHRIMP U-Pb ages and zircon oxygen isotopes of the granites with A-type affinity from the Western Carpathians (Turčok, Hrončok and Upohlav granite) and adjacent Pannonian area (Velence granite) with comparison to related acid volcanic rocks (Fig. 1). Our results contribute to constraining the petrogenetic history of these granitic rocks in the hosting tectonic units during Pangea supercontinent disintegration and Permian post-Variscan/early Alpine geological evolution of Europe.

Geological setting and review of petrographical and mineralogical data
Several small intrusions of post-Variscan anorogenic biotite leucogranites to granite porphyries with A-type affinity are also known in the Slovak Western Carpathians and the adjacent Transdanubian Central Range in Hungary. These granites are represented by the Turčok, Hrončok and Velence intrusive bodies and the widespread granite pebbles  Uher and Broska, 1996;Broska and Uher 2001) to boulders occur in the Upohlav-type Cretaceous to Palaeogene conglomerates of the Pieniny Klippen Belt (e.g., Uher and Broska 1996;Plašienka et al. 1997). The volcanic rocks of rhyolitic composition and A-type affinity are also widespread in the Western Carpathians (especially in the Veporic, Silicic and Gemeric tectonic units) and the Transdanubian Central Range Unit. The age of the rhyolites in the Silicic Unit were considered as Lower Triassic (e.g., Uher et al. 2002b;Ondrejka et al. 2015). However, recent in situ zircon U-Pb dating revealed Permian age of these felsic volcanic rocks (Lelkes-Felvári and Klötzli 2004;Vozárová et al. 2009Vozárová et al. , 2012Vozárová et al. , 2016Putiš et al. 2016;Pelech et al. 2017;Ondrejka et al. 2018b;Szemerédi et al. 2020a).

Analytical methods and sample locations
The multi-element lithogeochemistry of pulp samples has been performed by Bureau Veritas (AcmeLabs) in Vancouver, Canada, by X-ray fluorescence (XRF) for major elements, and the trace and rare earth elements (REE) were determined by inductively coupled plasma atomic emission spectrometry (ICP-AES) and inductively coupled plasma mass spectrometry (ICP-MS). Some older analyses have been performed by University of Ottawa, Ottawa, Memorial University of Newfoundland, St. John's, Canada and IGEM Moscow, Russia. For further details see Petrík et al. (1995), Putiš et al. (2000), Broska and Uher (2001), Uher et al. (2002aUher et al. ( , b, 2009) and Broska et al. (2004). All wholerock geochemical plots were performed by the R package GCDkit procedure (Janoušek et al. 2016). Zircon crystals were extracted from the rocks (Turčok, TU-3; Hrončok, HK-1; Upohlav, BP-1 and Velence VE-1 samples) by standard density and electromagnetic separation routine. The zircon crystals were mounted in epoxy, polished to expose the crystal interiors for analysis and imaged by cathodoluminescence (CL) and back-scattered electrons (BSE) to reveal their internal structure for analytical spot positioning. The highest quality zircon crystals in the studied samples were selected for measurement to avoid fractures, impurities and mineral inclusions. In situ U-Pb analysis was performed by SIMS SHRIMP-II apparatus at the Center of Isotopic Research (CIR) at the A.P. Karpinsky Russian Geological Research Institute (VSEGEI), St-Petersburg, Russia.
The results were acquired with a secondary electron multiplier in peak-jumping mode, following the standard procedure of Williams (1998) and Larionov et al. (2004). A primary O 2 − beam with 2 to 3nA ion current produced an approximately 25 × 20 μm elliptic analytical crater. Typical mass-resolution at 254 AMU ( 238 UO) was M/ΔM > 5000 (1% valley) and this enabled the resolution of isobaric interference. One-minute rastering over an approximately 65 × 50 μm rectangular area was then employed before each analysis to remove the gold coating and any surface Pb contamination.
The following ion species were measured in sequence: 196 (Zr 2 O)-204 Pb-background (204)(205)(206) Pb-238 U-248 ThO-254 UO) with integration times ranging from 2 to 30 s. Four cycles for each analysed spot were acquired, and each fourth measurement was made on the TEMORA zircon standard (Black et al. 2003) or 91500 as a secondary reference (Wiedenbeck et al. 1995). During the analytical session, 31 spots of TEMORA and 32 spots of 91500 as a concentration standard have been measured. The TEMORA zircons yielded a weighted mean of standard Pb/U calibration 0.01320, 1σ error of mean ± 0.71%, 1σ external spot-to-spot error 2.24%, MSWD 13.53 (with 204 Pb common lead correction).
The raw data were processed by the SQUID v1.13a software (Ludwig 2005a) and the ISOPLOT/Ex 3.22 (Ludwig 2005b) software with decay constants of Steiger and Jäger (1977) and common lead was corrected using measured 204 Pb/ 206 Pb and model values as in Stacey and Kramers (1975), and sample ages of the complex multi-stage evolution were processed by the ISOPLOT "Unmix Ages" tool to distinguish the main age groups.
Concordia diagrams show that almost all measured spots have concordant ages. Discordant results of multiple analyses from the same crystal were then employed to construct Discordia lines. An average of 10 zircon crystals for each rock sample were analysed and the resultant ages with 2σ error are shown in Figs. 8,9,10,11,12. Analysis of oxygen stable isotope composition of zircon was carried out at the Institute of Earth Sciences, Geology and Paleontology, K.F. University of Graz, Austria. For each sample, up to 1.5 to 2 mg of handpicked selected zircon crystals were heated with a 20 W CO 2 laser following the technique of Sharp (1990). Oxygen was extracted from silicate minerals by fluorination with BrF 5 and was measured directly on a Finnigan MAT Delta Plus mass spectrometer without combustion to CO 2 . Throughout the measurements 1-2 mg of material of several standards were analysed together with the samples. The reproducibility of the measurements was monitored using the garnet standard UWG-2 (Valley et al. 1995; mean value of 5.74 ‰, standard deviation of 0.15 ‰) for which a mean value of 5.8 ‰ and a standard deviation of 0.15 ‰ were obtained. Measurements on NBS 30 biotite gave an average value of 5.04 ‰ and a standard deviation of 0.2 ‰ (accepted value 5.1 ‰, standard deviation 0.2 ‰). The data are given on the VSMOW scale.
Locations of granite samples investigated by U-Pb SHRIMP and O isotope methods of zircon are given in Table 2.

Zircon characterization
Primary magmatic zircon belongs to the most common accessory minerals in the investigated A-type granites. The zircon crystals are usually transparent and most  are ~ 50-300 μm in length. Euhedral zircon crystals have predominantly P 4 -P 5 and D (sub)type morphology according to the typology classification (Pupin 1980). While the investigated zircon crystals' internal texture comprises magmatic fine oscillatory and/or sector zoning, irregular and most likely late-magmatic to subsolidus marginal domains, are present to a lesser extent. Some zircon crystals also contain small round inherited xenocrystic cores (Fig. 8).
The Hrončok granite zircon (HK-1) revealed 200-1100 ppm U and Th = 60-390 ppm Th, resulting in the relatively wide Th/U ratio between 0.15 and 0.90. Their U-Pb isotope analyses are concordant within the analytical error, yielding a Concordia age of 267 ± 2 Ma (MSWD of concordance = 0.70) (Fig. 10). Here, the HK-6-1 spot yielded a clearly older age of 632 ± 8 Ma (1σ) than most of the population, and the CL image highlighted a clear small resorbed core, thus indicating potential inheritance. The remaining nine zircons are mostly euhedral crystals with well-developed concentric or convolute irregular zoning.

Oxygen stable isotopes in zircon
A total of nine handpicked concentrates of transparent zircon crystals were analysed for δ 18 O values, the results being listed in Table 5 and plotted against SiO 2 in Fig. 13. One sample from the Turčok granite shows a value of 8.3 ‰, one sample from the Hrončok shows 8.1 ‰, four samples from Upohlav range between 7.5 and 8.5 ‰ and three samples from the Velence range between 8.0 and 8.5 ‰.

Geochemical and mineralogical characteristics
The investigated granites reveal geochemical and mineralogical characteristics which clearly reflect their A-type affinity based on those previously published (e.g., Broska 1994, 1996;Broska and Uher 2001) and our recent data as summarized in the following items: (i) Leucocratic and dominantly peraluminous, but also metaluminous, and even peralkaline character with high  Fig. 2 SiO 2 , K 2 O, Fe > Mg, K > Na and low TiO 2 , MgO, CaO and P 2 O 5 bulk-rock content. The very low phosphorus content is comparable with other similar granites worldwide (cf. Whalen et al. 1987), and it is the typical geochemical feature of A-type granites, reflected in the low amount of fluorapatite in these Ca, P-poor rocks. Low P content could be explained by the formation from melt-depleted lower crust after the main granite production during Variscan orogeny (Broska and Uher 2001). The R1-R2 classification diagram (de La Roche et al. 1980) indicates the transition of granite to alkali granite (Fig. 2). These are more alkali-calcic than alkalic (Fig. 3a) and more ferroan than magnesian (Fig. 3b) (according to Frost et al. 2001;Frost and Frost 2011). The relatively strong peraluminous character of some A-type rhyolites contrasts with the moderate peraluminous to metaluminous ± peralkaline granites (Fig. 4). They also have relatively higher Rb, Nb, Zr, Th, F, Ga/Al and occasional W, low Sr, Ba and V. However, two samples of Turčok granite show Rb/Sr ratio close to zero (Fig. 5b), because of the Rb loss during albitization connected with a late metamorphic overprint of the rock Kaur et al. 2012). All granite samples have high to moderate REE + Y content, seagull-shaped REE patterns with enriched LREE (La N /Sm N > 1), flat HREE (Gd N /Yb N ~ 1) and pronounced negative Eu anomaly (Fig. 7) indicating the feldspar fractionation.The assessment of trace Ga content compared to various major and trace element parameters provides strong indication of the A-type characteristics (Fig. 6). This is typical of the anorogenic origin of the investigated granites and rhyolites. The studied A-type granites are part of the A 2 group (Fig. 5), which is mainly derived from the continental crust (Eby 1992). This is confirmed by their low Nb/Ta ratio (av. ~ 12.2) consistent with the average composition of the continental crust (~ 11.4, Rudnick and Gao 2003). The empirical classification proposed by Bonin et al. (2020) places them in the KCG group (K-rich calc-alkaline granitoids) as the closest match. The major and trace element bulk rock composition of the granites was plotted with Permian rhyolites of the Silicic Unit because these have similar geochemistry and zircon ages, thus indicating a common geotectonic setting (Uher et al. 2002b;Ondrejka et al. 2018b). Nevertheless, some geochemical signatures, e.g. scattered A/CNK values: exceptionally high K in contrast to very low Ca and Na contents, indicate an obvious contribution of the hydrothermal processes to their geochemical evolution (Uher et al. 2002b); (ii) Biotite chemistry (annite > siderophyllite) with high Fe/ Mg ratio, significant F and Cl contents (max. 0.7 wt% F and 0.5 wt% Cl) and relatively low Al content (Uher and Broska 1996); (iii) Primary magmatic zircon with a high Zr/Hf ratio (generally over 50) and zircon typology with high I.A. parameter ~ 700 (Uher and Marschalko 1993;Broska 1994, 1996). Both these features indicate a high-alkaline crystallisation environment typical for alkaline (Pupin 1980(Pupin , 1992 and A-type granites (Breiter et al. 2014); (iv) relatively high Zr-saturation temperature (according to Boehnke et al. 2013): 780-920 °C for hypersolvus and 700-770 °C for subsolvus granites. These indicate crystallization from high-to mild-temperature   Barrat et al. 2012. Same symbols as in Fig. 2 and dry magma. However, leucocrate differentiates of the Hrončok subsolvus granite reveal low Zr-saturation temperatures (565-735 °C). A rough correlation between the saturation temperature and the Eu anomaly and SiO 2 is observed. Consistently, the temperature decreases as the Eu anomaly becomes more pronounced and the SiO 2 content increases, giving confidence to the values obtained using the Zr thermometer; (v) Presence of specific REE, Th and REE-Nb-Ta accessory minerals, such as allanite-(Ce), thorianite, gadolinite-hingganite-(Y), fergusonite-(Y) and aeschynite/ polycrase-(Y), which are characteristic for alkaline-rich      (Valley et al. 1998). Same symbols as in Fig. 2 magmatic suites (especially in the Turčok granite; Uher et al. 2009).
All (i-iv) geochemical and mineralogical features highlight the specific A-type characteristics of the studied granites (Uher and Broska 1996;Broska and Uher 2001;this study), which are typical in hot and dry post-orogenic to anorogenic granitic suites (cf. Whalen et al. 1987;Eby 1990;Frost and Frost 1997;Frost et al. 2001).

Isotopic composition
The isotopic compositions with whole rocks (WR) low to moderate 87 Sr/ 86 Sr (i) = 0.705-0.709 and moderate εNd (i) = -3.1 to + 1.9 indicate a variegated lower crustal meta-igneous protolith (Kohút et al. 1999;Kohút and Nabelek 2008; Table 1 in Magna et al. 2010). Positive zircon εHf (i) = + 0.2 to + 9.9 values are suggesting again a lower crustal meta-igneous protolith influenced by fluids from the metasomatized mantle (Kohút 2014, and Kohút unpublished data). However, lithium and sulphur whole-rock isotope signatures are not unambiguous. Mostly heavy Li isotope signatures with δ 7 Li = + 5.05 to + 6.67 ‰ of the Hrončok granite and whole-rock δ 34 S = − 0.69 ‰ of the Turčok granite probably reflect the derivation of these rocks from a mantle wedge modified by slab-derived fluids, whereas the modest value δ 7 Li = + 1.5 (Turčok) indicates rather a crustal source here (Kohút and Recio 2002;Magna et al. 2010). The isotope U-Sr-Nd-Hf-O signatures suggest the similar production from different crustal sources with varying contribution of mantle-derived basic materials to the post-collisional high-K calc-alkaline granitoid generation related to Arabian-Nubian Shield (Litvinovsky et al. 2021).
Zircon oxygen isotopic compositions show a low variation with a mean δ 18 O value of 8.3 ± 0.36 ‰. The values are markedly higher (2.5-3 ‰) than those of zircon from a primitive mantle-derived mafic magma (Valley 2003). For example, mantle zircon from the ultramafic rocks of Kimberley show a mean value of 5.3 ± 0.3‰ (Valley et al. 1998). For zircon, the range of measured isotopic compositions indicates the melting of igneous rocks characteristic, for example, for the lower crust (Taylor and Sheppard, 1986). The whole-rock δ 18 O values from Western Carpathian A-type granites 7.8-8‰ (Table 1 in Magna et al. 2010) are in good agreement with our zircon isotopic composition. The Δ(Gt-Zc) (representing the difference δ 18 O granite -δ 18 O zircon ) is close to 0 ‰, suggesting low or no crustal contamination of parental melt during crystallisation (Valley, 2003). In the present case, a biplot of wt% SiO 2 of granite versus δ 18 O zircon shows a low variation for the SiO 2 and δ 18 O values (Fig. 13). The δ 18 O values in zircon may be explained by the melting of igneous material of crustal origin and/or mantle basalts which interacted with low-temperature fluids (Gregory and Taylor, 1981). Both wt% SiO 2 and δ 18 O values are higher than, for example, those defined for the Neogene volcanic magmatism in the ALCAPA area (Seghedi et al. 2007), suggesting crustal and/or hydrothermal altered oceanic slab character of the melted material.

Age of emplacement, Permian magmatic activity and origin
The Permian magmatic crystallisation age interval of ~ 280-260 Ma presented herein is the first reported insitu isotopic age of A-type granitic rocks in the Western Carpathians and Pannonian region determined by U-Pb SHRIMP zircon dating. Previously obtained conventional multigrain U-Pb zircon geochronological results (Cambel et al. 1977;Uher and Pushkarev 1994;Putiš et al. 2000) indicated a very broad interval of Permian to Triassic ages (~ 285 to 240 Ma) for the Hrončok and Upohlav granites ( Table 6). The measurements of Hrončok gave discordant age with lower intercept of 238.6 ± 1.4 Ma and upper intercept of 1096 ± 44 Ma (Putiš et al. 2000). However, several examples show that the U-Pb zircon age calculated for the lower Discordia intersection for the multi-grain measurements of zircons is usually younger than the single-grain age of the same concordant zircons (Steiger et al 1993;Salnikova et al 1998). We attribute this rejuvenation to discrete Pb loss or U addition, particularly in cracked and partly damaged non-abraded zircon fraction. On the other hand, Triassic magmatic event (245 ± 3.3 Ma; U-Pb SIMS), most likely connected with the pre-oceanic advanced early Middle Triassic continental rifting, has been registered in zircon from calc-alkaline basalt intercalations in the Dobšiná accretionary wedge mélange (Meliatic Unit) in the southern part of the Inner Western Carpathians (Putiš et al. 2019b). Moreover, Lower Triassic ages of granite magmatism in Bulgaria (Peytcheva et al. 2005;Bonev et al. 2019) suggests the continuous Permian-Triassic magmatic activity.
There are also analogous, ~ 280-250 Ma ages documenting widespread magmatic activity in the broader Alpine-Carpathian area. Examples include the ~ 265 Ma diorite to granite porphyry of the Highiş Igneous Complex, Apuseni Mountains (Romanian Carpathians) which represent a halogen-rich A-type granite suite (Pană et al. 2002;Bonin and Tatu 2016). Permian leucogranites and associated spodumene-bearing granitic pegmatites from the Austroalpine Unit, Eastern Alps (Austria) yield the U-Pb and Sm-Nd ages of magmatic garnet in the interval of 245 to 280 Ma (Knoll et al. 2018;Putiš et al. 2019a). The LA-ICP-MS U-Pb dating of zircon from porphyric metagranites (Grobgneiss) of the Lower Austroalpine units (Austria) gave a narrow interval of 272.2 ± 1.2 to 267.6 ± 2.9 Ma; they represent high-K calc-alkaline to shoshonitic suite with S-type affinity, associated with gabbroic bodies (Yuan et al. 2020). High-K calc-alkaline lamprophyres of the Argentera-Mercantour Massif, Western Alps (Filippi et al. 2020) and the post-orogenic extensional basaltic andesites to rhyolites and granites of Athesian Volcanic Group (Southern Alps, Italy) have single-zircon U-Pb ages between ~ 290 and 275 Ma (Marocchi et al. 2008;Morelli et al. 2012). In addition, Permian granites of the Aya pluton and lamprophyre dykes in the Pyrenees were emplaced at ~ 270 Ma (Denèle et al. 2012).
The rare presence of 650-630 Ma inherited zircon cores in some samples indicates the admixture of Neoproterozoic material in the investigated A-type granite protolith. This systematic presence of older material is also supported by the sporadic occurrence of 670-640-Ma-old monazite-(Ce) domains in the Velence granite dated by EPMA in-situ chemical dating method (Sobocký et al. 2020).
Petrogenetic models for A-type granites commonly invoke igneous source (e.g., Collins et al. 1982;Creaser et al. 1991;Frost and Frost 1997) and peraluminous granites exhibiting A-type characteristics have been documented elsewhere (e.g. Huang et al. 2011;Sun et al. 2011;Dahlquist et al. 2014;Morales Cámera et al. 2018;Gao et al. 2020). These granites are usually derived from an infracrustal felsic source (King et al. 1997) alternatively with a dominant metasedimentary component rich in feldspars (e.g. Huang et al. 2011;Dahlquist et al. 2014). The optimal melting conditions are low pressure and high temperature, which can be created by the extensional setting (Frost and Frost 2011;Gao et al. 2020). In our case, partial melting of an Early Variscan, or perhaps of the Pan-African lower crustal meta-igneous rocks, possibly with a small contribution of meta-basic mantle-derived material provides a likely origin for our investigated A-type granite intrusions. The possible involvement of a more basic lower crustal to upper mantle protolith is also suggested by an occurrence of rare mafic enclaves of biotite tonalite composition at Velence (Uher and Broska 1996). Moreover, these global tectono-magmatic events are generally considered to have been related to the opening of the Meliata-Hallstatt marginal sea of the Neotethys Ocean (e.g., Kozur 1991;Ziegler and Stampfli 2001;Vai 2003;Muttoni et al. 2009;Cassinis et al. 2012) and to Pangea disintegration (Isozaki 2009;Putiš et al. 2019a, b).

Conclusions
The Western Carpathian A-type granite intrusions (Turčok, Hrončok and Upohlav) were emplaced at 270-260 Ma, and the Pannonian Velence A-type granite was emplaced earlier at ~ 280 Ma. This time gap (10 to 20 Ma) between magmatic solidification of the Transdanubian Central Range (Velence) granite and other West-Carpathian A-type granite occurrences most likely reflects their different palaeo-tectonic position during NW-ward prograding of rifting in the Neotethyan continental margin.
The geochemical and mineralogical data clearly indicate mostly ferroan and peraluminous A-type affinity with high total-alkali and FeO Tot /MgO, Ga/Al ratios. They are further classified as A 2 -type suite which is compositionally closer to average continental crust, and this is also supported by the relatively higher zircon δ 18 O isotope value (+ 7.5 to + 8.5).
The formation of these mostly peraluminous A-type granites is most likely related to the partial melting of the ancient lower crustal quartzo-feldspatic rocks with the possible small contribution of meta-basic material from the mantle. It was linked to the post-orogenic extensional tectonic regime within the Variscan orogenic belt and chronologically overlapping extension of the new (Alpine) Wilson cycle.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.