U-Pb geochronology and zircon composition of late Variscan S- and I -type granitoids from the Spanish Central System batholith David Orejana . Carlos Villaseca . Pablo V aIverde-Vaquero . Elena A. Belousova . Richard A. Annstrong Abstract The Spanish Central System (SCS) batholith. located in the Central Iberian Zone, is one of the largest masses of granite in the European Variscan Belt. This batholith is a composite unit of late- and post-kinematic granitoids dominated by S- and I-type series granite. with subordinate leucogranite and granodiorite. Zircon trace element contents, from two representative S-type and three I-type granitoids from the eastern portion of the SCS batholith, indicate a heterogeneous composition due to magma differentiation and co-crystallisation of other trace element-rich accessory phases. In situ. U-Pb dating of these zircons by SHRIMP and LA-ICP-MS shows 479-462-Ma inherited zircon ages in the I-type intrusions. indicating the involvement of an Ordovician rnetaigneolls protolith, while the S-type intrusions exclusively contain Cadornian and older zircon ages. The zircon crystallisation ages show that these granites have been emplaced at ca. 300 Ma with a time span between 303 ± 3 Ma and 298 ± 3 Ma. Precise dating by CA-ID-TIMS reveals a pulse at 305.7 ± 0.4 Ma and confirms the major pulse at D. Orejana (c:;]) . C. Villaseca Department of Petrology and Geochemistry (Complutense University of Madrid), Institute of Geosciences (CSIC), 28040 Madrid, Spain e-mail: dorejana@geo.ucm.es P. Valverde-Vaquero Spanish Geological Survey, Madrid, Spain E. A. Belousova Department of Earth and Planetary Sciences, GEMOC, Macquarie University, Sydney, NSW 2109, Australia R. A. Annstrong Research School of Earth Sciences, Australian National University, Canberra, ACT, Australia 300.7 ± 0.6 Ma. These ages match the Permo-Carbonif­ erous age for granulite-fades metamorphism of the lower crust under the SCS batholith and coincide with a wide­ spread granitic event throughout the Southern Variscides. Ti-in zircon thermometry indicates temperatures between 844 and 784°C for both the S- and I-type granites. rein­ forcing the hypothesis that these granites are derived from deep crustal sources. Keywords U-Pb zircon dating . Peralurninous granites' Spanish Central System· European Variscan Belt Introduction Widespread granite rnagrnatism is a common phenomenon in many continental collision zones. It usually occurs during late-orogenic crustal thinning and strongly over­ prints the previous metamorphic history. In this sense, precise dating of granites in this geodynarnic setting can help to constrain the evolution of the continental crust during orogenesis. The Iberian Massif is part of the European Variscan Belt. This collisional belt resulted from the oblique colli­ sion of Gondwana and Baltica-Laurentia during the Early Devonian to Mid-Carboniferous (e.g. Matte 1986. 2001). The Central Iberian Zone (CIZ; Julivert et a1. 1972) rep­ resents the ilUlermost region of the Iberian Massif. This zone is characterised by large volumes of granitic intru­ sions outcropping from central Spain to the NW of the Iberian Peninsula (Fig. 1). Initial attempts to date the Variscan plutonism in the CIZ by Rb-Sr and K-Ar tech­ niques provided a wide age range from 360 to 280 Ma (e.g. Serrano Pinto et a1. 1987; Villaseca et a1. 1995). Recent precise U-Pb studies have constrained the age range to a narrow span between 325 and 280 Ma (U-Pb zircon dating via TIMS or SHRIMP; e.g. Dias et a1. 1998; Femandez­ SUMez et a1. 2000; Valle Aguado et a1. 2005; Zeck et a1. 2007b; Sola et a1. 2009; Dfaz-Alvarado et a1. 2011). This time range coincides with other parts of the Variscan belt (e.g. Finger et a1. 1997; Ledm et a1. 2001; Ballevre et a1. 2009). In the CIZ, the plutonic events are late- and post­ kinematic (Ferreira et a1. 1987). The Spanish Central System batholith (SCS batholith; Villaseca and Herreros, 2000) has a dimension of 250 km x 90 km, which makes it one of the largest late-/post-Variscan batholith in the entire Variscan belt. This study has been focused on the eastern sector of the SCS batholith. We have sampled five representative granites ranging from 1- to S-type peralurninolls granites (Villaseca et a1. 1998). In order to constrain the age of emplacement and the inherited zircons, we have used a combination of U-Pb CA-ID-TIMS, SHRIMP and LA­ ICP-MS geochronology. The in situ dating techniques have the advantage of the fast data acquisition with the high spatial resolution, essential to unravel the inherited zircon component. The CA-ID-TIMS, on the other hand, allows to discriminate emplacement ages within a time span of 5 Ma, which otherwise would challenge the limits of pre­ cision and accuracy of the in situ techniques (see Black et al. 2004). The trace element composition of zircon analysed by LA-ICP-MS provides additional information on the magmatic evolution (e.g. Claibome et a1. 2006; Gagnevin et a1. 2010). We have also used the Ti-in zircon thermometer to relate zircon crystallisation with magma temperature (Ferry and Watson 2007), and in this manner constrain the granite genesis. Our data constitute the first precise geochronology results on the rnagrnatisrn in the eastern sector and add important information to constrain the age of emplacement of the SCS batholith and the crustal sources involved. Geological and geochronological background The Spanish Central System batholith is composed of more than 100 intrusive units that configure one of the largest exposures of granitic intrusions in the European Variscan Belt (Bea et a1. 1999; Villaseca and Herreros 2000). These felsic magmas were emplaced into an Upper Neoprotero­ zoic to Lower Palaeozoic sequence of metasedimentary and metaigneous rocks. According to their modal compo­ sition, these plutons can be classified mainly as granite sensu stricto, with minor proportions of leucogranites and granodiorites (e.g. Villaseca et a1. 1998). Almost, all of them are peraluminous and display silica contents chiefly from 67 to 76 wt% (Villaseca and Herreros 2000). Varia­ tions in their mineralogy and degree of Al saturation have led to subdivision in several series: (1) S-type peralurni­ nous cordierite-bearing granitoids, (2) I-type metaluminous amphibole-bearing granitoids and (3) transitional biotite granitoids of intermediate peraluminous affinity (Villaseca and Herreros 2000). The geochernical data do not display strong differences, but allow the distinction of crystal fractionation trends that rarely connect I-type granites with more differentiated S-type leucogranites (Villaseca et a1. 1998). The relatively similar composition between SCS S­ and I-type granites, both in major elements and isotopic ratios (Villaseca et a1. 1998), likely reflects the stability of the melting conditions during granite genesis and the par­ ticipation of similar source components (Villaseca and Herreros 2000). Nevertheless, various hypotheses regarding the origin of the SCS granitic batholith have been suggested: (a) hybridisation of crustal melts and mantle-derived mag­ mas (pinarelli and Rottura 1995; Moreno-Ventas et a1. 1995); (b) crustal assimilation by mantle-derived magmas (Ugidos and Recio 1993; Castro et a1. 1999); and (c) partial melting of essentially crustal sources, either from lower crustal derivation (Villaseca et a1. 1998, 1999) or from mid­ crustal levels (Bea et a1. 1999, 2003). These granitic bodies intrude into the metamorphic complexes of the Spanish Central System and locally into low-grade rocks. These metamorphic complexes constitute a high-grade infrastructure separated by major shear zones from a low-grade suprastructure (see Macaya et a1. 1991). Most of the granitoids from the SCS batholith intrude after the main phases of ductile deformation in the infrastructure (D2-D3 deformation of Macaya et a1. 1991), which are coeval with migmatisation, and low-pressure/high-tem­ perature M2 metamorphism, after an initial M1 Barrovian metamorphic event (Macaya et a1. 1991; Escuder-Viruete et a1. 1998). Peak metamorphism is constrained at around 340-330 Ma (U-Pb nrnz; Valverde-Vaquero 1998; Escu­ der-Viruete et a1. 1998; U-Pb zm SHRIMP, Castiiieiras et a1. 2008), and the low-pressure/high-temperature meta­ morphism is constrained at 322-320 Ma (U-Pb mnz, xnt, ttu; Valverde-Vaquero 1998). The high-grade gneisses have provided Ar-Ar muscovite and K-Ar biotite ages of 314-310 Ma (Bischoff et a1. 1978; Valverde-Vaquero et a1. 2007), which reflect cooling below 350°C. The fact that these granites have intruded in shallow level, epizonal conditions «2 Kb; e.g. Villaseca and Herreros 2000), and produced contact metamorphism in the high-grade gneis­ ses, indicates that the COlll1try rock in the eastern sector of the SCS batholith was already cold at the time of intrusion. This intrusive event is late-/post-kinematic with respect to the local D4 phase of deformation, which is equivalent to the regional D3 phase of upright folding in the CIZ. Many attempts have been made in order to determine the crystallisation ages of the SCS felsic magmas (Casillas et a1. 199 1; Moreno-Ventas et a1. 1995; Pinarelli and Fig. 1 a Sketch map of the Spanish Central System indicating available precise ages of granites from its western sector (l Zeck et al. 2007b, 2 Diaz-Alvarado et al. 201 1). b Detail of the SCS eastern sector showing the main granitoid intrusions. Stars indicate the sampling location of the five plutons considered in the present study a Spanish Central System b Guadarrama Sector N ! Vil/acaslill Rotura 1995; Villaseca et a1. 1995; Bea et a1. 1999). These studies were based on whole-rock Rb-Sr isochron and K-Ar in biotite analytical techniques, suggesting an extensive time period of 80 Ma for granite intrusion (360-280 Ma; see Bea et a1. 1999; Villaseca and Herreros 2000 and ref­ erences therein). In the western sector of the SCS batholith, recent U-Pb SHRIMP analyses on zircons yield ages in the 303-308 Ma range (Zeck et a1. 2007b; Dfaz-Alvarado et a1. 2011). The data in the eastern end of the SCS batholith are AI(,i(I)'G Real o 10 20 Km c=J Sedimentary rocks _ Metamorphic rocks CD Amph-bcaring granites � Biotitc granites r:::!:!] Crd-bcaring granites 1::;.:::·:::1 Lcucogranitcs •• , [ntcnncdiale and basic 40"21' limited to La Cabrera granite, which has been dated by Pb-Pb evaporation in zircon (301 Ma; Casquet et a1. 2003) and U-Pb ID-TIMS in monazite (293 ± 2 Ma; Valverde­ Vaquero 1998). The scarce ( ,......, 1 %) coeval basic to intermediate magmas which accompany the granite intrusions are represented by small gabbroic to quartzdioritic plutons (Orejana et a1. 2009; Scarrow et a1. 2009). Single-zircon dating by TIMS yielded intrusion ages in the range 312-305 Ma for these mafic bodies (Montero et a1. 2004), but recent ion micro­ probe U-Pb zircon studies constrain the range to 307-300 Ma (Bea et a1. 2006; Zeck et a1. 2007b; Villaseca et a1. 2011). The rest of igneous rocks cropping out in this region are restricted to post-orogenic dyke swarms of variable geochernical character (calc-alkaline, shoshonitic, alkaline and tholeiitic), which illustrates the transition to a ritting geodynarnic setting from Upper Perrnian to Jurassic (Villaseca et a1. 2004; Orejana et a1. 2008). The intrusion of calc-alkaline porphyry dykes is poorly constrained around 290 Ma (Rb-Sr whole-rock isochron; Galindo et a1. 1994), but data with higher accuracy confine the alkaline mag­ matism from 264 (Ar-Ar in arnphibole; Scarrow et a1. 2006) to 252 Ma (U-Pb in zircon; Femandez-Suarez et a1. 2006), and the Messejana-Plasencia tholeiitic dyke to 203 Ma (Ar-Ar in biotite; Dunn et a1. 1998). Field relations and petrography S-type intrusions (Alpedrete and Royo de Pinares) These plutons outcrop along areas near 350 km2 (Alpedr­ ete) and 130 km2 (Royo de Pinares) (Fig. 1) and show an irregular morphology due to emplacement of neighbouring younger granitic intrusions and tectonic contacts with the metamorphic wall-rocks. The development of contact aureoles is indicative of an epizonal emplacement level. They are mainly peraluminous medium-grained equigran­ uIar cordierite-bearing biotite granites and granodiorites. Leucogranites, which are also present in lower proportions, are generally interpreted as highly fractionated magmas derived from a rnonzogranitic parental melt (Villaseca et a1. 1998 and references there in). Porphyritic textures with K-feldspar phenocrysts up to 4 cm long can be found locally. Rounded or elongated mafic microgranular enclaves (biotite-rich quartz-diorites to tonalites) are abundant and may appear dispersed or concentrated in bands, showing fine grain texture and size up to 50 cm. It is also possible to find schlieren structures, micaceous enclaves and metamorphic xenoliths. Plagioclase is usually idiomorphic and displays normal zoning; it can be altered to secondary muscovite, epidote and clinozoisite. K-felspar is also idiomorphic and, similarly to plagioclase, shows biotite and quartz inclusions. Biotite is an early crystallis­ ing mineral, being sometimes altered to chlorite and dis­ playing a sagenitic texture, whereas muscovite is secondary and substitutes cordierite and feldspar. Cordierite appears as interstitial crystals highly transformed to pinnite. Its texture and composition suggest a magmatic origin (Villaseca and Barbero 1994). Typical accessory phases are apatite, zircon, monazite, xenotirne, ilrnenite and rare andalusite in aplo­ pegmatitic varieties. I-type intrusions (Villacastfn, Navas del Marques and Atalaya Real) Villacastfn and Navas del Marques plutons are large, rel­ atively irregular intrusions outcropping along 150 and 300 km2, respectively (Fig. 1). They intrude in low-grade metamorphic rocks, inducing contact aureoles in the wall­ rocks. The Villacasun granite seals the contact between the high-grade migmatitic goeisses and the low-grade rocks of the suprastructure (Macaya et a1. 1991). The Atalaya Real pluton is a small rounded body within the Alpedrete granite (Fig. 1), and the contacts are not exposed so the relation­ ship with the surrounding granite is uncertain. These I-type intrusions are mainly metaluminous to weakly peralumi­ nous amphibole-bearing biotite granites and granodiorites, with minor leucogranitic fades. Similarly to S-type intru­ sions, mafic microgranular enclaves can be found in these I-type granitoids. Plagioclase shows oscillatory zoning and, together with biotite, is the main major early crystallizing mineral. K-feldspar is normally interstitial and may poi­ kilitically include plagioclase, quartz and biotite. Arnphi­ bole is a green Mg-homblende. Biotite is a good discriminant for SCS I-type granites as it plots in subalu­ minous fields below more AI-rich S-type biotites (Fig. 2 of Villaseca et a1. 2009). Localised banding has been observed in Villacasun, whereas biotitic schlieren and metamorphic xenoliths near the contact are common in the three plutons. Flow textures appear in the most porphyritic facies. Occasionally, clinopyroxene (augite/diopside) is present as a residual mineral included in plagioclase or enclosed by amphibole. Apart from apatite, zircon, mona­ zite, xenotime and ilmenite, other important accessory phases in these rocks are allanite, titanite and magnetite. Field relations among the five studied intrusions are not straightforward. Nevertheless, according to its morphology and location, Atalaya Real seems to have intruded after Alpedrete pluton (Fig. 1). But both these granites are not in contact with the other three plutonic intrusions. Some deformation structures associated with the Variscan D4 deformation phase have been identified within the Royo de Pinares granite (Bellido et a1. 1990), but they are rare in the Navas del Marques and Villacastfn granites. Accordingly, the Royo de Pinares granite was likely intruded by the Navas del Marques pluton, whilst no clear relation exists between this latter intrusion and the Villacastin granite. Sample preparation and analytical procedures After standard separation techniques, a representative selection of zircons from each sample was handpicked and cast in epoxy resin for microanalytical analysis, together with some chips of reference zircons TEMORA 1 and Fig. 2 CL images of representative zircon grains from the SCS granite intrusions with the analysed spots and concordia ages SL13. The mOlll1t was ground down to expose the zircon central portions and imaged with transmitted and reflected light on a petrographic microscope, and with cathodolu­ minescence on a HITACHI S2250-N scanning electron microscope (housed at ANU-Canberra) to identify internal structure, inclusions, fractures and physical defects. Selected areas in the grains were analysed for U, Th and Pb isotopes on the Sensitive High Resolution Ion Micro­ probe (SHRIMP IT) at the Research School of Earth Sci­ ences (Australian National University) with each analysis consisting of 6 scans through the relevant mass range. A lO-kV negative O2 primary beam was focused to �20 [tm diameter. Positive secondary ions were extracted from the target spot of 0.5-1 [lll1 deep at 10 kV and mass analysed at c. R5000 on a single ETP multiplier by peak stepping through the isotopes of interest. One TEMORA zircon standard was analysed for every three unknown analyses. Data were reduced following Williams (1998) and refer­ ences therein, using the SQUID Excel Macro of Ludwig (2001), and PbfU ratios were normalised relative to a value of 0.06683 for the 206Pb!"'sU ratio of the TEMORA ref­ erence zircon, equivalent to an age of 417 Ma (Black et a1. 2003). Concentration data are normalised against zircon standard SL13 (210 ppm U, Black et a1. 2004). Uncer­ tainties given for individual analyses (ratios and ages) are at the l iT level; however, uncertainties in the calculated concordia ages are reported as 95% confidence limits. Concordia plots were carried out using Isoplot 3.0 soft­ ware (Ludwig 2003). Ages younger than 1,000 Ma are 204-corrected 206PbP38U, whereas one older age is 204-corrected 207Pbl"06Pb. Alpedrete sample was also selected for new U-Pb age determinations performed on laser ablation rCP-MS at the GEMOC in the Macquarie University of Sydney. Analyses were carried out in situ using a New Wave 213 laser ablation microprobe, attached to a Agilent 7500 quadrupole rCP-MS. The laser system delivers a beam of 213-nm UV light from a frequency-quintupled Nd: YAG laser. Analy­ ses were carried out with a beam diameter of 30 Mm, a 5 Hz repetition rate and energies of 60-100 rnJ/pulse. Typical ablation times are 100-120 s, resulting in pits 30-40 [lll1 deep. The time-resolved signals were processed using the GLITTER interactive software to select the portions of the grains that had suffered least lead loss or gain of common Pb and were thus closest to being con­ cordant. The standards used in this work are the GEMOC­ Gl-1 with an age of 608.5 Ma, the Mud Tank zircon (734 ± 32 Ma; Black and Gulson 1978) and 91500 zircon (1,064 Ma; Wiedenbeck et a1. 1995). Their measured mean values are within 2iT of the recommended values. Other analytical methods follow lackson et a1. (2004). Several zircon fractions from Villacastin and Atalaya Real samples were selected for ID-TIMS and pre-treated with the chemical abrasion (CA) method of Mattinson (2005). We have selected this analytical method (which has higher accuracy than microanalytical techniques) with the aim of comparing the results with SHRIMP data, and get a more constrained age. Significant discrepancies between micro-analytical and rD-TIMS U-Pb data in Variscan granites have already been described (Teixeira et a1. 2011). Zircon annealing was carried at 900°C for 48 h, and the chemical attack was done in Parrish-type minibombs inside Parr bombs at 180°C for 12 h. Final zircon dissolution was achieved after placing the bomb at 240°C for 72 h. The procedure for extraction and purification of Pb and U is a scale-down version of that of Krogh (1973). A 208Pb_235U spike was used to obtain the UlPb ratios by isotope dilution (ID) (Valverde-Vaquero et al. 2000). For test control, the 91500 (Wiedenbeck et al. 1995) and R33 zircon standards (Black et al. 2004) were dated with con cordia ages of 1,065 ± 2.3 Ma and 419.7 ± 1.9 Ma, respectively. Isoto­ pic ratios were measured with a Triton TIMS multi-col­ lector mass spectrorneter equipped with an axial secondary electron multiplier (SEM) ion counter. The instrument is set up to do measurements both in static and peak-jumping mode using the SEM. For static measurements, the 204Pb was measured with the calibrated SEM (92-93% Yield calibration). The Pb measurements were done in the 1,300-1,460°C range, and U was measured in the 1,420-1 ,500°C interval. Data reduction was done using the PbMacDat spreadsheet (Isachsen et al. http://www.earth­ tirne.org) and checked with PBDAT (Ludwig 1991). All isotopic ratios are corrected for mass fractionation, blank and initial common Pb after the model of Stacey and Kramers (1975). Ages and uncertainties were calculated with the decay constants of Jaffey et al. (1971) and are reported at the 20" level; "concordia ages" were calculated with Isoplot 3.0 (Ludwig 2003). Zircon trace element content was analysed using a New Wave 266 laser ablation microprobe coupled to the Agilent 7500 ICPMS system at the GEMOC. Analyses were carried with a 266-nm beam with a pulse rate of 5 Hz and energy of 1 rnJ/pulse, producing a spatial resolution of 30-50 [till. The NIST-610 glass was used as the external calibration standard. The precision and accuracy of the NIST-610 analyses are 2-5% for REE, Y, Nb, Hf, Ta, Th and U at the ppm concentration level, and from 8 to 10% for Mn, P, Ti and Pb. Other analytical description has been given by Belousova et al. (2006). Zircon description Size, external morphology and inner texture of zircons from distinct S- and I-type granites have been studied by transmitted and reflected microscopy and CL imaging. A total number of 93, 84, 99, 65 and 120 grains have been considered, respectively, in the following samples: Royo de Pinares (M21), Alpedrete (Y76), Villacastfn (Y24), Atalaya Real (Y78) and Navas del Marques (M9). CL internal structure from representative grains is shown in Fig. 2. When viewed in transmitted light, zircons from all samples are similar in the usual presence of broken crys­ tals, their euhedral to subhedral form, colourless to pale pink hue and abundant cracks and inclusions. They exhibit distinct primary morphologies: mainly elongated bipyra­ midal prisms with aspect ratios up to 1: 8, but stubby prisms predominate in Villacastin (Y24) and Navas del Marques (M9), and equant grains can be found in Atalaya Real (Y78). They show variable size, normally ranging from 80-100 to 450-500 [lIll, though in sample M9 and M21, they do not exceed 250 and 300 [lIll, respectively. A common feature of zircons from all samples is the presence of irregular cores separated from rims displaying CL zonation patterns by resorbtion surfaces (thin bright bands) (e.g. Fig. 2c, grain 7; Fig. 2e, grain 18). This texture appears associated with pre-V ariscan and Variscan cores. Sample M21 (Hoyo de Pinares) CL textures are dominated by fine euhedral oscillatory zoning, with a progressively darker response towards crystal rims (Fig. 2a). Nearly, 40% of the mounted grains present some kind of inner core, though half of these are homogeneous and concordant, and appear to be an integral part of the oscillatory-zoned rim. The rest are irregular and display variable zoning patterns (oscillatory, convoluted, sector zoning) truncated by the zircon rim. These inner areas sometimes exhibit widening of the zoning bands and bright CL response. Sample Y76 (AJpedrete) CL internal structure is dominated by oscillatory rims and mantles, but darker outer rims are also common. Less than 30% of the grains have inner irregular cores. These are characterised by homogeneous dark CL response, wide bright bands or local recrystallisation. A few almost equant grains with a dark thin rim and homogeneous or sector­ zoned core can also be found (Fig. 2b). Sample Y24 (Villacastin) According to CL imaging, elongated and stubby grains are characterised by oscillatory zoning. Independent of their habit, zircons may present texturally discordant cores with variable internal structures: oscillatory zoning, convoluted zoning and bright CL response (Fig. 2c). Sample Y78 (Atalaya Real) CL imaging reveals dominant oscillatory zoning in rims or the whole crystal (Fig. 2d). Widening of middle bright bands can occasionally appear. Zircon cores are usually homogeneous and concordant with the outer oscillatory zoning. Discordant corroded cores are rare and character­ ised by convoluted zoning or dark homogeneous CL response. Sample M9 (Navas del Marques) Fine oscillatory zoning is the dominant internal structure. The most elongated prisms rarely show inherited cores, but these are common in the stubby zircons. Cores might be divided between (1) texturally concordant inner areas with bright CL response, widened oscillatory bands or convoluted zoning (Fig. 2e; e.g. grains 1, 18) and (2) irregular grains discordant with respect to the outer rim (Fig. 2e; e.g. grains 3, 15). The latter are sometimes isolated from the external zircon by a thin bright irregular band, and their internal structure is usually characterised by a dark response, either homogeneous or displaying convoluted or sector zoning. Analytical results Trace element composition Table 1 reports selected zircon trace element analyses obtained by Laser Ablation rCP-MS from pre-Variscan inherited domains and Variscan zircons. A variety of CL textures have been chosen for analysis, which in most cases correspond with oscillatory-zoned rims or homogeneous and idiomorphic Variscan cores. All analyses (including inheritances) display large chemical variations ranging up to rather high concentrations: Hf (8,950-17,870 ppm), Y (760-5,000 ppm), U (92-3,500 ppm), Th (58-1,812 ppm), HREE (590-3,360 ppm), Nb (1-39 ppm) and Ta (0.32-12 ppm) (Table 1). Ti contents are relatively low, ranging mainly from 3.6 to 24.2 ppm. Chrondrite-normalised REE patterns are also similar in all cases (S- and r-types), with much higher HREE contents with respect to LREE and typical negative Eu anomaly and positive Ce anomaly (Fig. 3a, b). Extreme LREE enrich­ ment in 3 spots (Y24-02, Y24-09, M9-11) can be consid­ ered as anomalous values when compared with standard magmatic zircons (Hoskin and Schaltegger 2003). An evident similarity in composition exists between all Vari­ scan zircons and older inherited cores, with the latter showing higher Nb and Ta values (Fig. 3d). A positive correlation between Hf or U (both used in the figures as differentiation index) and Nb (Ta) (Fig. 3c), but negative with respect to Zr/Hf ratio (which ranges from 32 to 58) (Fig. 4), can be found for all Variscan zircons. On the other hand, the rest of trace elements behave differ­ entially depending on the intrusion considered (Fig. 4). Navas del Marques (M9) zircon is generally enriched in all elements towards higher Hf. Villacastin (Y24) and Atalaya Real (Y78) zircons show positive correlation for P, Y and HREE, whereas those from Alpedrete (Y76) and Hoyo de Pinares (M21) do not display a positive correlation neither for these latter elements nor for the LREE. U-Th-Pb zircon geochronology Ninety-four crystals were analysed by SHRIMP (19, 19, 22, 14 and 20 grains from samples M21, Y76, Y24, Y78 and M9, respectively), 21 of them in several places. In addition, 12 zircon cores from the Alpedrete (Y76) intru­ sion were analysed by Laser Ablation rCP-MS, resulting in a total of 127 in situ analyses (Table 2). CA-ID-TIMS was done in zircons from the Villacastin (Y24) and Atalaya Real (Y78) granites (Table 3). The data are plotted in Wetherill concordia diagrams with all error ellipses at 2rr (Figs. 5, 6, 7). Figure 2 shows representative CL images of crystals featuring location of the analysis craters and resulting ages. The in situ analyses yielded 238UF06Pb ages with relatively wide ranges. However, coherent concordia ages at 2rr confidence level have been calculated with Isoplot software using a selection of data (see "Discus­ sion" for further details on the criteria of selection). U and Th contents described bellow for all samples have been taken from the SHRIMP database, which is larger than the LA-rCP-MS data. Though some differences exist between both groups of analyses, Th/u ratios are generally similar irrespective of the technique considered. There is a chiefly continuous variation in U (74-2,037 ppm) and Th (25-1,067 ppm) contents. But much higher values (U = 2,000-6,000 ppm) can be found in a few analyses (Table 2). Th/u ratios are also fairly variable, ranging principally from 0.07 to 0.9. Haya de Pinares granite Twenty analyses from sample M21 provided Variscan ages, corresponding to oscillatory-zoned rims and inner cores with variable CL structure (leaving aside inheri­ tances). Four partially corroded zircon cores revealed concordant or slightly discordant 238UP06Pb pre-Variscan ages (mainly Upper Proterozoic): 828 ± 9 Ma (7.2), 638 ± 9 Ma (10.1), 619 ± 8 Ma (13.1) and 369 ± 6 Ma (18.1). The three first inheritances have variable CL response (bright to pale bands), whereas the latter displays convoluted zoning and is characterised by high common 206Pb contents. The Variscan ages display error ellipses (2a) overlaping the concordia curve to a greater or lesser extent (Fig. 5a). However, two analyses showing the highest degrees of discordance have not been considered for averaged age calculations (spots 6.1 and 17.1). Other data have also been rejected due to high U concentrations (analysis 1.1: 4,141 ppm) or common 206Pb (spot 16.1). The remaining 16 analyses yield 238UF°"pb ages in the range 306-281 Ma, with a narrower selection providing a con­ cordia age of 299.1 ± 1.8 Ma (MSWD = 0.99; n = 14) (Fig. 5b). Table 1 Representative trace element composition (ppm) of zircon from SCS granitic intrusions detennined by laser ablation ICP-MS Sample # analysis and description" P Ti Y Nb Ta Ht Pb 111 U La Ce Pr Nd Srn Eu Gd Tb Dy Ho Er Trn Yb Lu T CC)b S-type granites Hoyo de Pinares (M21) 2.1 V 276 14.0 814 1 .21 0.87 14,450 0.76 37.6 456 0.44 0.69 0.31 1.63 2.97 0.55 8.39 3.96 59.88 27.85 149.0 37.48 380.1 77.44 833 3 . 1 V 481 1 1 .9 2,021 2.93 0.63 10,958 4.33 177 340 0.38 3.73 0.31 5.75 10.13 1.00 49.07 17.14 195.1 67.44 302.1 60.06 562.7 105.5 8 1 5 4.2 V 706 14.8 2,221 3.07 0.80 13,106 6.04 243 5 1 6 0.55 3.68 0.38 2.66 10.38 0.90 38.57 17.08 205.9 76.72 345.2 73.64 684.7 122.1 839 6.1 V 379 19.0 1,104 1 . 16 0.39 1 1 ,570 3.13 163 213 0.37 3.08 0.28 1.91 6.08 0.70 29.20 9.56 115.2 38.59 166.5 35.31 331.0 57.99 868 18 PO-J 684 8.4 2,041 3 .18 3.03 13,221 2.22 53.3 649 0.32 1.78 0.22 1.46 2.34 0.57 16.23 9.67 150.3 68.83 357.4 84.00 863.3 160.1 779 Alpedrete (Y76) 1.2 V 909 16.0 3,483 2.09 0.90 17,864 1.66 126 243 0.73 3.52 0.52 5.09 13.58 1.05 78.56 28.54 328.5 117.4 493.8 100.9 913.4 163.7 847 4.1 V 732 14.6 3,780 4.56 2.41 13,178 27.0 827 944 1.71 21.07 1 .17 10.61 15.57 1.39 84.59 30.80 361.4 134.4 550.6 110.2 968.4 178.4 837 L3 V 306 6.63 3,652 2.07 0.49 10,388 6.02 375 332 0.39 1 1 . 1 5 0.92 10.03 17.71 2.29 98.66 33.43 396.5 134.2 554.6 107.3 969.2 159.6 756 lAY 178 5.87 2,014 1 .81 0.51 1 1,577 3.84 277 346 0.23 12.37 0.18 5.45 13.04 1.05 49.98 18.54 223.2 77.72 316.5 64.47 572.5 92.74 744 17.1 PO-J 758 12.1 2,902 3.19 1 . 04 16,842 3.93 106 576 0.58 2.65 0.40 3.00 4.63 0.82 38.37 16.36 243.3 98.89 468.5 107.1 1,074 195.3 817 Table 1 continued Sample # analysis and description" P Ti Y Nb Ta Ht Pb 111 U La Ce Pr Nd Srn Eu Gd Tb Dy Ho Er Trn Yb Lu T CC)b I-type granites Villacastin (Y24) 1 . 1 V 506 3.88 2,574 6.52 4.49 14,510 24.3 410 1,450 0.68 9.10 0.29 2.90 5.21 0.67 35.76 16.46 216.7 86.48 406.0 90.57 889.5 149.6 707 5.2 V 863 12.6 4,796 4.00 2.13 16,240 10.4 468 635 0.19 13.89 0.35 5.24 15.96 0.96 98.63 37.52 460.0 173.3 721.4 154.5 1,445 225.2 821 9.1 V 423 15.6 4,095 1 1 .6 7.71 14,687 4 1 .7 980 2,716 42.30 156.2 23.39 145.8 46.24 4.75 1 18.4 34.74 386.2 132.8 587.1 126.8 1,193 202.5 845 2.1 0-1 426 24.2 2,136 20.3 1 1 .8 13,168 44.4 366 1,809 32.70 91.86 13.23 74.08 3 1 .01 1.95 69.71 19.08 195.9 68.16 301.5 70.17 720.2 126.7 897 20.2 PO-I 264 1 1 .7 2,484 1 1 .8 3.96 16,131 30.9 262 374 0.56 21.69 0.35 2.85 9.83 0.87 46.68 19.23 233.7 86.96 369.2 79.61 738.0 124.9 814 0-1 ordovician inheritance, PO-1 pre-ordovician inheritance, V variscan zircon Atalaya Real (Y78) 2.2 V 348 3.64 1,830 3.15 2.07 16,189 4.26 205 878 0.39 5.44 0.14 1.22 4.07 0.28 26. 1 1 1 1 .62 155.1 63.58 302.2 67.09 678.4 112.3 701 3 .1 V 218 8.85 1,166 2.01 0.86 10,833 2.70 124 226 0.14 5.91 0.12 1.41 4.25 0.47 24. 1 5 8.63 105.5 40.44 182.8 38.39 371.5 61.55 784 13 . 1 V 291 13.6 893 1.04 0.32 1 1 ,202 1.05 58.9 92.9 0.12 1.61 0.12 1.87 3.92 0.23 19.32 6.91 85.01 30.93 136.1 28.60 265.6 45.96 829 N avas del Marques (M9) 7.2 V 101 16.2 1,030 2.85 1.22 8,954 2.82 98.5 251 0.97 5.41 0.54 4.34 5.82 0.99 17.42 7.10 88.9 37.17 161 . 1 36.62 361 . 1 71.39 849 1 1 . 1 V 385 8.51 2,841 16.7 6.93 12,857 46.3 1,812 3,415 9.74 52.28 5.56 30.53 20.32 3.63 73.28 22.86 282.7 98.17 413.8 89.08 912.1 135.3 780 15.1 V 157 7.59 1,620 3.10 1.25 9,543 8.92 377 616 1 . 19 14. 1 1 1 . 15 8.23 10.95 1.09 36.48 1 1 .85 155.8 53.32 254.3 59.71 597.8 93.64 769 b Temperatures estimated using the Ti-in-zrrcon geothennometer recalibrated by Ferry and Watson (2007). Temperature uncertainty for each data is ±4.5% 5.1 0-1 631 13.4 2,213 1 1 .2 2.55 12,713 5.58 194 450 1.89 10.78 1.44 8.57 10.71 0.64 35.84 14.30 192.57 78.71 337.56 76.59 707.91 1 16.07 827 5.2 PO-I 446 16.7 3,474 38.8 12.0 12,617 20.8 415 1,488 3.72 32.57 2.14 16.34 12.18 1 .38 75.94 27.60 339.9 121.91 536.5 1 15.8 1,064 144.3 852 Fig. 3 Trace element composition of zircons from the SCS granites: Chondrite­ nonnalised REE patterns of Variscan zircons a and 1000 a 1000 II :§ " 100 b zircons � pre-V ariscan cores b, and U concentrations versus Hf c and Nb d contents. Nonnalising values after Sun and McDonough (1 989) Variscan zircons DM9 .M21 0 Y24 .Y76 0Y78 � 8 10 i.:J Pre-Variscan cores \l Ordovician ... Pre-Ordovician La Pr Srn Gd Dy Er Yb Ce Nd Eu Tb Ho Trn Lu La Pr Srn Gd Dy Er Yb Ce Nd Eu Tb Ho Trn Lu � '-�--'-----�D�---------' C o 1000 D r"� / 'a, I 0 ''0 / 0 • D .. 0\ O . , '-p ... • : ..J .. . ,-�-------D o • o • 100 "---� __ �� ______________ cj 8000 10000 Hf Alpedrete granite Almost all SHRIMP analyses from sample Y76 represent crystallisation ages. Only two inherited ages have been found in this sample corresponding to corroded cores with faint oscillatory zoning surrounded by dark homogeneous irregular domains, which provide concordant Neoprotero­ zoic ages: 592 ± 7 Ma (analysis 17.1) and 581 ± 7 Ma (analysis L7). Exceedingly high common 206Pb contents have been found only in one zircon core (spot 19.1). The remaining analyses comprise a broad time span of 40 Ma with mostly concordant 238U/206Pb ages from 302 to 261 Ma (Fig. 5c). Th and U contents seem to correlate negatively with age for part of the data (not shown), probably suggesting some kind of secondary deviation of the calculated age. Analyses showing this behaviour (1.2, 2.1, 3.2, 4.2, 10.1, 11.1, 14.1 and 15.1) are seemingly concordant zircon cores with homogeneous bright CL luminescence, and display lower ages than their sur­ rounding rims. Thus, we have not considered these latter data in averaged age calculations, together with other spots with high discordancy (16.1, 18.1 and 19.1). The resulting 12 spots range from 288 to 302 Ma. Seven of these anal· yses provided a con cordia age of 297.1 ± 2.4 Ma (MSWD = 1.15) (Fig. 5d). LA·ICP·MS analyses yielded 11 crystallisation ages (Table 2). Two of them (L6 and L11) are excluded due to high common Pb or degree of discordance. The remaining 9 analyses scatter in a range very similar to that shown by 20000 u the SHIRMP results (Fig. 5e). A selection of six analyses provided a concordia age of 301.7 ± 3.4 Ma (MSWD = 2.7) (Fig. 5f), which overlap within analytical error with that calculated using SHRIMP data. Villacasttn granite Sample Y24 yielded 22 Permo·Carboniferous ages from a total of 26 analyses. Four pre-Variscan 238Upo6Pb ages have been found in three zircon grains. This pre-Variscan component is found in corroded cores with oscillatory zoning (sometimes partially convoluted) or decomposed zoning, enclosed by thin dark rims (Fig. 2c; grains 16 and 20). Three of those ages are Ordovician: 479 ± 5 Ma (2.1), 462 ± 5 Ma (16.1), 464 ± 5 Ma (20.1), and one is Lower Proterozoic, 1,956 ± 12 Ma (20.2), which corresponds to inheritance within an Ordovician core. Some of the analyses of dark zircon rims and blurred inner domains (spots 1.1, 3.1, 6.1, 7.2, 8.1, 11.1, 12.1, 14.1, 17.1, 18.1, 19.1, 21.1) show high levels of common 2°"pb lead with exceedingly young ages (271-292 Ma). We have excluded these analyses and one spot with high discor­ dancy (5.2) from averaged age calculations. The remaining 9 analyses, ranging from 288 to 300 Ma (Fig. 6a), corre· spond mainly with rims showing oscillatory zoning. The calculated concordia age for this sample yield 297.6 ± 2.8 Ma (MSWD = 0.16; n = 5) (Fig. 6b). The four zircon fractions analysed by CA·ID·TIMS overlap con cordia between 299 and 302 Ma and provide a Fig. 4 Trace element composition of Variscan zircons from the SCS granites. Hf concentration versus P, Th, Y, LREE and HREE contents and ZrIHf ratios. The field enclosed by the dotted line represents the composition of pre-Variscan inherited cores. Symbols as in legend of Fig. 3 2�,-�--,------------------, p I� 200 c---------------.. \0 ........ \ 3\ • o \ I! \ .� lilt> : • \ 0 i . "OP 0 : • 0 �".. 0 i ...... . : o 0 D 0 • . "'-----\. Inherited cores 5000 ,---�--,--__,_---�--�--=--�--_�_;o:_.-" v -----, y .\ •• • 0'\ . D 1000 D D , \ 0 0 // �:\.� ... \ 0 \ .. �--D • • 500,-�--�--------,_O ,-------, l:LREE lOO 10 D 00 o 2000 ,-��------�u--------� Th 1000 o • 100 D • 60,---�--,------------------, D 50 40 o ZrlHf D • Cfii----.. 0 8. .. ........ . �� dI. -...•••.• '\ \ . \ �\ \ \ .. 00 (\ \ .. .. _.�_.::./. 0' .. � .. � ..... -; • • 4000 ,---�--�------------------, l:HREE D 1000 D , ................ --....... --...... 0 \\ • 0 .. • ......... . .. \. \ 0 i \ 0 0 l • \. ..l �. \. <: ___ ...... -t:J o .... -... • • 8000 10000 2�0 8000 10000 20000 m m concordia age of 300.5 ± 0.55 Ma (Fig. 7a; MSWD = 0.0065), which we consider the most accurate estimate for the age of intrusion. The other analyses have error ellipses overlapping the concordia curve (Fig. 6c). Analysis 13.1 with high U contents, which yielded an anomalous high age of 313 ± 4 Ma, 4 data with high common 206Pb (7.1, 10.1, 14.1 and 17.2) and one spot with high degree of discor­ dance (18.1), have been excluded from the age calculation. The remaining 13 analyses display 238U;2""I>b ages in the range 306-280 Ma (Fig. 6c). A selection of eight spots allows the calculation of a concordia age of 300.3 ± 2.6 Ma (MSWD = 0.59) (Fig. 6d). Navas del Marques granite Sample M9 provided 19 crystallisation ages from a variety of zircon domains: rims with oscillatory zoning and vari­ able CL response and cores with oscillatory or convoluted zoning. Three zircon grains have pre-Variscan ages. One Upper Neoproterozoic age of 564 ± 6 Ma was obtained for an irregular homogeneous dark mineral domain (5.2) located within an Ordovician zircon core (5.1) (Fig. 2e), similarly to what was observed in the Villacastin granite (sample Y24). The other 3 pre-Variscan ages are Ordovi­ cian (Table 2; spots 3.1, 5.1 and 20.1) and have been obtained from anhedral to subhedral cores showing a var­ iable CL internal structure, and with a surrounding black rim. Atalaya real granite The 18 SHIRMP analyses from sample Y78 do not show inherited zircon. Five analyses corresponding to dark rims or inner domains with sector or irregular zoning exhibit high common 206Pb (analyses 5.1, 5.2, 8.1, 11.1 and 14.1). For the estimation of the crystallisation age, we have not considered these latter data, together with analyses Table 2 Ion microprobe (SHRIMP) and laser ablation ICP-MS U-Pb analytical data for zircons from SCS granitic intrusions Age (Ma) Spot number Common 206Pb (%) U (ppm) Th (ppm) Th/U Radiogenic ratios =-�--���------�-=---207Pb/235U ± er 206PbP38U ± er p 207PbP06Pb ± er 206PbP38U ± er 207Pb/206Pb ± er Hoyo de Pinares (M21 )-SHRIMP 1 . 1 0.03 2.1 3 .1 4.1 4.2 5 . 1 6 . 1 7 . 1 7.2 8 . 1 8.2 9.1 10.1 1 1 . 1 12.1 12.2 13 .1 14.1 14.2 15 .1 16.1 17.1 18 .1 19.1 0 .18 0.00 0.17 0.72 0.47 0.41 0.09 0.20 0.04 0.12 0.29 1.25 0.20 0.38 0.42 0.58 0.08 0.15 0.00 6.30 0.24 0 . 1 1 0.20 Alpedrete (Y76}-SHRlMP 1 . 1 0.08 1 .2 2.1 2.2 3 . 1 3.2 4.1 4.2 0.00 0.00 0.19 o.n 0.00 0.16 0.00 4,141 888 371 603 562 439 166 761 1 5 1 1,548 620 867 323 243 755 364 68 771 624 182 767 410 402 489 802 106 200 208 331 127 983 275 367 189 194 1 19 236 143 130 107 69 1,067 182 150 25 78 519 206 36 390 185 76 296 145 209 242 55 40 43 47 62 46 277 94 0.09 0.21 0.52 0.20 0.42 0.33 0.79 0.14 0.45 0.69 0.29 0.17 0.08 0.32 0.69 0.57 0.53 0.5 0.30 0.42 0.39 0.35 0.52 0.49 0.07 0.38 0.21 0.22 0.19 0.37 0.28 0.34 0.3727 ± 1 . 1 0.3370 ± 1.7 0.3434 ± 1.6 0.3382 ± 2.2 0.3540 ± 3.9 0.3160 ± 3.2 0.3410 ± 3.5 0.3474 ± 2.0 1.2193 ± 2.1 0.3501 ± 1.2 0.3417 ± 1.6 0.3410 ± 1.8 0.4590 ± 7.4 0.3423 ± 2.7 0.3344 ± 2.1 0.3400 ± 3.6 0.8740 ± 4.0 0.3378 ± 2.0 0.3381 ± 2.1 0.3334 ± 2.9 0.2830 ± 13.0 0.3339 ± 2.5 0.8470 ± 1.9 0.3282 ± 2.2 0.3437 ± 1.6 0.3181 ± 2.7 0.3292 ± 2.9 0.3309 ± 2.6 0.3225 ± 2.2 0.3241 ± 4.9 0.3286 ± 1.5 0.3037 ± 2.7 0.0517 ± 1 .0 0.0472 ± 1 .0 0.0479 ± 1 . 1 0.0472 ± 1 . 1 0.0482 ± 1 . 1 0.0446 ± 1 . 1 0.0488 ± 1 .2 0.0483 ± 1 . 1 0 .1370 ± 1 .2 0.0486 ± 1 .0 0.0477 ± 1 . 1 0.0477 ± 1 . 1 0.0589 ± 1.7 0.0483 ± 1.2 0.0459 ± 1 . 1 0.0474 ± 1 . 1 0 .1041 ± 1.4 0.0468 ± 1.2 0.0478 ± 1 . 1 0.0458 ± 1 .5 0.0403 ± 1 .3 0.0474 ± 1 .2 0.1008 ± 1 .3 0.0448 ± 1 .2 0.0480 ± 1 . 1 0.0447 ± 1 .3 0.0454 ± 2.4 0.0465 ± 1 .2 0.0458 ± 1 . 1 0.0443 ± 4.4 0.0459 ± 1 .0 0.0414 ± 2.3 0.93 0.0523 ± 0.4 0.61 0.0518 ± 1.4 0.68 0.0520 ± 1 .2 0.49 0.0520 ± 1 .9 0.28 0.0534 ± 3.8 0.35 0.0515 ± 3.0 0.36 0.0507 ± 3.2 0.52 0.0522 ± 1.7 0.57 0.0646 ± 1.7 0.83 0.0523 ± 0.7 0.65 0.0519 ± 1 .2 0.59 0.0519 ± 1.4 0.24 0.0565 ± 7.2 0.43 0.0514 ± 2.5 0.53 0.0528 ± 1 .8 0.31 0.0520 ± 3.4 0.36 0.0609 ± 3.7 0.59 0.0524 ± 1 .6 0.55 0.0513 ± 1.7 0.50 0.0528 ± 2.5 0.10 0.0509 ± 13.0 0.48 0.0511 ± 2.2 0.69 0.0610 ± 1.4 0.52 0.0532 ± 1 .9 0.68 0.0519 ± 1 . 1 0.49 0.0516 ± 2.4 0.83 0.0526 ± 1.7 0.45 0.0516 ± 2.3 0.51 0.0510 ± 1 .9 0.89 0.0530 ± 2.2 0.68 0.0520 ± 1 . 1 0.85 0.0533 ± 1 .5 325 ± 3.2 297 ± 3.0 302 ± 3.3 297 ± 3.1 303 ± 3.3 281 ± 3.0 307 ± 3.7 304 ± 3 . 1 828 ± 9 . 1 306 ± 3 . 1 300 ± 3 . 1 300 ± 3 . 1 369 ± 6.2 304 ± 3.5 289 ± 3.2 299 ± 3.3 638 ± 8.8 295 ± 3.4 301 ± 3.3 288 ± 4.1 255 ± 3.2 299 ± 3.4 619 ± 7.7 282 ± 3.2 302 ± 3 . 1 282 ± 3.7 286 ± 6.8 293 ± 3.4 289 ± 3.2 280 ± 12.0 289 ± 3.0 261 ± 5.9 300 ± 9 278 ± 3 1 285 ± 27 286 ± 44 345 ± 85 263 ± 68 227 ± 75 295 ± 40 760 ± 36 297 ± 16 282 ± 28 281 ± 33 474 ± 158 260 ± 56 322 ± 41 286 ± 79 636 ± 80 301 ± 37 254 ± 40 322 ± 58 235 ± 300 244 ± 50 637 ± 29 337 ± 43 282 ± 26 269 ± 54 3 1 1 ± 38 270 ± 54 242 ± 43 329 ± 50 284 ± 26 340 ± 33 Disc (%)a -8 -7 -6 -4 14 -7 -26 -3 -8 -3 -6 -6 28 - 14 1 1 -4 o 2 - 1 6 1 2 - 8 - 1 8 3 19 -7 -5 8 -8 - 1 6 1 8 - 2 30 Table 2 continued Age (Ma) Spot number Common 206Pb (%) U (ppm) Th (ppm) Th/U Radiogenic ratios =-�--���------�-=---207Pb/235U ± er 206PbP38U ± er p �w, PbrwoPb ± er 206PbP38U ± er 207Pb/206Pb ± er 5 . 1 6 . 1 7 . 1 7.2 8 . 1 9 . 1 10.1 1 1 . 1 12.1 13.1 14.1 15.1 16.1 17.1 18.1 19.1 0.04 0.09 0.00 0.13 0 . 1 1 0.14 0.64 0.00 0.37 0.00 0.18 0.45 0.34 0.18 0.31 2.21 Alpedrete (Y76}-LA-ICP-MS L l 0.00 L2 L3 L4 L5 L6 L7 L8 L9 LlO L l l Ll2 0.00 0.00 0.00 0.00 3.32 0.00 0.00 0.00 0.00 0.00 0.00 Villacastin (Y24)-SHRIMP 1 . 1 2.17 2.1 3 .1 4.1 0.79 8.60 0.15 1,009 780 496 3 1 9 327 249 74 143 976 278 527 82 201 270 2 1 1 778 585 1,614 589 559 1,333 677 912 638 297 274 370 644 1,871 817 1,716 1,130 276 241 103 109 69 57 78 58 69 99 220 26 207 43 55 287 218 635 479 354 787 741 350 238 125 66 148 232 495 40 479 332 0.27 0.31 0.21 0.34 0.21 0.23 1.05 0.41 0.07 0.36 0.42 0.32 1.03 0.16 0.26 0.37 0.37 0.39 0.81 0.63 0.59 1.09 0.38 0.37 0.42 0.24 0.40 0.36 0.26 0.05 0.28 0.29 0.3326 ± 1.4 0.3385 ± 1.6 0.3333 ± 1.6 0.3379 ± 2.2 0.3411 ± 2.1 0.3374 ± 2.4 0.3124 ± 7.0 0.3161 ± 3.6 0.3320 ± 1.8 0.3426 ± 1.8 0.3 1 1 3 ± 2.2 0.3071 ± 5.3 0.3270 ± 4.1 0.7779 ± 1.9 0.3240 ± 3.8 0.1733 ± 5.3 0.2926 ± 3.0 0.3478 ± 1.6 0.3480 ± 2.3 0.3541 ± 1.9 0.3586 ± 2.1 0.3508 ± 2.0 0.7719 ± 1.4 0.3499 ± 2.4 0.3239 ± 2.4 0.3611 ± 1.8 0.3661 ± 5.2 0.3361 ± 2.4 0.3373 ± 4.8 0.5976 ± 2.1 0.3458 ± 10.9 0.3364 ± 1.5 0.0466 ± 1 .0 0.0473 ± 1 . 1 0.0460 ± 1 . 1 0.0468 ± 1 . 1 0.0470 ± 1 . 1 0.0471 ± 1 .2 0.0449 ± 1 .7 0.0434 ± 1 .6 0.0463 ± 1 . 1 0.0477 ± 1 .2 0.0435 ± 1 .5 0.0454 ± 1 .4 0.0472 ± 1 .2 0.0962 ± 1 .2 0.0462 ± 1 .2 0.0249 ± 1 .2 0.0403 ± 1 .8 0.0478 ± 1 .4 0.0479 ± 1 .6 0.0482 ± 1 .4 0.0480 ± 1 .5 0.0467 ± 1 .4 0.0943 ± 1 .3 0.0472 ± 1 .6 0.0458 ± 1 .5 0.0490 ± 1 .3 0.0466 ± 2.3 0.0473 ± 1 .6 0.0464 ± 1 . 1 0.0771 ± 1 .0 0.0445 ± 1 .2 0.0468 ± 1 .0 0.77 0.0518 ± 0.9 0.68 0.0519 ± 1 . 1 0.72 0.0526 ± 1 . 1 0.51 0.0524 ± 1.9 0.53 0.0527 ± 1 .8 0.49 0.0519 ± 2.0 0.25 0.0505 ± 6.8 0.45 0.0529 ± 3.2 0.60 0.0520 ± 1.4 0.65 0.0521 ± 1.4 0.68 0.0519 ± 1 .6 0.27 0.0491 ± 5 . 1 0.30 0.0502 ± 3.9 0.64 0.0587 ± 1 .5 0.32 0.0509 ± 3.6 0.23 0.0504 ± 5 . 1 0.32 0.0528 ± 3.0 0.59 0.0528 ± 1.4 0.42 0.0528 ± 2.2 0.48 0.0533 ± 1.7 0.45 0.0542 ± 2.0 0.44 0.0545 ± 1 .9 0.57 0.0594 ± 1 .2 0.39 0.0539 ± 2.3 0.39 0.0514 ± 2.3 0.46 0.0535 ± 1 .6 0.16 0.0571 ± 5.4 0.39 0.0515 ± 2.3 0.22 0.0528 ± 4.6 0.51 0.0562 ± 1.8 0 . 1 1 0.0563 ± 10.8 0.70 0.0521 ± 1 . 1 293 ± 3.0 298 ± 3 . 1 290 ± 3.3 295 ± 3.3 296 ± 3.3 297 ± 3.4 283 ± 4.8 274 ± 4.3 292 ± 3.0 300 ± 3.4 275 ± 4.0 286 ± 4.0 297 ± 3.5 592 ± 7.0 291 ± 3.4 159 ± 1.9 254 ± 4 301 ± 4 301 ± 5 304 ± 4 302 ± 4 294 ± 4 581 ± 7 297 ± 5 288 ± 4 308 ± 4 294 ± 7 298 ± 5 292 ± 3.0 479 ± 4.8 281 ± 3.4 295 ± 3.0 277 ± 20 282 ± 26 310 ± 25 301 ± 44 316 ± 41 283 ± 47 218 ± 157 323 ± 73 287 ± 32 290 ± 3 1 280 ± 36 153 ± 1 19 205 ± 90 555 ± 33 236 ± 84 214 ± 1 19 319 ± 70 318 ± 32 319 ± 5 1 341 ± 40 381 ± 45 391 ± 43 581 ± 27 365 ± 54 257 ± 53 348 ± 38 494 ± 121 265 ± 54 319 ± 106 462 ± 39 465 ± 239 290 ± 24 Disc (%)a -6 -5 7 2 7 -5 -23 18 - 1 -4 2 -46 - 3 1 -6 - 1 9 35 21 6 6 1 1 21 25 o 19 - 1 3 1 2 42 - 1 3 8 -4 40 -2 Table 2 continued Age (Ma) Spot number Common 206Pb (%) U (ppm) Th (ppm) Th/U Radiogenic ratios =-�--���------�-=---207Pb/235U ± er 206PbP38U ± er p �w, PbrwoPb ± er 206PbP38U ± er 207Pb/206Pb ± er 5 . 1 5.2 6.1 7.1 7.2 8 . 1 9 . 1 10.1 1 1 . 1 12.1 13.1 13.2 14.1 15.1 16.1 17.1 18.1 19.1 20.1 20.2 21 . 1 22.1 0.03 4.23 29.55 1 . 1 6 3.47 5.85 0.65 1.00 15.44 18 . 13 0.00 0.08 12.65 0.07 0.16 4.50 8.00 2.33 0.05 0.01 4.74 0.04 Atalaya Real (Y78)-SHRlMP 1 . 1 0.01 2.1 2.2 3 . 1 4 . 1 5 . 1 5.2 6.1 6.2 7.1 7.2 0.54 0.09 0.24 0.12 5.47 1 1 .42 0.00 0.19 0.58 0.23 1,005 288 92 1,204 709 178 1,101 996 905 983 271 2037 1 ,115 1,305 455 942 1,915 952 271 377 215 1,804 6,139 575 1,222 210 280 866 349 449 1,048 574 5,022 243 124 30 340 169 69 335 241 214 247 91 461 298 241 37 274 475 230 10 215 82 251 2,802 159 245 76 1 12 270 188 153 637 199 390 0.24 0.43 0.33 0.28 0.24 0.39 0.30 0.24 0.24 0.25 0.34 0.23 0.27 0.18 0.08 0.29 0.25 0.24 0.04 0.57 0.38 0.14 0.46 0.28 0.20 0.36 0.40 0.31 0.54 0.34 0.61 0.35 0.08 0.3382 ± 1.4 0.3653 ± 18.0 0.3583 ± 58.4 0.3450 ± 3.2 0.3342 ± 8.1 0.3895 ± 1 1 .5 0.3349 ± 2.2 0.3465 ± 2.5 0.2977 ± 24.5 0.3197 ± 29.8 0.3324 ± 1.9 0.3418 ± 1.3 0.3482 ± 14.6 0.3438 ± 1.4 0.5754 ± 1.6 0.3509 ± 12.0 0.3285 ± 15.2 0.3277 ± 5.2 0.5742 ± 1.7 5 .5240 ± 2.4 0.3075 ± 1 1 .0 0.3289 ± 1.3 0.3624 ± 1 . 1 0.3161 ± 2.8 0.3366 ± 1.4 0.3427 ± 2.7 0.3456 ± 1.9 0.3 1 1 5 ± 7.3 0.2047 ± 18.7 0.3439 ± 1.5 0.3415 ± 1.6 0.3411 ± 3.0 0.3450 ± 1.2 0.0468 ± 1 . 1 0.0482 ± 2.2 0.0432 ± 3.8 0.0477 ± 1 .0 0.0460 ± 1 .2 0.0449 ± 1 .5 0.0465 ± 1 .0 0.0471 ± 1 . 1 0.0432 ± 1.7 0.0444 ± 2.2 0.0462 ± 1 . 1 0.0474 ± 1 .0 0.0447 ± 1 .4 0.0476 ± 1 . 1 0.0743 ± 1 . 1 0.0452 ± 1 .3 0.0455 ± 1 .3 0.0449 ± 1 . 1 0.0746 ± 1 .2 0.3339 ± 2.3 0.0430 ± 1 .5 0.0458 ± 1 .0 0.0501 ± 1 .0 0.0427 ± 1 . 1 0.0468 ± 1 . 1 0.0485 ± 1 .3 0.0479 ± 1 . 1 0.0433 ± 1 . 1 0.0284 ± 1 .6 0.0471 ± 1 . 1 0.0479 ± 1 .0 0.0470 ± 1 . 1 0.0477 ± 1 .0 0.76 0.0524 ± 0.9 0.12 0.0550 ± 17.9 0.06 0.0601 ± 58.3 0.33 0.0524 ± 3.0 0.14 0.0527 ± 8.0 0.13 0.0629 ± 1 1 .4 0.48 0.0522 ± 1 .9 0.44 0.0534 ± 2.2 0.07 0.0499 ± 24.4 0.07 0.0522 ± 29.7 0.59 0.0522 ± 1 .6 0.79 0.0523 ± 0.8 0.10 0.0565 ± 14.5 0.82 0.0523 ± 0.8 0.66 0.0562 ± 1 .2 0.11 0.0563 ± 1 1 .9 0.09 0.0523 ± 15.1 0.21 0.0529 ± 5 . 1 0.69 0.0558 ± 1 .2 0.96 0.1200 ± 0.7 0.13 0.0519 ± 10.9 0.80 0.0521 ± 0.8 0.94 0.0524 ± 0.4 0.38 0.0537 ± 2.6 0.73 0.0521 ± 1 .0 0.49 0.0512 ± 2.4 0.57 0.0523 ± 1 .6 0.16 0.0522 ± 7.2 0.09 0.0524 ± 18.6 0.71 0.0530 ± 1 . 1 0.65 0.0517 ± 1 .2 0.36 0.0527 ± 2.8 0.88 0.0525 ± 0.6 295 ± 3 . 1 303 ± 6.5 273 ± 10.1 300 ± 3.1 290 ± 3.3 283 ± 4.1 293 ± 3.0 297 ± 3.1 273 ± 4.5 280 ± 5.9 291 ± 3.3 298 ± 3.0 282 ± 3.8 300 ± 3.3 462 ± 4.8 285 ± 3.5 287 ± 3.7 283 ± 3.0 464 ± 5.2 1857 ± 37.4 271 ± 3.9 289 ± 2.9 315 ± 3 . 1 270 ± 2.8 295 ± 3.0 305 ± 4.0 302 ± 3.3 273 ± 3.0 180 ± 2.8 297 ± 3.2 302 ± 3 . 1 296 ± 3 . 1 300 ± 3.0 305 ± 21 412 ± 400 607 ± 1260 305 ± 68 316 ± 181 703 ± 244 296 ± 44 345 ± 5 1 193 ± 568 293 ± 679 293 ± 36 300 ± 18 472 ± 321 300 ± 18 459 ± 27 465 ± 263 300 ± 344 326 ± 1 16 445 ± 27 1956 ± 12 280 ± 249 289 ± 17 305 ± 8 357 ± 59 291 ± 22 251 ± 55 298 ± 36 295 ± 164 301 ± 424 327 ± 25 271 ± 28 314 ± 64 305 ± 13 Disc (%)a 3 26 55 8 60 14 -42 4 o 40 o - 1 39 5 13 -4 5 3 o -3 24 - 1 -22 - 1 7 40 9 - 1 1 6 2 Table 2 continued Age (Ma) Spot number Common 206Pb (%) U (ppm) Th (ppm) Th/U Radiogenic ratios =-�--���------�-=---207Pb/235U ± er 206PbP38U ± er p �w, PbrwoPb ± er 206PbP38U ± er 207Pb/206Pb ± er 8 . 1 9 . 1 10.1 1 1 . 1 12.1 13.1 14.1 4.76 0.22 0.21 9.53 0.00 0.67 7.45 N avas del Marques (M9)-SHRIMP 1 . 1 0.00 2.1 3 .1 4.1 5 .1 5 .2 6.1 7.1 7.2 8 . 1 9 . 1 10.1 1 1 . 1 12.1 13.1 14.1 15.1 16.1 17.1 17.2 18 .1 19.1 20.1 0.38 0.00 1.26 0.25 0.59 0.30 6.08 0.25 0.17 0.90 2.86 0.19 0.38 0.33 5.00 1 . 14 0.61 0.09 5.93 0.43 0.76 0.00 322 1.406 271 298 263 149 227 219 280 204 133 306 2537 232 667 272 1.464 1,203 937 2035 405 3227 371 473 773 5 1 3 994 185 493 354 1 12 289 65 124 122 109 89 71 102 97 64 23 861 79 386 82 438 267 229 734 169 1049 122 277 151 257 876 102 156 26 0.35 0.21 0.24 0.41 0.46 0.73 0.39 0.32 0.36 0.47 0.48 0.08 0.34 0.34 0.58 0.30 0.30 0.22 0.24 0.36 0.42 0.33 0.33 0.59 0.20 0.50 0.88 0.56 0.32 om 0.2594 ± 9.4 0.3532 ± 1.6 0.3446 ± 2.7 0.3173 ± 22.1 0.3423 ± 2.2 0.3314 ± 7.0 0.3392 ± 16.1 0.3340 ± 2.1 0.3499 ± 4.1 0.5947 ± 1.8 0.3254 ± 7.9 0.5749 ± 2.7 0.7430 ± 1.5 0.3353 ± 3.3 0.3547 ± 16.8 0.3430 ± 3.2 0.3437 ± 1.6 0.3349 ± 2.4 0.3311 ± 7.3 0.3425 ± 1.6 0.3407 ± 3.9 0.3615 ± 1.6 0.3327 ± 17.8 0.3383 ± 4.4 0.3335 ± 2.3 0.3495 ± 2.0 0.3272 ± 8.5 0.3389 ± 3.8 0.3349 ± 2.9 0.6113 ± 1.6 0.0368 ± 1.3 0.0486 ± 1.0 0.0485 ± 1.2 0.0449 ± 1.6 0.0428 ± 1.2 0.0490 ± 1.3 0.0431 ± 1.9 0.0459 ± 1.3 0.0486 ± 1.3 0.0747 ± 1.3 0.0445 ± 1.5 0.0751 ± 1.3 0.0915 ± 1.2 0.0466 ± 1.3 0.0483 ± 1.5 0.0479 ± 1.3 0.0478 ± 1.2 0.0467 ± 1.2 0.0470 ± 1.3 0.0479 ± 1.2 0.0473 ± 1.3 0.0498 ± 1.2 0.0459 ± 1.4 0.0472 ± 1.3 0.0463 ± 1.2 0.0482 ± 1.2 0.0455 ± 1.3 0.0483 ± 1.4 0.0470 ± 1.3 0.0769 ± 1.3 0.13 0.0511 ± 9.3 0.66 0.0527 ± 1 .2 0.42 0.0515 ± 2.5 0.07 0.0513 ± 22.0 0.52 0.0580 ± 1 .9 0.19 0.0490 ± 6.9 0.12 0.0571 ± 16.0 0.63 0.0527 ± 1 .6 0.33 0.0522 ± 3.9 0.71 0.0577 ± 1 .3 0.19 0.0530 ± 7.7 0.48 0.0555 ± 2.3 0.80 0.0589 ± 0.9 0.40 0.0521 ± 3 . 1 0.09 0.0532 ± 16.7 0.41 0.0519 ± 2.9 0.75 0.0522 ± 1 . 1 0.51 0.0520 ± 2.0 0.17 0.0511 ± 7.2 0.75 0.0519 ± 1 . 1 0.33 0.0522 ± 3.7 0.76 0.0527 ± 1 . 1 0.08 0.0525 ± 17.7 0.29 0.0520 ± 4.2 0.52 0.0522 ± 2.0 0.62 0.0526 ± 1 .6 0.16 0.0521 ± 8.4 0.36 0.0509 ± 3.6 0.43 0.0517 ± 2.7 0.78 0.0577 ± 1 .0 233 ± 2.9 306 ± 3 . 1 305 ± 3.4 283 ± 4.4 270 ± 3 . 1 308 ± 3.9 272 ± 5 . 1 289 ± 3.7 306 ± 4.0 465 ± 5.8 281 ± 4.1 467 ± 5.8 564 ± 6.4 294 ± 3.8 304 ± 4.5 302 ± 3.9 301 ± 3.5 294 ± 3.5 296 ± 3.6 302 ± 3.5 298 ± 3.7 313 ± 3.7 290 ± 4.0 297 ± 3.7 292 ± 3.5 303 ± 3.7 287 ± 3.7 304 ± 4.1 296 ± 3.6 478 ± 5.8 Uncertainties given at the one sigma level. Common Pb corrected using measured 204Pb. p error correlation between 206pbP38U and 207pbP35U ratios a Disc U-Pb discordance, difference between the 204Pb_corrected 206PbP38U and 207PbP06Pb ages 245 ± 214 315 ± 27 265 ± 57 254 ± 506 531 ± 42 149 ± 161 494 ± 352 318 ± 37 296 ± 88 519 ± 28 330 ± 176 433 ± 52 564 ± 19 291 ± 70 338 ± 379 282 ± 67 293 ± 24 285 ± 47 244 ± 167 279 ± 24 296 ± 84 315 ± 24 308 ± 404 287 ± 97 296 ± 45 313 ± 36 292 ± 192 235 ± 82 271 ± 61 517 ± 22 Disc (%)a 5 3 - 1 5 - 1 1 49 - 106 45 9 -3 10 15 -8 o - 1 10 -7 -3 -3 -21 -8 - 1 6 -4 3 2 -30 -9 8 Table 3 CA-ID-TIMS U-Pb analytical data for zircons from SCS granitic intrusions Fraction Weight Concentration Measured" (mg) U Pb' Pbo 206PbP04pb* 2°SpbP06Pb* (ppm) (ppm) (pg) Villacastin (Y24) ZI 10 Xtls. CA 0.04 328 17.7 61.0 493 0.1260 Z2 5 Xtls. CA 0.02 102 5.18 9.5 361 0.1015 Z3 15 SM xtls. CA 0.07 303 15.3 41.0 1,022 0.1126 Z4 10 SM xtls. CA 0.03 170.84 8.7 22 540 0.1188 Atalaya Real (Y78) ZI 3 xtls. CA 0.04 168 9.9 49.9 3 1 1 0.1340 Z2 15 SM xtls. CA 0.05 288 14.0 14.6 2,147 0.1035 Z3 6 Xtls. CA 0.03 210 10.8 15.3 1,015 0.1185 Z4 14 Xtls. CA 0.04 260 14.4 49.5 454 0.1280 Z5 6 Xtls. CA 0.05 245 12.5 20.3 1,341 0. 1240 Corrected atomic ratios 2°tpbP38U ± er 0.04769 ± 0.42 0.04784 ± 0.32 0.04773 ± 0.22 0.0475 ± 0.43 0.04857 ± 0.17 0.04855 ± 0.34 0.04862 ± 0.20 0.04838 ± 0.24 0.04860 ± 0.29 207pbP35U ± er 0.34418 ± 2.42 0.34531 ± 0.92 0.34436 ± 1.02 0.34270 ± 0.77 0.35159 ± 0.49 0.35124 ± 0.44 0.35216 ± 0.30 0.34991 ± 0.62 0.35151 ± 1.50 P 0.18 0.48 0.39 0.63 0.46 0.80 0.69 0.41 0.44 207PbP06Pb ± er 0.05234 ± 2.38 0.05235 ± 0.82 0.05233 ± 0.95 0.052326 ± 0.60 0.05250 ± 0.44 0.05247 ± 0.27 0.05253 ± 0.22 0.05246 ± 0.57 0.05245 ± 1.40 Age (Ma) 2°"pb;''''u 207pbP35U 300.3 300.3 301.2 301.2 300.5 300.5 299.2 299.2 305.7 305.9 305.6 305.7 306.0 306.3 304.6 304.7 305.9 305.9 Xtls euhedral zircon prisms, 1:3-1:5 widthllength ratio. SM small « 100 mm), CA chemically abraded (Mattison 2005). Weight estimated before CA. Pb (pg): total Pb blank 207PbP06Pb 300.4 300.9 299.9 299.7 307.2 306.0 308.7 305.3 305.2 " Measured ratio corrected for blank and fractionation. Atomic ratios corrected for fractionation (0.11 ± 0.02% AMU Pb; 0.10 ± 0.02% AMU U), spike e08Pb-235U) laboratory blanks (6 pg Pb; 0.1 pg U) and initial common Pb after Stacey and Kramers (1975). Errors are at the 2-sigma level. Data reduced with PbMacDat (Isachsen et al. http://www.earth-time.org). Pb* radiogenic lead, Pbc total common lead. p error correlation between 206PbP38U and 207PbP35U ratios a 0_049 :0 0_047 � :0 "" � 0_045 0_043 0.27 0_25 0_049 p t 0.047 R 0_045 e 0.30 0_31 0_29 0_35 0_39 SHRIMP data ALPEDRETE( Y76) 0.33 0_37 �--'IA-ICP-MS data ALPEDRETE( Y76) 0.42 Fig. 5 a, c, e Concordia diagrams showing the SHRIMP and LA­ ICP-MS U-Pb data for Variscan zircons from samples M21 and Y76, except a few analyses with exceedingly high or low concordia ages. displaying high U and four spots with high %discorcance (2.1, 3.1, 12.1 and 13.1). The rest of data (7 analyses) represent ages from 295 to 306 Ma (Fig. 6e). Five of these data were selected and provided a con cordia age of 302.3 ± 2.9 Ma (MSWD = 0.002) (Fig. 6f). This age is consistent with the probability density function, which shows a maximum of the bin distribution towards 302-304 Ma (Fig. 6g). The five fractions analysed by CA-ID-TlMS form a cluster on con cordia between 304 and 307 Ma and pro­ vide a concordia age of 305.68 ± 0.41 Ma (Fig. 7b; 0.051 Concordia age= 299. 1±1.8r---:.l'�---' MSWD=Q .99[ n=141 b 0_049 0.047 0_045 0.30 0_0495 0_0485 0_0475 0_0465 0_0455 0_31 0.32 HOYO DE PINARES (M21) 0_34 0_36 0.38 d SHRIMP data ALPEDRETE (Y76) 0_32 0_33 0_34 0_35 0_36 Concordia age= 301.7±3.4 MSWD=2 .7 [n=6] O_04951--"�=--"--'-'-=--'"�5;%?\'--. f 0_0485 0_0475 0_0465 0_0455 0.30 0.32 LA-/CP-MS data ALPEDRETE (Y76) 0.38 Concordia age plots (b, d, f) in the right consider a restricted group of selected spots. Error ellipses are given at the 2a level MSWD = 0.33), which is considered the most accurate estimate for the age of intrusion. Discussion Zircon composition and Ti-in-zircon thermometry The steep patterns from MREE to HREE normalised values shown by zircons from the sampled SCS granites (Fig. 3a, b), in conjunction with dominant oscillatory zoning, Fig. 6 a, c, e Concordia diagrams showing the SHRIMP U-Pb data for Variscan zircons from samples Y24, Y78 and M9, except a few analyses with exceedingly low concordia ages. Concordia age plots (b, d, f) in the right consider a restricted group of selected spots. g Probability density plot made for sample Y78 (2a errors used for calculation). Error ellipses are given at the 2a level 0.052 �-------�-----, a 0.050 310 0.048 P � � 0.046 � 0.044 0.042 VILLACASTiN (Y24) 0.05 0.15 0.25 0.35 0.45 0.55 0.051 e 32 0.049 � 0.047 il r 0.045 0.043 0.041 0.10 ATALAYA REAL (Y78) 0.2 0.3 0.4 2D7Pb/235U 0.5 5 ,------------------------, ATALAYA REAL (Y78) 9 4 260 280 300 320 340 SHRIMP Age 0.0490 0.0480 0.0470 0.0460 Concordia age= 297.6±2.8 MSWD=O.16[ n=51 b VILLACASTiN (Y24) 0.31 0.33 0.35 0.37 0.051 Concordia age: 300.3±2.6 MSWD=0.59[ n=81 0.050 l�--------��====:f!-� d 0.049 0.048 0.047 0.046 0.29 0.31 0.33 0.35 0.37 0.39 0.050 Concordia age: 302.3±2.9 MSWD=O.OO2 [n=5] f 0.049 0.048 0.047 0.046 ATALAYA REAL (Y78) 0.315 0.375 evidence the igneous ongm of Variscan zircons. The anomalous high LREE contents in 3 spots: Y24-02, Y24-09 and M9-11 (Fig. 3a), which have been analysed in dark bands from oscillatory-zoned domains, are likely associ­ ated with secondary modifications at late magmatic stages. Preferential concentration of LREE, U, Th and other large cations has been described in zircons with similar CL textures and explained as alteration phenomena during magma cooling (pidgeon et a1. 1998). The trace element variability of Variscan zircons (Figs. 3, 4) suggests the possibility of crystallisation from several magmas, an evolving melt or a combination of both Fig. 7 Concordia diagrams showing the TIMS U-Pb data for Variscan zircons from samples Y24 a and Y78 Concordia age = 300.5 ± 0.55 Concordia age = 305.68 ± 0.41 f-__ , 0.0481 MSWD = 0.0065 [n=4] MSWD = 0.33 [n=5] b b. Error ellipses are given at the 26 level � it. If � 0.0479 0.0477 0.0475 0.0473 a 0.0487 0.0485 0.0483 VILLACASTiN (Y24) 0.0481 AT ALAYA REAL (Y78) 0.336 0.344 0.352 0.360 0.345 0.349 0.353 0.357 0.361 207PbP5U factors. Significant intra-mineral chemical heterogeneity between sectors separated by resorption surfaces has been described in magmatic zircons and ascribed to Zr under­ saturation during recharge with hotter ascending magma (Belousova et a1. 2006; Claiborne et a1. 2006; Gagnevin et a1. 2010). These abrupt chauges in trace element con­ centrations, however, can coexist with a long-term inter­ mineral chemical variation related to melt evolution (e.g. Claiborne et a1. 2006). Variscau zircons aualysed for the present study show textural features in agreement with resorption stages during magma cooling (e.g. Fig. 2c, grain 7; Fig. 2e, grain 18). But the spots selected for laser ablation analysis do not allow the discussion of intra­ mineral chemistry and magma mixing processes. Anyway, the trace element contents of SCS zircons are characteristic of grauitoid rocks (Nb = 1-17 ppm, Ta = 0.32-12 ppm aud EulEu* < 0.3) (Belousova et a1. 2002) aud exhibit crustal signatures (Zr/Hf = 32-58) (pupin 2000), so mix­ ing with felsic magmas is preferred rather than implication of mantle-derived melts. The degree of magma differentiation can correlate positively with the abundance of Hf in igneous zircon (Hoskin and Schaltegger 2003 aud references therein). Experiments by Linnen and Keppler (2002) concluded that this effect is related to modifications of the Zr/Hf ratio during crystal fractionation of granitic melts. SCS grauite zircons display a positive correlation between Hf and V-Nb (Ta) (Fig. 3b, c), whilst Zr/Hf ratio decreases (Fig. 4). These characteristics are in accordance with a progressive increase in the degree of magma evolution and an incom­ patible behaviour of U, Nb and Ta with respect to the fractionating mineral assemblage. However, the response shown by other trace elements is fairly variable and dependent on the intrusion considered, pointing to influ­ ence of co-precipitating trace element-rich accessory pha­ ses. These minerals would withdraw highly incompatible elements and compensate for their enrichment in the melt during differentiation. Common accessory phases in these 207PbP5U rocks are REE-rich phosphates (apatite, monazite aud xenotime). Their involvement during crystal fractionation can be envisaged in S-type granites zircons [Alpedrete (Y76) aud Hoyo de Pinares (M21)] on the basis of the lack of P increase with higher Hf (Fig. 4). These intrusions do not show significant trends for REE, Y and Th, thus sup­ porting that monazite and xenotime are likely the main accessory minerals associated with magma differentiation. On the other haud, P, Y aud HREE from Villacastfn (Y24) aud Atalaya Real (Y78) zircons are positively correlated with Hf (Fig. 4). This characteristic is likely related to the "xenotime" coupled substitution mechanism in zircon (e.g. Hoskin and Schaltegger 2003), which adds to the effect caused by magma evolution. However, LREE contents in these zircons do not correlate significantly with the degree of differentiation (Fig. 4). Fractionation of a LREE-rich silicate, such as allanite, might be responsible for this behaviour. Monazite and allanite are the predominant frac­ tionating LREE-rich minerals in the SCS S- aud I-type granitoids, respectively (Villaseca et a1. 1998). This agrees well with the proposed implication of these accessory phases during magma fractionation. Moreover, a recent study by Perez-Soba et a1. (2007) has highlighted a similar influence of accessory minerals in the zircon chemistry of several granitic intrusions of eastern SCS (including Alpedrete aud Atalaya Real). The fact that Navas del Marques zircons present a general increase in all trace elements towards higher Hf or V supports that no accessory minerals have significautly modified zircon chemistry during its growth. The analysed pre-Variscan zircons, irrespective of their crystallisation age, show a trace element composition overlapping that of the Variscan grains (Figs. 3, 4; Table 1). This similarity indicates that the inherited zircons likely originated from a granitic melt, which would be in accordance with the predominance of felsic peraluminous magrnatism occurring in the ClZ during Lower Palaeozoic (e.g. Bea et a1. 2003; Zeck et a1. 2004; Montero et a1. 2009). 1000 Y24 M21 M9 Y76 Y78 900 0 { + t t "--- i" 800 " lii Q) 700 CL E � 600 0 I-type granites • S-type granites 500 -'-_________________ -' Fig. 8 Mean temperatures of the five S- and I-type granite samples calculated using the recalibrated Ti-in-zircon geothennometer of Ferry and Watson (2007). The grey bar represents the error associated with each data. Fairly similar high temperatures have been obtained for either S- or I-type granites We have estimated the temperature of crystallisation of each intrusion on the basis of Ti concentrations using the recalibrated Ti-in-zircon equation of Ferry and Watson (2007). Due to absence of rutile and presence of ilmenite in these granites, we have assnrned an a(Ti02) value of 0.6, typical of silicic melts (Ferry and Watson 2007). The cal­ culated mean temperatures are quite similar in S-type granites: 844°C (Royo de Pinares) and 788°C (Alpedrete), and I-type granites: 784°C (Villacastin), 787°C (Atalaya Real) and 804°C (Navas del Marques) (Fig. 8). Tempera­ ture subestimation or overestimation in the order of ,......, 40°C would result from the application of a(Ti02) = 1 or 0.5, respectively. The T range obtained for the SCS granite zircons (784-844°C) represents relatively elevated tem­ peratures in all cases (Fig. 8). These temperatures are equivalent to those proposed for "hot" granites (e.g. Miller et a1. 2003) and are indicative of a deep level of melting, likely within the lower crust. The low number of zircon inheritances preserved in the five samples is in agreement with the above high-T data. Constraints on the protolith nature Only 14 pre-Variscan ages have been recorded in several zircon cores from 4 of the analysed samples, defining a total age range of c. 369-1,956 Ma (Table 2). Although this dataset is very limited and should be taken with cau­ tion, it is interesting to note that two groups of ages might be distinguished: (1) six analyses define a fairly restricted cluster of 462-479 Ma and (2) seven inherited cores recorded ages older than 564 Ma (1,956-564 Ma). The age of 369 Ma (sample M21) from an inherited core with convoluted zoning is younger than any possible source rock in the region (either magmatic or sedimentary), so it is considered to be the result of secondary processes. The 12 10 8 6 4 2 6 4 SCS 2 6 4 2 400 600 SCS granulite 800 1000 Zircon age (Ma) 1800 a b c 2000 Fig. 9 Relative probability plots of pre-Variscan ages registered in zircons from SCS granitic intrusions (grey fields). Similar density curves with U-Pb zircon ages from a SCS granulite xenoliths, b SCS migmatites derived from metaigneous protoliths (Sotosalbos samples; Castifieiras et al. 2008) and c Anatectic Complex of Toledo (ACT) migmatites derived from metasedimentary protoliths (Castifieiras et al. 2008), have also been plotted for comparison. See that the inheritances in the SCS granites match better the curves of igneous­ derived rocks a and b density curve that represents these data has been plotted in Fig. 9, in comparison with inherited ages from deep-seated SCS granulite xenoliths and outcropping metamorphic rocks from central Spain. The six Ordovician ages are all recorded in I-type granites (Villacastin and Navas del Marques). These two samples also show Neoproterozoic (564 Ma) and lower Proterozoic (1,956 Ma) inheritances, but both data repre­ sent texturally discordant cores within Ordovician zircon domains (Fig. 2c, grain 20; Fig. 2e, grain 5). S-type gran­ ites (Royo de Pinares and Alpedrete) present pre-Variscan cores older than 580 Ma (828-581 Ma). This clear contrast in the age of inheritances suggests that two distinct source rocks were involved in the genesis of the two SCS granitic series. Different sources for SCS S- and I -type granites were not obvious using classical geochemical approaches (Sr, Nd, 0, Pb isotopes) (e.g. Villaseca et a1. 1998, 2009; Villaseca and Rerreros 2000). Thus, the study of zircon inheritances by U-Pb geochronology is a powerful tool in discriminating igneous protoliths. D CIZ granites (n=59) D Granites from the SCS Batholith (n=9) Late-/post -Variscan t::! granites of Southern Variscides 0 8 � :'" " " • c W .,. � • Cl> 0 • "' -� '" • " i" u.. • .� E 6 '" " (5 .c "- 0 , -1;5 0 0 ", 4 % . .g)