TY - JOUR
AU - Rohrmüller, Johann
AB - Abstract Cenozoic primitive basanites, nephelinites and melilitites from the Heldburg region, SE Germany, are high-MgO magmas (8·5–14·1 wt % MgO), with low SiO2 (34·2–47·1 wt %) and low to moderately high Al2O3 (9·0–15·5 wt %) and CaO (8·7–12·7 wt %). The Ni and Cr contents of most samples are up to 470 ppm and 640 ppm, respectively, and match those inferred for primary melts. In multi-element diagrams, all samples are highly enriched in incompatible trace elements with chondrite-normalised La/Yb = 19–45, strongly depleted in Rb and K, with primitive mantle normalised K/La = 0·15–0·72, and moderately depleted in Pb. The initial Sr–Nd–Hf isotope compositions (87Sr/86Sr = 0·7033–0·7051, 143Nd/144Nd = 0·51279–0·51288 and 176Hf/177Hf = 0·28284–0·28294) fall within the range observed for other Tertiary volcanic rocks of the Central European Volcanic Province, whereas 208Pb/204Pb and 206Pb/204Pb (38·42–38·88 and 18·49–18·98) are distinctly lower at comparable 207Pb/204Pb (15·60–15·65). Trace element modelling and pressure–temperature estimates based on major element compositions and experimental data suggest that the nephelinites/melilitites formed within the lowermost lithospheric mantle, close to the lithosphere–asthenosphere boundary, by ∼3–5% partial melting of a highly enriched, metasomatised, carbonated phlogopite-bearing garnet–lherzolite at temperatures <1250 °C and pressures of ∼2·8 GPa. This corresponds to a melting depth of less than ∼85 km. Formation and eruption of these magmas, based on 40Ar/39Ar dating, started in the late Eocene (38·0 Ma) and lasted until the late Oligocene (25·4 Ma). Basanite eruptions occurred in the same area in the middle Miocene, about 7·7 Myr after nephelinite/melilitite generation has ceased, and lasted from 17·7 to 13·1 Ma. The basanites were generated at lower pressures (2·2–1·7 GPa) at similar temperatures (∼1220–1250 °C) within the spinel stability field in the lithospheric mantle by 2–6% partial melting. Isotope and trace element systematics indicate that the lithospheric mantle source of the Heldburg magmas was affected by metasomatism associated with long-lasting subduction of oceanic and continental crust during the Variscan orogeny. Aqueous or supercritical fluids that formed at temperatures <1000 °C and pressures of likely >4 GPa infiltrated the thermal boundary layer at the base of the lithospheric mantle and imprinted a crustal lead isotope, and to a minor extent crustal Sr, Nd and Hf isotope signatures. They also reduced Nb/U, Ce/Pb, Lu/Hf, Sm/Nd, U/Pb and Th/Pb, but increased Rb/Sr and Nb/Ta and amplified the enrichment of LILE and LREE relative to HREE. This lead to the highly-enriched trace element patterns observed in both sample suites, and to overall less radiogenic 206Pb/204Pb and 208Pb/204Pb compared to other continental basalts in Central Europe, and to less radiogenic 176Hf/177Hf and 143Nd/144Nd that plot distinctly below the terrestrial mantle array. Temporal evolution of magmatism in the Heldburg region coincides with the changing Tertiary intraplate stress field in Central Europe, which developed in response to the Alpine orogeny. Magmatism was most probably caused in response to lithosphere deformation and perturbation of the thermal boundary layer, and not by actively upwelling asthenosphere. INTRODUCTION The volume and composition of magmas erupted in response to continental rifting varies as a function of the lithospheric thickness, the temperature in the lithospheric and underlying asthenospheric mantle, and the mineralogical and chemical composition of the melt source region(s). The principal mechanisms that control the first order chemical composition of intraplate volcanic rocks, such as depth and degree of melting, fractional crystallisation and crustal assimilation processes are relatively well understood (e.g. DePaolo, 1981; McKenzie & Bickle, 1988; Wilson & Downes, 1991; Spera & Bohrson, 2001), but controversy still exists about the composition, origin and evolution of the distinct source regions involved in magma genesis (Wilson et al., 1995a; Jung et al., 2005; Wilson & Downes, 2006; Lustrino & Wilson, 2007; Pfänder et al., 2012; Lustrino & Anderson, 2015; Ulrych et al., 2016). The Heldburg region is part of the Tertiary Central European Volcanic Province (CEVP; Fig. 1a), whose activity is associated with the European Continental Rift System (ECRIS; e.g. Ziegler & Dèzes, 2006). Major rift zones and associated volcanic centres are the Limagne and Bresse grabens within the Massif Central in France, the Upper and Lower Rhine grabens with the Kaiserstuhl, Eifel, Siebengebirge, Westerwald, Vogelsberg, Hessian depression and Rhön volcanic areas in Germany, and the Eger graben with the volcanic areas of the Bohemian Massif in the Czech Republic and Poland (Fig. 1a). The Heldburg area, along with the Urach and Hegau volcanic fields in Southern Germany, has an exceptional position within the German part of the CEVP, as these volcanic areas are located more distant from the major rift axes (Fig. 1a). Furthermore, the Heldburg area is located between the prominent volcanic centres of the Vogelsberg and Rhön areas to the NW, and the volcanic centres of the Upper Palatinate to the east (Fig. 1a). It therefore might provide a link between magmatism in the northern part of the Upper Rhine graben and magmatism in the southwestern part of the Eger graben. Fig. 1. View largeDownload slide (a) Location of the Heldburg area within the Central European Volcanic Province (black, volcanic centres; grey, major rift zones). Numbers in parentheses next to location names denote the age of magmatic activity in Ma (K–Ar ages from Lippolt 1982; map modified from Wedepohl et al., 1994). Area limited by dashed lines denotes the Saxothuringian Zone of the Variscan Orogen (SAX). Thick dashed line denotes the Franconian Lineament (FL). (b) Geological map of the study area with sample locations (yellow asterisks) and 40Ar/39Ar ages in Ma (±1σ) (map adopted from the Geologische Übersichtskarte 1: 200 000, sheet CC 6326 Bamberg, Bundesanstalt für Geowissenschaften und Rohstoffe, Hannover, Bundesrepublik Deutschland, 1994). Fig. 1. View largeDownload slide (a) Location of the Heldburg area within the Central European Volcanic Province (black, volcanic centres; grey, major rift zones). Numbers in parentheses next to location names denote the age of magmatic activity in Ma (K–Ar ages from Lippolt 1982; map modified from Wedepohl et al., 1994). Area limited by dashed lines denotes the Saxothuringian Zone of the Variscan Orogen (SAX). Thick dashed line denotes the Franconian Lineament (FL). (b) Geological map of the study area with sample locations (yellow asterisks) and 40Ar/39Ar ages in Ma (±1σ) (map adopted from the Geologische Übersichtskarte 1: 200 000, sheet CC 6326 Bamberg, Bundesanstalt für Geowissenschaften und Rohstoffe, Hannover, Bundesrepublik Deutschland, 1994). Basanites and alkali basalts dominate the CEVP, highly SiO2-undersaturated lithologies (nephelinites and in particular melilitites) are more scarce but prevail in some regions (e.g. Tertiary Eifel, Urach, Hegau; Fig. 1a). Differentiated to felsic lithologies also occur, mostly in bimodal associations and less frequently as comprehensive differentiation series such as in the Cantal (Massif Central; Wilson et al., 1995a), Siebengebirge (Jung et al., 2012; Kolb et al., 2012; Schneider et al., 2016), Eifel (Jung et al., 2006) and Rhön volcanic regions (Jung et al., 2013), as well as in the České Středohoří volcanic complex in the Eger graben (Ackerman et al., 2015). Voluminous tholeiitic series that underwent differentiation and crustal assimilation within magmatic plumbing systems occur in the large volcanic complex of the Vogelsberg and within the Hessian depression in Central Germany (Jung & Masberg, 1998; Wedepohl, 2000; Bogaard & Wörner, 2003). A variety of distinct asthenospheric and lithospheric mantle sources have been invoked to play a role during magma genesis within the CEVP. Several authors favour asthenospheric melting in response to deep rooted large-scale upwelling mantle plumes (e.g. Hoernle et al., 1995; Goes et al., 2000; Ritter et al., 2001; Buikin et al., 2005), others suggest more local, small-scale diapiric upwellings (plumelets) as the source of melt formation and a more pivotal role of lithosphere extension (e.g. Wilson & Downes, 1991; Haase & Renno, 2008; Kolb et al., 2012). A few studies suggest that intraplate volcanic rocks are solely derived from the (metasomatised) lithospheric mantle (Gallagher & Hawkesworth, 1992), either by melting of pyroxene- and/or amphibole-bearing veins with a lower solidus than the ambient mantle (Jung et al., 2005; Pilet et al., 2008), or by low degree melting of highly metasomatised domains to form strongly SiO2-undersaturated melilitites (Hegner et al., 1995; Wilson et al., 1995b; Blusztajn & Hegner, 2002). First-order chemical and isotopic similarities such as overall enriched trace element patterns, mostly unradiogenic 87Sr/86Sr, but radiogenic 143Nd/144Nd and 206Pb/204Pb, have lead to the suggestion of a common asthenospheric mantle source for the CEVP magmas, the European Asthenospheric Reservoir (EAR; Cebria & Wilson, 1995). Deviations and compositional variations from the inferred EAR composition are commonly explained as the result of mixing of EAR-derived melts with melts from heterogeneous lithospheric mantle sources. A variety of chemically distinct lithospheric components has been suggested (amphibole-rich veins, phlogopite-rich domains, carbonatite impregnations) that may either melt directly (Blusztajn & Hegner, 2002; Wilson et al., 1995b) or serve as assimilants (Jung et al., 2005; Pfänder et al., 2012). The heterogeneous lithospheric composition is thought to originate from earlier silicate/carbonate melt or fluid-related metasomatism (Hegner & Vennemann, 1997; Jung et al., 2005; Pilet et al., 2008; Pfänder et al., 2012; Ackerman et al., 2013; Brandl et al., 2015; Puziewicz et al., 2015), tentatively related to subduction processes (Wilson & Downes, 1991; Hegner et al., 1995; Hegner & Vennemann, 1997; Lustrino & Wilson, 2007; Lustrino & Anderson, 2015). Besides the ongoing discussion on potential melt source regions and their origin, and the relatively well known petrogenetic evolution of the Cenozoic magmas of the CEVP, the temporal evolution of individual volcanic regions is known in much less detail. This is due to the much smaller number of available geochronological data compared to geochemical and petrological data. Most available age data are K–Ar ages that have been published more than 25 years ago, and these may be biased due to sample alteration, argon loss or the presence of excess argon. Finally, datasets that include geochemical and geochronological data for the same samples are rare, and thus the relation of petrogenetic processes to the geodynamic framework remains enigmatic in most cases. In this study, we combine high-resolution 40Ar/39Ar dating of a series of SiO2-undersaturated volcanic rocks from the Heldburg area (Fig. 1a and c) with trace element and isotope investigations in order to constrain the origin and evolution of primary magmas and their sources in intraplate settings, and to relate this information to the time and duration of their formation. Investigating the type and origin of magma, the degree of melting and the source composition as a function of time, indirectly provides information on the dynamics of the lithospheric and asthenospheric mantle in such settings. Our new data improve the age resolution of volcanic activity in the Heldburg region where the nephelinites/melilitites are significantly older than the alkali basalts/basanites. We show that both rock suites have been derived from a similar, strongly metasomatised and hence highly enriched, lithospheric mantle sources at different pressures. GEOLOGICAL SETTING The Heldburger Gangschar covers an area of ∼1500 km2 SE of the Rhön volcanic region and is located SW of the Franconian Lineament within the Saxothuringian zone of the Variscan orogen (Fig. 1a). This orogen formed during the late Paleozoic due to the collision of the converging plates of Gondwana and Laurussia. Spatially and temporally variable subduction–accretion processes, including the subduction of continental crust after the closure of the Rheic Ocean, led to the amalgamation of heterogeneous lithospheric blocks (Kroner & Romer, 2013). Numerous Tertiary basaltic plugs and predominantly NNE striking sub-parallel mafic dike swarms cross-cut the Mesozoic cover of the Saxothuringian zone in the Heldburg area (Fig. 1b). The distance between individual dikes varies between 0·2 to 2 km; the dikes have a length of up to several km, and a thickness in the order of 1 m (Carlé, 1952; Schröder, 1966; Tanyeri, 2006). In contrast to several other Tertiary volcanic areas of the CEVP, the Heldburg region lacks voluminous and areally extensive intrusive complexes or layers of volcanic rocks. Erupted magma volumes in this region were likely dominated by aligned scoria cones associated with local fissure eruptions (Schröder, 1966), similar to the Quaternary West Eifel volcanic field. The majority of volcanic rocks are alkali basalts, basanites, olivine nephelinites, and melilitite-olivine nephelinites (Huckenholz & Schröder, 1981; Huckenholz & Werner, 1990). They either occur as massive rocks or as volcanic tuffs and breccias, associated with wall-rock fragments. Differentiated lithologies are absent, with the exception of the prominent Heldburg phonolite plug, a subvolcanic intrusion containing abundant mantle xenocrysts and micro-xenoliths (Grant et al., 2013; Abratis et al., 2015). Other prominent edifices are the Zeilberg (nepheline basanite) in the south and the Gleichberg and Steinsburg in the north of the Heldburg region (Fig. 1b). Available K–Ar ages for the Heldburg area span a range of 42 to 16 Ma, suggesting relatively long-lasting magmatic activity in this region with an older phase from ∼ 42 to ∼24 Ma that produced more SiO2-undersaturated magmas, and a younger phase between ∼16 and ∼14 Ma that produced less SiO2-undersaturated melts (Pohl & Soffel, 1977; Abratis et al., 2007; Hofbauer, 2007). The age of the Heldburg phonolite was determined by K–Ar dating to 11 Ma (Lippolt, 1982) and 13·3 Ma (Kaiser & Pilot, 1986). A more recent 40Ar/39Ar investigation provided a more precise and slightly older age of 15·0 ± 0·2 Ma for the phonolite (Abratis et al., 2015), and documented substantial Ar-loss of the groundmass, which might explain the younger K–Ar ages. The Central European lithosphere was in isostatic equilibrium with the asthenosphere at the end of the Mesozoic and lithosphere thicknesses were ∼100–120 km for internal parts of the Variscan orogen (Ziegler et al., 2004). The present-day lithosphere thickness in Central Europe varies between 70–80 and ∼90–110 km (Geissler et al., 2010) and the overall extension accommodated by the Central European Rift System does not exceed 5–7 km in its central parts (Dèzes et al., 2004; Ziegler & Dèzes, 2006). The lowest values of lithosphere thickness occur along the NNE-trending Upper Rhine graben (Fig. 1a). From there, thickness increases to the east, reaching about ∼90–100 km in the Heldburg area (Geissler et al., 2010). The Moho discontinuity beneath the Heldburg region is only slightly deeper than beneath the Rhine Graben, about 28–32 km, but deepens towards the east beneath the Bohemian Massif (Ziegler et al., 2004; Ziegler & Dèzes, 2006). SAMPLES AND ANALYTICAL METHODS Nineteen samples from an existing sample collection of the Bayerisches Landesamt für Umwelt (in part described in Huckenholz & Schröder, 1981), complemented by our own field sampling, were investigated. Sample locations are shown in Fig. 1b and listed in Supplementary DataTable S1; supplementary data are available for downloading at http://www.petrology.oxfordjournals.org. The areal extent of outcrops in the Heldburg region is limited and, in part, samples have been collected as rubble stones from the field, based on geological maps, locally supported by geomagnetic investigations. Macroscopically the majority of samples are moderately to strongly altered. Therefore, after crushing, the least altered fragments from each sample were selected, further crushed and finally ground in an agate mill for chemical and isotopic analyses. Major and some trace element concentrations (Table 1) were determined on fused lithium–tetraborate glass beads using standard X-ray fluorescence (XRF) techniques (Vogel & Kuipers, 1987) and a Pananalytical Magix PRO X-ray fluorescence spectrometer at the Mineralogisch-Petrographisches Institut, Universität Hamburg, Germany. Loss on ignition (LOI) was determined gravimetrically after fusion of a sample aliquot at 1050 °C (Lechler & Desilets, 1987). The uncertainty for major element contents are 1–2%, and 5–20% for trace elements, depending on concentration. Table 1: Major element, trace element and isotope data of SiO2-undersaturated rocks from the Heldburg region Sample Method S097511 S097512 S097519 S107048 S107052 S097513 S097514 S097515 S107045 Rock type Mel Mel Mel Mel Mel Neph Neph Neph Neph SiO2 XRF 35·5 34·9 35·4 34·5 34·2 38·4 41·0 41·3 40·0 Al2O3 XRF 9·00 9·42 9·09 9·60 9·52 9·6 11·1 11·1 10·3 Fe2O3 XRF 12·9 12·2 11·8 13·4 11·6 12·0 12·6 13·1 13·3 FeO 11·6 10·9 10·6 12·0 10·4 10·8 11·4 11·8 11·9 MnO XRF 0·20 0·18 0·17 0·18 0·17 0·17 0·20 0·20 0·20 MgO XRF 13·3 12·8 14·0 13·3 14·1 13·6 12·3 12·6 13·7 CaO XRF 12·0 11·6 12·5 11·6 12·7 11·2 11·8 11·3 11·9 Na2O XRF 3·39 3·16 3·47 3·11 3·82 2·92 4·04 3·44 3·26 K2O XRF 1·35 1·25 0·98 1·16 0·74 0·79 1·08 1·14 0·95 TiO2 XRF 2·85 2·84 2·67 2·90 2·53 2·80 2·84 2·94 2·89 P2O5 XRF 1·00 0·99 0·84 0·76 0·82 0·61 0·82 0·77 0·86 SO3 XRF 1·11 0·84 0·71 0·28 0·46 0·31 0·03 0·05 0·06 L·O·I XRF 7·60 9·86 8·94 9·48 9·92 8·42 2·04 2·12 2·96 SUM XRF 100·3 99·9 100·5 100·3 100·6 100·8 99·9 100·1 100·3 Li ICPMS 36·9 31·6 45·2 15·5 54·6 20·0 41·5 21·1 27·2 Sc ICPMS 20·7 20·1 22·3 21·9 26·2 21·3 21·9 24·8 24·3 V XRF 265 265 228 272 246 262 279 281 289 Cr XRF 494 458 409 367 539 451 476 527 643 Co ICPMS 72·8 53·5 66·5 70·0 61·0 74·4 64·9 69·0 74·6 Ni XRF 424 206 354 309 379 385 312 325 468 Cu XRF 188 254 133 146 144 185 113 158 178 Zn XRF 173 193 117 137 104 163 129 163 185 Ga ICPMS 16·4 17·8 15·9 18·0 16·2 16·6 17·8 18·0 17·1 Rb ICPMS 33·1 42·4 25·0 30·8 22·1 14·7 18·0 32·1 16·3 Sr XRF 1361 1104 959 815 988 807 1105 1083 982 Y ICPMS 29·0 30·0 23·5 23·4 23·5 21·1 27·3 26·9 25·2 Zr XRF 321 307 266 299 225 230 264 264 286 Nb ICPMS 106·3 117·6 91·1 96·5 99·9 72·5 97·5 91·7 92·3 Cs ICPMS 0·677 0·485 0·392 0·302 0·299 2·50 1·32 1·06 0·976 Ba XRF 828 778 797 2961 719 657 870 828 788 La ICPMS 109·2 109·3 80·5 75·3 77·2 62·4 96·0 89·1 81·6 Ce ICPMS 208 208 146 147 141 125 182 168 154 Pr ICPMS 23·8 23·9 16·8 16·8 15·3 15·1 20·6 19·2 17·7 Nd ICPMS 88·5 88·3 65·5 63·9 55·6 59·1 76·6 73·0 67·0 Sm ICPMS 14·3 14·5 11·1 11·2 10·4 9·8 12·5 12·1 11·2 Eu ICPMS 4·47 4·55 3·54 3·53 3·24 3·25 3·97 3·88 3·53 Gd ICPMS 11·88 12·18 9·48 9·31 8·44 8·90 10·58 10·36 9·45 Tb ICPMS 1·55 1·60 1·23 1·23 1·08 1·18 1·42 1·36 1·20 Dy ICPMS 8·09 8·50 6·45 6·35 5·81 6·15 7·51 7·24 6·48 Ho ICPMS 1·32 1·40 1·09 1·02 0·969 1·02 1·25 1·24 1·08 Er ICPMS 3·09 3·31 2·54 2·48 2·35 2·36 3·11 2·98 2·63 Tm ICPMS 0·362 0·406 0·301 0·274 0·279 0·286 0·379 0·358 0·331 Yb ICPMS 1·73 1·89 1·48 1·40 1·54 1·32 1·84 1·78 1·67 Lu ICPMS 0·238 0·262 0·199 0·188 0·214 0·180 0·252 0·245 0·238 Hf ICPMS 7·11 7·17 6·18 7·03 5·46 5·62 6·10 6·06 6·44 Ta ICPMS 5·57 5·92 4·72 5·31 4·83 4·11 4·92 4·79 5·05 Pb ICPMS 13·36 11·99 8·72 6·73 4·89 8·20 7·75 7·93 8·77 Th ICPMS 14·02 14·00 10·30 8·26 9·34 7·74 12·40 11·07 9·40 U ICPMS 4·01 3·27 3·29 2·16 1·41 2·36 3·69 3·33 2·93 (206Pb/204Pb)m 18·72 18·54 18·87 18·57 18·92 18·56 18·95 18·67 18·49 (206Pb/204Pb)i 18·61 18·45 18·73 18·47 18·82 18·47 18·77 18·55 18·41 (207Pb/204Pb)m 15·62 15·60 15·61 15·61 15·61 15·62 15·64 15·61 15·62 (207Pb/204Pb)i 15·61 15·60 15·60 15·61 15·61 15·62 15·63 15·60 15·62 (208Pb/204Pb)m 38·65 38·42 38·77 38·49 38·78 38·49 38·88 38·57 38·42 (208Pb/204Pb)i 38·52 38·29 38·62 38·36 38·57 38·40 38·68 38·44 38·33 (176Hf/177Hf)m 0·282881±4 0·282879±4 0·282894±5 0·282868±5 0·282922±4 – 0·282858±7 0·282867±6 0·282917±6 (176Hf/177Hf)i 0·282878 0·282876 0·282891 0·282865 0·282919 0·282854 0·282864 0·282914 eHf(0) 3·9 3·8 4·3 3·4 5·3 – 3·0 3·4 5·1 eHf(i) 3·7 3·7 4·2 3·3 5·2 2·9 3·3 5·0 (143Nd/144Nd)m 0·512858±7 0·512846±7 0·512866±10 0·512856±10 0·512879±8 – 0·512838±9 0·512844±8 0·512881±8 (143Nd/144Nd)i 0·512839 0·512825 0·512841 0·512834 0·512855 – 0·512813 0·512824 0·512864 eNd(0) 4·3 4·1 4·5 4·2 4·7 – 3·9 4·0 4·7 eNd(i) 4·7 4·5 4·9 4·6 5·1 4·4 4·4 5·0 (87Sr/86Sr)m 0·705187±12 0·704482±11 0·703682±13 0·703687±13 0·703970±32 – 0·703686±11 0·703740±11 0·703517±21 (87Sr/86Sr)i 0·70515 0·70443 0·70364 0·703637 0·703939 0·70366 0·70370 0·70350 Sample Method S097511 S097512 S097519 S107048 S107052 S097513 S097514 S097515 S107045 Rock type Mel Mel Mel Mel Mel Neph Neph Neph Neph SiO2 XRF 35·5 34·9 35·4 34·5 34·2 38·4 41·0 41·3 40·0 Al2O3 XRF 9·00 9·42 9·09 9·60 9·52 9·6 11·1 11·1 10·3 Fe2O3 XRF 12·9 12·2 11·8 13·4 11·6 12·0 12·6 13·1 13·3 FeO 11·6 10·9 10·6 12·0 10·4 10·8 11·4 11·8 11·9 MnO XRF 0·20 0·18 0·17 0·18 0·17 0·17 0·20 0·20 0·20 MgO XRF 13·3 12·8 14·0 13·3 14·1 13·6 12·3 12·6 13·7 CaO XRF 12·0 11·6 12·5 11·6 12·7 11·2 11·8 11·3 11·9 Na2O XRF 3·39 3·16 3·47 3·11 3·82 2·92 4·04 3·44 3·26 K2O XRF 1·35 1·25 0·98 1·16 0·74 0·79 1·08 1·14 0·95 TiO2 XRF 2·85 2·84 2·67 2·90 2·53 2·80 2·84 2·94 2·89 P2O5 XRF 1·00 0·99 0·84 0·76 0·82 0·61 0·82 0·77 0·86 SO3 XRF 1·11 0·84 0·71 0·28 0·46 0·31 0·03 0·05 0·06 L·O·I XRF 7·60 9·86 8·94 9·48 9·92 8·42 2·04 2·12 2·96 SUM XRF 100·3 99·9 100·5 100·3 100·6 100·8 99·9 100·1 100·3 Li ICPMS 36·9 31·6 45·2 15·5 54·6 20·0 41·5 21·1 27·2 Sc ICPMS 20·7 20·1 22·3 21·9 26·2 21·3 21·9 24·8 24·3 V XRF 265 265 228 272 246 262 279 281 289 Cr XRF 494 458 409 367 539 451 476 527 643 Co ICPMS 72·8 53·5 66·5 70·0 61·0 74·4 64·9 69·0 74·6 Ni XRF 424 206 354 309 379 385 312 325 468 Cu XRF 188 254 133 146 144 185 113 158 178 Zn XRF 173 193 117 137 104 163 129 163 185 Ga ICPMS 16·4 17·8 15·9 18·0 16·2 16·6 17·8 18·0 17·1 Rb ICPMS 33·1 42·4 25·0 30·8 22·1 14·7 18·0 32·1 16·3 Sr XRF 1361 1104 959 815 988 807 1105 1083 982 Y ICPMS 29·0 30·0 23·5 23·4 23·5 21·1 27·3 26·9 25·2 Zr XRF 321 307 266 299 225 230 264 264 286 Nb ICPMS 106·3 117·6 91·1 96·5 99·9 72·5 97·5 91·7 92·3 Cs ICPMS 0·677 0·485 0·392 0·302 0·299 2·50 1·32 1·06 0·976 Ba XRF 828 778 797 2961 719 657 870 828 788 La ICPMS 109·2 109·3 80·5 75·3 77·2 62·4 96·0 89·1 81·6 Ce ICPMS 208 208 146 147 141 125 182 168 154 Pr ICPMS 23·8 23·9 16·8 16·8 15·3 15·1 20·6 19·2 17·7 Nd ICPMS 88·5 88·3 65·5 63·9 55·6 59·1 76·6 73·0 67·0 Sm ICPMS 14·3 14·5 11·1 11·2 10·4 9·8 12·5 12·1 11·2 Eu ICPMS 4·47 4·55 3·54 3·53 3·24 3·25 3·97 3·88 3·53 Gd ICPMS 11·88 12·18 9·48 9·31 8·44 8·90 10·58 10·36 9·45 Tb ICPMS 1·55 1·60 1·23 1·23 1·08 1·18 1·42 1·36 1·20 Dy ICPMS 8·09 8·50 6·45 6·35 5·81 6·15 7·51 7·24 6·48 Ho ICPMS 1·32 1·40 1·09 1·02 0·969 1·02 1·25 1·24 1·08 Er ICPMS 3·09 3·31 2·54 2·48 2·35 2·36 3·11 2·98 2·63 Tm ICPMS 0·362 0·406 0·301 0·274 0·279 0·286 0·379 0·358 0·331 Yb ICPMS 1·73 1·89 1·48 1·40 1·54 1·32 1·84 1·78 1·67 Lu ICPMS 0·238 0·262 0·199 0·188 0·214 0·180 0·252 0·245 0·238 Hf ICPMS 7·11 7·17 6·18 7·03 5·46 5·62 6·10 6·06 6·44 Ta ICPMS 5·57 5·92 4·72 5·31 4·83 4·11 4·92 4·79 5·05 Pb ICPMS 13·36 11·99 8·72 6·73 4·89 8·20 7·75 7·93 8·77 Th ICPMS 14·02 14·00 10·30 8·26 9·34 7·74 12·40 11·07 9·40 U ICPMS 4·01 3·27 3·29 2·16 1·41 2·36 3·69 3·33 2·93 (206Pb/204Pb)m 18·72 18·54 18·87 18·57 18·92 18·56 18·95 18·67 18·49 (206Pb/204Pb)i 18·61 18·45 18·73 18·47 18·82 18·47 18·77 18·55 18·41 (207Pb/204Pb)m 15·62 15·60 15·61 15·61 15·61 15·62 15·64 15·61 15·62 (207Pb/204Pb)i 15·61 15·60 15·60 15·61 15·61 15·62 15·63 15·60 15·62 (208Pb/204Pb)m 38·65 38·42 38·77 38·49 38·78 38·49 38·88 38·57 38·42 (208Pb/204Pb)i 38·52 38·29 38·62 38·36 38·57 38·40 38·68 38·44 38·33 (176Hf/177Hf)m 0·282881±4 0·282879±4 0·282894±5 0·282868±5 0·282922±4 – 0·282858±7 0·282867±6 0·282917±6 (176Hf/177Hf)i 0·282878 0·282876 0·282891 0·282865 0·282919 0·282854 0·282864 0·282914 eHf(0) 3·9 3·8 4·3 3·4 5·3 – 3·0 3·4 5·1 eHf(i) 3·7 3·7 4·2 3·3 5·2 2·9 3·3 5·0 (143Nd/144Nd)m 0·512858±7 0·512846±7 0·512866±10 0·512856±10 0·512879±8 – 0·512838±9 0·512844±8 0·512881±8 (143Nd/144Nd)i 0·512839 0·512825 0·512841 0·512834 0·512855 – 0·512813 0·512824 0·512864 eNd(0) 4·3 4·1 4·5 4·2 4·7 – 3·9 4·0 4·7 eNd(i) 4·7 4·5 4·9 4·6 5·1 4·4 4·4 5·0 (87Sr/86Sr)m 0·705187±12 0·704482±11 0·703682±13 0·703687±13 0·703970±32 – 0·703686±11 0·703740±11 0·703517±21 (87Sr/86Sr)i 0·70515 0·70443 0·70364 0·703637 0·703939 0·70366 0·70370 0·70350 Table 1: Major element, trace element and isotope data of SiO2-undersaturated rocks from the Heldburg region Sample Method S097511 S097512 S097519 S107048 S107052 S097513 S097514 S097515 S107045 Rock type Mel Mel Mel Mel Mel Neph Neph Neph Neph SiO2 XRF 35·5 34·9 35·4 34·5 34·2 38·4 41·0 41·3 40·0 Al2O3 XRF 9·00 9·42 9·09 9·60 9·52 9·6 11·1 11·1 10·3 Fe2O3 XRF 12·9 12·2 11·8 13·4 11·6 12·0 12·6 13·1 13·3 FeO 11·6 10·9 10·6 12·0 10·4 10·8 11·4 11·8 11·9 MnO XRF 0·20 0·18 0·17 0·18 0·17 0·17 0·20 0·20 0·20 MgO XRF 13·3 12·8 14·0 13·3 14·1 13·6 12·3 12·6 13·7 CaO XRF 12·0 11·6 12·5 11·6 12·7 11·2 11·8 11·3 11·9 Na2O XRF 3·39 3·16 3·47 3·11 3·82 2·92 4·04 3·44 3·26 K2O XRF 1·35 1·25 0·98 1·16 0·74 0·79 1·08 1·14 0·95 TiO2 XRF 2·85 2·84 2·67 2·90 2·53 2·80 2·84 2·94 2·89 P2O5 XRF 1·00 0·99 0·84 0·76 0·82 0·61 0·82 0·77 0·86 SO3 XRF 1·11 0·84 0·71 0·28 0·46 0·31 0·03 0·05 0·06 L·O·I XRF 7·60 9·86 8·94 9·48 9·92 8·42 2·04 2·12 2·96 SUM XRF 100·3 99·9 100·5 100·3 100·6 100·8 99·9 100·1 100·3 Li ICPMS 36·9 31·6 45·2 15·5 54·6 20·0 41·5 21·1 27·2 Sc ICPMS 20·7 20·1 22·3 21·9 26·2 21·3 21·9 24·8 24·3 V XRF 265 265 228 272 246 262 279 281 289 Cr XRF 494 458 409 367 539 451 476 527 643 Co ICPMS 72·8 53·5 66·5 70·0 61·0 74·4 64·9 69·0 74·6 Ni XRF 424 206 354 309 379 385 312 325 468 Cu XRF 188 254 133 146 144 185 113 158 178 Zn XRF 173 193 117 137 104 163 129 163 185 Ga ICPMS 16·4 17·8 15·9 18·0 16·2 16·6 17·8 18·0 17·1 Rb ICPMS 33·1 42·4 25·0 30·8 22·1 14·7 18·0 32·1 16·3 Sr XRF 1361 1104 959 815 988 807 1105 1083 982 Y ICPMS 29·0 30·0 23·5 23·4 23·5 21·1 27·3 26·9 25·2 Zr XRF 321 307 266 299 225 230 264 264 286 Nb ICPMS 106·3 117·6 91·1 96·5 99·9 72·5 97·5 91·7 92·3 Cs ICPMS 0·677 0·485 0·392 0·302 0·299 2·50 1·32 1·06 0·976 Ba XRF 828 778 797 2961 719 657 870 828 788 La ICPMS 109·2 109·3 80·5 75·3 77·2 62·4 96·0 89·1 81·6 Ce ICPMS 208 208 146 147 141 125 182 168 154 Pr ICPMS 23·8 23·9 16·8 16·8 15·3 15·1 20·6 19·2 17·7 Nd ICPMS 88·5 88·3 65·5 63·9 55·6 59·1 76·6 73·0 67·0 Sm ICPMS 14·3 14·5 11·1 11·2 10·4 9·8 12·5 12·1 11·2 Eu ICPMS 4·47 4·55 3·54 3·53 3·24 3·25 3·97 3·88 3·53 Gd ICPMS 11·88 12·18 9·48 9·31 8·44 8·90 10·58 10·36 9·45 Tb ICPMS 1·55 1·60 1·23 1·23 1·08 1·18 1·42 1·36 1·20 Dy ICPMS 8·09 8·50 6·45 6·35 5·81 6·15 7·51 7·24 6·48 Ho ICPMS 1·32 1·40 1·09 1·02 0·969 1·02 1·25 1·24 1·08 Er ICPMS 3·09 3·31 2·54 2·48 2·35 2·36 3·11 2·98 2·63 Tm ICPMS 0·362 0·406 0·301 0·274 0·279 0·286 0·379 0·358 0·331 Yb ICPMS 1·73 1·89 1·48 1·40 1·54 1·32 1·84 1·78 1·67 Lu ICPMS 0·238 0·262 0·199 0·188 0·214 0·180 0·252 0·245 0·238 Hf ICPMS 7·11 7·17 6·18 7·03 5·46 5·62 6·10 6·06 6·44 Ta ICPMS 5·57 5·92 4·72 5·31 4·83 4·11 4·92 4·79 5·05 Pb ICPMS 13·36 11·99 8·72 6·73 4·89 8·20 7·75 7·93 8·77 Th ICPMS 14·02 14·00 10·30 8·26 9·34 7·74 12·40 11·07 9·40 U ICPMS 4·01 3·27 3·29 2·16 1·41 2·36 3·69 3·33 2·93 (206Pb/204Pb)m 18·72 18·54 18·87 18·57 18·92 18·56 18·95 18·67 18·49 (206Pb/204Pb)i 18·61 18·45 18·73 18·47 18·82 18·47 18·77 18·55 18·41 (207Pb/204Pb)m 15·62 15·60 15·61 15·61 15·61 15·62 15·64 15·61 15·62 (207Pb/204Pb)i 15·61 15·60 15·60 15·61 15·61 15·62 15·63 15·60 15·62 (208Pb/204Pb)m 38·65 38·42 38·77 38·49 38·78 38·49 38·88 38·57 38·42 (208Pb/204Pb)i 38·52 38·29 38·62 38·36 38·57 38·40 38·68 38·44 38·33 (176Hf/177Hf)m 0·282881±4 0·282879±4 0·282894±5 0·282868±5 0·282922±4 – 0·282858±7 0·282867±6 0·282917±6 (176Hf/177Hf)i 0·282878 0·282876 0·282891 0·282865 0·282919 0·282854 0·282864 0·282914 eHf(0) 3·9 3·8 4·3 3·4 5·3 – 3·0 3·4 5·1 eHf(i) 3·7 3·7 4·2 3·3 5·2 2·9 3·3 5·0 (143Nd/144Nd)m 0·512858±7 0·512846±7 0·512866±10 0·512856±10 0·512879±8 – 0·512838±9 0·512844±8 0·512881±8 (143Nd/144Nd)i 0·512839 0·512825 0·512841 0·512834 0·512855 – 0·512813 0·512824 0·512864 eNd(0) 4·3 4·1 4·5 4·2 4·7 – 3·9 4·0 4·7 eNd(i) 4·7 4·5 4·9 4·6 5·1 4·4 4·4 5·0 (87Sr/86Sr)m 0·705187±12 0·704482±11 0·703682±13 0·703687±13 0·703970±32 – 0·703686±11 0·703740±11 0·703517±21 (87Sr/86Sr)i 0·70515 0·70443 0·70364 0·703637 0·703939 0·70366 0·70370 0·70350 Sample Method S097511 S097512 S097519 S107048 S107052 S097513 S097514 S097515 S107045 Rock type Mel Mel Mel Mel Mel Neph Neph Neph Neph SiO2 XRF 35·5 34·9 35·4 34·5 34·2 38·4 41·0 41·3 40·0 Al2O3 XRF 9·00 9·42 9·09 9·60 9·52 9·6 11·1 11·1 10·3 Fe2O3 XRF 12·9 12·2 11·8 13·4 11·6 12·0 12·6 13·1 13·3 FeO 11·6 10·9 10·6 12·0 10·4 10·8 11·4 11·8 11·9 MnO XRF 0·20 0·18 0·17 0·18 0·17 0·17 0·20 0·20 0·20 MgO XRF 13·3 12·8 14·0 13·3 14·1 13·6 12·3 12·6 13·7 CaO XRF 12·0 11·6 12·5 11·6 12·7 11·2 11·8 11·3 11·9 Na2O XRF 3·39 3·16 3·47 3·11 3·82 2·92 4·04 3·44 3·26 K2O XRF 1·35 1·25 0·98 1·16 0·74 0·79 1·08 1·14 0·95 TiO2 XRF 2·85 2·84 2·67 2·90 2·53 2·80 2·84 2·94 2·89 P2O5 XRF 1·00 0·99 0·84 0·76 0·82 0·61 0·82 0·77 0·86 SO3 XRF 1·11 0·84 0·71 0·28 0·46 0·31 0·03 0·05 0·06 L·O·I XRF 7·60 9·86 8·94 9·48 9·92 8·42 2·04 2·12 2·96 SUM XRF 100·3 99·9 100·5 100·3 100·6 100·8 99·9 100·1 100·3 Li ICPMS 36·9 31·6 45·2 15·5 54·6 20·0 41·5 21·1 27·2 Sc ICPMS 20·7 20·1 22·3 21·9 26·2 21·3 21·9 24·8 24·3 V XRF 265 265 228 272 246 262 279 281 289 Cr XRF 494 458 409 367 539 451 476 527 643 Co ICPMS 72·8 53·5 66·5 70·0 61·0 74·4 64·9 69·0 74·6 Ni XRF 424 206 354 309 379 385 312 325 468 Cu XRF 188 254 133 146 144 185 113 158 178 Zn XRF 173 193 117 137 104 163 129 163 185 Ga ICPMS 16·4 17·8 15·9 18·0 16·2 16·6 17·8 18·0 17·1 Rb ICPMS 33·1 42·4 25·0 30·8 22·1 14·7 18·0 32·1 16·3 Sr XRF 1361 1104 959 815 988 807 1105 1083 982 Y ICPMS 29·0 30·0 23·5 23·4 23·5 21·1 27·3 26·9 25·2 Zr XRF 321 307 266 299 225 230 264 264 286 Nb ICPMS 106·3 117·6 91·1 96·5 99·9 72·5 97·5 91·7 92·3 Cs ICPMS 0·677 0·485 0·392 0·302 0·299 2·50 1·32 1·06 0·976 Ba XRF 828 778 797 2961 719 657 870 828 788 La ICPMS 109·2 109·3 80·5 75·3 77·2 62·4 96·0 89·1 81·6 Ce ICPMS 208 208 146 147 141 125 182 168 154 Pr ICPMS 23·8 23·9 16·8 16·8 15·3 15·1 20·6 19·2 17·7 Nd ICPMS 88·5 88·3 65·5 63·9 55·6 59·1 76·6 73·0 67·0 Sm ICPMS 14·3 14·5 11·1 11·2 10·4 9·8 12·5 12·1 11·2 Eu ICPMS 4·47 4·55 3·54 3·53 3·24 3·25 3·97 3·88 3·53 Gd ICPMS 11·88 12·18 9·48 9·31 8·44 8·90 10·58 10·36 9·45 Tb ICPMS 1·55 1·60 1·23 1·23 1·08 1·18 1·42 1·36 1·20 Dy ICPMS 8·09 8·50 6·45 6·35 5·81 6·15 7·51 7·24 6·48 Ho ICPMS 1·32 1·40 1·09 1·02 0·969 1·02 1·25 1·24 1·08 Er ICPMS 3·09 3·31 2·54 2·48 2·35 2·36 3·11 2·98 2·63 Tm ICPMS 0·362 0·406 0·301 0·274 0·279 0·286 0·379 0·358 0·331 Yb ICPMS 1·73 1·89 1·48 1·40 1·54 1·32 1·84 1·78 1·67 Lu ICPMS 0·238 0·262 0·199 0·188 0·214 0·180 0·252 0·245 0·238 Hf ICPMS 7·11 7·17 6·18 7·03 5·46 5·62 6·10 6·06 6·44 Ta ICPMS 5·57 5·92 4·72 5·31 4·83 4·11 4·92 4·79 5·05 Pb ICPMS 13·36 11·99 8·72 6·73 4·89 8·20 7·75 7·93 8·77 Th ICPMS 14·02 14·00 10·30 8·26 9·34 7·74 12·40 11·07 9·40 U ICPMS 4·01 3·27 3·29 2·16 1·41 2·36 3·69 3·33 2·93 (206Pb/204Pb)m 18·72 18·54 18·87 18·57 18·92 18·56 18·95 18·67 18·49 (206Pb/204Pb)i 18·61 18·45 18·73 18·47 18·82 18·47 18·77 18·55 18·41 (207Pb/204Pb)m 15·62 15·60 15·61 15·61 15·61 15·62 15·64 15·61 15·62 (207Pb/204Pb)i 15·61 15·60 15·60 15·61 15·61 15·62 15·63 15·60 15·62 (208Pb/204Pb)m 38·65 38·42 38·77 38·49 38·78 38·49 38·88 38·57 38·42 (208Pb/204Pb)i 38·52 38·29 38·62 38·36 38·57 38·40 38·68 38·44 38·33 (176Hf/177Hf)m 0·282881±4 0·282879±4 0·282894±5 0·282868±5 0·282922±4 – 0·282858±7 0·282867±6 0·282917±6 (176Hf/177Hf)i 0·282878 0·282876 0·282891 0·282865 0·282919 0·282854 0·282864 0·282914 eHf(0) 3·9 3·8 4·3 3·4 5·3 – 3·0 3·4 5·1 eHf(i) 3·7 3·7 4·2 3·3 5·2 2·9 3·3 5·0 (143Nd/144Nd)m 0·512858±7 0·512846±7 0·512866±10 0·512856±10 0·512879±8 – 0·512838±9 0·512844±8 0·512881±8 (143Nd/144Nd)i 0·512839 0·512825 0·512841 0·512834 0·512855 – 0·512813 0·512824 0·512864 eNd(0) 4·3 4·1 4·5 4·2 4·7 – 3·9 4·0 4·7 eNd(i) 4·7 4·5 4·9 4·6 5·1 4·4 4·4 5·0 (87Sr/86Sr)m 0·705187±12 0·704482±11 0·703682±13 0·703687±13 0·703970±32 – 0·703686±11 0·703740±11 0·703517±21 (87Sr/86Sr)i 0·70515 0·70443 0·70364 0·703637 0·703939 0·70366 0·70370 0·70350 Trace element analyses by inductively coupled plasma mass spectrometry (ICP-MS) were carried out at the Fachbereich Geowissenschaften, Universität Bremen, Germany. About 50 mg of sample material were digested with ultrapure-grade acids (2 ml HF + 2 ml HNO3 + 1 ml HCl) in Savillex™ beakers on a hot plate at 150 °C for 12 hours, evaporated at 110 °C, and re-dissolved in HNO3. Analyte solutions had a final dilution of 1: 25 000 (40 µg/ml of total dissolved solid) and were spiked with 0·25 ng/ml indium as internal standard. The analyses were carried out on a Thermo Element2 with an ApexQ™ introduction system, using high resolution (10 000) for middle to heavy rare earth elements (REE) and Hf, medium (4000) for transition metals, and low (300) for other elements. Precision and accuracy were monitored by analyzing replicates and USGS standard reference materials BCR-2 and BHVO-2 along with the samples (Supplementary DataTable S2). For Nd–Sr–Hf isotope analysis about 100 mg of sample powder was leached for 15 minutes in cold 6 M HCl. After repeated washing in deionised water, the samples were digested in hot (∼120 °C) HF/HNO3 (∼5: 1) for >12 h. After drying down 2–3 times in hot 6 M HCl, the residues were dissolved in 3 M HCl and loaded on Eichrom Ln-Spec ion exchange resin for Hf isolation, following the procedure described in Münker et al. (2001). The Sr–Nd cut was dried and dissolved in 3 M HCl to separate Sr from the rare earth elements (REE) using AG50X8 ion exchange resin. Nd was separated on Ln-Spec using diluted HCl as eluent. Isotope measurements were performed on a Thermo Neptune multicollector ICP-MS at the joint Cologne-Bonn facility, Germany. Measured Nd and Sr isotope data were mass bias corrected using 146Nd/144Nd = 0·7219 and 86Sr/88Sr = 0·1194. All measured values are reported relative to a 143Nd/144Nd value of 0·511859 for the La Jolla Nd-standard and an 87Sr/86Sr value of 0·710240 for the NBS 987 Sr-standard. Hf isotope data were normalised to 179Hf/177Hf = 0·7325 and 176Hf/177Hf values are reported relative to a value of 0·282160 for the AMES Hf standard that is isotopically indistinguishable from the JMC-475 Hf standard. Long term external reproducibilities (2 r.s.d.) were better than ±100 ppm for 87Sr/86Sr and ±40 ppm for 143Nd/144Nd and 176Hf/177Hf. Pb isotope compositions were determined at GeoForschungsZentrum (GFZ) Potsdam, Germany, following procedures described in Romer & Hahne (2010). Samples were leached using the same procedure as for the trace element analysis and then dissolved in concentrated HF at 160 °C for four days on a hot plate, dried at low temperature, taken up in 2 N HNO3 to destroy fluorides, and dried again. Pb was separated on anion exchange resin Bio Rad AG1-X8 (100–200 mesh) in teflon columns and using HCl–HBr eluents. Purification was achieved by a second pass over the same column. Pb was loaded together with H3PO4 and silica gel on single Re-filaments. The isotopic composition was determined at 1200–1250 °C on a Thermo Scientific Triton multicollector mass spectrometer using static multicollection. Instrumental fractionation was corrected with 0·1%/a.m.u. as determined from repeated measurement of lead reference material NBS 981. Accuracy and precision of reported Pb isotope ratios are better than 0·1% at the 2σ level. Total procedural blanks for whole-rock samples are better than 15–30 pg Pb. Initial Sr–Nd–Hf–Pb isotope compositions were calculated based on parent–daughter ratios calculated from measured element concentrations, atomic masses, natural isotope abundances and the corresponding 40Ar/39Ar age. For Pb, the isotope abundances and atomic masses for each sample were calculated from the measured Pb isotope composition. 40Ar/39Ar dating was performed at Argonlab Freiberg (ALF), TU Freiberg, Germany. About 50 mg of phenocryst-free whole-rock fragments with a size of 500–1000 µm were repeatedly washed in deionised water in an ultrasonic bath. After drying, the samples were wrapped in Al-foil and placed along with fluence monitors in holes (∼5x5 mm diameter x depth) on Al-disks with ∼33 mm diameter for irradiation. This was done without Cd-shielding for 3 hours at the LVR-15 research reactor of the Nuclear Research Institute in Řež, Czech Republic. The reactor was operated at a power of 9·7 MW, providing a thermal neutron fluence of ∼8x1013 n/cm2s at a thermal to fast neutron ratio of ∼1·1. Irradiated samples were unwrapped and loaded into small Mo-crucibles for furnace step heating experiments. Step heating was performed using a Createc® high-temperature cell (HTC) controlled by an Eurotherm 3504 controller (for details, see Pfänder et al., 2014). Gas purification was achieved by two AP10N getter pumps, one at room temperature and the other at 400 °C. Heating time was 7 minutes, cleaning time 10 minutes per step. Argon isotope compositions were measured in static mode on a GV Instruments ARGUS noble gas mass spectrometer equipped with five faraday cups and 1012 ohm resistors on mass positions 36–39 and a 1011 ohm resistor on mass position 40. Typical blank levels are 2·5x10–16 mol 40Ar and 8·1x10–18 mol 36Ar. Measurement time was 7·5 minutes per step acquiring 45 scans at 10 seconds integration time each. Mass bias was corrected assuming linear mass dependent fractionation and using an atmospheric 40Ar/36Ar ratio of 295·5. For raw data reduction and time-zero intercept calculation in-house developed Matlab® software was used, and isochron, inverse isochron and plateau ages were calculated using ISOPLOT 3.7 (Ludwig, 2008). All ages were calculated using Fish Canyon sanidine as flux monitor (28·305 ± 0·036 Ma; Renne et al., 2010) and by applying the decay constants of Renne et al. (2010). Reported errors on ages are 1σ. Corrections for interfering Ar isotopes were done using (36Ar/37Ar)Ca = 0·000245 ± 0·000012, (39Ar/37Ar)Ca = 0·000932 ± 0·000035, (38Ar/39Ar)K = 0·01211 ± 0·00061 and (40Ar/39Ar)K = 0·00183 ± 0·00009. RESULTS Petrography All samples are porphyritic with abundant olivine and in most cases subordinate clinopyroxene phenocrysts. In most samples, olivine is partly to completely altered to calcite ± clay minerals ± serpentine, or locally transformed into iddingsite, whereas clinopyroxene is fresh to mildly altered. Some samples contain anhedral olivine with elongated subgrains and orthopyroxene with reaction rims, which are interpreted as xenocrysts derived from mantle peridotite. The groundmass dominantly consists of clinopyroxene, opaques (likely Ti-magnetite), interstitial nepheline and/or plagioclase, and in some cases altered nosean or haüyne, apatite, and alkali feldspar (cf. Huckenholz & Schröder, 1981). Although melilite was not found in any sample, it is possible that it had been a groundmass phase but was replaced during alteration; note that melilite has been described for other samples from the Heldburg area (Huckenholz & Schröder, 1981). Secondary groundmass minerals include calcite, zeolite (phillipsite, analcime), clay minerals and biotite. Most vesicles are filled with calcite ± zeolites. Major and trace element compositions Major and trace element compositions are listed in Table 1 and plotted in Figs 2–4. All samples are primitive, silica-undersaturated, mafic rocks with high MgO contents between 8·5 and 14·1 wt % (with one exception; Fig. 3a) and Ni and Cr concentrations of 141–468 ppm and 181–643 ppm, respectively (Fig. 3f and g). Nickel and Cr concentrations are positively correlated with MgO (not shown). SiO2 varies between 34·2 and 47·1 wt % (Fig. 3) and all samples are sodic with Na2O = 2·92–4·74 wt %, K2O = 0·69–2·15 wt % and Na2O/K2O = 2·2–6·3. According to the TAS classification, the samples plot within the foidite, basanite/tephrite and hawaiite fields (Fig. 2a). Al2O3 contents are between 9·0 and 15·5 wt % (Fig. 3b), and CaO contents between 8·7 and 12·7 wt % (Fig. 3c). A classification based on CaO + Na2O + K2O vs SiO2 + Al2O3 (Le Bas, 1989) subdivides the investigated samples into melilitites and nephelinites, as well as basanites (Fig. 2b). Although no melilite is observed in thin sections, the strongly SiO2-undersaturated rock group is termed nephelinites/melilitites in the following discussion and the less undersaturated rocks are termed basanites. The CaO contents in the nephelinites/melilitites are lower (11·2–12·7 wt %) than in melilitites from other regions (e.g. Urach, Hegau and Eger provinces: >14 up to ∼25 wt % in a few olivine melilitite pegmatoids; Hegner et al., 1995; Wilson et al., 1995b; Ulrych & Pivec, 1997; Ulrych et al., 1999), whereas the Al2O3 contents (9·0–11·1 wt %) are similar (7·0–14·7 wt % to other provinces). Compared to the basanites, CaO is higher in the nephelinites/melilitites (Fig. 3c) and Al2O3 is distinctly lower (Fig. 3b). CaO/Al2O3 in the nephelinites/melilitites is mostly lower than in melilitites from elsewhere, but similar to nephelinites from other regions and distinctly higher than in basanites from this study and from elsewhere (Fig. 3d). TiO2 concentrations tend to be slightly higher in the nephelinites/melilitites than in the basanites, with an overall variation from 2·13 to 3·23 wt % (Fig. 3e), well within the range of melilitites, nephelinites and basanites from other regions (e.g. Hegner et al., 1995; Jung & Masberg, 1998; Jung & Hoernes, 2000; note that SiO2-undersaturated rocks from the Bohemian Massif tend to have higher TiO2 concentrations up to 5·8 wt %; see data compilation in Lustrino & Wilson, 2007). Fig. 2. View largeDownload slide (a) Total alkalis vs silica diagram for Heldburg mafic rocks (TAS; Le Bas et al., 1986). (b) CaO + Na2O + K2O vs SiO2 + Al2O3 diagram (Le Bas, 1989). Compositions of melilitites and nephelinites from other volcanic provinces are shown for comparison (data from Wilson et al., 1995b and references therein). Asterisks: composition of experimental liquids from melting peridotite + CO2 at 3 GPa; white: 2·5 wt % CO2, grey: 1·0 wt % CO2 (Dasgupta et al., 2007), black: 2·5 wt % CO2 (Hirose, 1997). Oxide contents are re-normalised to a volatile free composition. Fig. 2. View largeDownload slide (a) Total alkalis vs silica diagram for Heldburg mafic rocks (TAS; Le Bas et al., 1986). (b) CaO + Na2O + K2O vs SiO2 + Al2O3 diagram (Le Bas, 1989). Compositions of melilitites and nephelinites from other volcanic provinces are shown for comparison (data from Wilson et al., 1995b and references therein). Asterisks: composition of experimental liquids from melting peridotite + CO2 at 3 GPa; white: 2·5 wt % CO2, grey: 1·0 wt % CO2 (Dasgupta et al., 2007), black: 2·5 wt % CO2 (Hirose, 1997). Oxide contents are re-normalised to a volatile free composition. Fig. 3. View largeDownload slide Major element and Cr and Ni contents of samples from the Heldburg region. Also shown are the compositions of melilitites (small circles: various locations including Urach, data from Wilson et al., 1995b, Hegner et al., 1995; rhombs: Eger rift, data from Ulrych et al., 2013) as well as nephelinites (triangles) and basanites (squares) from the Hocheifel, Vogelsberg and Rhön areas (Jung, 1995; Jung & Masberg, 1998; Jung & Hoernes, 2000; Jung et al., 2006). Composition of experimental melts are from Hirose (1997) for melting dry and carbonated peridotite (KLB-1) at different temperatures (in °C) and pressures. Fig. 3. View largeDownload slide Major element and Cr and Ni contents of samples from the Heldburg region. Also shown are the compositions of melilitites (small circles: various locations including Urach, data from Wilson et al., 1995b, Hegner et al., 1995; rhombs: Eger rift, data from Ulrych et al., 2013) as well as nephelinites (triangles) and basanites (squares) from the Hocheifel, Vogelsberg and Rhön areas (Jung, 1995; Jung & Masberg, 1998; Jung & Hoernes, 2000; Jung et al., 2006). Composition of experimental melts are from Hirose (1997) for melting dry and carbonated peridotite (KLB-1) at different temperatures (in °C) and pressures. Fig. 4. View largeDownload slide (a, b) Normalised rare earth and (c, d) multi-element concentration diagrams for nephelinites/melilitites and basanites from the Heldburg region (normalising data for chondrite from Sun & McDonough (1989), for PRIMA from Hofmann (1988)). Shown for reference are the compositions of basanites and nephelinites from the Hocheifel (Jung et al., 2006), Vogelsberg (Jung & Masberg, 1998) and Rhön volcanic provinces (Jung & Hoernes, 2000), melilitites from Urach (Hegner et al., 1995) and nephelinites and melilitites from the Eger graben (see data compilation in Lustrino & Wilson, 2007). Fig. 4. View largeDownload slide (a, b) Normalised rare earth and (c, d) multi-element concentration diagrams for nephelinites/melilitites and basanites from the Heldburg region (normalising data for chondrite from Sun & McDonough (1989), for PRIMA from Hofmann (1988)). Shown for reference are the compositions of basanites and nephelinites from the Hocheifel (Jung et al., 2006), Vogelsberg (Jung & Masberg, 1998) and Rhön volcanic provinces (Jung & Hoernes, 2000), melilitites from Urach (Hegner et al., 1995) and nephelinites and melilitites from the Eger graben (see data compilation in Lustrino & Wilson, 2007). Chondrite and primitive mantle normalised incompatible trace element patterns are shown in Fig. 4. All samples are highly enriched in LREE relative to HREE. Lanthanum concentrations are slightly higher in the nephelinites/melilitites (∼360-times chondritic) than in the basanites (∼300-times chondritic), whereas Yb is slightly lower (9·5 vs 11·1-times chondritic). Chondrite normalised (La/Yb)n ranges from 45 to 34 in the nephelinites/melilitites and from 33 to 19 in the basanites. (Gd/Yb)n is consistently higher in the nephelinites/melilitites (3·1–2·5) than in the basanites (2·3–1·9). The REE patterns resemble those from other locations within the CEVP, including the Eger graben (Fig. 4a and b). The nephelinites/melilitites have slightly higher REE concentrations than the basanites and primitive mafic rocks from other regions and are very similar to melilitites from Urach, with the exception of more enriched HREE (Fig. 4b). In primitive mantle normalised multi-element diagrams, both rock groups display pronounced depletions of Rb, K and Pb compared to trace elements with similar incompatibility (Fig. 4c and d), with (Rb/Ba)PM and particularly (K/La)PM being lower in the nephelinites/melilitites (0·15–0·62 and 0·16–0·29, respectively) than in the basanites (0·32–0·72 and 0·30–0·61, respectively, Fig. 5c). Such strong depletions are also observed in melilitites from Urach and the Eger graben ((K/La)PM = 0·05–0·48 and 0·27–0·64, respectively; Hegner et al., 1995; Ulrych et al., 2013). The most enriched trace elements in both rock groups relative to the primitive mantle composition (PRIMA) are Ba, Nb, Ta and U, with Nb/U being on average slightly higher in the basanites (42 ± 4) than in the nephelinites/melilitites (32 ± 5) and Nb/Ta being significantly more variable in the nephelinites/melilitites (17·7–20·7) than in the basanites (18·3–19·4; Fig. 5e). Th/U is nearly identical in both rock groups with values between 3·1 and 4·3 (with one exception having 6·6). In terms of Nb/Ta vs Zr/Hf (not shown), the samples plot within the compositional fields of continental basalts from elsewhere in central Germany (Pfänder et al., 2012). Notably, the variable Nb/Ta in the nephelinites/melilitites corresponds to nearly invariant Zr/Hf (41–45), whereas for the basanites nearly invariant Nb/Ta correspond to more variable Zr/Hf (40–50). Fig. 5. View largeDownload slide SiO2 and selected trace element ratios vs 40Ar/39Ar ages. (a) Polzenite: average SiO2 content (n = 20) of polzenites from Ulrych et al. (1999); OM, MON and ON: range of olivine melilitites, melilite olivine nephelinites and olivine nephelinites from the Eger graben (Ulrych et al., 2013) and the Upper Rhine graben (Hegner et al., 1995; Wilson et al., 1995b). (b) Chondrite normalised La/Yb. Arrows indicate increasing (nephelinites/melilitites) and decreasing (basanites) degrees of melting (F) with time (F values calculated using the parameters outlined in the caption of Fig. 15, see text for details). (c, d) Primitive mantle normalised ratios (phlog, phlogopite; sp, spinel; gt, garnet). (e) Grey area: range of Nb/Ta in primitive continental basalts from central Germany (Pfänder et al., 2012). Arrow indicates decreasing contributions from a metasomatised source from the initial to later stages of melting for the nephelinites/melilitites. Note that Nb/Ta in the continental crust is ∼12–13 (Barth et al., 2000). (f) Decreasing Lu/Hf with time in the basanites reflects decreasing degrees of melting (F). Normalising values as in Fig. 4. Fig. 5. View largeDownload slide SiO2 and selected trace element ratios vs 40Ar/39Ar ages. (a) Polzenite: average SiO2 content (n = 20) of polzenites from Ulrych et al. (1999); OM, MON and ON: range of olivine melilitites, melilite olivine nephelinites and olivine nephelinites from the Eger graben (Ulrych et al., 2013) and the Upper Rhine graben (Hegner et al., 1995; Wilson et al., 1995b). (b) Chondrite normalised La/Yb. Arrows indicate increasing (nephelinites/melilitites) and decreasing (basanites) degrees of melting (F) with time (F values calculated using the parameters outlined in the caption of Fig. 15, see text for details). (c, d) Primitive mantle normalised ratios (phlog, phlogopite; sp, spinel; gt, garnet). (e) Grey area: range of Nb/Ta in primitive continental basalts from central Germany (Pfänder et al., 2012). Arrow indicates decreasing contributions from a metasomatised source from the initial to later stages of melting for the nephelinites/melilitites. Note that Nb/Ta in the continental crust is ∼12–13 (Barth et al., 2000). (f) Decreasing Lu/Hf with time in the basanites reflects decreasing degrees of melting (F). Normalising values as in Fig. 4. Trace element characteristics of both rock groups display distinct differences that, in part, are correlated with age, indicating a time-related evolution of the magmas (Fig. 5). Chondrite normalised (La/Yb)n are higher in the nephelinites/melilitites and decrease with decreasing age (Fig. 5b). In the basanites, conversely, increasing (La/Yb)n strongly correlates with decreasing age (Fig. 5b, inset). (K/La)PM and (Nb/La)PM are higher in the basanites than in the nephelinites/melilitites (Fig. 5c and d), whereas Nb/Ta ratios overlap for both groups (Fig. 5e). In terms of Lu/Hf, the basanites positively correlate with age, and have overall higher values than the nephelinites/melilitites (Fig. 5f). Nd–Sr–Hf–Pb isotope compositions Isotope compositions are given in Table 1 and plotted in Figs 6–8. In (143Nd/144Nd)i vs (87Sr/86Sr)i space, most samples plot within the range observed for other continental basalts from the CEVP. Both sample groups have overlapping ranges of initial 143Nd/144Nd and the majority of them plot along an array with a negative slope (Fig. 6, inset). The basanites on average are offset from the nephelinites/melilitites to slightly less radiogenic (87Sr/86Sr)i (Fig. 6 inset). Highly variable (87Sr/86Sr)i in nephelinites/melilitites contrast with a limited range in (143Nd/144Nd)i. Fig. 7 shows the covariation of Hf and Nd isotope compositions. The samples plot below the mantle array, which is defined by MORB and OIB (except EM1), and below the compositional range of basanites and nephelinites from other volcanic provinces within the CEVP. At a given εNd, the nephelinites/melilitites have lower εHf than the basanites (Fig. 7 inset). Fig. 6. View largeDownload slide Initial 143Nd/144Nd vs initial 87Sr/86Sr of nephelinites/melilitites and basanites from the Heldburg region along with data for other primitive mafic alkaline volcanic rocks from adjacent volcanic provinces. Composition of the Heldburg rocks mostly falls within the range of volcanic rocks from other regions, with a trend to slightly more radiogenic Sr isotope compositions that is also known from mafic rocks from the Rhön, the Eger graben and the transition zone between the Siebengebirge and Westerwald volcanic fields (labelled SG-WW). Slightly higher 87Sr/86Sr in the nephelinites/melilitites relative to the basanites (inset) is consistent with a higher contribution from a metasomatised source in the petrogenesis of the more undersaturated lavas. EAR (European Asthenospheric Reservoir) field from Wilson & Downes (2006; present day values). Dashed lines labelled LCC and UCC are calculated mixing lines between the most primitive magmas and lower and upper crustal compositions. LCC and UCC trace element compositions were taken from Rudnick & Fountain (1995), and Sr and Nd isotope compositions have been assumed to be 0·7170/0·51107 and 0·7365/0·51222, respectively. The εNd values given on the right-hand side axis are calculated for an average age of 20 Ma and vary by about ±0·4 εNd units with respect to the age range of the samples. Data sources: Hocheifel: Jung et al. (2006); Siebengebirge: Jung et al. (2012), Kolb et al. (2012); Westerwald: pers. comm. S. Jung; SG-WW: Schubert et al. (2015); Vogelsberg: Jung et al. (2011), Bogaard & Wörner (2003); Rhön: Mayer et al. (2013); Eger graben: Haase & Renno (2008); Lower Silesian: Blusztajn & Hart (1989); Urach and Hegau: Blusztajn & Hegner (2002). Initial values for the Hegau melilitites are calculated based on 87Rb/86Sr and 147Sm/144Nd values calculated from concentration values given in Hegner et al. (1995). Fig. 6. View largeDownload slide Initial 143Nd/144Nd vs initial 87Sr/86Sr of nephelinites/melilitites and basanites from the Heldburg region along with data for other primitive mafic alkaline volcanic rocks from adjacent volcanic provinces. Composition of the Heldburg rocks mostly falls within the range of volcanic rocks from other regions, with a trend to slightly more radiogenic Sr isotope compositions that is also known from mafic rocks from the Rhön, the Eger graben and the transition zone between the Siebengebirge and Westerwald volcanic fields (labelled SG-WW). Slightly higher 87Sr/86Sr in the nephelinites/melilitites relative to the basanites (inset) is consistent with a higher contribution from a metasomatised source in the petrogenesis of the more undersaturated lavas. EAR (European Asthenospheric Reservoir) field from Wilson & Downes (2006; present day values). Dashed lines labelled LCC and UCC are calculated mixing lines between the most primitive magmas and lower and upper crustal compositions. LCC and UCC trace element compositions were taken from Rudnick & Fountain (1995), and Sr and Nd isotope compositions have been assumed to be 0·7170/0·51107 and 0·7365/0·51222, respectively. The εNd values given on the right-hand side axis are calculated for an average age of 20 Ma and vary by about ±0·4 εNd units with respect to the age range of the samples. Data sources: Hocheifel: Jung et al. (2006); Siebengebirge: Jung et al. (2012), Kolb et al. (2012); Westerwald: pers. comm. S. Jung; SG-WW: Schubert et al. (2015); Vogelsberg: Jung et al. (2011), Bogaard & Wörner (2003); Rhön: Mayer et al. (2013); Eger graben: Haase & Renno (2008); Lower Silesian: Blusztajn & Hart (1989); Urach and Hegau: Blusztajn & Hegner (2002). Initial values for the Hegau melilitites are calculated based on 87Rb/86Sr and 147Sm/144Nd values calculated from concentration values given in Hegner et al. (1995). Fig. 7. View largeDownload slide εHf vs εNd in nephelinites/melilitites and basanites from the Heldburg region, along with compositional fields of oceanic and intracontinental basalts (modified from Pfänder et al., 2007; see this reference and Pfänder et al., 2012, for data sources). In comparison to other CEVP basalts (Rhön, Vogelsberg, Eifel), the Heldburg samples are less radiogenic in Nd and Hf. This is consistent with the derivation of these magmas from a metasomatised lithospheric mantle source (see text for details). Both sample series form distinct trends (inset), with the nephelinites/melilitites being less radiogenic in Hf at a given Nd isotope composition, reflecting a higher contribution from a metasomatised source in these melts. Note that mixing with continental crust would produce similar trends, but such a process is in conflict with the primitive nature of both sample suites and their high Nb/Ta ratios. Fig. 7. View largeDownload slide εHf vs εNd in nephelinites/melilitites and basanites from the Heldburg region, along with compositional fields of oceanic and intracontinental basalts (modified from Pfänder et al., 2007; see this reference and Pfänder et al., 2012, for data sources). In comparison to other CEVP basalts (Rhön, Vogelsberg, Eifel), the Heldburg samples are less radiogenic in Nd and Hf. This is consistent with the derivation of these magmas from a metasomatised lithospheric mantle source (see text for details). Both sample series form distinct trends (inset), with the nephelinites/melilitites being less radiogenic in Hf at a given Nd isotope composition, reflecting a higher contribution from a metasomatised source in these melts. Note that mixing with continental crust would produce similar trends, but such a process is in conflict with the primitive nature of both sample suites and their high Nb/Ta ratios. Fig. 8. View largeDownload slide Pb isotope composition of nephelinites/melilitites and basanites from the Heldburg area along with data for other primitive mafic alkaline volcanic rocks from adjacent volcanic provinces. (206Pb/204Pb)i and (208Pb/204Pb)i in the Heldburg samples are markedly lower than in rocks from other regions and indicate the involvement of a metasomatised source carrying a Pb isotope signature from subducted continental crust. Pb isotope evolution lines for mantle and upper continental crust along with compositional fields (black dashed lines and grey dashed fields, tick marks represent 100 Ma steps) are from Zartman & Haines (1988). Stippled black lines denoted ‘ml UCC - LCC (350 Ma)’ in (a) and ‘ml’ in (b) are mixing lines between upper and lower continental crust compositions 350 Myr ago (composition of lower continental crust is out of range). The samples plot close to these mixing lines due to very low 238U/204Pb and low 232Th/204Pb after metasomatism of the lithospheric mantle and thus restricted ingrowth of radiogenic 206Pb and 208Pb after metasomatism. Data sources as in Fig. 6. SG-WW includes samples from the area between the Siebengebirge and Westerwald volcanic fields. NHRL, Northern Hemisphere Reference Line (Hart, 1984). Fig. 8. View largeDownload slide Pb isotope composition of nephelinites/melilitites and basanites from the Heldburg area along with data for other primitive mafic alkaline volcanic rocks from adjacent volcanic provinces. (206Pb/204Pb)i and (208Pb/204Pb)i in the Heldburg samples are markedly lower than in rocks from other regions and indicate the involvement of a metasomatised source carrying a Pb isotope signature from subducted continental crust. Pb isotope evolution lines for mantle and upper continental crust along with compositional fields (black dashed lines and grey dashed fields, tick marks represent 100 Ma steps) are from Zartman & Haines (1988). Stippled black lines denoted ‘ml UCC - LCC (350 Ma)’ in (a) and ‘ml’ in (b) are mixing lines between upper and lower continental crust compositions 350 Myr ago (composition of lower continental crust is out of range). The samples plot close to these mixing lines due to very low 238U/204Pb and low 232Th/204Pb after metasomatism of the lithospheric mantle and thus restricted ingrowth of radiogenic 206Pb and 208Pb after metasomatism. Data sources as in Fig. 6. SG-WW includes samples from the area between the Siebengebirge and Westerwald volcanic fields. NHRL, Northern Hemisphere Reference Line (Hart, 1984). Lead isotope compositions of the basanites and nephelinites/melilitites broadly overlap, with values of (206Pb/204Pb)i = 18·41–18·92, (207Pb/204Pb)i = 15·59–15·65 and (208Pb/204Pb)i = 38·29–38·70 (Table 1; Fig. 8). In (207Pb/204Pb)i vs (206Pb/204Pb)i space all samples plot above the Northern Hemisphere Reference Line (NHRL; Hart, 1984) and are distinctly offset from other primitive intraplate basalts from the CEVP (Fig. 8a), having lower (206Pb/204Pb)i at similar (207Pb/204Pb)i. On average, (206Pb/204Pb)i tends to be slightly lower in the nephelinites/melilitites than in the basanites. In terms of (208Pb/204Pb)i vs (206Pb/204Pb)i, both rock groups plot distinctly above the NHRL and are significantly less radiogenic than the other primitive alkaline lavas from the CEVP (Fig. 8b). Each rock group defines a positive correlation, with the nephelinites/melilitites being offset from the basanites to slightly higher (208Pb/204Pb)i values. 40Ar/ 39Ar ages 40Ar/39Ar ages are summarized in Table 2; the underlying data, age spectra and inverse isochron diagrams are given as Supplementary DataTable S3. The majority of samples have complex age spectra that in most cases hamper a straightforward interpretation and age derivation. We therefore provide a comprehensive description and interpretation of the data in the Supplementary Data (Interpretation Ar–Ar data), and outline the criteria that lead to the reported ages. Table 2: Summary of 40Ar/39Ar ages of whole rock samples from the Heldburg area Sample Location Exp.-No. WPA [Ma] MSWD Steps [%] 39Ar IIA [Ma] (40Ar/36Ar)i MSWD steps Comment S097511 Schwanhausen 3025 – – – – – – – – No age, likely < 62 Ma S097512 Schwanhausen 2764 – – – – 33·1 ± 0·6 294·0 ± 1·7 2·1 7–9 & 11–12 One obvious outlier was taken out S097513 Zimmerau 3029 – – – – 29·9 ± 1·8 297·0 ± 3·6 16 1–20 S097514 Zimmerau 2780 – – – – 37·1 ± 3·2 293·9 ± 2·6 1·2 10–14 S097515 Schwanhausen 2843 29·5 ± 0·2 0·6 17–20 35·2 28·9 ± 2·0 297·3 ± 5·2 0·7 17–20 Distinct loss–excess combination at LT S097519 Nassach 3042 38·0 ± 0·2 1·0 13–19 32·0 37·5 ± 1·1 373·1 ± 6·4 0·8 13–16 WPA calculated with (40Ar/36Ar)i = 373 S107045 Wolfshügel 2807 25·4 ± 0·4 0·6 7–20 89·1 25·1 ± 2·3 298·9 ± 1·5 0·7 7–20 WPA calculated with (40Ar/36Ar)i = 298·9 S107046 Baunach 2811 30·6 ± 0·3 0·2 17–20 38·8 32·6 ± 4·2 301·3 ± 6·2 127 1–23 (40Ar/36Ar)i of plateau steps 17-20: 295 ± 42 S107047 Schweinshaupten 2862 51·9 ± 0·2 7·7 9–10 50·2 33·1 ± 1·5 anchored 0·1 1–8 Anchored to 295·5 S107048 Manau 2815 31·9 ± 0·2 0·4 10–13 30·8 32·0 ± 0·5 295·1 ± 7·8 0·7 10–13 S107051 Schwanhausen 3010 – – – – 34·5 ± 3·1 298·1 ± 4·7 6·7 1–20 S107052 Fuchslöcher 2776 33·1 ± 0·2 0·4 7–12 68·3 33·1 ± 0·8 298·3 ± 1·7 0·8 7–12 WPA calculated with (40Ar/36Ar)i = 298·3 S097510 Zimmerau 2839 14·1 ± 0·1 0·61 10–20 52·9 14·0 ± 0·6 296·0 ± 2·0 0·66 10-20 S097517 Alsleben 2858 17·5 ± 0·2 2·7 16–20 25·4 19·6 ± 3·8 314 ± 16 346 1–24 WPA likely an upper limit S107049 Rehsalm 2827 17·7 ± 0·2 0·7 15–20 26·6 18·1 ± 1·0 295·0 ± 1·2 0·8 15–20 S107050 Langer Grund 2799 14·8 ± 0·1 0·5 9–18 42·5 15·3 ± 0·6 294·1 ± 1·9 0·4 9–18 S107053 Voccawind 2831 15·1 ± 0·1 0·1 16–18 33·3 15·2 ± 0·9 295·0 ± 7·5 0·2 16–18 S107054 Zeilberg 2803 13·1 ± 0·2 0·2 8–12 25·1 12·9 ± 0·9 295·8 ± 1·1 0·1 8–12 S107055 Bramberg 2753 – – – – 17·1 ± 0·5 292·5 ± 1·3 13 1–13 & 17–20 Sample Location Exp.-No. WPA [Ma] MSWD Steps [%] 39Ar IIA [Ma] (40Ar/36Ar)i MSWD steps Comment S097511 Schwanhausen 3025 – – – – – – – – No age, likely < 62 Ma S097512 Schwanhausen 2764 – – – – 33·1 ± 0·6 294·0 ± 1·7 2·1 7–9 & 11–12 One obvious outlier was taken out S097513 Zimmerau 3029 – – – – 29·9 ± 1·8 297·0 ± 3·6 16 1–20 S097514 Zimmerau 2780 – – – – 37·1 ± 3·2 293·9 ± 2·6 1·2 10–14 S097515 Schwanhausen 2843 29·5 ± 0·2 0·6 17–20 35·2 28·9 ± 2·0 297·3 ± 5·2 0·7 17–20 Distinct loss–excess combination at LT S097519 Nassach 3042 38·0 ± 0·2 1·0 13–19 32·0 37·5 ± 1·1 373·1 ± 6·4 0·8 13–16 WPA calculated with (40Ar/36Ar)i = 373 S107045 Wolfshügel 2807 25·4 ± 0·4 0·6 7–20 89·1 25·1 ± 2·3 298·9 ± 1·5 0·7 7–20 WPA calculated with (40Ar/36Ar)i = 298·9 S107046 Baunach 2811 30·6 ± 0·3 0·2 17–20 38·8 32·6 ± 4·2 301·3 ± 6·2 127 1–23 (40Ar/36Ar)i of plateau steps 17-20: 295 ± 42 S107047 Schweinshaupten 2862 51·9 ± 0·2 7·7 9–10 50·2 33·1 ± 1·5 anchored 0·1 1–8 Anchored to 295·5 S107048 Manau 2815 31·9 ± 0·2 0·4 10–13 30·8 32·0 ± 0·5 295·1 ± 7·8 0·7 10–13 S107051 Schwanhausen 3010 – – – – 34·5 ± 3·1 298·1 ± 4·7 6·7 1–20 S107052 Fuchslöcher 2776 33·1 ± 0·2 0·4 7–12 68·3 33·1 ± 0·8 298·3 ± 1·7 0·8 7–12 WPA calculated with (40Ar/36Ar)i = 298·3 S097510 Zimmerau 2839 14·1 ± 0·1 0·61 10–20 52·9 14·0 ± 0·6 296·0 ± 2·0 0·66 10-20 S097517 Alsleben 2858 17·5 ± 0·2 2·7 16–20 25·4 19·6 ± 3·8 314 ± 16 346 1–24 WPA likely an upper limit S107049 Rehsalm 2827 17·7 ± 0·2 0·7 15–20 26·6 18·1 ± 1·0 295·0 ± 1·2 0·8 15–20 S107050 Langer Grund 2799 14·8 ± 0·1 0·5 9–18 42·5 15·3 ± 0·6 294·1 ± 1·9 0·4 9–18 S107053 Voccawind 2831 15·1 ± 0·1 0·1 16–18 33·3 15·2 ± 0·9 295·0 ± 7·5 0·2 16–18 S107054 Zeilberg 2803 13·1 ± 0·2 0·2 8–12 25·1 12·9 ± 0·9 295·8 ± 1·1 0·1 8–12 S107055 Bramberg 2753 – – – – 17·1 ± 0·5 292·5 ± 1·3 13 1–13 & 17–20 Samples are given in an order with increasing SiO2 content as in Table 1. All errors are 1σ. Errors on plateau ages, where given, are weighted mean errors and commonly underestimate the external reproducibility which is in the order of ±1%. Recommended ages are given in bold, critical data are given in italics. WPA, weighed plateau age, IIA, inverse isochron age, LT, low temperature. Table 2: Summary of 40Ar/39Ar ages of whole rock samples from the Heldburg area Sample Location Exp.-No. WPA [Ma] MSWD Steps [%] 39Ar IIA [Ma] (40Ar/36Ar)i MSWD steps Comment S097511 Schwanhausen 3025 – – – – – – – – No age, likely < 62 Ma S097512 Schwanhausen 2764 – – – – 33·1 ± 0·6 294·0 ± 1·7 2·1 7–9 & 11–12 One obvious outlier was taken out S097513 Zimmerau 3029 – – – – 29·9 ± 1·8 297·0 ± 3·6 16 1–20 S097514 Zimmerau 2780 – – – – 37·1 ± 3·2 293·9 ± 2·6 1·2 10–14 S097515 Schwanhausen 2843 29·5 ± 0·2 0·6 17–20 35·2 28·9 ± 2·0 297·3 ± 5·2 0·7 17–20 Distinct loss–excess combination at LT S097519 Nassach 3042 38·0 ± 0·2 1·0 13–19 32·0 37·5 ± 1·1 373·1 ± 6·4 0·8 13–16 WPA calculated with (40Ar/36Ar)i = 373 S107045 Wolfshügel 2807 25·4 ± 0·4 0·6 7–20 89·1 25·1 ± 2·3 298·9 ± 1·5 0·7 7–20 WPA calculated with (40Ar/36Ar)i = 298·9 S107046 Baunach 2811 30·6 ± 0·3 0·2 17–20 38·8 32·6 ± 4·2 301·3 ± 6·2 127 1–23 (40Ar/36Ar)i of plateau steps 17-20: 295 ± 42 S107047 Schweinshaupten 2862 51·9 ± 0·2 7·7 9–10 50·2 33·1 ± 1·5 anchored 0·1 1–8 Anchored to 295·5 S107048 Manau 2815 31·9 ± 0·2 0·4 10–13 30·8 32·0 ± 0·5 295·1 ± 7·8 0·7 10–13 S107051 Schwanhausen 3010 – – – – 34·5 ± 3·1 298·1 ± 4·7 6·7 1–20 S107052 Fuchslöcher 2776 33·1 ± 0·2 0·4 7–12 68·3 33·1 ± 0·8 298·3 ± 1·7 0·8 7–12 WPA calculated with (40Ar/36Ar)i = 298·3 S097510 Zimmerau 2839 14·1 ± 0·1 0·61 10–20 52·9 14·0 ± 0·6 296·0 ± 2·0 0·66 10-20 S097517 Alsleben 2858 17·5 ± 0·2 2·7 16–20 25·4 19·6 ± 3·8 314 ± 16 346 1–24 WPA likely an upper limit S107049 Rehsalm 2827 17·7 ± 0·2 0·7 15–20 26·6 18·1 ± 1·0 295·0 ± 1·2 0·8 15–20 S107050 Langer Grund 2799 14·8 ± 0·1 0·5 9–18 42·5 15·3 ± 0·6 294·1 ± 1·9 0·4 9–18 S107053 Voccawind 2831 15·1 ± 0·1 0·1 16–18 33·3 15·2 ± 0·9 295·0 ± 7·5 0·2 16–18 S107054 Zeilberg 2803 13·1 ± 0·2 0·2 8–12 25·1 12·9 ± 0·9 295·8 ± 1·1 0·1 8–12 S107055 Bramberg 2753 – – – – 17·1 ± 0·5 292·5 ± 1·3 13 1–13 & 17–20 Sample Location Exp.-No. WPA [Ma] MSWD Steps [%] 39Ar IIA [Ma] (40Ar/36Ar)i MSWD steps Comment S097511 Schwanhausen 3025 – – – – – – – – No age, likely < 62 Ma S097512 Schwanhausen 2764 – – – – 33·1 ± 0·6 294·0 ± 1·7 2·1 7–9 & 11–12 One obvious outlier was taken out S097513 Zimmerau 3029 – – – – 29·9 ± 1·8 297·0 ± 3·6 16 1–20 S097514 Zimmerau 2780 – – – – 37·1 ± 3·2 293·9 ± 2·6 1·2 10–14 S097515 Schwanhausen 2843 29·5 ± 0·2 0·6 17–20 35·2 28·9 ± 2·0 297·3 ± 5·2 0·7 17–20 Distinct loss–excess combination at LT S097519 Nassach 3042 38·0 ± 0·2 1·0 13–19 32·0 37·5 ± 1·1 373·1 ± 6·4 0·8 13–16 WPA calculated with (40Ar/36Ar)i = 373 S107045 Wolfshügel 2807 25·4 ± 0·4 0·6 7–20 89·1 25·1 ± 2·3 298·9 ± 1·5 0·7 7–20 WPA calculated with (40Ar/36Ar)i = 298·9 S107046 Baunach 2811 30·6 ± 0·3 0·2 17–20 38·8 32·6 ± 4·2 301·3 ± 6·2 127 1–23 (40Ar/36Ar)i of plateau steps 17-20: 295 ± 42 S107047 Schweinshaupten 2862 51·9 ± 0·2 7·7 9–10 50·2 33·1 ± 1·5 anchored 0·1 1–8 Anchored to 295·5 S107048 Manau 2815 31·9 ± 0·2 0·4 10–13 30·8 32·0 ± 0·5 295·1 ± 7·8 0·7 10–13 S107051 Schwanhausen 3010 – – – – 34·5 ± 3·1 298·1 ± 4·7 6·7 1–20 S107052 Fuchslöcher 2776 33·1 ± 0·2 0·4 7–12 68·3 33·1 ± 0·8 298·3 ± 1·7 0·8 7–12 WPA calculated with (40Ar/36Ar)i = 298·3 S097510 Zimmerau 2839 14·1 ± 0·1 0·61 10–20 52·9 14·0 ± 0·6 296·0 ± 2·0 0·66 10-20 S097517 Alsleben 2858 17·5 ± 0·2 2·7 16–20 25·4 19·6 ± 3·8 314 ± 16 346 1–24 WPA likely an upper limit S107049 Rehsalm 2827 17·7 ± 0·2 0·7 15–20 26·6 18·1 ± 1·0 295·0 ± 1·2 0·8 15–20 S107050 Langer Grund 2799 14·8 ± 0·1 0·5 9–18 42·5 15·3 ± 0·6 294·1 ± 1·9 0·4 9–18 S107053 Voccawind 2831 15·1 ± 0·1 0·1 16–18 33·3 15·2 ± 0·9 295·0 ± 7·5 0·2 16–18 S107054 Zeilberg 2803 13·1 ± 0·2 0·2 8–12 25·1 12·9 ± 0·9 295·8 ± 1·1 0·1 8–12 S107055 Bramberg 2753 – – – – 17·1 ± 0·5 292·5 ± 1·3 13 1–13 & 17–20 Samples are given in an order with increasing SiO2 content as in Table 1. All errors are 1σ. Errors on plateau ages, where given, are weighted mean errors and commonly underestimate the external reproducibility which is in the order of ±1%. Recommended ages are given in bold, critical data are given in italics. WPA, weighed plateau age, IIA, inverse isochron age, LT, low temperature. The nephelinites/melilitites span an age range from 38·0 ± 0·2 Ma to 25·4 ± 0·4 Ma, the basanites from 17·7 ± 0·2 to 13·1 ± 0·2 Ma (Fig. 9). These ages not only confirm previous findings that magmatism in the Heldburg region occurred predominantly in the late Oligocene to Miocene (Lippolt, 1978, 1982; Abratis et al., 2007), but also that volcanic activity had already started in the late Eocene. Previously published 40Ar/39Ar ages for a tholeiitic basalt (14·6 ± 0·2 Ma), a basanite (14·3 ± 0·1 Ma) and two alkali basalts (16·3 ± 0·2 and 14·4 ± 0·2 Ma; Abratis et al., 2007) from the northern part of the Heldburg region (Fig. 1b) fall well within the younger age range observed in this study for the basanites. Fig. 9. View largeDownload slide Distribution of 40Ar/39Ar ages from the Heldburg area. Nephelinites/melilitites and basanites are separated by an age gap of ∼7·7 Myr. Time range for rapid convergence between Europe and Africa taken from Rosenbaum et al. (2002). Details see text. Fig. 9. View largeDownload slide Distribution of 40Ar/39Ar ages from the Heldburg area. Nephelinites/melilitites and basanites are separated by an age gap of ∼7·7 Myr. Time range for rapid convergence between Europe and Africa taken from Rosenbaum et al. (2002). Details see text. The spatial distribution of ages (Fig. 1b) suggests that magmatism in the Heldburg region began with the eruption of nephelinites/melilitites in the central area and from there migrated southward, to terminate at about 25 Ma. After an amagmatic phase of ∼7·7 Myr, basanite magmatism started at around 18 Ma in the south, then migrated northeastward and terminated there after ∼5 Ma. DISCUSSION Alteration of samples The most SiO2-undersaturated rocks exhibit relatively large loss-on-ignition values (LOI) up to 9·9 wt % (Table 1). High LOI (up to 6 wt %) is a common feature in strongly SiO2-undersaturated volcanic rocks (particularly in melilitites; Hegner et al., 1995; Ulrych et al., 2013) and reflects the presence of hydrous phases (phlogopite, amphibole) and/or primary carbonate. They may also reflect, however, the presence of alteration-related secondary minerals such as zeolites and calcite. The CaO contents of our samples are only weakly correlated with the LOI values, suggesting that calcite, either primary or secondary, is not the dominant phase in controlling their volatile content. We note, however, that CaO in general is sensitive to alteration (e.g. Nesbitt & Wilson, 1992) and that the two samples having the highest CaO content (12·7 and 12·5 wt %) also have very high LOI (9·9 and 8·9 wt %). Shifts in major element compositions in basalts during alteration and weathering strongly depend on the intensity of these processes. Under tropical climate conditions, large shifts have been described for all major elements within completely decomposed rocks (Patino et al., 2003), but within temperate climate zones compositional variations may not exceed ∼3–5% with respect to the original values (Patino et al., 2003). Potassium, Rb and U are regarded as alteration sensitive elements. A comparison of the concentration distribution of these elements with immobile elements having a similar compatibility allows us to test potential shifts caused by alteration. The linear correlation between U and Th (R2=0·948), with only two exceptions (samples S097512 and S107052), shows that the effect of alteration on U was limited as Th is highly immobile (Staudigel et al., 1996; Kelley et al., 2003). Rubidium and K are correlated (R2=0·805), but do not correlate with Ba or Th (R2<0·5 and <0·1, respectively). This may indicate shifts in Rb and K concentrations, however, which are regarded as being minor with respect to the variations observed in normalised multi-element diagrams (Fig. 4). High field strength elements (HFSE) along with middle and heavy REE and hence parent–daughter ratios derived from them (Lu/Hf, Sm/Nd), as well as transition metals (Ni, Cr) are known to behave as immobile elements during alteration. Strontium and Pb are moderately fluid-mobile during low-temperature alteration (Staudigel et al., 1996; Kelley et al., 2003). Comparatively large variations in 87Sr/86Sr at restricted variations in 143Nd/144Nd (Fig. 6) might reflect secondary input of radiogenic Sr, even though all samples have been leached to remove such artefacts. An alternative interpretation for the radiogenic Sr isotopic composition in some of the samples is that this signature already was present in the mantle source and reflects fluid-mediated Sr addition during mantle metasomatism. The correlation between 206Pb/204Pb and 143Nd/144Nd (Fig. 10) indicates that secondary effects on Pb concentrations and isotope compositions are restricted. Fig. 10. View largeDownload slide 143Nd/144Nd vs 206Pb/204Pb in Heldburg samples. Unradiogenic 206Pb/204Pb at a given 143Nd/144Nd in comparison to other lavas from the CEVP is indicative of derivation of the lavas from a metasomatised, predominantly lithospheric mantle source. Similar trends have been previously interpreted as reflecting the admixture of lithospheric melts with primary asthenospheric magmas (Wilson & Downes, 2006). Compositional field for EAR taken from Cebria & Wilson (1995), CHUR value from Blichert-Toft & Albarède (1997), recalculated to 25 Ma. Data for the Pannonian Basin and Lower Silesian are present day values from Wilson & Downes (2006) and references therein. SG-WW denotes the area between the Siebengebirge and Westerwald volcanic provinces. Other data sources as in Fig. 6. Fig. 10. View largeDownload slide 143Nd/144Nd vs 206Pb/204Pb in Heldburg samples. Unradiogenic 206Pb/204Pb at a given 143Nd/144Nd in comparison to other lavas from the CEVP is indicative of derivation of the lavas from a metasomatised, predominantly lithospheric mantle source. Similar trends have been previously interpreted as reflecting the admixture of lithospheric melts with primary asthenospheric magmas (Wilson & Downes, 2006). Compositional field for EAR taken from Cebria & Wilson (1995), CHUR value from Blichert-Toft & Albarède (1997), recalculated to 25 Ma. Data for the Pannonian Basin and Lower Silesian are present day values from Wilson & Downes (2006) and references therein. SG-WW denotes the area between the Siebengebirge and Westerwald volcanic provinces. Other data sources as in Fig. 6. Overall, since the most fresh fragments available from the pre-crushed samples have been selected for analysis, we suppose that even the few samples which underwent relatively strong alteration have modified chemical composition that are minor compared to the variations imprinted by petrogenetic processes. Fractional crystallisation and crustal assimilation The high MgO contents (>8·5 wt %, with the exception of one sample having 5·5 wt % MgO) and moderate to high Ni and Cr concentrations (141–468 ppm and 181–643 ppm, respectively; Fig. 11) in all samples suggest only limited fractionation of the parental magmas prior to eruption. Primary magmas in equilibrium with lherzolitic mantle have ∼15–30 ppm Sc and ∼30–80 ppm Co at MgO contents between 10–15 wt % (Roeder & Emslie, 1970; Frey et al., 1978; Hess, 1992). They also have about 400–500 ppm Ni and 750–1100 ppm Cr in equilibrium with garnet peridotite, and similar Ni but markedly lower Cr contents (400–540 ppm) in equilibrium with spinel peridotite (Fig. 11). The most primitive sample from the Heldburg area (S107045) falls well within these ranges and likely represents a near-primary melt (Fig. 11), but decreasing Cr with decreasing Ni concentrations suggest that the other nephelinites/melilitites as well as the basanites underwent some crystal fractionation. Calculated fractionation trajectories for olivine, clinopyroxene and magnetite suggest limited olivine (<10%), but clinopyroxene and/or magnetite fractionation (Fig. 11). Scandium and Cr are both compatible in clinopyroxene with about identical partition coefficients (D∼4; Bédard, 1994). In contrast, Cr is highly compatible in magnetite, whereas mag/meltDSc is about unity (∼1·3; Bédard, 1994). Therefore, the nearly invariant Sc concentrations in both rock groups at decreasing Cr contents favour magnetite over limited clinopyroxene fractionation. This is consistent with the nearly constant CaO at decreasing MgO in both rock groups. Given a D-value for Cr between magnetite and melt (∼20; Bédard, 1994), magnetite fractionation in the order of 6–8% is required. Increasing TiO2 in nephelinites/melilitites and about constant TiO2 in basanites with decreasing MgO and increasing SiO2 (Fig. 3e) limit the potential amount of fractionated Ti-augite. In summary, both rock groups underwent restricted olivine (<10%) and clinopyroxene (<10%) crystallisation, accompanied by up to 8% magnetite fractionation. Fig. 11. View largeDownload slide Chromium and Ni concentrations in the Heldburg rocks. Lines indicate the composition of the remaining liquid after crystal fractionation of olivine (ol), clinopyroxene (cpx), magnetite (mag) and spinel (sp), using the most primitive sample as the starting composition. Numbers on circles denote the amount of fractionated phase in %. Grey fields: approximate composition of primary magmas in equilibrium with garnet (gt) and spinel peridotite (sp-per; Frey et al., 1978; Hess, 1992; and calculated using: CrD(gt-per/melt) = 3·8, CrD(sp-per/melt) = 7·0, NiD(gt-per/melt) = 5·0, NiD(sp-per/melt) = 5·3; D-values: Bédard, 1994; Adam & Green, 2006; modes: Salters, 1996; initial peridotite composition: McDonough & Sun, 1995). Fig. 11. View largeDownload slide Chromium and Ni concentrations in the Heldburg rocks. Lines indicate the composition of the remaining liquid after crystal fractionation of olivine (ol), clinopyroxene (cpx), magnetite (mag) and spinel (sp), using the most primitive sample as the starting composition. Numbers on circles denote the amount of fractionated phase in %. Grey fields: approximate composition of primary magmas in equilibrium with garnet (gt) and spinel peridotite (sp-per; Frey et al., 1978; Hess, 1992; and calculated using: CrD(gt-per/melt) = 3·8, CrD(sp-per/melt) = 7·0, NiD(gt-per/melt) = 5·0, NiD(sp-per/melt) = 5·3; D-values: Bédard, 1994; Adam & Green, 2006; modes: Salters, 1996; initial peridotite composition: McDonough & Sun, 1995). Intracontinental and oceanic basalts have high Nb/U (>37, average 47 ± 10), Ce/Pb (>20, average 25 ± 5; Hofmann et al., 1986) and Nb/Ta (>16; Pfänder et al., 2012), whereas continental crust has Nb/U ∼25, Ce/Pb ∼5 and Nb/Ta ∼12–13 (Barth et al., 2000; Rudnick & Gao, 2003; Hawkesworth & Kemp, 2006). Hence, substantial assimilation of continental crust lowers these ratios in mantle-derived magmas. The Nb/U and Ce/Pb ratios of the Heldburg basanites plot within the range for undifferentiated oceanic and continental intraplate basalts, whereas the nephelinites/melilitites tend to have lower Ce/Pb and particularly lower Nb/U (Fig. 12). Although this suggests some crustal assimilation, such a process is expected to produce a correlated shift in Ce/Pb vs Nb/U, as well as visibly lower Nb/Ta and εNd in these lavas, which is not observed (Figs 5e, 6, 12, 13). Instead, the εNd values of both groups are identical, with a maximum variation of 1·3 εNd units, and within the range of other primitive lavas from the CEVP. In addition, εNd is not correlated with Nb/U, Ce/Pb or Nb/Ta and therefore the observed variations of these ratios are regarded as reflecting source heterogeneities. Fig. 12. View largeDownload slide Ce/Pb vs Nb/U in the Heldburg samples. Lower values than in typical primitive oceanic and intraplate basalts (grey field) suggest a fluid-metasomatised source, here predominantly preserved in the nephelinites/melilitites. Small white squares: composition of primitive basanites and alkali basalts from the Eger graben and Siebengebirge (data from Jung et al., 2012; Ulrych et al., 2013). Small black squares: melilitites from Urach (Hegner et al., 1995). Dashed lines: average values in ocean island basalts (Hofmann et al., 1986). Fig. 12. View largeDownload slide Ce/Pb vs Nb/U in the Heldburg samples. Lower values than in typical primitive oceanic and intraplate basalts (grey field) suggest a fluid-metasomatised source, here predominantly preserved in the nephelinites/melilitites. Small white squares: composition of primitive basanites and alkali basalts from the Eger graben and Siebengebirge (data from Jung et al., 2012; Ulrych et al., 2013). Small black squares: melilitites from Urach (Hegner et al., 1995). Dashed lines: average values in ocean island basalts (Hofmann et al., 1986). P–T conditions of melting and source mineralogy Melting experiments using carbonated garnet lherzolite (e.g. Hirose, 1997; Keshav & Gudfinnsson, 2004; Gudfinnsson & Presnall, 2005; Dasgupta & Hirschmann, 2006; Dasgupta et al., 2007) have shown that low degrees of melting produce melilititic to nephelinitic melts that resemble the composition of the Heldburg nephelinites/melilitites with respect to their SiO2 and Al2O3 contents (Fig. 3b). This suggests melting temperatures of 1450–1475 °C at ∼3 GPa (Hirose, 1997). In contrast to the experimental melts, and to melilitites from elsewhere, the nephelinites/melilitites from the Heldburg region display lower CaO contents at a given SiO2 content (Fig. 3c). This might be an effect of pressure or lower fCO2, i.e. a lower modal abundance of carbonates in their source. Dasgupta & Hirschmann (2006) reported a much lower solidus temperature for carbonated natural peridotite of ∼1050–1100 °C at 3 GPa. Their near-solidus melts, however, are carbonatitic with SiO2 <10 wt %, and only at higher temperatures (>1325–1350 °C) and melting degrees of ∼1–5% develop to silicate melts with >25 wt % SiO2. These experiments along with variable SiO2 and MgO (e.g. Fig. 3a) suggest that the mantle source of the Heldburg nephelinites/melilitites was a carbonated lherzolite with about 0·1–0·25 wt % CO2 (Dasgupta et al., 2007) that underwent melting at pressures ≤3 GPa and temperatures ≥1325 °C. This lherzolite likely contained residual K-bearing phases such as phlogopite and/or amphibole (K-richterite), as is suggested by strongly negative K anomalies and low primitive mantle normalised K/La (Figs 4d and 5c; Wilson & Downes, 1992; Konzett & Fei, 2000). Melting temperatures therefore must have been lower than those derived from melting experiments, which are far above the stability limit of these phases (1100–1200 °C at 2–3 GPa; Class & Goldstein, 1997; Mallik et al., 2015). A possible explanation for this might be provided by higher water contents in the source of the Heldburg nephelinites/melilitites than used in melting experiments. The SiO2, Al2O3 and CaO contents of the basanites resemble those of experimental melts generated from dry peridotite (KLB-1) at 3 GPa and ∼1500 °C (Fig. 3b and c; Hirose & Kushiro, 1993; Hirose, 1997). Moderate negative K anomalies and primitive mantle normalised K/La < 1 (Figs 4c and 5c), however, also indicate the presence of residual amphibole/phlogopite in the source of the basanites, excluding such high melting temperatures and a volatile-free source composition. Melting temperatures in the range of 1150–1175 °C, as suggested for the melting of amphibole-rich metasomes within the lithospheric mantle to produce nephelinites and basanites (Pilet, 2015), likely provide a more realistic estimate for the formation of the Heldburg basanites. Lee et al. (2009) have developed a thermobarometer based on the pressure (and temperature) sensitive silica activity in natural basalts with SiO2 contents >42 wt %. Applying this thermobarometer to the basanites with MgO >8 wt % and without adding olivine (Mayer et al., 2013) suggests a pressure range for melting of 2·2–1·7 GPa at temperatures of 1255–1222 °C, i.e. within the stability range of amphibole in the spinel–peridotite field (Fig. 14). This suggests that the formation of the basanites took place at lower pressures and slightly lower temperatures than the formation of the nephelinites/melilitites. As the basanites are significantly younger than the nephelinites/melilitites, this observation is consistent with the assumption of an ongoing process of decompression melting with the formation of successively less undersaturated magmas over time. This inference is in agreement with the negative correlation between SiO2 and age (Fig. 5a). In summary, major element compositions and negative K anomalies suggest that initial melting in the Heldburg region took place within or close to the lithosphere-asthenosphere boundary to form the nephelinites/melilitites from a carbonated, phlogopite/amphibole-bearing lherzolite at ∼2·8 GPa and a maximum temperature of ∼1250 °C. Ongoing melting lead to a chemical change in source composition and, likely, to the consumption of carbonates and in part hydrous phases, before the basanites were generated at lower pressures and slightly lower temperatures within the spinel stability field of the mantle. Lithospheric vs asthenospheric mantle source Phlogopite and/or amphibole are not stable in the convecting sub-lithospheric upper mantle (e.g. Green et al., 2010) or within thermally upwelling mantle plumes (Class & Goldstein, 1997). Negative K and Rb anomalies in continental and oceanic intraplate basalts, reflecting the presence of these phases during melting, are therefore taken as pivotal evidence for the involvement of colder lithospheric mantle during their formation (Wilson & Downes, 1992; Class & Goldstein, 1997). Continental basalts, however, tend to have distinctly higher Nb/Ta but lower Zr/Sm than oceanic basalts, and these elevated Nb/Ta are a distinct feature of metasomatised domains within the subcontinental lithospheric mantle and have been ascribed to carbonatite infiltration (Aulbach et al., 2008; Pfänder et al., 2012). The Heldburg rocks have distinctly higher Nb/Ta and lower Zr/Sm than melts derived from the convecting asthenospheric mantle (i.e. MORB and OIB; Münker et al., 2003; Pfänder et al., 2007; Fig. 13), underlining a dominant role of the lithospheric mantle during magma genesis. This is supported by low 206Pb/204Pb relative to 143Nd/144Nd in both rock groups (Fig. 10); such isotopic characteristics have been previously described as being indicative of the addition of lithospheric mantle melts to EAR-derived magmas (Wilson & Downes, 2006). Higher and more variable Nb/Ta in the nephelinites/melilitites likely indicate a higher contribution of metasomatised domains during melting within the subcontinental lithospheric mantle and a more heterogeneous source than for the basanites. Differences in (K/La)PM, (Nb/La)PM and Lu/Hf between the nephelinites/melilitites and the basanites (Fig. 5c, d, f) underline that the two rock groups have been derived from different sources. The systematic change of these ratios over time (Fig. 5) furthermore indicates that melt contributions from these two sources changed systematically over time, possibly from a carbonate–phlogopite-rich to a carbonate–phlogopite-poor, but amphibole rich, source. Fig. 13. View largeDownload slide (a) Nb/Ta vs Lu/Hf and (b) Zr/Sm vs Zr concentration in nephelinites/melilitites and basanites from the Heldburg region. Elevated and variable Nb/Ta, but low Lu/Hf and Zr/Sm in the nephelinites/melilitites is consistent with the assumption of a metasomatised, carbonatitic lithospheric mantle source (details see text). Also shown are compositions of basanites and nephelinites (B&N) from the Rhön, Vogelsberg and Eifel regions, and of tholeiites (Th) from the Vogelsberg (modified from Pfänder et al., 2012). Fig. 13. View largeDownload slide (a) Nb/Ta vs Lu/Hf and (b) Zr/Sm vs Zr concentration in nephelinites/melilitites and basanites from the Heldburg region. Elevated and variable Nb/Ta, but low Lu/Hf and Zr/Sm in the nephelinites/melilitites is consistent with the assumption of a metasomatised, carbonatitic lithospheric mantle source (details see text). Also shown are compositions of basanites and nephelinites (B&N) from the Rhön, Vogelsberg and Eifel regions, and of tholeiites (Th) from the Vogelsberg (modified from Pfänder et al., 2012). In summary, there is evidence that the Heldburg magmas have been derived from a heterogeneous carbonate-bearing lithospheric mantle source within the stability field of phlogopite and K-richterite, consistent with inferred pressure–temperature conditions (Fig. 14). However, due to less pronounced negative K and Rb anomalies in the basanites as well as lower and more constant Nb/Ta and slightly higher Zr/Sm, the involvement of an asthenosphere (EAR)- derived melt in the petrogenesis of the basanites cannot be ruled out. This is underlined by slightly higher initial 143Nd/144Nd and 206Pb/204Pb, but lower 87Sr/86Sr in the basanites compared to the nephelinites/melilitites. Fig. 14. View largeDownload slide Pressure and temperature conditions for the formation of the Heldburg basanites calculated from silica activity (Lee et al., 2009). The asterisk denotes pressure and temperature estimate of initial nephelinite/melilitite formation as deduced from major element compositions and experimental results (see text). These data along with estimates of lithosphere thicknesses refine the source region of the Heldburg primary magmas to the lowermost lithosphere, i.e. the thermal boundary layer (TBL) extending to a depth of ∼90 km beneath Central Europe (∼3 GPa; Wilson et al., 1995b; Geissler et al., 2010). LAB, Lithosphere–asthenosphere boundary, estimated based on data from Geissler et al. (2010). Siebengebirge and Westerwald fields are shown for comparison. Base figure modified after Schubert et al. (2015), see this reference for underlying data sources. Note the low solidus of CO2-saturated mantle, which is from Falloon & Green (1990) for water and CO2-saturated natural fertile peridotite. Water and CO2-undersaturated systems share a similar topology, but are shifted towards higher temperatures (Hammouda & Keshav, 2015). Fig. 14. View largeDownload slide Pressure and temperature conditions for the formation of the Heldburg basanites calculated from silica activity (Lee et al., 2009). The asterisk denotes pressure and temperature estimate of initial nephelinite/melilitite formation as deduced from major element compositions and experimental results (see text). These data along with estimates of lithosphere thicknesses refine the source region of the Heldburg primary magmas to the lowermost lithosphere, i.e. the thermal boundary layer (TBL) extending to a depth of ∼90 km beneath Central Europe (∼3 GPa; Wilson et al., 1995b; Geissler et al., 2010). LAB, Lithosphere–asthenosphere boundary, estimated based on data from Geissler et al. (2010). Siebengebirge and Westerwald fields are shown for comparison. Base figure modified after Schubert et al. (2015), see this reference for underlying data sources. Note the low solidus of CO2-saturated mantle, which is from Falloon & Green (1990) for water and CO2-saturated natural fertile peridotite. Water and CO2-undersaturated systems share a similar topology, but are shifted towards higher temperatures (Hammouda & Keshav, 2015). Petrogenesis from melting modelling To further confine the nature and chemical composition of the magma source(s) and to estimate the degree of partial melting, non-modal batch melting modelling has been applied (Shaw, 1970). The amphibole-bearing spinel and garnet peridotite source modes and melting reactions of Jung et al. (2012) (Supplementary DataTable S5) with 5% modal amphibole, along with the averaged chemical composition of the most enriched peridotite xenoliths (H-group) from the Hessian depression (Hartmann & Wedepohl, 1990) have been used as starting conditions. The results are shown in Fig. 15. In general, considering amphibole during partial melting provides a good match to the observed compositional pattern of the Heldburg lavas, but underestimates the negative K and Rb anomalies. The model source was therefore adapted by replacing 4% of the amphibole by phlogopite. Both elements, along with Ba, are compatible in phlogopite, with partition coefficients in equilibrium with basanitic melts of ∼3·7 (K), ∼2·5–8 (Rb) and ∼3·1–4·1 (Ba), whereas all three elements are moderately incompatible in amphibole (partition coefficients ∼0·6 for K and ∼0·1–1·0 for Rb and Ba; La Tourrette et al., 1995; Green et al., 2000; Adam & Green, 2003, 2006; see also Supplementary DataTable S4). At melting degrees between 3% and 5% in the presence of phlogopite and garnet, a good fit with respect to the LILE, including Rb and K, LREE, MREE and Nb, is achieved, but the concentrations of HREE, Y, Hf and Ti in the melt are still underestimated (Fig. 15a, dotted lines labelled HW3 and HW5). As the amount of these elements in the melt is a little sensitive to the degree of melting, the source of the Heldburg lavas must have been more enriched in these elements than the averaged H-group xenoliths of the Hessian depression. The model source was therefore adapted by increasing the concentrations of these elements such that an overall acceptable fit in a multi-element concentration diagram was achieved for melting degrees between 2% and 6% (Fig. 15a, dashed lines labelled F = 2%, 4%, 6%). The inferred source composition along with enrichment factors for individual elements relative to the primitive mantle, and the average composition of the H-group xenoliths of Hartmann & Wedepohl (1990) and the continental lithospheric mantle (McDonough, 1990) are listed in Table 3. Partial melting in the garnet stability field fractionates Dy/Yb significantly more strongly than melting in the spinel stability field, whereas La/Yb is primarily a function of the degree of melting, as shown in Fig. 15b. The nephelinites/melilitites plot at higher Dy/Yb than the basanites, underlining the presence of residual garnet in their source. The overall trend of the data points, however, cross-cuts the calculated melting curve for phlogopite/amphibole-bearing garnet peridotite, with a tendency to spinel field compositions. This likely reflects melting over a considerable pressure range within or across the garnet/spinel peridotite transition. The lower Dy/Yb in the basanites (Fig. 15b) reflects their origin predominantly from a spinel-bearing peridotite, and the overall variation in La/Yb is consistent with the estimated degrees of melting between ∼2% and 6% for the inferred source composition (Fig. 15b). Table 3: Inferred source composition and enrichment factors in comparison to other ‘reservoirs’ Inferred source [ppm] PRIMA [ppm] H&W90 H-group [ppm] Ack13 Eger graben [ppm] McD90 SCLM [ppm] EF relative to PRIMA EF relative to H-group (H&W1990) EF relative to McD90 Rb 4·0 0·54 7·4 2·4 1·9 7·5 0·5 2·1 Ba 130 6·00 74 – 33 22 1·8 3·9 Th – 0·08 – 0·09 0·71 – – – U – 0·02 – 0·11 0·12 – – – K 2090 258 2090 492 448 8·1 1·0 4·7 Nb 4·1 0·62 4·1 2·2 4·8 6·6 1·0 0·9 La 4·7 0·61 4·7 1·1 2·6 7·7 1·0 1·8 Ce 10·6 1·60 11 2·4 6·3 6·6 1·0 1·7 Pb 0·28 0·18 0·28 0·95 0·16 1·6 1·0 1·8 Nd 5·8 1·19 5·8 1·2 2·7 4·9 1·0 2·2 Sr 83 18·2 83 51 49 4·6 1·0 1·7 Sm 1·3 0·39 1·3 0·22 0·47 3·4 1·0 2·8 Hf 0·60 0·27 0·39 0·20 0·27 2·2 1·5 2·2 Zr 19 9·71 16 8·6 21 2·0 1·2 0·9 Ti 3500 1085 1120 420 539 3·2 3·1 6·5 Eu 0·50 0·15 0·36 0·07 0·16 3·4 1·4 3·1 Gd 1·4 0·51 1·0 0·22 0·60 2·7 1·4 2·3 Dy 1·1 0·64 0·66 0·14 0·51 1·7 1·7 2·2 Y 4·5 3·94 2·5 0·53 4·4 1·1 1·8 1·0 Er 0·50 0·42 0·34 0·08 0·30 1·2 1·5 1·7 Yb 0·40 0·41 0·27 0·07 0·26 1·0 1·5 1·5 Lu 0·06 0·06 0·04 0·01 0·04 0·9 1·5 1·4 Zr/Hf 31·7 36·2 41·0 43·4 77·8 0·9 0·8 0·4 Zr/Sm 14·6 25·1 12·3 39·7 44·7 0·6 1·2 0·3 Inferred source [ppm] PRIMA [ppm] H&W90 H-group [ppm] Ack13 Eger graben [ppm] McD90 SCLM [ppm] EF relative to PRIMA EF relative to H-group (H&W1990) EF relative to McD90 Rb 4·0 0·54 7·4 2·4 1·9 7·5 0·5 2·1 Ba 130 6·00 74 – 33 22 1·8 3·9 Th – 0·08 – 0·09 0·71 – – – U – 0·02 – 0·11 0·12 – – – K 2090 258 2090 492 448 8·1 1·0 4·7 Nb 4·1 0·62 4·1 2·2 4·8 6·6 1·0 0·9 La 4·7 0·61 4·7 1·1 2·6 7·7 1·0 1·8 Ce 10·6 1·60 11 2·4 6·3 6·6 1·0 1·7 Pb 0·28 0·18 0·28 0·95 0·16 1·6 1·0 1·8 Nd 5·8 1·19 5·8 1·2 2·7 4·9 1·0 2·2 Sr 83 18·2 83 51 49 4·6 1·0 1·7 Sm 1·3 0·39 1·3 0·22 0·47 3·4 1·0 2·8 Hf 0·60 0·27 0·39 0·20 0·27 2·2 1·5 2·2 Zr 19 9·71 16 8·6 21 2·0 1·2 0·9 Ti 3500 1085 1120 420 539 3·2 3·1 6·5 Eu 0·50 0·15 0·36 0·07 0·16 3·4 1·4 3·1 Gd 1·4 0·51 1·0 0·22 0·60 2·7 1·4 2·3 Dy 1·1 0·64 0·66 0·14 0·51 1·7 1·7 2·2 Y 4·5 3·94 2·5 0·53 4·4 1·1 1·8 1·0 Er 0·50 0·42 0·34 0·08 0·30 1·2 1·5 1·7 Yb 0·40 0·41 0·27 0·07 0·26 1·0 1·5 1·5 Lu 0·06 0·06 0·04 0·01 0·04 0·9 1·5 1·4 Zr/Hf 31·7 36·2 41·0 43·4 77·8 0·9 0·8 0·4 Zr/Sm 14·6 25·1 12·3 39·7 44·7 0·6 1·2 0·3 EF, enrichment factor. PRIMA (primitive mantle) composition from Hofmann (1988), H&W90 (H-group) denotes the averaged composition of enriched spinel–peridotite xenolites from the Hessian depression (Hartmann & Wedepohl, 1990). McD90 (SCLM) denotes the average composition of the continental lithospheric mantle derived from xenoliths hosted in intraplate basalts (McDonough 1990). Ack13 denotes the average composition of xenoliths from the western termination of the Eger graben (Ackerman et al., 2013)· Table 3: Inferred source composition and enrichment factors in comparison to other ‘reservoirs’ Inferred source [ppm] PRIMA [ppm] H&W90 H-group [ppm] Ack13 Eger graben [ppm] McD90 SCLM [ppm] EF relative to PRIMA EF relative to H-group (H&W1990) EF relative to McD90 Rb 4·0 0·54 7·4 2·4 1·9 7·5 0·5 2·1 Ba 130 6·00 74 – 33 22 1·8 3·9 Th – 0·08 – 0·09 0·71 – – – U – 0·02 – 0·11 0·12 – – – K 2090 258 2090 492 448 8·1 1·0 4·7 Nb 4·1 0·62 4·1 2·2 4·8 6·6 1·0 0·9 La 4·7 0·61 4·7 1·1 2·6 7·7 1·0 1·8 Ce 10·6 1·60 11 2·4 6·3 6·6 1·0 1·7 Pb 0·28 0·18 0·28 0·95 0·16 1·6 1·0 1·8 Nd 5·8 1·19 5·8 1·2 2·7 4·9 1·0 2·2 Sr 83 18·2 83 51 49 4·6 1·0 1·7 Sm 1·3 0·39 1·3 0·22 0·47 3·4 1·0 2·8 Hf 0·60 0·27 0·39 0·20 0·27 2·2 1·5 2·2 Zr 19 9·71 16 8·6 21 2·0 1·2 0·9 Ti 3500 1085 1120 420 539 3·2 3·1 6·5 Eu 0·50 0·15 0·36 0·07 0·16 3·4 1·4 3·1 Gd 1·4 0·51 1·0 0·22 0·60 2·7 1·4 2·3 Dy 1·1 0·64 0·66 0·14 0·51 1·7 1·7 2·2 Y 4·5 3·94 2·5 0·53 4·4 1·1 1·8 1·0 Er 0·50 0·42 0·34 0·08 0·30 1·2 1·5 1·7 Yb 0·40 0·41 0·27 0·07 0·26 1·0 1·5 1·5 Lu 0·06 0·06 0·04 0·01 0·04 0·9 1·5 1·4 Zr/Hf 31·7 36·2 41·0 43·4 77·8 0·9 0·8 0·4 Zr/Sm 14·6 25·1 12·3 39·7 44·7 0·6 1·2 0·3 Inferred source [ppm] PRIMA [ppm] H&W90 H-group [ppm] Ack13 Eger graben [ppm] McD90 SCLM [ppm] EF relative to PRIMA EF relative to H-group (H&W1990) EF relative to McD90 Rb 4·0 0·54 7·4 2·4 1·9 7·5 0·5 2·1 Ba 130 6·00 74 – 33 22 1·8 3·9 Th – 0·08 – 0·09 0·71 – – – U – 0·02 – 0·11 0·12 – – – K 2090 258 2090 492 448 8·1 1·0 4·7 Nb 4·1 0·62 4·1 2·2 4·8 6·6 1·0 0·9 La 4·7 0·61 4·7 1·1 2·6 7·7 1·0 1·8 Ce 10·6 1·60 11 2·4 6·3 6·6 1·0 1·7 Pb 0·28 0·18 0·28 0·95 0·16 1·6 1·0 1·8 Nd 5·8 1·19 5·8 1·2 2·7 4·9 1·0 2·2 Sr 83 18·2 83 51 49 4·6 1·0 1·7 Sm 1·3 0·39 1·3 0·22 0·47 3·4 1·0 2·8 Hf 0·60 0·27 0·39 0·20 0·27 2·2 1·5 2·2 Zr 19 9·71 16 8·6 21 2·0 1·2 0·9 Ti 3500 1085 1120 420 539 3·2 3·1 6·5 Eu 0·50 0·15 0·36 0·07 0·16 3·4 1·4 3·1 Gd 1·4 0·51 1·0 0·22 0·60 2·7 1·4 2·3 Dy 1·1 0·64 0·66 0·14 0·51 1·7 1·7 2·2 Y 4·5 3·94 2·5 0·53 4·4 1·1 1·8 1·0 Er 0·50 0·42 0·34 0·08 0·30 1·2 1·5 1·7 Yb 0·40 0·41 0·27 0·07 0·26 1·0 1·5 1·5 Lu 0·06 0·06 0·04 0·01 0·04 0·9 1·5 1·4 Zr/Hf 31·7 36·2 41·0 43·4 77·8 0·9 0·8 0·4 Zr/Sm 14·6 25·1 12·3 39·7 44·7 0·6 1·2 0·3 EF, enrichment factor. PRIMA (primitive mantle) composition from Hofmann (1988), H&W90 (H-group) denotes the averaged composition of enriched spinel–peridotite xenolites from the Hessian depression (Hartmann & Wedepohl, 1990). McD90 (SCLM) denotes the average composition of the continental lithospheric mantle derived from xenoliths hosted in intraplate basalts (McDonough 1990). Ack13 denotes the average composition of xenoliths from the western termination of the Eger graben (Ackerman et al., 2013)· Fig. 15. View largeDownload slide Results of melting modelling. (a) 3–5% partial melting of a highly enriched phlog-amph-gt-lherzolite mantle closely matches the composition of the Heldburg nephelinites/melilitites, and approximately mimics the negative K anomaly observed in the lavas. Thick dashed lines are melt compositions calculated for 2%, 4% and 6% non-modal batch melting (Shaw 1970; partition coefficients used are given in Supplementary DataTable S4) using the following modal composition: ol: opx: cpx: gt: amph: phlog = 0·58: 0·15: 0·20: 0·02: 0·01: 0·04 in the source and 0·10: 0·20: 0·40: 0·10: 0·04: 0·16 in the melt. The trace element composition of the inferred source is given in Table 3. HW3 and HW5 are melt compositions calculated from 3% and 5% partial melting of a source with the same mineralogy, but assuming a trace element composition according to the H-group xenoliths of Hartmann & Wedepohl (1990). For details see text. Urach: melilitites from the Urach area (Hegner et al., 1995). Inset: REE pattern of the inferred source, along with the composition of the H-group xenoliths (H) from Hartmann & Wedepohl (1990) and the source composition of primitive lavas from the Siebengebirge (SGS; Jung et al., 2012). (b) Dy/Yb vs La/Yb in nephelinites/melilitites and basanites. Also shown are non-modal batch melting curves for garnet and spinel peridotite sources containing various proportions of phlogopite and/or amphibole, and using the following modes: am-sp, amphibole spinel peridotite with ol: opx: cpx: sp: amph = 0·58: 0·15: 0·20: 0·02: 0·05 (source) and 0·10: 0·20: 0·40: 0·10: 0·20 (melt); am-gt, amphibole garnet peridotite with ol: opx: cpx: gt: amph = 0·58: 0·15: 0·20: 0·02: 0·05 (source) and 0·10: 0·20: 0·40: 0·10: 0·20 (melt); phl-am-gt, phlogopite amphibole garnet peridotite with ol: opx: cpx: gt: amph: phlog = 0·58: 0·15: 0·20: 0·02: 0·01: 0·04 (source) and 0·10: 0·20: 0·40: 0·10: 0·04: 0·16 (melt). Letters in parentheses following the modes denote the starting composition used for modelling: HW, composition of the most enriched (H-group) xenoliths from Hartmann & Wedepohl (1990); inf, inferred source composition as given in Table 3. In contrast to the nephelinites/melilitites that were produced in the garnet-stability field, the basanites are best explained by assuming 2–6% partial melting predominantly in the spinel stability field (details see text). Grey arrows denote magma evolution over time with ages stated in rectangles. Fig. 15. View largeDownload slide Results of melting modelling. (a) 3–5% partial melting of a highly enriched phlog-amph-gt-lherzolite mantle closely matches the composition of the Heldburg nephelinites/melilitites, and approximately mimics the negative K anomaly observed in the lavas. Thick dashed lines are melt compositions calculated for 2%, 4% and 6% non-modal batch melting (Shaw 1970; partition coefficients used are given in Supplementary DataTable S4) using the following modal composition: ol: opx: cpx: gt: amph: phlog = 0·58: 0·15: 0·20: 0·02: 0·01: 0·04 in the source and 0·10: 0·20: 0·40: 0·10: 0·04: 0·16 in the melt. The trace element composition of the inferred source is given in Table 3. HW3 and HW5 are melt compositions calculated from 3% and 5% partial melting of a source with the same mineralogy, but assuming a trace element composition according to the H-group xenoliths of Hartmann & Wedepohl (1990). For details see text. Urach: melilitites from the Urach area (Hegner et al., 1995). Inset: REE pattern of the inferred source, along with the composition of the H-group xenoliths (H) from Hartmann & Wedepohl (1990) and the source composition of primitive lavas from the Siebengebirge (SGS; Jung et al., 2012). (b) Dy/Yb vs La/Yb in nephelinites/melilitites and basanites. Also shown are non-modal batch melting curves for garnet and spinel peridotite sources containing various proportions of phlogopite and/or amphibole, and using the following modes: am-sp, amphibole spinel peridotite with ol: opx: cpx: sp: amph = 0·58: 0·15: 0·20: 0·02: 0·05 (source) and 0·10: 0·20: 0·40: 0·10: 0·20 (melt); am-gt, amphibole garnet peridotite with ol: opx: cpx: gt: amph = 0·58: 0·15: 0·20: 0·02: 0·05 (source) and 0·10: 0·20: 0·40: 0·10: 0·20 (melt); phl-am-gt, phlogopite amphibole garnet peridotite with ol: opx: cpx: gt: amph: phlog = 0·58: 0·15: 0·20: 0·02: 0·01: 0·04 (source) and 0·10: 0·20: 0·40: 0·10: 0·04: 0·16 (melt). Letters in parentheses following the modes denote the starting composition used for modelling: HW, composition of the most enriched (H-group) xenoliths from Hartmann & Wedepohl (1990); inf, inferred source composition as given in Table 3. In contrast to the nephelinites/melilitites that were produced in the garnet-stability field, the basanites are best explained by assuming 2–6% partial melting predominantly in the spinel stability field (details see text). Grey arrows denote magma evolution over time with ages stated in rectangles. The melting calculations support qualitative constraints that the source of the nephelinites/melilitites was a highly enriched phlogopite- and minor amphibole-bearing garnet peridotite. To explain the strong negative K and Rb anomalies, at least 4–5% phlogopite is required. Compared to the sources of other SiO2-undersaturated lavas from the CEVP (e.g. Rhön, Vogelsberg, Eifel, Siebengebirge, Urach), where source enrichments to a level represented by the most enriched spinel peridotite xenoliths from the Hessian depression mark an upper limit (Hegner et al., 1995; Pfänder et al., 2012; Schubert et al., 2015; see also Fig. 15a inset), the source of the Heldburg mafic rocks is the most enriched one that has been detected beneath Central Europe so far (Table 3). For example, REE contents are enriched by a factor of ∼3 compared to the source of primitive mafic lavas from the Siebengebirge volcanic region (Jung et al., 2012; Kolb et al., 2012; Schubert et al., 2015; Fig. 15a inset). Particularly with respect to the HREE, Y, Hf and Ti, the concentrations in the Heldburg mantle source exceed even those of the H-group xenoliths from Hartmann & Wedepohl (1990) and those in the average continental lithospheric mantle (McDonough, 1990) by a factor of ∼1·5–2 and ∼3–6 for Ti, respectively. The source of the basanites was similarly enriched to the source of the nephelinites/melilitites, but predominantly underwent partial melting in the spinel stability field in the presence of amphibole but only minor phlogopite. This is suggested by the less pronounced negative K and Rb anomalies (Fig. 4c) in normalized trace element patterns, and thus by primitive mantle normalised K/La values that are closer to unity (Fig. 5c). The positive correlation between La/Yb and the age of the nephelinites/melilitites (Figs 5b and 15b) suggests lower degrees of melting during the initial stages of melt formation (∼38 Ma), and increasing degrees of melting over time. This contrasts with the negative correlation between La/Yb and age for the basanites, that suggests higher degrees of melting during the initial phase of basanite formation (∼18 Ma) and decreasing degrees of melting over time. Assuming adiabatic decompression melting at a given heat budget and steady state conditions, higher degrees of melting may relate to higher decompression rates, and decreasing degrees of melting may reflect either decreasing decompression rates or heat consumption and cooling due to melt formation. Role and type of metasomatism Epsilon Nd and εHf values, as well as 206Pb/204Pb and 208Pb/204Pb, for the Heldburg samples are lower than in all other primitive mafic lavas from the CEVP (Figs 7 and 8). This may be a feature inherited from their source prior to melting via melt- or fluid-driven metasomatism, or may reflect long-term evolution of this source at lower time-integrated parent daughter ratios. In addition, 208Pb/204Pb is slightly elevated relative to 206Pb/204Pb in the nephelinites/melilitites compared to the basanites, and both groups display significantly higher 208Pb/204Pb at a given 206Pb/204Pb compared to other CEVP lavas (Fig. 8b). This indicates that the Heldburg magmas were derived from a source with a distinctly higher time-integrated Th/U than the source of other CEVP lavas. The melilitites from the Urach/Hegau region plot also in the range of elevated 208Pb/204Pb at given 206Pb/204Pb (Fig. 8b), suggesting that such a feature might be common to the sources of strongly SiO2-undersaturated magmas in general. Metasomatic processes that affect the mantle source of primary magmas prior to melting are a viable mechanism to explain trace element enrichments and parent/daughter fractionation and related isotopic anomalies, in melts derived from such sources after aging. Very low degree silicate or carbonatite melts released from the convecting asthenospheric mantle have been frequently suggested to represent the metasomatic agents that infiltrate the lithospheric mantle, leading to domains with enriched trace element patterns and fractionated trace element ratios such as Nb/Ta, Zr/Hf, Rb/Sr, Sm/Nd, Lu/Hf, Th/U and U/Pb (e.g. Green & Wallace, 1988; McKenzie 1989; Dupuy et al., 1992; Ionov et al., 1993; Harry & Leeman, 1995; Jung et al., 2005; Aulbach et al., 2008; Pilet et al., 2008; Pfänder et al., 2012). For example, elevated Th/U and Nb/Ta, as well as low Lu/Hf may be characteristic of low-degree silicate melts derived from garnet peridotite or garnet pyroxenite sources, in which U and Ta are less incompatible in garnet than Th and Nb, though all are strongly incompatible (Beattie, 1993; Asmerom, 1999; van Westrenen et al., 1999; Green et al., 2000). In contrast, metasomatism by carbonatite melts will increase Nb/Ta along with Lu/Hf (Pfänder et al., 2012). As the heat content of low melt volumes is low, melt movements are likely to terminate within the thermal boundary layer at the asthenosphere–lithosphere transition zone (McKenzie, 1989), or the endothermal phase transition zone where garnet peridotite reacts to spinel peridotite (Asimow et al., 1995). In any case, infiltration, cooling and freezing of low-degree melts within the lithospheric mantle will produce trace element enriched domains and, dependent on their composition and fluid content, they will react with mantle phases to form hydrous minerals such as amphibole and/or phlogopite (Jung et al., 2005; Pilet et al., 2008). Melting of mantle domains metasomatised predominantly by hydrous low-degree silicate melts may explain many of the features observed in the Heldburg lavas, in particular the pronounced enrichment of highly incompatible trace elements, high LREE/HREE ratios as well as elevated Nb/Ta at low Lu/Hf (Fig. 13a). However, low degree metasomatic melts derived from an EAR-like source are unlikely to lower Ce/Pb and Nb/U in the lithospheric mantle to the level observed in the Heldburg nephelinites/melilitites (Fig. 12), nor they can explain the unradiogenic 206Pb/204Pb and 208Pb/204Pb and low εNd and εHf of the samples. The same is valid if recycled oceanic crust is considered as an additional source component. We therefore suggest an alternative process linked to subduction-related metasomatism to explain the observed features. At high pressures (5–6 GPa) and moderate temperatures (900–1000 °C), aqueous fluids and silicate melts derived from subducted crustal lithologies become completely miscible to form supercritical fluids with high solubilities for a number of trace elements, including LILE, L + MREE and HFSE (Kelemen et al., 2004; Schmidt et al., 2004; Kessel et al., 2005). Such supercritical fluids, and to a lesser extent their precursor aqueous fluids, preferably enrich Pb over U and Ce, and these over Nb (Kessel et al., 2005); at temperatures below 1000 °C and pressures between 4 and 6 GPa these elements are compatible in aqueous fluids and supercritical fluids in equilibrium with clinopyroxene/garnet-bearing residues (DPb>DU∼DCe>DNb, with DPb∼31, DU∼6·0, DCe∼5·2 and DNb∼3·7 at 900 °C and 4 GPa; Kessel et al., 2005; note that D = Cfluid / Csolid). Therefore, aqueous fluids and supercritical fluids released from subducting crust during the Variscan Orogeny between ∼420–300 Ma would have low Nb/U and Ce/Pb at relatively high elemental concentrations, and metasomatic enrichment of the lithospheric mantle by such liquids is a viable mechanism to explain the low Nb/U and Ce/Pb ratios in the source of the Heldburg magmas. Pressures >4·0 GPa are required for supercritical fluids to form. This confines the depth of metasomatism to more than ∼125 km, and also provides a minimum value for the lithosphere thickness during the Variscan Orogeny. Dehydration of subducted continental crust, which has been invoked to play a role during the Variscan Orogeny based on the occurrence of coesite- and diamond-bearing UHP rocks (Kroner & Romer, 2013), will imprint a crustal lead isotope signature to derivative liquids (Fig. 8) and relatively high Pb concentrations will dominate the Pb isotope budget of the metasomatised source. Correspondingly low U/Pb and Th/Pb will restrict the ingrowth of radiogenic Pb after metasomatism, such that the metasomatised source will remain close to the Paleozoic crustal Pb isotope composition, a feature observed in all samples (Fig. 8). Sm/Nd and Lu/Hf will also be low in high-pressure aqueous and supercritical fluids with DSm/DNd∼0·2 and DLu/DHf∼0·02 at 900 °C and 4 GPa (Kessel et al., 2005) and hence the metasomatised domains will also carry some crustal Nd and Hf and will evolve to less radiogenic Nd and Hf isotope compositions than the ambient mantle. This is consistent with low Lu/Hf (Fig. 13a) in the samples as well as low εHf and εNd that plot below the mantle array (Fig. 7). Rb/Sr will be high in aqueous and supercritical fluids, which may account for the moderately high 87Sr/86Sr in some of the samples and the approximately horizontal variation in terms of 143Nd/144Nd vs 87Sr/86Sr, which tends to deviate from the crust–mantle mixing array (Fig. 6). Elevated Nb/Ta as observed predominantly in some of the nephelinites/melilitites is also consistent with fluid-dominated metasomatism, as Nb and Ta are both compatible in supercritical fluids with DNb/DTa>1 (Kessel et al., 2005). Silicate melts from subducted continental crust as a metasomatic component would imprint similar features to metasomatised domains within the lithospheric mantle as supercritical fluids, but are difficult to reconcile with the strongly SiO2-undersaturated nature of the investigated rocks. Additionally, Nb/Ta in the continental crust is low (∼12–13; Barth et al., 2000), the opposite of what is observed in the Heldburg magmas. In summary, derivation of the Heldburg lavas from a lithospheric mantle source that had been metasomatised by slab-derived supercritical fluids during the Variscan orogeny explains the overall source enrichment in LILE, LREE and MREE as constrained by melting modelling, as well as the combined trace element and isotope systematics. Whereas melting started at depths of less than ∼85 km, metasomatism occurred significantly deeper, within the formation range of supercritical fluids, i.e. deeper than ∼125 km. Although a preceding or subsequent metasomatic event caused by low-degree asthenospheric silicate or carbonatite melts cannot be ruled out, the crustal-like Pb isotopic composition of the magmatic rocks in conjunction with elevated Nb/Ta favours metasomatism by supercritical fluids. The signature of this metasomatised source is stronger in the melilitites/nephelinites, as these represent the initial stages of melting with a higher contribution from metasomatised domains. Geodynamic implications The interrelationships between rifting and intraplate magmatism in central Europe, although subject to intense research over the past few decades, remain enigmatic. Seismic tomography suggests plume-like thermal anomalies rooting into the deep mantle and fingering upwards beneath e.g. the Rhenish Massif (Eifel region) and Massif Central (Granet et al., 1995; Goes et al., 2000; Ritter et al., 2001). However, there is some consensus that these anomalies are not the major driving forces for the evolution of the European Cenozoic Rift System (ECRIS; Dèzes et al., 2004; Ziegler & Dèzes, 2006). Instead, its development more probably was triggered by compressional stresses in the foreland of the Alps and the Pyrenees during the collision of Apulia and Iberia, respectively, with the European continent (e.g. Lustrino & Carminati, 2007). This hints at a complex interplay between foreland compressive stresses with gravitational forces caused by thermally upwelling asthenosphere. The main thrust systems in Central Europe with their NW–SE orientations as well as smaller scaled deformation structures indicate a pronounced phase of NNE–SSW shortening during Late Cretaceous to Palaeogene times (Kley & Voigt, 2008). This tectonic event, interpreted as being the result of the collision between Iberia and Europe, might have induced the first NNE-trending fractures in the Heldburg region. In the course of the Alpine Orogeny the convergence between Africa and Europe slowed down due to continental collision of Adria with the southern palaeo-margin of Europe in the Paleocene and early Eocene, between ∼67 to ∼55 Ma, accompanied by a phase of marked intraplate compression in the Alpine foreland from ∼61 to ∼54 Ma (Rosenbaum et al., 2002). Possibly at the same time an extended thermal anomaly developed at the base of the central European lithosphere (Goes et al., 2000). These events preceded the development of the ECRIS (Rosenbaum et al., 2002; Dèzes et al., 2004; Ziegler & Dèzes, 2006), but melilitite and nephelinite magmatism at ∼61 Ma as described from the Vosges and the Black Forest regions flanking the Upper Rhine graben (Keller et al., 2002) and in the Bohemian Massif (pre-rift period: 79–49 Ma; Ulrych et al., 2011) likely mark the onset of graben formation in central Europe. The main rifting phase of the ECRIS began in late Eocene times at ∼40 Ma (Fig. 9; Schumacher, 2002; Dèzes et al., 2004; Ziegler & Dèzes, 2006) and was associated with a resumed increase in convergence between Africa and Europe that lasted from the middle Eocene to the late Oligocene (∼52 to ∼19 Ma; Rosenbaum et al., 2002). The time of nephelinite/melilitite eruption in the Heldburg region (∼38 to ∼25 Ma) coincides with this time span and it is therefore suggested that this north–south foreland compression reactivated the pre-existing NNE–SSW trending fractures as tension gashes with associated magmatism. Most likely, this event also triggered an intense phase of magmatic activity in the nearby Bohemian Massif that took place between ∼34 Ma and ∼28 Ma (Ulrych et al., 1999). Melilitite/nephelinite magmatism ceased in the Heldburg region at ∼25 Ma. Possible causes are either a decrease in rifting activity or an increasing solidus temperature in the lithospheric mantle at that time, e.g. caused by the consumption of hydrous phases. After an amagmatic phase, the onset of basanite magmatism at ∼17·7 Ma coincides with increasing NW–SE extension starting at ∼18 Ma across the nearby Eger graben (Adamovic & Coubal, 1999). This event was accompanied by contemporaneous uplift of the northern Bohemian massif, which occurred in response to lithosphere folding in the region east of the Heldburg area (Dèzes et al., 2004). With respect to the overall coincidence of magmatic activity in the Heldburg region with the temporal evolution of the intraplate stress field in central Europe, there seems to be no need for an extended plume-like thermal anomaly beneath the Heldburg region that caused magmatic activity. It is rather proposed that this area represents a failed attempt to disrupt the lithosphere along pre-existing Late Cretaceous NNE–SSW trending fractures, where limited extension caused local lithospheric stretching and restricted melting. This is consistent with the overall low degrees of partial melting and with the low volumes of erupted magmas. CONCLUSIONS The major and trace element and isotope compositions, along with 40Ar/39Ar dating, of mostly primitive nephelinites/melilitites and basanites from the Heldburg region, part of the CEVP in SE Germany, provide important implications for their mantle source, petrogenesis, time and sequence of eruption and the related geodynamic framework. Major element compositions in conjunction with data from melting experiments and the pressure and temperature dependent silica activity indicate that the source region of the Heldburg magmas was metasomatised lithospheric mantle, likely the thermal boundary layer at a depth of ∼80–90 km for the nephelinites/melilitites, and the lithospheric mantle at a depth of ∼50–70 km for the basanites. The mantle source of both rock suites was strongly enriched in highly incompatible as well as fluid-mobile trace elements (LILE, LREE, U, Th, Ba, Pb), but also in moderately incompatible HREE. Enrichment factors for the latter exceed those of other sources of continental basalts within the Central European Volcanic Province. Negative K and in part Rb anomalies indicate that phlogopite and minor amphibole were residual phases during melting. Melting modelling has revealed that the nephelinites/melilitites are the products of 3–5% partial melting of a carbonated garnet–phlogopite–amphibole bearing lherzolite. The basanites share a similar source, which underwent melting predominantly in the spinel stability field at a lower phlogopite to amphibole ratio. Low Ce/Pb and Nb/U ratios observed in most of the nephelinites/melilitites are not caused by crustal assimilation as they are not correlated with low Nb/Ta, but reflect a source feature caused by metasomatism of the lowermost lithospheric mantle. This metasomatism was by supercritical fluids released from subducting oceanic and continental crust during the Variscan orogeny. Subduction of continental crust is evident from the occurrence of coesite- and diamond-bearing UHP rocks within the Variscan belt (Kroner & Romer, 2013). Such a metasomatic event is consistent with the less radiogenic εHf and εNd of the Heldburg samples compared to other mafic rocks from the CEVP, as supercritical fluids (like crustal melts) lower Lu/Hf and Sm/Nd in the metasomatised domains, but without adding high amounts of SiO2. Metasomatism also imprinted the Pb isotopic signature of the subducted Palaeozoic continental crust to the source of the Heldburg magmas, along with low U/Pb and Th/Pb ratios, which limited the ingrowth of radiogenic Pb after Variscan metasomatism. Due to this, the Heldburg magmatic rocks have markedly lower 208Pb/204Pb and 206Pb/204Pb than primitive magmas from elsewhere in the CEVP. A similar scenario was suggested by Lustrino et al. (2000), who explained low 208Pb/204Pb and 206Pb/204Pb in alkaline lavas from Sardinia by the direct entrainment of lower continental crust into the source of these rocks several 100 Myr prior to melting. Subduction-related metasomatism in the Heldburg lithospheric mantle is consistent with plate tectonic reconstructions for the Variscan subduction system. The nephelinites/melilitites erupted significantly earlier than the basanites, i.e. between 38·0 ± 0·2 Ma and 25·4 ± 0·4 Ma vs 17·7 ± 0·2 to 13·1 ± 0·2 Ma. These eruption periods coincide with major periods of intraplate deformation in this region, caused by the convergence between Africa and Europe. From this we conclude that limited extension along pre-existing fractures caused local lithospheric stretching, decompression and restricted melting. Acknowledgements Sample handling during irradiation by the reactor services team in Řež, Czech Republic, is greatly appreciated. Many thanks also to the Bayerisches Landesamt für Umwelt for providing most of the samples. Stefan Sopke is thanked for support during trace element analyses in Bremen. Constructive reviews by Lukas Ackerman and Michele Lustrino, and editorial handling by Marjorie Wilson helped to improve the manuscript and are greatly appreciated. SUPPLEMENTARY DATA Supplementary data are available at Journal of Petrology online. FINANCIAL SUPPORT Financial support for this project came indirectly through the Argonlab Freiberg (ALF), initially funded by the Bundesministerium für Bildung und Forschung, Bundesrepublik Deutschland, and the Freistaat Saxony. The TU Bergakademie Freiberg is thanked for covering the costs associated to maintain operation of a noble gas mass spectrometry laboratory. REFERENCES Abratis M. , Viereck L. , Pfänder J. A. , Hentschel R. ( 2015 ). Geochemical composition, petrography and 40Ar/39Ar age of the Heldburg phonolite: implications on magma mixing and mingling . International Journal of Earth Sciences 104 , 2033 – 2055 . Google Scholar CrossRef Search ADS Abratis M. , Mädler J. , Hautmann S. , Leyk H. J. , Meyer R. , Lippolt H. J. , Viereck-Götte L. ( 2007 ). Two distinct Miocene age ranges of basaltic rocks from the Rhön and Heldburg areas (Germany) based on 40 Ar/39 Ar step heating data . Chemie der Erde-Geochemistry 67 , 133 – 150 . Google Scholar CrossRef Search ADS Ackerman L. , Špaček P. , Magna T. , Ulrych J. , Svojtka M. , Hegner E. , Balogh K. ( 2013 ). Alkaline and carbonate-rich melt metasomatism and melting of subcontinental lithospheric mantle: evidence from mantle xenoliths, NE Bavaria, Bohemian Massif . Journal of Petrology 54 , 2597 – 2633 . Google Scholar CrossRef Search ADS Ackerman L. , Ulrych J. , Řanda Z. , Erban V. , Hegner E. , Magna T. , Balogh K. , Frána J. , Lang M. , Novák J. K. ( 2015 ). Geochemical characteristics and petrogenesis of phonolites and trachytic rocks from the České Středohoří Volcanic Complex, the Ohře Rift, Bohemian Massif . Lithos 224–225 , 256 – 271 . Google Scholar CrossRef Search ADS Adam J. , Green T. ( 2003 ). The influence of pressure, mineral composition and water on trace element partitioning between clinopyroxene, amphibole and basanitic melts . European Journal of Mineralogy 15 , 831 – 841 . Google Scholar CrossRef Search ADS Adam J. , Green T. ( 2006 ). Trace element partitioning between mica-and amphibole-bearing garnet lherzolite and hydrous basanitic melt: 1. Experimental results and the investigation of controls on partitioning behaviour . Contributions to Mineralogy and Petrology 152 , 1 – 17 . Google Scholar CrossRef Search ADS Adamovic J. , Coubal M. ( 1999 ). Intrusive geometries and Cenozoic stress history of the northern part of the Bohemian Massif . Geolines 9 , 5 – 14 . Asimow P. D. , Hirschmann M. M. , Ghiorso M. S. , O’Hara M. J. , Stolper E. M. ( 1995 ). The effect of pressure-induced solid-solid phase transitions on decompression melting of the mantle . Geochimica et Cosmochimica Acta 59 , 4489 – 4506 . Google Scholar CrossRef Search ADS Asmerom Y. ( 1999 ). Th–U fractionation and mantle structure . Earth and Planetary Science Letters 166 , 163 – 175 . Google Scholar CrossRef Search ADS Aulbach S. , O’Reilly S. Y. , Griffin W. L. , Pearson N. J. ( 2008 ). Subcontinental lithospheric mantle origin of high niobium/tantalum ratios in eclogites . Nature Geoscience 1 , 468 – 472 . Google Scholar CrossRef Search ADS Barth M. G. , McDonough W. F. , Rudnick R. L. ( 2000 ). Tracking the budget of Nb and Ta in the continental crust . Chemical Geology 165 , 197 – 213 . Google Scholar CrossRef Search ADS Beattie P. ( 1993 ). Uranium–thorium disequilibria and partitioning on melting of garnet peridotite . Nature 363 , 63 – 65 . Google Scholar CrossRef Search ADS Bédard J. H. ( 1994 ). A procedure for calculating the equilibrium distribution of trace elements among the minerals of cumulate rocks, and the concentration of trace elements in the coexisting liquids . Chemical Geology 118 , 143 – 153 . Google Scholar CrossRef Search ADS Blichert-Toft J. , Albarède F. ( 1997 ). The Lu-Hf isotope geochemistry of chondrites and the evolution of the mantle-crust system . Earth and Planetary Science Letters 148 , 243 – 258 . Google Scholar CrossRef Search ADS Blusztajn J. , Hart S. R. ( 1989 ). Sr, Nd, and Pb isotopic character of Tertiary basalts from southwest Poland . Geochimica et Cosmochimica Acta 53 , 2689 – 2696 . Google Scholar CrossRef Search ADS Blusztajn J. , Hegner E. ( 2002 ). Osmium isotopic systematics of melilitites from the Tertiary Central European Volcanic Province in SW Germany . Chemical Geology 189 , 91 – 103 . Google Scholar CrossRef Search ADS Bogaard P. J. F. , Wörner G. ( 2003 ). Petrogenesis of basanitic to tholeiitic volcanic rocks from the Miocene Vogelsberg, Central Germany . Journal of Petrology 44 , 569 – 602 . Google Scholar CrossRef Search ADS Brandl P. A. , Genske F. S. , Beier C. , Haase K. M. , Sprung P. , Krumm S. H. ( 2015 ). Magmatic evidence for carbonate metasomatism in the lithospheric mantle underneath the Ohře (Eger) Rift . Journal of Petrology 56 , 1743 – 1774 . Google Scholar CrossRef Search ADS Buikin A. , Trieloff M. , Hopp J. , Althaus T. , Korochantseva E. , Schwarz W. H. , Altherr R. ( 2005 ). Noble gas isotopes suggest deep mantle plume source of late Cenozoic mafic alkaline volcanism in Europe . Earth and Planetary Science Letters 230 , 143 – 162 . Google Scholar CrossRef Search ADS Carlé W. ( 1952 ). Die Hegau-Heldburg-Zone, ein rheinisch streichendes Lineament in Süddeutschland . Jahrbuch der geologischen Abteilung des Württembergischen statistischen Landesamtes 2 , 14 – 26 . Cebria J. M. , Wilson M. ( 1995 ). Cenozoic mafic magmatism in Western/Central Europe: a common European asthenospheric reservoir . Terra Nova 7 , 162 . Class C. , Goldstein S. L. ( 1997 ). Plume–lithosphere interactions in the ocean basins: constraints from the source mineralogy . Earth and Planetary Science Letters 150 , 245 – 260 . Google Scholar CrossRef Search ADS Dasgupta R. , Hirschmann M. M. ( 2006 ). Melting in the Earth’s deep upper mantle caused by carbon dioxide . Nature 440 , 659 – 662 . Google Scholar CrossRef Search ADS PubMed Dasgupta R. , Hirschmann M. M. , Smith N. D. ( 2007 ). Partial melting experiments of peridotite+CO2 at 3 GPa and genesis of alkalic ocean island basalts . Journal of Petrology 48 , 2093 – 2124 . Google Scholar CrossRef Search ADS DePaolo D. J. ( 1981 ). Trace element and isotopic effects of combined wallrock assimilation and fractional crystallization . Earth and Planetary Science Letters 53 , 189 – 202 . Google Scholar CrossRef Search ADS Dèzes P. , Schmid S. M. , Ziegler P. A. ( 2004 ). Evolution of the European Cenozoic Rift System: interaction of the Alpine and Pyrenean orogens with their foreland lithosphere . Tectonophysics 389 , 1 – 33 . Google Scholar CrossRef Search ADS Dupuy C. , Liotard J. M. , Dostal J. ( 1992 ). Zr/Hf fractionation in intraplate basaltic rocks: carbonate metasomatism in the mantle source . Geochimica et Cosmochimica Acta 56 , 2417 – 2423 . Google Scholar CrossRef Search ADS Falloon T. J. , Green D. H. ( 1990 ). Solidus of carbonated fertile peridotite under fluid-saturated conditions . Geology 18 , 195 – 199 . Google Scholar CrossRef Search ADS Frey F. A. , Green D. H. , Roy S. D. ( 1978 ). Integrated models of basalt petrogenesis: a study of quartz tholeiites to olivine melilitites from south eastern Australia utilizing geochemical and experimental petrological data . Journal of Petrology 19 , 463 – 513 . Google Scholar CrossRef Search ADS Gallagher K. , Hawkesworth C. ( 1992 ). Dehydration melting and the generation of continental flood basalts . Nature 358 , 57 – 59 . Google Scholar CrossRef Search ADS Geissler W. H. , Sodoudi F. , Kind R. ( 2010 ). Thickness of the central and eastern European lithosphere as seen by S receiver functions . Geophysical Journal International 181 , 604 – 634 . Goes S. , Govers R. , Vacher P. ( 2000 ). Shallow mantle temperatures under Europe from P and S wave tomography . Journal of Geophysical Research: Solid Earth 105 , 11153 – 11169 . Google Scholar CrossRef Search ADS Granet M. , Achauer U. , Wilson M. ( 1995 ). Nachweis eines Mantelplumes unter Frankreichs Zentralmassiv durch seismische Sondierung . Spektrum der Wissenschaft 10 , 28 – 33 . Grant T. B. , Milke R. , Pandey S. , Jahnke H. ( 2013 ). The Heldburg Phonolite, Central Germany: reactions between phonolite and xenocrysts from the upper mantle and lower crust . Lithos 182–183 , 86 – 101 . Google Scholar CrossRef Search ADS Green D. H. , Wallace M. E. ( 1988 ). Mantle metasomatism by ephemeral carbonatite melts . Nature 336 , 459 – 462 . Google Scholar CrossRef Search ADS Green D. H. , Hibberson W. O. , Kovács I. , Rosenthal A. ( 2010 ). Water and its influence on the lithosphere-asthenosphere boundary . Nature 467 , 448 – 451 . Google Scholar CrossRef Search ADS PubMed Green T. H. , Blundy J. D. , Adam J. , Yaxley G. M. ( 2000 ). SIMS determination of trace element partition coefficients between garnet, clinopyroxene and hydrous basaltic liquids at 2–7.5 GPa and 1080–1200°C . Lithos 53 , 165 – 187 . Google Scholar CrossRef Search ADS Gudfinnsson G. H. , Presnall D. C. ( 2005 ). Continuous gradations among primary carbonatitic, kimberlitic, melilititic, basaltic, picritic, and komatiitic melts in equilibrium with garnet lherzolite at 3–8 GPa . Journal of Petrology 46 , 1645 – 1659 . Google Scholar CrossRef Search ADS Haase K. M. , Renno A. D. ( 2008 ). Variation of magma generation and mantle sources during continental rifting observed in Cenozoic lavas from the Eger Rift, Central Europe . Chemical Geology 257 , 192 – 202 . Google Scholar CrossRef Search ADS Hammouda T. , Keshav S. ( 2015 ). Melting in the mantle in the presence of carbon: review of experiments and discussion on the origin of carbonatites . Chemical Geology 418 , 171 – 188 . Google Scholar CrossRef Search ADS Harry D. L. , Leeman W. P. ( 1995 ). Partial melting of melt metasomatized subcontinental mantle and the magma source potential of the lower lithosphere . Journal of Geophysical Research: Solid Earth 100 , 10255 – 10269 . Google Scholar CrossRef Search ADS Hart S. R. ( 1984 ). A large-scale isotope anomaly in the Southern Hemisphere mantle . Nature 309 , 753 – 757 . Google Scholar CrossRef Search ADS Hartmann G. , Wedepohl K. H. ( 1990 ). Metasomatically altered peridotite xenoliths from the Hessian Depression (Northwest Germany) . Geochimica et Cosmochimica Acta 54 , 71 – 86 . Google Scholar CrossRef Search ADS Hawkesworth C. J. , Kemp A. I. S. ( 2006 ). Evolution of the continental crust . Nature 443 , 811 – 817 . Google Scholar CrossRef Search ADS PubMed Hegner E. , Vennemann T. W. ( 1997 ). Role of fluids in the origin of Tertiary European intraplate volcanism: evidence from O, H, and Sr isotopes in melilitites . Geology 25 , 1035 – 1038 . Google Scholar CrossRef Search ADS Hegner E. , Walter H. J. , Satir M. ( 1995 ). Pb-Sr-Nd isotopic compositions and trace element geochemistry of megacrysts and melilitites from the Tertiary Urach volcanic field: source composition of small volume melts under SW Germany . Contributions to Mineralogy and Petrology 122 , 322 – 335 . Google Scholar CrossRef Search ADS Hess P. C. ( 1992 ). Phase equilibria constraints on the origin of ocean floor basalts . Geophysical Monograph 71 , 67 – 102 . Hirose K. ( 1997 ). Partial melt compositions of carbonated peridotite at 3 GPa and role of CO2 in alkali-basalt magma generation . Geophysical Research Letters 24 , 2837 – 2840 . Google Scholar CrossRef Search ADS Hirose K. , Kushiro I. ( 1993 ). Partial melting of dry peridotites at high pressures: determination of compositions of melts segregated from peridotite using aggregates of diamond . Earth and Planetary Science Letters 114 , 477 – 489 . Google Scholar CrossRef Search ADS Hoernle K. A. J. , Zhang Y. S. , Graham D. ( 1995 ). Seismic and geochemical evidence for large-scale mantle upwelling beneath the eastern Atlantic and western and central Europe . Nature 374 , 34 – 39 . Google Scholar CrossRef Search ADS Hofbauer G. ( 2007 ). Der Vulkan von Oberleinleiter: Spuren eines Maars in der Nördlichen Frankenalb . Jahresmitteilungen der Naturhistorischen Gesellschaft Nürnberg - Natur und Mensch 2007 , 69 – 87 . Hofmann A. W. ( 1988 ). Chemical differentiation of the Earth: the relationship between mantle, continental crust, and oceanic crust . Earth and Planetary Science Letters 90 , 297 – 314 . Google Scholar CrossRef Search ADS Hofmann A. W. , Jochum K. P. , Seufert M. , White W. M. ( 1986 ). Nb and Pb in oceanic basalts: new constraints on mantle evolution . Earth and Planetary Science Letters 79 , 33 – 45 . Google Scholar CrossRef Search ADS Huckenholz H. G. , Schröder B. ( 1981 ). Die Alkalibasaltassoziation der Heldburger Gangschar (Exkursion I am 25. April 1981) . Jahresberichte und Mitteilungen des Oberrheinischen Geologischen Vereins 63 , 125 – 138 . Google Scholar CrossRef Search ADS Huckenholz H. G. , Werner C. D. ( 1990 ). Die tertiären Vulkanite der Heldburger Gangschar (Bayerisch-thüringisches Grabfeld) . European Journal of Mineralogy 2 , 1 – 42 . Ionov D. A. , Dupuy C. , O’Reilly S. Y. , Kopylova M. G. , Genshaft Y. S. ( 1993 ). Carbonated peridotite xenoliths from Spitsbergen: implications for trace element signature of mantle carbonate metasomatism . Earth and Planetary Science Letters 119 , 283 – 297 . Google Scholar CrossRef Search ADS Jung S. ( 1995 ). Geochemistry and petrogenesis of rift-related Tertiary alkaline rocks from the Rhon area (central Germany) . Neues Jahrbuch fur Mineralogie-Abhandlungen 169 , 193 – 226 . Jung S. , Hoernes S. ( 2000 ). The major- and trace-element and isotope (Sr, Nd, O) geochemistry of Cenozoic alkaline rift-type volcanic rocks from the Rhön area (central Germany): petrology, mantle source characteristics and implications for asthenosphere–lithosphere interactions . Journal of Volcanology and Geothermal Research 99 , 27 – 53 . Google Scholar CrossRef Search ADS Jung S. , Masberg P. ( 1998 ). Major- and trace-element systematics and isotope geochemistry of Cenozoic mafic volcanic rocks from the Vogelsberg (central Germany): constraints on the origin of continental alkaline and tholeiitic basalts and their mantle sources . Journal of Volcanology and Geothermal Research 86 , 151 – 177 . Google Scholar CrossRef Search ADS Jung C. , Jung S. , Hoffer E. , Berndt J. ( 2006 ). Petrogenesis of Tertiary mafic alkaline magmas in the Hocheifel, Germany . Journal of Petrology 47 , 1637 – 1671 . Google Scholar CrossRef Search ADS Jung S. , Pfänder J. A. , Brauns M. , Maas R. ( 2011 ). Crustal contamination and mantle source characteristics in continental intra-plate volcanic rocks: Pb, Hf and Os isotopes from central European volcanic province basalts . Geochimica et Cosmochimica Acta 75 , 2664 – 2683 . Google Scholar CrossRef Search ADS Jung S. , Pfänder J. A. , Brügmann G. , Stracke A. ( 2005 ). Sources of primitive alkaline volcanic rocks from the Central European Volcanic Province (Rhön, Germany) inferred from Hf, Os and Pb isotopes . Contributions to Mineralogy and Petrology 150 , 546 – 559 . Google Scholar CrossRef Search ADS Jung S. , Mezger K. , Hauff F. , Pack A. , Hoernes S. ( 2013 ). Petrogenesis of rift-related tephrites, phonolites and trachytes (Central European Volcanic Province, Rhön, FRG): constraints from Sr, Nd, Pb and O isotopes . Chemical Geology 354 , 203 – 215 . Google Scholar CrossRef Search ADS Jung S. , Vieten K. , Romer R. L. , Mezger K. , Hoernes S. , Satir M. ( 2012 ). Petrogenesis of Tertiary alkaline magmas in the Siebengebirge, Germany . Journal of Petrology 53 , 2381 – 2409 . Google Scholar CrossRef Search ADS Kaiser G. , Pilot J. ( 1986 ). Weitere K–Ar-Datierungen an jungen Vulkaniten . Zeitschrift für Geologische Wissenschaften 14 , 121 – 124 . Kelemen P. B. , Hanghj K. , Greene A. R. ( 2004 ). One view of the geochemistry of subduction-related magmatic arcs, with an emphasis on primitive andesite and lower crust . Treatise on Geochemistry 3 , 593 – 659 . Keller J. , Kraml M. , Henjes-Kunst F. ( 2002 ). 40Ar/39Ar single crystal laser dating of early volcanism in the Upper Rhine Graben and tectonic implications . Schweizerische Mineralogische und Petrographische Mitteilungen 82 , 121 – 130 . Kelley K. A. , Plank T. , Ludden J. , Staudigel H. ( 2003 ). Composition of altered oceanic crust at ODP Sites 801 and 1149 . Geochemistry, Geophysics, Geosystems 4 , doi:10.1029/2002GC000435 Keshav S. , Gudfinnsson G. H. ( 2004 ). Silica-poor, mafic alkaline lavas from ocean islands and continents: petrogenetic constraints from major elements . Journal of Earth System Science 113 , 723 – 736 . Google Scholar CrossRef Search ADS Kessel R. , Schmidt M. W. , Ulmer P. , Pettke T. ( 2005 ). Trace element signature of subduction-zone fluids, melts and supercritical liquids at 120–180 km depth . Nature 437 , 724 – 727 . Google Scholar CrossRef Search ADS PubMed Kley J. , Voigt T. ( 2008 ). Late Cretaceous intraplate thrusting in central Europe: effect of Africa-Iberia-Europe convergence, not Alpine collision . Geology 36 , 839 – 842 . Google Scholar CrossRef Search ADS Kolb M. , Paulick H. , Kirchenbaur M. , Münker C. ( 2012 ). Petrogenesis of mafic to felsic lavas from the Oligocene Siebengebirge volcanic field (Germany): implications for the origin of intracontinental volcanism in Central Europe . Journal of Petrology 53 , 2349 – 2379 . Google Scholar CrossRef Search ADS Konzett J. , Fei Y. ( 2000 ). Transport and storage of potassium in the Earth’s upper mantle and transition zone: an experimental study to 23 GPa in simplified and natural bulk compositions . Journal of Petrology 41 , 603 . 583. Kroner U. , Romer R. L. ( 2013 ). Two plates—many subduction zones: the Variscan orogeny reconsidered . Gondwana Research 24 , 298 – 329 . Google Scholar CrossRef Search ADS La Tourrette T. , Hervig R. L. , Holloway J. R. ( 1995 ). Trace element partitioning between amphibole, phlogopite, and basanite melt . Earth and Planetary Science Letters 135 , 13 – 30 . Google Scholar CrossRef Search ADS Le Bas M. J. ( 1989 ). Nephelinitic and basanitic rocks . Journal of Petrology 30 , 1299 – 1312 . Google Scholar CrossRef Search ADS Le Bas M. J , Le Maitre R. W. , Streckeisen A. , Zanettin B. ( 1986 ). A chemical classification of volcanic rocks based on the total alkali-silica diagram . Journal of Petrology 27 , 745 – 750 . Google Scholar CrossRef Search ADS Lechler P. J. , Desilets M. O. ( 1987 ). A review of the use of loss on ignition as a measurement of total volatiles in whole-rock analysis . Chemical Geology 63 , 341 – 344 . Google Scholar CrossRef Search ADS Lee C. T. A. , Luffi P. , Plank T. , Dalton H. , Leeman W. P. ( 2009 ). Constraints on the depths and temperatures of basaltic magma generation on Earth and other terrestrial planets using new thermobarometers for mafic magmas . Earth and Planetary Science Letters 279 , 20 – 33 . Google Scholar CrossRef Search ADS Lippolt H. ( 1978 ). K–Ar-Untersuchungen zum Alter des Rhoen-Vulkanismus . Fortschritte Mineralogie 56 , 85 . Lippolt H. ( 1982 ). K–Ar age determinations and the correlation of Tertiary volcanic activity in Central Europe . Geologisches Jahrbuch Hannover D52 , 113 – 135 . Ludwig K. R. ( 2008 ). Isoplot 3.70. A geochronological toolkit for Microsoft Excel . Berkeley Geochronology Center Special Publication 4 , 1 – 76 . Lustrino M. , Anderson D. L. ( 2015 ). The mantle isotopic printer: basic mantle plume geochemistry for seismologists and geodynamicists . Geological Society of America Special Papers 514 , 257 – 279 . Google Scholar CrossRef Search ADS Lustrino M. , Carminati E. ( 2007 ). Phantom plumes in Europe and the circum-Mediterranean region . Geological Society of America Special Papers 430 , 723 – 745 . Lustrino M. , Wilson M. ( 2007 ). The circum-Mediterranean anorogenic Cenozoic igneous province . Earth Science Reviews 81 , 1 – 65 . Google Scholar CrossRef Search ADS Lustrino M. , Melluso L. , Morra V. ( 2000 ). The role of lower continental crust and lithospheric mantle in the genesis of Plio–Pleistocene volcanic rocks from Sardinia (Italy) . Earth and Planetary Science Letters 180 , 259 – 270 . Google Scholar CrossRef Search ADS Mallik A. , Nelson J. , Dasgupta R. ( 2015 ). Partial melting of fertile peridotite fluxed by hydrous rhyolitic melt at 2–3 GPa: implications for mantle wedge hybridization by sediment melt and generation of ultrapotassic magmas in convergent margins . Contributions to Mineralogy and Petrology 169 , 1 – 24 . Google Scholar CrossRef Search ADS Mayer B. , Jung S. , Romer R. L. , Stracke A. , Haase K. M. , Garbe S. C. D. ( 2013 ). Petrogenesis of Tertiary hornblende-bearing lavas in the Rhön, Germany . Journal of Petrology 54 , 2095 – 2123 . Google Scholar CrossRef Search ADS McDonough W. F. ( 1990 ). Constraints on the composition of the continental lithospheric mantle . Earth and Planetary Science Letters 101 , 1 – 18 . Google Scholar CrossRef Search ADS McDonough W. F. , Sun S. S. ( 1995 ). The composition of the Earth . Chemical Geology 120 , 223 – 253 . Google Scholar CrossRef Search ADS McKenzie D. ( 1989 ). Some remarks on the movement of small melt fractions in the mantle . Earth and Planetary Science Letters 95 , 53 – 72 . Google Scholar CrossRef Search ADS McKenzie D. , Bickle M. J. ( 1988 ). The volume and composition of melt generated by extension of the lithosphere . Journal of Petrology 29 , 625 – 679 . Google Scholar CrossRef Search ADS Münker C. , Weyer S. , Scherer E. , Mezger K. ( 2001 ). Separation of high field strength elements (Nb, Ta, Zr, Hf) and Lu from rock samples for MC‐ICPMS measurements . Geochemistry, Geophysics, Geosystems 2 , doi:10.1029/2001GC000183. Münker C. , Pfänder J. A. , Weyer S. , Büchl A. , Kleine T. , Mezger K. ( 2003 ). Evolution of planetary cores and the Earth-Moon system from Nb/Ta systematics . Science 301 , 84 – 87 . Google Scholar CrossRef Search ADS PubMed Nesbitt H. W. , Wilson R. E. ( 1992 ). Recent chemical weathering of basalts . American Journal of Science 292 , 740 – 777 . Google Scholar CrossRef Search ADS Patino L. C. , Velbel M. A. , Price J. R. , Wade J. A. ( 2003 ). Trace element mobility during spheroidal weathering of basalts and andesites in Hawaii and Guatemala . Chemical Geology 202 , 343 – 364 . Google Scholar CrossRef Search ADS Pfänder J. A. , Münker C. , Stracke A. , Mezger K. ( 2007 ). Nb/Ta and Zr/Hf in ocean island basalts—implications for crust-mantle differentiation and the fate of Nb . Earth and Planetary Science Letters 254 , 158 – 172 . Google Scholar CrossRef Search ADS Pfänder J. A. , Jung S. , Münker C. , Stracke A. , Mezger K. ( 2012 ). A possible high Nb/Ta reservoir in the continental lithospheric mantle and consequences on the global Nb budget–Evidence from continental basalts from Central Germany . Geochimica et Cosmochimica Acta 77 , 232 – 251 . Google Scholar CrossRef Search ADS Pfänder J. A. , Sperner B. , Ratschbacher L. , Fischer A. , Meyer M. , Leistner M. , Schaeben H. ( 2014 ). High‐resolution 40Ar/39Ar dating using a mechanical sample transfer system combined with a high‐temperature cell for step heating experiments and a multicollector ARGUS noble gas mass spectrometer . Geochemistry, Geophysics, Geosystems 15 , 2713 . Google Scholar CrossRef Search ADS Pilet S. ( 2015 ). Generation of low-silica alkaline lavas: petrological constraints, models, and thermal implications . In Foulger G. R. , Lustrino M. , King S. D. (eds) The Interdisciplinary Earth: A Volume in Honor of Don L. Anderson: Geological Society of America Special Paper 514 and American Geophysical Union Special Publication 71 , 281 – 304 . Pilet S. , Baker M. B. , Stolper E. M. ( 2008 ). Metasomatized lithosphere and the origin of alkaline lavas . Science 320 , 916 – 919 . Google Scholar CrossRef Search ADS PubMed Pohl J. , Soffel H. ( 1977 ). Paleomagnetic and rockmagnetic investigations of Tertiary volcanics in Northern Bavaria . Journal of Geophysics 42 , 459 – 474 . Puziewicz J. , Matusiak-Małek M. , Ntaflos T. , Grégoire M. , Kukuła A. ( 2015 ). Subcontinental lithospheric mantle beneath Central Europe . International Journal of Earth Sciences 104 , 1913 – 1924 . Google Scholar CrossRef Search ADS Renne P. R. , Mundil R. , Balco G. , Min K. , Ludwig K. R. ( 2010 ). Joint determination of 40K decay constants and 40Ar∗/40K for the Fish Canyon sanidine standard, and improved accuracy for 40Ar/39Ar geochronology . Geochimica et Cosmochimica Acta 74 , 5349 – 5367 . Google Scholar CrossRef Search ADS Ritter J. R. , Jordan M. , Christensen U. R. , Achauer U. ( 2001 ). A mantle plume below the Eifel volcanic fields, Germany . Earth and Planetary Science Letters 186 , 7 – 14 . Google Scholar CrossRef Search ADS Roeder P. L. , Emslie R. ( 1970 ). Olivine-liquid equilibrium . Contributions to Mineralogy and Petrology 29 , 275 – 289 . Google Scholar CrossRef Search ADS Romer R. L. , Hahne K. ( 2010 ). Life of the Rheic Ocean: scrolling through the shale record . Gondwana Research 17 , 236 – 253 . Google Scholar CrossRef Search ADS Rosenbaum G. , Lister G. S. , Duboz C. ( 2002 ). Relative motions of Africa, Iberia and Europe during Alpine orogeny . Tectonophysics 359 , 117 – 129 . Google Scholar CrossRef Search ADS Rudnick R. L. , Fountain D. M. ( 1995 ). Nature and composition of the continental crust: a lower crustal perspective . Reviews of Geophysics 33 , 267 – 309 . Google Scholar CrossRef Search ADS Rudnick R. L. , Gao S. ( 2003 ). Composition of the continental crust . Treatise on Geochemistry 3 , 1.64 . Salters V. J. ( 1996 ). The generation of mid-ocean ridge basalts from the Hf and Nd isotope perspective . Earth and Planetary Science Letters 141 , 109 – 123 . Google Scholar CrossRef Search ADS Schmidt M. W. , Vielzeuf D. , Auzanneau E. ( 2004 ). Melting and dissolution of subducting crust at high pressures: the key role of white mica . Earth and Planetary Science Letters 228 , 65 – 84 . Google Scholar CrossRef Search ADS Schneider K. P. , Kirchenbaur M. , Fonseca R. O. C. , Kasper H. U. , Münker C. , Froitzheim N. ( 2016 ). Role of crustal assimilation and basement compositions in the petrogenesis of differentiated intraplate volcanic rocks: a case study from the Siebengebirge Volcanic Field, Germany . Contributions to Mineralogy and Petrology 171 , 1 – 30 . Google Scholar CrossRef Search ADS Schröder B. ( 1966 ). Deckgebirgstektonik und vulkanische Förderzonen in Nordost-Bayern . Geologische Rundschau 55 , 530 – 541 . Google Scholar CrossRef Search ADS Schubert S. , Jung S. , Pfänder J. A. , Hauff F. , Garbe-Schönberg D. ( 2015 ). Petrogenesis of Tertiary continental intra-plate lavas between Siebengebirge and Westerwald, Germany: constraints from trace element systematics and Nd, Sr and Pb isotopes . Journal of Volcanology and Geothermal Research 305 , 84 – 99 . Google Scholar CrossRef Search ADS Schumacher M. E. ( 2002 ). Upper Rhine Graben: role of preexisting structures during rift evolution . Tectonics 21 , doi:10.1029/2001TC900022. Shaw D. M. ( 1970 ). Trace element fractionation during anatexis . Geochimica et Cosmochimica Acta 34 , 237 – 243 . Google Scholar CrossRef Search ADS Spera F. J. , Bohrson W. A. ( 2001 ). Energy-constrained open-system magmatic processes I: general model and energy-constrained assimilation and fractional crystallization (EC-AFC) formulation . Journal of Petrology 42 , 999 – 1018 . Google Scholar CrossRef Search ADS Staudigel H. , Plank T. , White B. , Schmincke H. U. ( 1996 ). Geochemical fluxes during seafloor alteration of the basaltic upper oceanic crust: DSDP Sites 417 and 418 . Geophysical Monograph 96 , 19 – 38 . Sun S. S. , McDonough W. S. ( 1989 ). Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes . Geological Society, London, Special Publications 42 , 313 – 345 . Google Scholar CrossRef Search ADS Tanyeri J. ( 2006 ). Geochemische Untersuchungen der tertiären Vulkanite der Hocheifel, der Heldburger Gangschar und der Fichtelgebirgszone. Ph.D. thesis, Ludwig-Maximilians-Universität München. Ulrych J. , Dostal J. , Adamovič J. , Jelínek E. , Špaček P. , Hegner E. , Balogh K. ( 2011 ). Recurrent Cenozoic volcanic activity in the Bohemian Massif (Czech Republic) . Lithos 123 , 133 – 144 . Google Scholar CrossRef Search ADS Ulrych J. , Krmíček L. , Tomek Č. , Lloyd F. E. , Ladenberger A. , Ackerman L. , Balogh K. ( 2016 ). Petrogenesis of Miocene alkaline volcanic suites from western Bohemia: whole rock geochemistry and Sr–Nd–Pb isotopic signatures . Chemie der Erde-Geochemistry 76 , 77 – 93 . Google Scholar CrossRef Search ADS Ulrych J. , Pivec E. ( 1997 ). Age-related Contrasting Alkaline Volcanic Series in North Bohemia . Chemie der Erde-Geochemistry 57 , 311 – 336 . Ulrych J. , Pivec E. , Lang M. , Balogh K. , Kropáček V. ( 1999 ). Cenozoic intraplate volcanic rock series of the Bohemian Massif: a review . Geolines 9 , 123 – 129 . Ulrych J. , Ackerman L. , Balogh K. , Hegner E. , Jelínek E. , Pécskay Z. , Přichystal A. , Upton B. G. J. , Zimák J. , Foltýnová R. ( 2013 ). Plio-Pleistocene basanitic and melilititic series of the Bohemian Massif: K-Ar ages, major/trace element and Sr–Nd isotopic data . Chemie der Erde-Geochemistry 73 , 429 – 450 . Google Scholar CrossRef Search ADS van Westrenen W. , Blundy J. , Wood B. ( 1999 ). Crystal-chemical controls on trace element partitioning between garnet and anhydrous silicate melt . American Mineralogist 84 , 838 – 847 . Google Scholar CrossRef Search ADS Vogel W. , Kuipers G. ( 1987 ). A pre-calibrated program for geological applications, Phillips New Developments . X-Ray Spectrometry 11 , 2 – 8 . Wedepohl K. H. ( 2000 ). The composition and formation of Miocene tholeiites in the Central European Cenozoic plume volcanism (CECV) . Contributions to Mineralogy and Petrology 140 , 180 – 189 . Google Scholar CrossRef Search ADS Wedepohl K. H. , Gohn E. , Hartmann G. ( 1994 ). Cenozoic alkali basaltic magmas of western Germany and their products of differentiation . Contributions to Mineralogy and Petrology 115 , 253 – 278 . Google Scholar CrossRef Search ADS Wilson M. , Downes H. ( 1991 ). Tertiary-quaternary extension-related alkaline magmatism in western and central Europe . Journal of Petrology 32 , 811 – 849 . Google Scholar CrossRef Search ADS Wilson M. , Downes H. ( 1992 ). Mafic alkaline magmatism associated with the European Cenozoic rift system . Tectonophysics 208 , 173 – 182 . Google Scholar CrossRef Search ADS Wilson M. , Downes H. ( 2006 ). Tertiary-quaternary intra-plate magmatism in Europe and its relationship to mantle dynamics. In: Gee, D. G. & Stephenson, R. (eds.) European Lithosphere Dynamics . Geological Society, London, Memoirs 32 , 147 – 166 . Google Scholar CrossRef Search ADS Wilson M. , Downes H. , Cebriá J. M. ( 1995a ). Contrasting fractionation trends in coexisting continental alkaline magma series; Cantal, Massif Central, France . Journal of Petrology 36 , 1729 – 1753 . Wilson M. , Rosenbaum J. M. , Dunworth E. A. ( 1995b ). Melilitites: partial melts of the thermal boundary layer? Contributions to Mineralogy and Petrology 119 , 181 – 196 . Google Scholar CrossRef Search ADS Zartman R. E. , Haines S. M. ( 1988 ). The plumbotectonic model for Pb isotopic systematics among major terrestrial reservoirs—a case for bi-directional transport . Geochimica et Cosmochimica Acta 52 , 1327 – 1339 . Google Scholar CrossRef Search ADS Ziegler P. A. , Dèzes P. ( 2006 ). Crustal evolution of western and central Europe. In: Gee, D. G. & Stephenson, R. (eds.) European Lithosphere Dynamics . Geological Society, London, Memoirs 32 , 43 – 56 . Google Scholar CrossRef Search ADS Ziegler P. A. , Schumacher M. E. , Dèzes P. , Van Wees J. D. , Cloetingh S. A. P. L. ( 2004 ). Post-Variscan evolution of the lithosphere in the Rhine Graben area: constraints from subsidence modelling . Geological Society, London, Special Publications 223 , 289 – 317 . Google Scholar CrossRef Search ADS © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
TI - Recurrent Local Melting of Metasomatised Lithospheric Mantle in Response to Continental Rifting: Constraints from Basanites and Nephelinites/Melilitites from SE Germany
JF - Journal of Petrology
DO - 10.1093/petrology/egy041
DA - 2018-05-03
UR - https://www.deepdyve.com/lp/oxford-university-press/recurrent-local-melting-of-metasomatised-lithospheric-mantle-in-S3Pq6rsrN0
SP - 1
EP - 694
VL - Advance Article
IS - 4
DP - DeepDyve
ER -