Die Variszischen Granitoide Mitteleuropas: Typologie, Potentielle Quellen Und Tektonothermische Zusammenh�nge

Transcript

Mineralogy and Petrology (1997) 61:67-96 Mineralogy ~"KI Petrology © Springer-Verlag 1997 Printed in Austria Variscan granitoids of central Europe: their typology, potential sources and tectonothermal relations E Finger 1, M. P. Roberts I , B. H a u n s c h m i d 1, A. Schermaier 1, and H. P. Steyrer 2 1 Institut for Mineralogie, Universit/it Salzburg, Salzburg, Austria 2 Institut ffir Geologic und Pal~iontologie, Universitfit Salzburg, Salzburg, Austria With 6 Figures Received November 12, 1996; accepted September 24, 1997 Summary During the Variscan orogenic cycle, central Europe was intruded by numerous granitoid plutons. Typological and age relationships show that the characteristics of the granitoid magmatism changed during the course of the Variscan orogeny. Five genetic groups of granitoids may be distinguished: 1. Late Devonian to early Carboniferous "Cordilleran" 1-type granitoids (ca. 370-340 Ma): These early Variscan granitoids are mainly tonalites and granodiorites. They often have hornblende and occur in association with diorites and gabbros. They form plutonic massifs in the Saxothuringian unit, in Central Bohemia and the intra-Alpine Variscides. In terms of existing models, they can be interpreted as volcanic arc granites, being related to the subduction of early Variscan oceans. Models involving mantle sources and AFC may be feasible. 2. Early Carboniferous, deformed S-type granite/migmatite associations (ca. 340 Ma): These occur in the footwall of a thick thrust in Southern Bohemia (Gf6hl nappe) and seem to represent a phase of water-present, syn-collisional crustal melting related to nappe stacking. 3. Late Visean and early Namurian S-type and high-K, I-type granitoids (ca. 340-310 Ma): These granitoids are mainly granitic in composition and particularly abundant along the central axis of the orogen (Moldanubian unit). This zone experienced a high heat flow at this time, probably as a consequence of post-collisional extension and magmatic underplating. Most of group 3 granitoids formed through high-T fluid-absent melting in the lower crust. Enriched mantle melts interacted with some crustal magmas on a local scale to form durbachites. Partial melting events in the middle 68 F. Finger et al. crust produced a number of high-T/low-P, S- and I-type diatexites and some S-type granite magmas. 4. Post-collisional, epizonal 1-type granodiorites and tonalites (ca. 310-290 Ma): These plutons can be found throughout the Central European Variscides. However, most of them occur in the Alps (near the southern flank of the orogen). Such late I-type plutons could be related to renewed subduction along the southern fold belt flank, and/or to extensional decompression melting near the crust/mantle boundary. Post-collisional mantle or slab melting may have occurred in connection with remnant subduction zones below the orogen undergoing thermal relaxation and dehydration. 5. Late Carboniferous to Permian leucogranites (ca. 300-250 Ma): Many of these rocks are similar to sub-alkaline A-type granites. Potential sources for this final stage of plutonism could have been melt-depleted lower crust or lithospheric mantle. Zusammenfassung Die variszischen Granitoide Mitteleuropas: Typologie, potentielle Quellen und tektonothermische Zusammenhiinge Im Verlauf der variszischen Orogenese intrudierten im mitteleuropfiischen Raum groBe Massen von Granitoiden. Eine Bewertung geochronologischer und granittypologischer Daten zeigt, dab sich die Magmencharakteristik mit der Zeit ver~indert hat. FiJnf Hauptgruppen von Granitoiden k6nnen unterschieden werden: 1.1-Typ Granitoide des spiiten Devon und friihen Karbon (ca. 370-340 Ma): Es handelt sich dabei durchwegs um I-Typ Tonalite und Granodiorite, welche h~iufig Hornblende fi,ihren. Typisch ffir diese Plutone ist die Pr~isenz gabbroischer oder dioritischer Endglieder. Eine Magmenentstehung aus Mantelquellen mit Modifikation durch AFC und eine genetische Verbindung zu friihvariszischen Subduktionszonen ist denkbar. 2. Syntektonische S-Typ Granite und Migmatite (ca. 340 Ma): GroBe Massen solcher Granitoide treten im Deckenstapel der stidlichen B6hmischen Masse auf. Sie repr~isentieren wasserges~ittigte, syn-kollisionale Krustenschmelzen, die sich in der N~ihe von tektonischen Uberschiebungsbahnen gebildet haben. 3. S-Typ und kalireiche l-Typ Granitoide des spiiten Vis~ und friihen Namur (ca. 340310 Ma): Diese Plutone haben in der Regel granitische Zusammensetzung und intrudierten vornehmlich in der moldanubischen Zentralzone des Orogens. Die dortige kontinentale Kruste war zu dieser Zeit einem extrem hohen W~irmefluB ausgesetzt, der vermutlich durch postkollisionale Extension mit rascher Krustenhebung und magmatischem ,,underplating" verursacht wurde. Die meisten dieser Granite bildeten sich durch Dehydratationsschmelzen der Unterkruste aus Paragneisen und eventuell auch intermedi~iren kaliumreichen Orthogneisen. Einige wenige Plutone zeigen Interaktionen mit mafischen Magmen, die aus einem angereicherten Lithosph~irenmantel stammen (Durbachite). Schmelzprozesse in der mittleren Kruste ftihrten weitr~iumig zur Bildung von Migmatiten mit groBen Anteilen an S-Typ und I-Typ Diatexiten. 4. Postkollisionale, epizonale l-Typ Granodiorite und Tonalite (ca. 310-290 Ma): Die Hauptverbreitung dieser Plutone liegt in den Alpen. Eine genetische Verbindung zu Variscan granitoids of central Europe 69 einer sp~itvariszischen Subduktionszone am Variszikums-Stidrand erscheint m6glich. Andererseits k6nnte auch die bloBe Reaktivierung und Dehydratation von alten (friihvariszischen) Subduktionszonen unter dem Orogen die Produktion entsprechender I-Typ Magmen bewirkt haben, ebenso wie ein postkollisionales Druckentlastungsschmelzen von I-Typ Quellen im Bereich der Krusten-Mantel Grenze ohne Subduktionzusammenhang. 5. Leukogranite des spiiten Karbon und Perm (ca. 300-250 Ma): Viele dieser Plutone zeigen Eigenschaften von A-Typ Graniten. Die entsprechenden Magmen sind vermutlich durch Schmelzprozesse in einer restitischen Unterkruste oder im lithosph~irischen Mantel entstanden. Introduction The ascent of granitoid magmas and their emplacement in the middle and upper crust is a typical feature of orogenic events. Research has shown that these magmas may form from contrasting source regions, and that each source rock/source region is likely to produce fairly distinct granitoid types and granitoid suites (e.g. White and Chappell, 1983; Pitcher, 1983). A particularly high melt productivity is commonly attributed to lower crustal melting of prograde metamorphosed pelite and greywacke sources (Clemens and Vielzeuf, 1987; Conrad et al., 1988; Vielzeuf and Holloway, 1988; Pati~o-Douce and Johnston, 1991; Vielzeuf and Montel, 1994). Also, granitoid magmas may form through the partial melting of hydrated oceanic crust, e.g. slab melting in subduction zones (Peacock et al., 1993), lower crustal amphibolite melting (Clemens and Vielzeuf, 1987) or remelting of underplated basaltic crust (Atherton and Petford, 1993). If temperatures are high enough, felsic magmas may be produced by remelting of restitic lower crust (Collins et al., 1982). Furthermore, it is commonly believed that granitoid magmas may evolve through fractional crystallisation, or assimilation and fractional crystallisation (AFC) processes from mantle-derived basaltic magmas, associated predominantly with subduction zones and divergent plate boundaries (Brown, 1981; DePaolo, 1981; Brown et al., 1984; Pearce et al., 1984; Hildreth and Moorbath, 1988). The timing and localities of partial melt formation during an orogenic cycle depend on the crustal composition, as well as the thermal and tectonic state within the orogen (Thompson et al., 1984; England and Thompson, 1984; Harris et al., 1984; White and Chappell, 1983). Thus, the granite inventory of an orogenic belt may be seen as an important reflection of deep structures and processes that are commonly the least accessible and the worst exposed to direct geological observation. This paper is a preliminary attempt to establish a relationship between the very extensive Variscan plutonism of central Europe in terms of granitoid types, probable source rocks and source regions. The aim is to present a coherent largescale model that relates the timing of plutonism, and the regional distribution of different granitoid types, to the main geological structures and the geodynamic evolution of the Variscan orogen. An important basis is the compilation of high quality zircon and monazite ages that have become available over the past years from different areas. It is hoped that the presentation of this model will stimulate 70 F. Finger et al. considerable discussion as to its viability and draw further attention to the significance of the abundant granitoid rocks distributed throughout Variscan central Europe. Geological and tectonothermal background The Variscides of central Europe are commonly interpreted as a collision orogen that resulted from the docking of Gondwana and Laurasia in the late Devonian to early Carboniferous (Figs. 1 and 2). The pre-collisional framework may be visualised as a ridge-trough system of continental microplates, intervening basins and partly oceanic rifts, which evolved in the early Palaeozoic along the fragmented margin of northern Gondwana (e.g., Pin, 1990; Finger and Steyrer, 1995). It is commonly believed that two major pre-collisional subduction systems developed within this ridge-trough system (Franke, 1989; Matte, 1986). These were a northern, southward-dipping system (probably a tandem of two staggered subduction zones that led to the closing of the Saxothuringian and Rhenohercynian basins - Fig. 1), and a southern, northward-dipping system that led to the closing of a Massif Central - intra-Alpine ocean(s). During the Variscan collision, the Precambrian basement of the ridges amalgamated with the thick sedimentary filling and some ocean floor of the early ~North Seo~ ? __ ,. / AM ~ _ b, ~ L-'.>..'x:.-. _..~¢~ ' -,.~.SQXOthur i ~ ' "~ iao "-¢" ::.-: ~ .. v ...:¢~-~ / Fig. 1. Simplified sketch-map showing the main geological and tectonic units of Variscan Europe (modified after Matte, 1986; Franke, 1989). AM Amorican massif; BM Bohemian massif; H Harz; MC Massif Central; MOR Moravo-Silesian Unit; RS Rheinisches Schiefergebirge; SW Schwarzwald; V Vosges Variscan granitoids of central Europe 71 o,.~ ~ ~.~ ~ ~.~ °oo ".~, oO O0 0~5~ .~,~ • o o ;~ 0 -~ ~ ~ . n ='~ ~ •~,~ ~.~ • o o o o o ~ ~.~ =o 0 .0 ,.~ o . ~ 72 F. Finger et al. Palaeozoic rift basins. Thrusting concomitant with crustal stacking was most intense in the late Devonian and early Carboniferous, leading to widespread Barrowian regional metamorphism, particularly in the internal Moldanubian unit. Parts of the Moldanubian and Saxothuringian realm were additionally affected by (subduction-zone related?) late Devonian/early Carboniferous eclogite/granulitefacies metamorphism (O'Brien and Carswell, 1993). Nappe stacking ceased in the late Visean and gave way to transpressional and transtensional tectonics, in response to the dextral wrenching of Gondwana relative to Laurasia (Arthaud and Matte, 1977; Schaltegger and Corfu, 1995). During this late Visean period, large parts of the Moldanubian unit experienced intense high-T/ low-P regional metamorphism and anatexis. In the late Carboniferous and Permian, the Variscan crust underwent postorogenic uplift and extension. However, renewed subduction of oceanic lithosphere and single terrane accretions might have occurred in the late Carboniferous along the southern flank of the fold belt (Fig. 2), after this had sheared off from the westward drifting Gondwana supercontinent (opening of a Pangea gulf - see Finger and Steyrer, 1990). Due to the involvement of these southern Variscan units in the Mesozoic to Tertiary Alpine orogen, their original relationships to each other before this later event are unclear. Granitoid groups and their geographical distribution The spatial distribution of Variscan (late Devonian to Permian) granitoids in central Europe is given in Fig. 3. This map shows that intrusions were concentrated along the central axis of the fold belt (the Moldanubian unit), where granitoid rocks make up more than half of the crystalline basement. The largest Moldanubian batholiths are the Southern Bohemian Batholith, the Central Bohemian Batholith, the Oberpfalz Batholith and the Schwarzwald batholithic terrain. Variscan plutons are very abundant in the intra-Alpine/Carpathian realm also, which represents the former southern flank of the orogen. About 20-30% of all prePermian rocks of these areas are recorded as "Variscan granitoids" on the official geological maps. Large Variscan batholiths occur in different tectonic positions of the Alpine nappe system. For instance, the Aar Batholith is part of the Helvetic unit, the Hohe Tauern Batholith belongs to the Penninic unit, and the Bernina, Schladming, Seckau/B6senstein and Semmering/Raabalpen Batholiths are part of the Austroalpine basement. The structural position of the Carpathian granitoid plutons corresponds to the Austroalpine nappe system also (Neubauer, 1994). Variscan plutons are present in the northern external zones of the fold belt as well. Several extensive plutonic massifs have been mapped in the Saxothuringian unit (Odenwald, Fichtelgebirge/Erzgebirge, Sudetic Batholith), and a few plutons occur in the Rhenohercynian unit (Harz). The principal purpose of this paper is to demonstrate that the Variscan granitoid rocks of central Europe can be loosely combined into five main groups based on age distribution, chemistry, mineralogy and structural relationships. The characteristics of the five granitoid groups are described below and outlined in Table 1. Their geographical distribution is shown in Fig. 3. The backbone of this tentative granite classification has resulted from studies on various granitoid occurrences in Variscan granitoids of central Europe 73 O , S:~ % co ~D "x, ° ~ O x @~..., xx ~ ") ~ ~i~ •~ = x 4 t~ o~ O t~ O ¢'q k~ o 2 /"/5 \ @ @ z o ' \\ ; %\ .~ %'x 0 ~,% 5 Z Z \ i "o ~ \ O ~ co I 74 F. Finger et al. ~a aa &" =- Ta ca ca 9 ca ca © © e~ ca O > ~ ' 4 == ca ca ca .9 g~ "=9 ca ca ca ¢% D'- r--- ¢-1 D'- <5 $ $ t~ D-- D-"4 D,- D-- r--- ~5 ca ©,,o O,,o +~ ca ca ~.4 ¢-i~. +~ II ~ ii ~ J~ II ~ ~.4 9 (0 {o .,..~ ~a ca ~a .~ g em O © .9 9 -d .9 9 © ca .,..~ ~9 -- ,~6 ca .,-.~ ,-~ em n:~n ¢,0 ca ~ ~= °~ ~ ~ ~a ~9 O r~ O ~$ q= ca ¢'-I ~ ¢,% ~a= m . o Variscan granitoids of central Europe 75 Austria carried out over recent years by our research group at Salzburg University. These studies have revealed an interesting systematic variation in the style of granitoid magmatism through time. In this paper, the Austrian data are combined with other published data from the western Carpathians (Petrik et al., 1994), Oberpfalz and Fichtelgebirge (Siebel, 1994), Aar Massif and Gotthard Massif (Schaltegger, 1990; Schaltegger et al., 1991); Schwarzwald (Langer et al., 1995), Odenwald (Henes-Klaiber, 1992), Bayerischer Wald (Troll, 1964), Bernina (Spillmann and Biichi, 1993), Central Bohemian Batholith (Kosler et al., 1993; Janousek et al., 1995; Holub et al., 1997) and Erzgebirge (Stemprok, 1994). Examples for which reliable geochronological, petrographic and geochemical data are available are summarised in Table 2. Unfortunately, little conclusive data are available from the Sudetic Batholith, which means that it cannot be fully included in our petrogenetic model for the time being. Mineral abbreviations after Kretz (1983) are used throughout. Group 1: The early-Variscan I-type plutons (ca. 370-340 Ma) Evaluation of literature data clearly shows that the oldest Variscan granitoid massifs are Upper Devonian to Lower Carboniferous in age. These typically include metaluminous (I-type) Hbl-bearing diorites, tonalites and granodiorites with relatively low K20, Rb/K20 ratios, and primitive isotopic compositions (Figs. 4-6). Coeval high-K I-type granodiorites with more evolved isotopic signatures are also present. An important occurrence of early Variscan I-type plutons is in the Odenwald within the Saxothuringian zone. These rocks are considered as having chemical signatures indicative of volcanic arc-type settings and include a few gabbro bodies, diorites, tonalites and large masses of normal and high-K granodiorites (Liew and Hofmann, 1988; Henes-Klaiber, 1989; 1992). In a corresponding tectonic position are granitoids and lower Carboniferous volcanics in the northern Vosges and the northern Schwarzwald. These have similar geochemical features to those of the Odenwald (Holl and Altherr, 1987; Volker and Altherr, 1987; Langer et al., 1995), and may be tentatively combined to give a northern Variscan belt of group 1 magmatism. It appears that the northern belt of group 1 plutons continues eastwards into the tectonically complex and highly disturbed Bohemian Massif. Granitoids of similar type and age have been recorded in the Central Bohemian Batholith. These include the Stare Sedlo and Mirotice orthogneisses, which are small occurrences of isotopically primitive I-type tonalite/granodiorite gneisses (Table 2), recently dated at ca. 370 Ma (Kosler et al., 1993; Kosler and McFarrow, 1994), and the metaluminous tonalite-granodiorite S~zava suite with a zircon evaporation age of roughly 350 Ma (Holub et al., 1996). The available zircon evaporation age data indicate that most of the other granitoids of the Central Bohemian Batholith intruded also around 350 Ma. These comprise a high-K, I-type suite (Blatn~ granodiorite), as well as some slightly younger peraluminous granitoids with superficial S-type features. An interesting example of an intra-Alpine, normal-K granitoid suite of earlyVariscan age are the so-called "Cetic granitoids" (Frasl and Finger, 1988). These are exotic boulders of sodium-rich, Hbl-bearing tonalites to granodiorites with very 76 F. Finger et al. ~-._ ~&~ ~,~ • ~:~ ~ .~ ~ ~ ¢) ;>-. d~ © © © 6. © =~ © 0 0 ~ ....~ . ~ ~m ~ o ~ m 0 0 ~ ~ 0 0 0 0 0 -H -H -H -H -H -H -H t'q ~ Cq C',,I C~I 't"',l ,'~ r,~ © © © ,1 < .+..2 o'h 0,1 I ~ ~.= o c¢3 .~ ~~ m~ ~ ~0 ~ ~ . Variscan granitoids of central Europe 77 oO © °~ O~ Ox ..~ 0 o ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~'~ M M ~ M ~ "T' ~ • .~ ~.~ M >, © ~o ~ b.~ . . . 0 ~ ~'= 0 ~:" 0 b~ b ~ . < < ~: ~: ~: < < < < < ~ ~-1 -H 44 -H +1 +1 -H -H -H -H t"q tq 0 0 0 ~ c ~ c ~ c ~ Cq ~ c~ 0 c~ 0 c~ ~- ~ • ~-H r "~ ~-1 t"¢3 ~.~ ~ ~ ~ cq t",l -H c ~ ,---~ ,.-.~ t",l ~ cq t"¢3 t'¢3 ~"¢3 I -~ '~ r<3 t'~ ~. ° ,.~0 t",l t'¢3 t'~ t~ ~ •~ ~ .~ ~ ~5 ~ N ~ ~ " ~ ... :: ~.~ •= ' ~ .~ o ~ ~ E~ .-~ 78 F. Finger et al. ..=aE C~ ~ i: • ~ © =,..~ 0 ~e << < < < ~oo ~o ~ . ~~ e. ~~_~ .~ ~-~[~ ~;~ oo= ~. ~-~ _~.n ~ II o ,= o N~H -H + ~ & o "~ Variscan granitoids of central Europe 79 300 e~ ne 250 E Q. e.l 200 \ 150 100 3< I .'1 I .l :: "- " 50 cZx 0 ___ I ~,- wt% K 2 0 I I I I I 1 2 3 4 5 6 Fig. 4. Plot of wt% K20 versus ppm Rb showing the variability in compositions of Variscan I-type granitoid suites (authors' data from Austria). The Cetic granitoids are the least evolved, with the lowest Rb/K20 ratio, and Rb and K20 contents. The Weinsberg granite, as a representative of the infra-crustal group 3 granitoids, has the highest Rb/K20 ratio, and Rb and K20 contents. Group 4 I-types fall between the Cetic granitoid and the Weinsberg granite fields. CG Cetic granitoids; FGD Freistadt granodiorite; HP Hochalm pluton; WBG Weinsberg granite; ZV Zillertal-Venediger pluton primitive Sr-isotope characteristics (0.703-0.705) embedded in the northern Alpine flysch. The rocks represent pieces of a former "Cetic" massif (Cetic ridge), now buried below the Alps. They were long considered as late Precambrian in age, by comparison with petrographically similar granitoids in the Moravo-Silesian Brno Massif (Frasl and Finger, 1988). However, an early Variscan age has been obtained recently (Table 2). Unfortunately, associating the Cetic granitoids with a northern or southern belt of group 1 granitoids is no simple matter, due to the yet unknown tectonic effects of the Alpine Orogeny. Isotopically primitive, Hbl-bearing I-type granitoids in the southern Vosges (Langer et al., 1995) might represent structural and temporal equivalents to the Cetic granitoid suite. Group 2: Syn-orogenic S-type granites (ca. 340 Ma) Relatively old (Lower Carboniferous) formation ages are also reported for deformed granites in the southeastern Bohemian Massif (Gf6hl and Wolfshof gneisses; Table 2 and Fig. 3). However, these rocks are clearly distinct from group 1 granitoids in their typology. Firstly, they are all S-type, leucocratic and have A/ CNK between 1.1 and 1.3. There are no I-type granitoids associated with the bodies and the isotopic signatures are strictly crustal (Table 1). Secondly, the rocks appear to be closely related to thrusting, because they occur concordantly, mainly at the base of a hot, granulite nappe (e.g., Fuchs and Matura, 1976) and are 8O E Finger et al. .3 ., • • Ill wtN I Si02 / HFG3 __ .01 ~ .002 68 I I I I I 69 70 71 72 73 74 I 75 Fig. 5. Plot of wt% SiO2 versus log(Y/Ba) showing the unique geochemical characteristics of the group 5 A-type granites compared to group 1 to 4 within the same high range of silica contents. • Eastern Alpine A-types (Finger et al., 1993; Schermaier et al., 1997), • Central Aar granite (Schaltegger, 1990), • Brocken granite, Harz (Baumann et al., 1991), "k average A-type of White and Chappell (1983). Note the overlap of highly fractionated group 3 granitoids (HFG3) from the Southern Bohemian Batholith (Plochwald granite; Haunschmid, 1989) with the very felsic group 5 granitoids. However, the HFG3 granitoids are of little volumetric importance among group 3 granitoids. The typical range of group 3 I- and S-type granitoids is at lower SiO2 contents and Y/Ba ratios. Field SBB3 comprises the bulk of Southern Bohemian Batholith group 3 I- and S-type granitoids. Data sources for all fields are given in Table 2. SBB Southern Bohemian Batholith. For other abbreviations see Fig. 4 commonly deformed with structures having formed in the magmatic state and continuing down to sub-solidus conditions. Unlike group 1 granitoids, they show ubiquitous transitions into syn-deformational migmatites, e.g., in Lower Austria. Group 3: Late Vis~an to early Namurian S-type and high-K 1-type granitoids (ca. 340-310 Ma) Granitoids of this age represent the largest volume of Variscan plutonism. Unlike those of group 2, they are weakly or undeformed and cross-cut structures developed during Variscan regional metamorphism. They are particularly abundant within the Moldanubian unit (see Fig. 3); in the Southern Bohemian Batholith, Variscan granitoids of central Europe I X X I~ Ix x x x x x I i x x 1 Cetic granitoids Odenwald Central Bohemian Batholith I x 81 ;~ I GfOhl gneiss I \ \ ",I x x \ I Southern Bohemian Batholith Sehwarzwald Oberpfalz I k Harz Aar Massif Hohe Tauern Batholith I I [\ Hohe Tauern Batholith Southern Bohemian Batholith iliiiiiiiiliIII;iiiiiii;;; ....... ,,H, -10 I'".'- I I I I I I 0 \ \ \ \ \ \1 h \ \ \ \ "~1 \ h "~ \ \ % \ \ \1 mmmm~ llIll Ill IIIIIV, III" I 11 =~i I I I I I I l i I I ~Ndi "," .'-" . " ' 1 Sr, I I I I I l l l l 0.710 Fig. 6. Diagram showing the ranges of initial eNd and initial 875r/86Sr of the granitoid groups calculated at their respective ages. Note that group 1 granitoids have consistently the most primitive isotope compositions. For data sources see Table 2 and Siebel (1994), Langer et al. (1995) and Liew et al. (1988) these granitoids make up more than two thirds of the outcrop area. Also, in the Schwarzwald, most granitoid plutons appear to be of late Visean/early Namurian age (Todt, 1976; Langer et al., 1995; Schaltegger, 1995). S-type granitoids of a similar age are present in the Alps and Carpathians. However, these are of minor importance compared to younger calc-alkaline and A-type granitoids (groups 4 and 5). The group 3 plutons may be broadly divided into four subgroups. The first subgroup (3.1) consists of moderately peraluminous, S-type granites with some muscovite or cordierite and sometimes garnet as additional Al-oversaturated phases. These include the Eisgarn granite of the Southern Bohemian Batholith, the Falkenberg and Flossenburg granites of NE-Bavaria (Siebel, 1994), the B~haldeSchluchsee granites of the Schwarzwald (Emmermann, 1977), and the Granatspitz granite of the Hohe Tauern (Finger et al., 1993). It is generally agreed that these peraluminous granites were derived mainly from partial melting of paragneisses in the lower to mid crust. The second sub-group (3.2) includes highly potassic, weakly peraluminous, sometimes metaluminous, K-feldspar megacrystic biotite-granitoids. Important examples are the Weinsberg granite in the Southern Bohemian Batholith (Frasl and Finger, 1991; Finger and Clemens, 1995), the Leuchtenberg granite in the Oberpfalz (Siebel, 1995a), the Kristallgranit I in Bavaria (Kraus, 1962), the 0.720 82 E Finger et al. Oberkirch granite in the Schwarzwald (Emmermann, 1977) and the Knorrkogeland Hochweif3enfeldgneiss in the Hohe Tauern (Schermaier, 1993). These widespread megacrystic Kfs granitoids have a considerable silica range from felsic ( ~ 70 wt% SiO2) to more mafic quartz-monzonite end-members with silica contents of around 60 wt%. Although some of the more felsic variants approach S-type granite compositions, most of the megacrystic K-feldspar granitoids are I-types in view of their low A/CNK ratios between 0.95 and 1.10, and intermediate Sr isotope values (mostly 0.706 to 0.710). The presence of large amounts of magmatic zircons in this subgroup suggests high magma temperatures in the range of 800 to 900 °C (Broska et al., 1995). The third subgroup (3.3) can be termed magnesio-potassic plutons. As with other group 3 granites, they have high K20 contents and mostly contain megacrystic K-feldspar. However, unlike the other group 3 granites, they are characterised by high compatible element contents such as Mg and Cr. They are commonly Hbl-bearing melagranites with transitions into monzonite and syenite. Typically, they are associated with small stocks and dikes of coeval mafic to intermediate rocks. Commonly, mafic-felsic magma mingling phenomena are present. The most famous example are the "durbachite" plutons of central and eastern Bohemia (Holub, 1977). These exhibit geochemical and isotopic mixing patterns between crustal partial melts and enriched upper mantle mafic magmas (Gerdes et al., 1995; Gerdes, 1997). Other examples of magnesio-potassic group 3 plutons are the Meigen syenite pluton in eastern Germany (Wenzel et al., 1991), the "granite de Crete" of the Vosges (Langer et al., 1995), the Guiv metasyenite of the Aar Massif (Schaltegger et al., 1991) and the Romate syenite gneiss of the eastern Tauern Window (Finger et al., 1993; Haunschmid, 1993). The fourth subgroup (3.4) are leucocratic, near-minimum peraluminous (S-type) granites, which are commonly fine-grained with primary muscovite. These are slightly younger and often cross-cut other group 3 granitoids. The Altenberg granite in the South Bohemian Batholith with a formation age of ca. 310-320 Ma (Frasl and Finger, 1991) is an example of this type, and seems to be spatially associated with a major mid-crustal shear zone. Fine-grained granites of the Oberpfalz (e.g., the 310 Ma Mitterteich granite; Siebel, 1995b) appear to be petrologic and temporal analogues. Group 4 granitoids: Late Variscan calc-alkaline plutonism (ca. 310-290 Ma) Group 4 granitoids are epizonal metaluminous to weakly peraluminous tonalites, granodiorites and granites with late-Namurian/Stephanian ages between ca. 290 and 310 Ma. Most of the group 4 granitoids can be clearly classified as I-types on the basis of their high sodium contents and low to moderate A/CNK ratios (Chappell and White, 1974). A few of the plutons are normal-K and display a wide variation in silica content down to diorite and gabbro end-members typical of Cordilleran I-type plutons (e.g., Silver and Chappell, 1988). However, most are high-K biotite granodiorites without obvious genetic relationships to mafic rocks. Textures are typically fine- to medium-grained, equigranular; porphyritic or coarse-grained variants are rare. Group 4 granitoids are most common in the Alpine-Carpathian chain (Finger et al., 1993; Petrik et al., 1994). Granitoids of the same age and type have also been Variscan granitoids of central Europe 83 recognised in the Southern Bohemian Batholith (Mauthausen/Freistadt group; Frasl and Finger, 1991). Here, these granitoids intrude older (group 3) granitoids, forming small high level plutons, stocks and dikes. Contacts are sharp and discordant, indicating that the group 3 granitoids had already been uplifted and cooled by this time. Also, in the Hohe Tauern Batholith, group 4 granitoids show field relations typical of high level plutons. In the few cases where contacts with group 3 granitoids and migmatites are exposed, the group 4 I-type plutons were always found to be significantly younger (Schermaier, 1993). A similar situation has also been documented in the field and by high-precision U-Pb geochronology in the Aar Massif (Schaltegger and Corfu, 1992). Here, high-K granitoid intrusions of late VisOan age (group 3 granitoids) are followed by I-type granitoid plutonism in the late Namurian (e.g., Dtissi diorite - see Table 2). However, calc-alkaline plutonism in the Western Alps appears to be volumetrically less important than in the Eastern Alps and Carpathians. In the northern part of the Variscan Fold Belt, late-stage I-type intrusions of considerable size are documented. However, age data for these rocks are scarce. Hornblende-bearing I-type granites at the western end of the Karkonose pluton (Sudetic Batholith) were recently dated at roughly 300 Ma (KrOner et al., 1994). Also, late Variscan granodiorites with hornblende and cognate mafic inclusions from southern Poland have been described (e.g. Lorenc, 1994; Pin et al., 1989). In contrast to the rest of central Europe, late Carboniferous calc-alkaline plutonism appears to have played no role in the Schwarzwald. Group 5 granitoids: Late stage leucogranites (ca. 300-250 Ma) Recently, a number of small leucocratic plutons with high-Y A-type granite characteristics in the broad sense of Eby (1990) have been recognised in the Alps and Carpathians (Finger et al., 1993; Petrik and Broska, 1994). The rocks are generally weakly peraluminous or metaluminous with intrusion ages that appear to be between 300 and 250 Ma. As with many other more typical A-types, they are characterised by high K, Rb, and Th contents, a high FeO/MgO ratio, distinct negative Ba and Eu anomalies, and high HREE and Y abundances (Fig. 5). Fluorite is a common accessory mineral, implying that the magmas were rich in E On Pearce-type discrimination diagrams (Pearce et al., 1984), the rocks plot mostly in the "within-plate-granite" field. In the Hohe Tauern Batholith it can be clearly demonstrated that the leucocratic A-types are always the youngest intrusions, and frequently they intrude bodies of group 4 granitoids (Haunschmid, 1993). The intrusion event probably occurred during the latest Carboniferous or Permian (see Vavra and Hansen, 1991). However, for these Hohe Tauern Batholith rocks, precise geochronological data are still missing. An early Permian age of 274 • 13 Ma (U-Pb zircon) has been obtained recently for a late Variscan high-Y A-type granite from the Carpathians (Uher and Pushkarev, 1994). The ca. 290 Ma Brocken granite from the Harz mountains (Baumann et al., 1991) is reminiscent of A-type granites in terms of its chemistry (Fig. 5). The Aar massif contains large bodies of granites chemically similar to A-types. These have been dated as Stephanian by Schaltegger and Corfu (1992). Unlike the A-types of the Eastern Alps and Carpathians, those of the Aar Massif are characterised by low initial Sr isotope ratios (Fig. 6). 84 E Finger et al. In the Moldanubian unit, similar high-Y A-type leucogranites appear to be rare. Some very small bodies of leucocratic high Nb-Ta granites, that could be broadly classified as A-types in the sense of Eby (1990), have been found near the AustroCzech border (Homolka granite; Matejka and Klecka, 1992). However, because of their chemistry (low Y, low HREE) and slightly older age (ca. 310 Ma; Scharbert, 1987) these high-Nb "A-types" are not really equivalent to the intra-Alpine high-Y A-types, and appear to represent examples of extremely fractionated, late group 3 (subgroup 3.4) S-type granitoids. Likewise, a small body of similar high-Nb granite ("K1-Gneis"; Finger et al., 1985) is exposed near the major group 3 S-type granite in the Hohe Tauern Batholith (Granatspitz granite). Both granites have the same zircon age of about 330 Ma (Eichhorn et al., 1993; Cliff, 1981), implying that they are cogenetic. The famous, heavily mineralised "younger leucogranites" of the Saxothuringian unit (Fichtelgebirge, Erzgebirge) also may have had an origin via fractional crystallisation from S-type granite parents (see Breiter and Siebel, 1995; Siebel, 1994). However, the age of these granites is well constrained at between 290 to 305 Ma, and thus a genetic relation to group 3 S-types is not likely. It could be that the younger granites of the Fichtelgebirge/Erzgebirge represent an independent group of late Variscan leucogranites. Discussion The temporal and geographical distribution of the different types of Variscan granitoid rocks provide insights into possible tectono-thermal processes that may have operated in the crust throughout the Variscan orogenic evolution of the central European area. In the past, many geologists have speculated that the bulk of Variscan granite plutonism could have been magmatic arc related, associated with early-Variscan subduction systems and their successive maturation (Anderson, 1975; Nicolas, 1972; Liew and Hofmann, 1988; Finger and Steyrer, 1988). However, new geochronological and geological data have shown that intrusion of most of the Variscan plutons occurred long after the closure of the early-Variscan oceans. This implies that early-Variscan subduction processes probably played no major role in granite formation. Group 1: the occurrences of late Devonian to Lower Carboniferous group 1 I-type granitoids appear to be the only ones that might have formed contemporaneously with early Variscan subduction processes. They may represent magmas derived from subduction-modified mantle or primitive underplated crust that were further modified by AFC-type processes during ascent and emplacement. Both the Odenwald and the Central Bohemian Batholith contain isotopically evolved high-K granodiorite, which suggests that crustal material was available to make a significant contribution in group 1 granitoids. Within the scope of existing tectonic models (e.g., Franke, 1989), the northern Variscan belt of group 1 I-type granitoids (Odenwald, northern Schwarzwald and Central Bohemian Batholith) may be related to the southwards subduction of northern Variscan oceans. The fact that the Odenwald and Central Bohemian Batholith are similar in age and typology, lends support to the tectonic model of Variscan granitoids of central Europe 85 Edel and Weber (1995) based on geophysical data, according to which both areas were in a similar structural zone before dissection by large strike-slip faults. In the south of the Variscan fold belt in central Europe, the northward closure of the southern Variscan oceanic realm(s) may have caused some early-Variscan subduction-related plutonism in the eastern and western Alps (Bonin et al., 1993). However, the exact locations of the appropriate subduction zones are still in debate (Frisch and Neubauer, 1989; Schermaier et al., 1997). Group 2: In contrast to group 1, the deformed S-type leucogranite/migmatite suites of group 2 that occur adjacent to major thrust planes in the southeastern Bohemian Massif appear to represent examples of syn-collisional granites. Indeed, thrust tectonics is commonly seen as one of the most potential mechanisms for producing near-minimum S-type granitic partial melts (Harris et al., 1984). Hot, quartzo-feldspathic amphibolite-facies rocks occurring near the base of a thick nappe pile may start to melt, when water is introduced from the overridden rocks of the footwall that are undergoing prograde metamorphism. An additional motor for this syn-tectonic melting may be local shear-heating, which can raise temperatures by some tens of degrees (England and Thompson, 1984). Relatively low-T melts, generated in this fashion, commonly lack sufficient thermal energy to rise into higher crustal levels and, therefore, will mostly solidify within the stress field where they were produced. Most syn-collision granites will therefore commonly appear in the field as deformed leucogranite/ migmatite associations near major thrusts. For the zone in which the group 2 granitoids occur, petrological data indicate that temperatures of at least 700 °C were reached during the collision stage (O'Brien and Carswell, 1993). This means that the temperature of the water-saturated granite-solidus was definitely overstepped. Moreover, the lower overthrusted units contain early Palaeozoic sediments, that were able to supply sufficient quantities of water through the dehydration and recrystallisation of phyllosilicates. The fluids thus released would be capable of fluxing melting under suitable thermal regimes during the collision event (Finger and Steyrer, 1995). Petrologic evidence for the low temperatures of the group 2 magmas is the generally high amount of inherited zircons (Friedl, 1997). It is likely that further water-saturated syn-collision granites are present in other parts of the central European Variscan orogen, but have yet to be recognised. Group 3: In terms of mineralogy and geochemistry, group 3 is the most diverse of all the five groups, and involves quite different suites of I- and S-type granitoid rocks , which can be further divided into four sub-groups (see above). However, from petrologic and geochemical studies in the South Bohemian Batholith (Finger and Clemens, 1995; Gerdes, 1997) it may be concluded that, in nearly all cases, crustal material appears to have been the major source for the magmas (Fig. 6). Judging from the high thermal gradient evidenced by mid-crustal metamorphic rocks in the realm of the Southern Bohemian Batholith (Knop et al., 1995), the lower crust was hot enough to permit fluid-absent melting of both I- and S-type sources. Suitable I-type sources could have been intermediate, K-rich orthogneisses or low-A1 greywackes. Thus, it is not necessary to invoke complex mantle 86 E Finger et al. magma-crustal magma AFC type mixing processes for the formation of the group 3 I-type granitoids (Finger and Clemens, 1995). The particularly high-K chemistry of some of the granitoids (e.g., the megacrystic K-feldspar-bearing I-types of subgroup 3.2) may result from the fractionation of K-free minerals (orthopyroxene, plagioclase, some garnet and cordierite) in the lower crust, which formed as a consequence of the breakdown of biotite + quartz + plagioclase assemblages. Evidence for this is the presence of orthopyroxene or garnet/cordierite-rich cumulates brought up locally with the melts (Haunschmid, 1989; Haunschmid and Finger, 1994). A significant contribution of mantle melts is verified only for the magnesiopotassic plutons (the "durbachites"), which are of minor importance within group 3. The mantle component in these plutons appears to have been derived from an enriched lithospheric mantle domain (Wenzl et al., 1991; Gerdes, 1997). Although we consider the lower crust as the main magma source region, the role of the middle crust in the formation of group 3 granitoids should not be ignored. Several mid-crustal melting events can be documented in the Moldanubian unit between 340 and 310 Ma. The most typical rocks produced by late-Vis6an mid-crustal melting are ca. 340 to 320 Ma S-type diatexites. These are restite-rich, little homogenised and unfractionated, in-situ magmas that formed mainly through the partial-melting of metapelites and high-A1 metagreywackes (Mehnert, 1968; Thiele, 1962). The most relevant melting reaction was probably Bt + Sil + Qtz = melt + Crd + Grt + Kfs (e.g., Bliimel and Schreyer, 1976). Most of the diatexites were formed in connection with Moldanubian low-P regional metamorphism that occurred slightly earlier than the intrusion of the granitoid magmas derived from the lower crust (Friedl et al., 1993). An interesting example of a group 3 diatexite is the "Schlierengranite" in the South Bohemian Batholith (Finger, 1986). This I-type magma with partly liquidus hornblende apparently formed through local fluid-present melting of mildly peraluminous biotite - plagioclase - quartz gneisses in the contact aureole of the Weinsberg granite. Melting occurred because the Weinsberg granite transferred extra heat and water exsolved from the decompressed crystallising magma, into the middle crust. Thus, it appears that temperatures in the Moldanubian middle crust were not high enough for fluid-absent melt formation by Bt + P1 + Qtz breakdown (Finger and Clemens, 1995). Coupled with further heat input from the hot granitoid magmas, the middle crust seems to have remained in a state of anatexis for a fairly long time span as isotopic cooling ages seem to suggest (Scharbert et al., 1997). The latest, mid-crustal melting events that formed the Altenberg granite in the South Bohemian Batholith could have been triggered by the local introduction of meteoric water into the still hot middle crust along shear zones as suggested by Frasl and Finger (1991). There have been intense discussions over the past years concerning the tectonothermal significance of the group 3 granitoids and possible mechanisms that may have caused the widespread crustal fusion during the late Vis6an/early Namurian. It is doubtful that the high magma-temperatures of the granitoids can be reached through simple burial and radioactive heat production in a thickened postcollisional crust. A more feasible mechanism to promote such high-T partial melting in the crust is extra heat-input from the mantle. Therefore, it would seem Variscan granitoids of central Europe 87 that mantle melting and subsequent intra- and underplating of the crust by hot mafic magmas was the ultimate cause of the extensive late Vistan/early Namurian plutonism in the internal Variscan units. Rapid isothermal exhumation of the hot crust after the collision phase (Biittner and Kruhl, 1997) may have been responsible for the late Variscan low-P/high-T regional metamorphism and the initial stages of mid-crustal partial melting. Given the likelihood of large scale magmatic underplating, the question arises of why there is so little evidence of mantle magmatism in connection with group 3 granitoids. One reason for this could be that the underplated mafic magmas may have lost much of their buoyant energy through providing heat for crustal melting. Also, the reduced density of the partially molten lower crust may have been an additional factor that caused the mafic magmas to pond at or near the crust/mantle boundary. Thus, it would appear that a "fertile" (non-granulitic) lower crust may constitute an effective barrier against uprising mantle melts in regimes of enhanced mantle heat flow. Group 4: The presence of some zoned normal-K plutons with gabbro and diorite members suggests the involvement of mantle-derived melts in group 4. However, a decision as to what extent mantle melts were actually involved in the formation of the granitoids in this group is not possible on the basis of the available data. Most of the high-K, I-type, granodiorite and granite plutons of group 4 lack any obvious relationship to mafic mantle melts. The Sr and Nd isotope values of these rocks are such that they would theoretically allow derivation solely from crust (Fig. 6). Possible sources of these magmas could be basaltic to basaltic andesitic high-K amphibolites (e.g., Clemens and Vielzeuf, 1987; Roberts and Clemens, 1993). Conversely, it may be argued that slab-melting or mantle-melting, combined with extensive AFC, may produce similar granodiorite melts. In terms of a tectonothermal event, an important question is, whether it is likely from the geological point of view that widespread anatexis, related to group 4 granitoid formation, occurred at ca. 300 Ma in the lower crust of the central European Variscan orogen. Several points appear to argue against this. Firstly, almost no S-type plutons were formed at that time (an exception might be the Erzgebirge/ Fichtelgebirge area - see below). Secondly, the melt productivity (fertility) of the Variscan lower crust was probably quite limited in the late Carboniferous, since widespread granulite-facies metamorphism (i.e., dehydration of the lower crust) and melt extraction had occurred previously with the formation of group 3 granitoids. Thirdly, no major phases of metamorphism are recorded at 300 Ma, suggesting that there was no substantial regional reheating of the crust at that time. Finally, almost no cogenetic migmatites, or lower crustal restites (that may evidence "dry" lower crustal fractionation of the melts) are brought up with the magmas as in case of group 3 granitoids. Therefore, it is reasonable to assume that the initial melting sites of the group 4 magmas may have been beneath the crust, with the formation of parental basalts or andesites. To explain this, several different tectonothermal scenarios may be invoked. A late-Variscan active-plate-margin model: Because of the extraordinary abundance of post-collisional I-type granitoids in the Alps and Carpathians, and 88 E Finger et al. their northward transition in eNd towards more crustal values (see comparison for southern Bohemian and Hohe Tauern group 4 I-types in Fig. 6), Finger and Steyrer (1990) proposed that an important convergent ocean-continental margin might have developed along the southern flank of the Variscan orogen in the late Carboniferous, related to the westerly drift of Gondwana and the opening of the Pangea gulf (Fig. 2). However, this model is not totally supported by other geological constraints (Neubauer, 1991). Also, it does not explain the considerable amounts of late Variscan I-type granites in the north of the fold belt. A remnant subduction-zone activation model: Early-Variscan subduction zones may have survived below the fold belt. These could have been reactivated at about 300 Ma producing volcanic arc type melts (Finger and Steyrer, 1991). This model is particularly effective in explaining the post-collisional I-type plutons in the northern part of the fold belt, as there is no evidence for a 300 Ma old ocean in this area. Melting or dehydration of remnant oceanic crust beneath the collision zone may have been triggered by further descent of the slabs after detachment from the overlying orogen. Thus, renewed subduction of oceanic crust is not required. A similar model has been recently suggested for post-collisional I-type plutons of the Alpine orogeny (von Blankenburg and Davies, 1995). Alternatively, simple thermal equilibration of previously subducted oceanic crust with the surrounding mantle would probably cause dehydration reactions, and hence, slab melting and/or melting in the overlying mantle wedge. Theoretically, such a process would be able to produce post-collisional magmas with volcanic arc-type chemical signatures, comparable to those of active subduction settings, without any requirement for renewed subduction. Magmas derived therefrom may well have been modified during ascent through collision thickened continental crust. Perhaps, the final result could be I-type granitoids with dominantly crustal isotopic and geochemical signatures. Decompression melting of mantle unrelated to subduction zones: Based on their studies in the Aar massif, Schaltegger and Corfu (1995) proposed that the main source of the late-Variscan plutonism was enriched lithospheric mantle, that began to melt as a consequence of post-orogenic uplift and adiabatic decompression. They drew this conclusion mainly from the observation that a number of late Variscan transtensional basins opened in the Aar Massif during the time span when the granites were intruded. However, unlike the Eastern Alps and Carpathians, the Aar Massif contains only minor amounts of typical group 4 I-types (Fig. 3). Most of the Aar granites appear, in our opinion, to be A-types, because of their geochemical characteristics (Fig. 5). Therefore, models that describe the situation in the Aar Massif are not necessarily applicable to the late Variscan calc-alkaline plutonism in the rest of the Alps and Carpathians. Bearing in mind the dextral westward wrenching of Gondwana relative to Laurasia in the Upper Carboniferous, and the opening of the Palaeotethys gulf in the east (Fig. 2), it is possible that transpressional-transtensional tectonics between continental blocks persisted in the more western parts of the Variscides (e.g. in the Aar Massif), whilst at the same time, Palaeotethys oceanic crust was already subducted in the east (see model 1). The scarcity of group 4 I-types in the Aar massif and the Black Forest and their Variscan granitoids of central Europe 89 relative abundance in the Eastern Alps and Carpathians, seems to support this interpretation. Group 5: Some of the late stage leucogranites of this group appear to be highly fractionated S-type granites (Fichtelgebirge and Erzgebirge). These might evidence late stage, local crustal fusion events, possibly related to thrust-reactivation and further magmatic underplating as suggested from seismic data from the margins of Rhenohercynian and Saxothuringian zones (Tomek, personal communication). However, most of the group 5 leucogranites, particularly those in the Alps, are distinct high Y magmas, which probably require a separate magma source (c.f., Harris et al., 1984). Since the high Y A-types appear to be generally younger than the group 4 I-types, an explanation through fractional crystallisation from the latter type is not feasible. One possibility is that they formed from an enriched subcontinental mantle source, that started to melt as a consequence of orogenic collapse (Schaltegger and Corfu, 1995). This model is particularly reasonable for the A-types of the Aar Massif with their low Sr initial ratios. Cognate basalt intrusions, that could support such a model have not been reported from the Aar Massif as yet. However, mafic members of the suite might have preferred to pond at the crust-mantle boundary, due to their higher density (Bonin, 1992). The fact that group 5 A-type granites occur preferably in the Alps and Carpathians and rarely in the central Moldanubian unit, is similar to the situation in the Australian Lachlan fold belt. Here, felsic A-type plutonism was most intense in the outer I-type granite belt, and of much less importance in the inner S-type granite province (White and Chappell, 1983). A-type granite formation in the Lachlan fold belt has been interpreted as high-T remelting of a chemically distinct restitic lower crust, formed after large volumes of 1-type melts had been extracted (Collins et al., 1982). It is possible that a similar genetic interpretation may hold true for some of the late stage Variscan A-type granites in the Eastern Alps and Carpathians. However, just as in case of group 4 1-types, the exact sources of the late Variscan A-types are still not well constrained, and it would seem that they were derived from a variety of different sources with some of the leucogranites representing highly fractionated S-type melts (e.g., those in the Fichtelgebirge and Erzgebirge; Breiter and Siebel, 1995). To resolve this problem, a thorough regional geochemical and isotopic study is urgently required. Conclusions The Variscan granitoid plutonism of central Europe is distinctly heterogeneous in terms of age and regional distribution of the different granite types. The suggested subdivision of the plutons into five groups is, in some ways, similar to the granite classification schemes proposed for collisional orogens such as the Himalayas (e.g., Harris et al., 1984). Here, remnants of pre- to early collisional subductionrelated plutons are preserved along sutures, syn-collisional granites have formed near major thrust planes, and some post-collisional I- and A-type granites occur during the late extensional phase of the orogeny. However, there are several unique features of the Variscan granitoids that do not fit with such a simple "Himalayan" continent-continent collision type model. 90 E Finger et al. Firstly, unlike the Himalayas, pre-collision I-type rocks, that might represent the former arc above a subduction zone, are not dominant. This may indicate that subduction played little role in the pre-collisional evolution of the Variscan crust, and supports the common hypothesis that any presumed (intra-Variscan) oceans (Rhenohercynian, Saxothuringian, Massif Central, Intra-Alpine oceans) were small. Secondly, a major characteristic of the European Variscides is widespread postcollisional melting of the lower crust and production of voluminous high temperature S-type and high-K I-type granite magmas. Such conditions are special, requiring a large heat input, possibly through mantle-derived mafic magmas, into a lower crust apparently devoid of older, infertile granitic or granulitic material. Models that invoke delamination of a granulitic lower-crust (orogenic root) could explain the high heat flow at this stage. Thirdly, there is a particularly large number of late-stage calc-alkaline, I-type plutons within Variscan Europe. This may point towards a protracted evolutionary history of transpressive and transtensive crustal dynamics with or without renewed subduction of ocean crust. The tectonothermal information provided by the granitoids is important and should not be overlooked when large-scale tectonic models for the Variscan Orogen are constructed, as they are "windows" through which the composition of the pre-Variscan lower crust can be indirectly viewed. Acknowledgements This paper is largely based on results of various regional granite research projects that were performed over the last years at Salzburg University. G. Frasl is particularly thanked for his ideas and samples. We would also like to thank the Bundesministerium fiir Wissenschaft und Forschung and the Fonds zur F6rderung der wissenschaftlichen Forschung (FWF) for financial support through grants OWP 69, P-9434, P-7353 and P10708. This work benefited from discussions with many others too numerous to mention during various excursions and meetings. J. Kosler (Prague) is thanked for providing us with a useful abstract recently published by the local Czech geological community. A thought provoking review by A. White made a significant contribution to improving the clarity of the article. References Anderson TA (1975) Carboniferous subduction complex in the Harz Mountains, Germany. Geol Soc Am Bull 86:77-82 Arthaud F, Matte P (1977) Late Paleozoic strike-slip faulting in southern Europe and northern Africa: result of a right-lateral shear-zone between the Appalachians and the Urals. Geol Soc Am Bull 88:1305-1320 Atherton MP, Petford N (1993) Generation of sodium-rich magmas from newly underplated basaltic crust. Nature 362:144-146 Baumann A, Grauert B, Mecklenburg S, ½"nx R (1991) Isotopic age determinations of crystalline rocks of the Upper Harz Mountains, Germany. Geol Rundsch 80:669-690 Bliimel P, Schreyer W (1976) Progressive regional low-pressure metamorphism in Moldanubian metapelites of the northern Bavarian Forest, Germany. Krystallinikum 12:7-30 Variscan granitoids of central Europe 91 Bonin B (1992) The role of crust in the development of A-type alkali-feldspar granites in within-plate bimodal alkaline magmatism. Ind Geol Assoc Bull 25:11-27 Bonin B, Briindlein P, Bussy F, Desmons J, Eggenberger U, Finger F, Graf K, Marro C, Mercolli I, Oberhiinsli R, von Quadt A, von Raumer J, Schaltegger U, Steyrer H P, Visona D, Vivier G (1993) Late Variscan magmatic evolution of the Alpine basement. In: von Raumer JF, Neubauer F (eds) The pre-Mesozoic geology in the Alps. Springer, Berlin Heidelberg New York Tokyo, pp171-202 Breiter K, Seibel W (1995) Granitoids in the Rozvadov Pluton Southern Bohemia and Oberpfalz. Geol Rundsch 84:506-519 Broska I, Gerdes A, Haunschmid B, Schindlmayr AO, Finger F (1995) Magma temperatures in the Southern Bohemian Batholith estimated on the basis of zircon solubility. Terra Nova 7 [Suppl 1]: 143 Brown GC (1981) Space and time in granite plutonism. Phil Trans Roy Soc Lond A301: 321-336 Brown GC, Thorpe RS, Webb PC (1984) The geochemical characteristics of granitoids in contrasting arcs and comments on magma sources. J Geol Soc Lond 141: 413-426 Biichi HJ (1994) Der variskische Magmatismus in der 6stlichen Bernina (Graubfinden, Schweiz). Schweiz Mineral Petrograph Mitt 74:361-364 Biittner S, Kruhl JH (1997) The evolution of a late-Variscan high-T/low-P region: the south-eastern margin of the Bohemian Massif. Geol Rundsch 86:21-38 Chappell BW, White A JR (1974) Two contrasting granite types. Pacific Geol 8:173-174 Choubert G, Faure-Muret A (1980) Atlas g6ologique du monde feuille 9 (scale 1 10 000 000). Int Geol Congr Paris Clemens JD, VielzeufD (1987) Constraints on melting and magma production in the crust. Earth Planet Sci Lett 86:287-306 Cliff RA (1981) Pre-Alpine history of the Pennine zone in the Tauern window Austria: U-Pb and Rb-Sr geochronology. Contrib Mineral Petrol 77:101-104 Collins WJ, Beams SD, Chappell BW, White A JR (1982) Nature and origin of A-type granites with particular reference to south-eastern Australia. Contrib Mineral Petrol 80: 189-200 Conrad WK, Nicholls IA, Wall VJ (1988) Water-saturated and -undersaturated melting of metaluminous and peraluminous crustal compositions at 10 kbar: evidence for the origin of silicic magmas in the Taupo Volcanic Zone New Zealand and other occurrences. J Petrol 29:765-803 DePaolo DJ (1981) Neodymium isotopes in the Colorado Front Range and crust-mantle evolution in the Proterozoic. Nature 291:193-196 Eby GN (1990) The A-type granitoids: a review of their occurrence and chemical characteristics and speculations on their petrogenesis. Lithos 26:115-134 Edel JB, Weber K (1995) Cadomian terranes, wrench faulting and thrusting in the central European Variscides: geophysical and geological evidence. Geol Rundsch 84:412-432 Eichhorn R, Schiirer U, Jagoutz E, HOll R (1993) Die Scheelit-Lagerst~itte Felbertal - das Alter der Vererzung und der metamorphen Oberpr~igungen. Ber Dtsch Mineral Ges 5/1:122 Emmermann R (1977) A petrogenetic model for the origin and evolution of the Hercynian granite series of the Schwarzwald. N Jb Mineral Abh 128:219-253 England P, Thompson AB (1984) Pressure-temperature-time paths of regional metamorphism 1. Heat transfer during the evolution of regions of thickened continental crust. J Petrol 25:894-928 Finger F (1986) Die synorogenen Granitoide und Gneise des Moldanubikums im Gebiet der Donauschlingen bei Obermtihl (Ober6sterreich). Jb Geol Bundesanst 128:383-402 92 E Finger et al. Finger F, Steyrer HP (1988) Granite-types in the Hohe Tauern (Eastern Alps Austria) some aspects on their correlation to Variscan plate tectonic processes. Geodinam Acta 2:75-87 Finger F, Steyrer HP (1990) I-type granitoids as indicators of a late-Paleozoic convergent ocean-continent margin along the southern flank of the Central European Variscan orogen. Geology 18:1207-1210 Finger F, Steyrer H P (1991) Comments and replies on "I-type granitoids as indicators of a late Paleozoic convergent ocean-continent margin along the southern flank of the central Variscan orogen". Geology 20:1245-1248 Finger F, yon Quadt A (1993) Genauere U/Pb Alter ffir Granitgneise durch sorgf~ltige Zirkonselektion unter dem Durchlichtmikroskop - Der Knorrkogelgneis der Hohen Tauern als Beispiel. Ber Deutsch Mineral Ges 5/1:118 Finger F, Clemens JD (1995) Migmatization and "secondary" granitic magmas: effect of emplacement and crystallization of "primary" granitoids in Southern Bohemia, Austria. Contrib Mineral Petrol 120:311-326 Finger F, Steyrer HP (1995) A tectonic model for the eastern Variscides: indications from a chemical study of amphibolites in the south-eastern Bohemian Massif. Geol Carpath 46:137-150 Finger F, Kraiger H, Steyrer HP (1985) Zur Geochemie des K1-Gneises der Scheelitlagerst~itte Felbertal (Pinzgau/Salzburg) - ein Vorbericht. Karinthin 92:225-235 Finger F, FrasI G, Haunschmid B, Lettner H, von Quadt A, Schermaier A, Schindlmayr AO, Steyrer HP (1993) The Zentralgneise of the Tauern window (Eastern Alps): insight into an intra-Alpine Vafiscan batholith. In: von Raumer JF, Neubauer F (eds) The pre-Mesozoic geology in the Alps. Springer, Berlin Heidelberg New York, pp 375-392 Franke W (1989) Tectonostratigraphic units in the Variscan belt of central Europe. Geol Soc Am Spec Pap 230:67-87 Frasl G, Finger F (1988) The "Cetic Massif" below the Eastern Alps - characterized by its granitoids. Schweiz Mineral Petrograph Mitt 68:433-439 Frasl G, Finger F (1991) Geologisch-petrographische Exkursion in den 6sterreichischen Teil des Sfidb6hmischen Batholiths. Eur J Mineral 3/2:23-40 Friedl G (1997) U/Pb-Datierungen an Zirkonen und Monaziten aus Gesteinen vom 6sterreichischen Anteil der B6hmischen Masse. Thesis, University of Salzburg, pp 242 FriedI G, Frasl G, von Quadt A, Finger F (1992) Neue U/Pb Altersdaten aus der sfidlichen B6hmischen Masse. Frankfurter Geowiss Arb A l l : 217 Friedl G, yon Quadt A, Ochsner A, Finger F (1993) Timing of the Variscan orogeny in the Southern Bohemian Massif (NE-Austria) deduced from new U-Pb zircon and monazite dating. Terra Abstr 5/1:235-236 Friedl G, yon Quadt A, Finger F (1996) Timing der Intrusionst~itigkeit im Sfidb6hmischen Batholith. In: Amann G, Handler R, Kurz W, Steyrer HP (eds) 6 Symposium Tektonik Strukturgeologie - Kristallingeologie, Salzburg 1996. Extended Abstracts, pp 127-130 Frisch W, Neubauer F (1989) Pre-Alpine terranes and tectonic zoning in the Eastern Alps. Geol Soc Am Spec Pap 230:91-100 Gerdes A (1997) Geochemische und thermische Modelle zur Frage der sp~itorogenen Granitgenese am Beispiel des Sfidb6hmischen Batholiths: Basaltisches Underplating oder Krustenstapelung. Thesis, University of G6ttingen, pp 113 Gerdes A, W6rner G, Finger F (1995) Quellen und Prozesse der Genese des Sfidb6hmischen Batholiths. Terra Nova 95/8:97 Harris NBW, Pearce JA, Tindle AG (1984) Geochemical characteristics of collision-zone magmatism. In: Ries AC, Coward MP (eds) Collision tectonics. Geol Soc Lond Spec Pub 19:67-81 Variscan granitoids of central Europe 93 Haunschmid B (1989) Das Granitgebiet um Plochwald zwischen Sandl und Windhaag im nordOstlichen OberOsterreich mit besonderer Beriicksichtigung des dortigen Plochwalder Granit-Typs und des Pseudokinzigits. Thesis, University of Salzburg, pp 168 Haunschmid B (1993) Zentralgneisgenerationen im 6stlichen Tauernfenster - Geologie, Petrographie, Zirkontypologie, Geochemie. Thesis, University of Salzburg, pp 159 Haunschmid B, Finger F (1994) Der Quarzmonzodiorit von Sarleinsbach eine Kumulatvariante des Weinsberger Granits. Mitt Osterr Mineral Gesell 139:310-312 Henes-Klaiber U (1989) Der Odenwald: Hinweise auf einen variszischen ArcMagmatismus? Beih Eur Jb Mineral 1:166 Henes-Klaiber U (1992) Zur Geochemie der variszischen Granitoide des Bergstr~isser Odenwalds. Thesis, University of Karlsruhe, pp 264 Hildreth W, Moorbath S (1988) Crustal contributions to arc magmatism in the Andes of Central Chile. Contrib Mineral Petrol 98:455-489 Holl A, Altherr R (1987) Hercynian I-type granitoids of northern Vosges: documents of increasing arc maturity. Terra Cognita 7:174 Holub FV (1977) Petrology of inclusions as a key to petrogenesis of the durbachitic rocks from Czechoslovakia. Tschermaks Mineral Petrogr Mitt 24:133-150 Holub FV, Rossi P, Cocherie A (1997) Radiometric dating of granitic rocks from the Central Bohemian Plutonic Complex (Czech Republic): constraints on the chronology of thermal and tectonic events along the Modanubian-Barrandian boundary. CR Acad Sci Paris 325:19-26 Janousek V, Rogers G, Bowes DR (1995) Sr-Nd isotopic constraints on the petrogenesis of the Central Bohemian Pluton, Czech Republic. Geol Rundsch 84:520-534 Kirsch H, Kober B, Lippolt HJ (1988) Age of intrusion and rapid cooling of the Frankenstein gabbro (Odenwald, SW-Germany) evidenced by Ar4°/Ar39 and single zircon pb2°7/pb 209 measurements. Geol Rundsch 77:693-711 Knop E, Biittner S, Haunschmid B, Finger F, Mirwald PW (1995) P-T conditions of Variscan metamorphism and migmatization in the Sauwald, southern Bohemian Massif. Terra Nova 7/1:316 Kosler J, McFarrow CM (1994) Mid-late Devonian arc-type magmatism in the Bohemian Massif: Sr and Nd isotope and trace element evidence from the Stare Sedlo and Mirotice Gneiss Complexes, Czech Republic. J Czech Geol Soc 39:56-58 Kosler J, Afialion M, Bowes DR (1993) Mid-late Devonian plutonic activity in the Bohemian Massif: U-Pb zircon isotopic evidence from the Stare Sedlo and Mirotice gneiss complex, Czech Republic. N Jb Mineral Abh 9:417-431 Kraus G (1962) GefiJge, KristallgrOl3en und Genese des Kristallgranits 1 im vorderen Bayrischen Wald. N Jb Mineral Abh 97:357-434 Kretz R (1983) Symbols for rock-forming minerals. Am Mineral 68:277-279 KrOner A, Hegner E, Hammer J, Haase G, Bielicki K-H, Krauss M, Eidam J (1994) Geochronology and Nd-Sr systematics of Lusitian granitoids: significance for the evolution of the Variscan orogen in east-central Europe. Geol Rundsch 83:357-376 Langer C, Hegner E, Altherr R, Satir M, Henjes-Kunst F (1995) Carboniferous granitoids from the Odenwald, the Schwarzwald and the Vosges - constraints on magma sources. Terra Nostra 95/8:114 Linnet M (1992) Metamorphose der Paragneise in der Monotonen Serie (SE Moldanubikum). Thesis, University of Vienna Liew TC, Hofmann AW (1988) Precambrian crustal components, plutonic associations, plate environment of the Hercynian Fold Belt of Central Europe: indications from a Nd and Sr isotopic study. Contrib Mineral Petrol 98:129-138 Liew TC, Finger F, H6ck V (1989) The Moldanubian granitoids of Austria: chemical and isotopic studies bearing on their environmental setting. Chem Geol 76:41-55 94 E Finger et al. Lorenc MW (1994) Rola magm zasadowych w ewolucji intruzji granitoidowych (studium porownawcze wybranych masywow Hercynskich). Geol Sudetica 28:3-130 Matejka D, Klecka M (1992) Petrogenese und geodynamische Stellung des Moldanubischen Batholiths, B6hmische Masse. Rundgespr~ich 8:49 Matte P (1986) Tectonics and plate tectonics model for the Variscan Belt of Europe. Tectonophysics 126:329-374 Mehnert KR (1968) Migmatites and the origin of granitic rocks. Elsevier, New York, pp 393 Neubauer F (1991) Comment on "I-type granitoids as indicators of a late-Paleozoic convergent ocean-continent margin along the southern flank of the Central European Variscan orogen". Geology 19:1246 Neubauer F (1994) Kontinentkollision in den Ostalpen. Geowissensch 12:136-140 Neugebauer J (1988) The Variscan plate tectonic evolution: an improved "Iapetus model" Schweiz Mineral Petrogr Mitt 68; 313-333 Nicolas A (1972) Was the Hercynian orogenic belt of Europe of the Andean type? Nature 236:221-223 O'Brien PJ, Carswell DA (1993) Tectonometamorphic evolution of the Bohemian Massif: evidence from high pressure metamorphic rocks. Geol Rundsch 82:531-555 Pati~o-Douce AE, Johnston AD (1991) Phase equilibria and melt productivity in the pelitic system: implications for the origin of peraluminous granitoids and aluminous granulites. Contrib Mineral Petrol 107:202-218 Peacock SM, Rushmer T, Thompson AB (1994) Partial melting of subducting oceanic crust. Earth Planet Sci Lett 121:227-244 Pearce JA, Harris NBW, Tindle AG (1984) Trace element discrimination diagrams for the tectonic interpretation of granitic rocks. J Petrol 25:956-983 Petrik I, Broska I, Uher P (1994) Evolution of the Western Carpathian granite magmatism: age source rock, geotectonic setting and relation to the Variscan structure. Geol Carpath 45:283-29 Pin C (1990) Variscan oceans, ages, origins and geodynamic implications inferred from geochemical and radiometric data. Tectonophysics 177:215-227 Pin C, Puziewicz J, Duthou JL (1989) Ages and origins of a composite granite massif in the Variscan belt: a Rb-Sr study of the Strzegom-Sobotka Massif, W Sudetes (Poland). N Jb Mineral Abh 160:71-82 Pitcher WS (1983) Granite types and tectonic environment. In: Hsti K (ed) Mountain building processes. Academic Press, London, pp 19-40 Roberts MP, Clemens JD (1993) Origin of high-potassium, calc-alkaline, I-type granitoids. Geology 21:825-828 Schaltegger U (1990) The Central Aar granite: highly differentiated calc-alkaline magmatism in the Aar Massif (Central Alps, Switzerland). Eur J Mineral 2:254-259 Schaltegger U (1994) Unravelling the pre-Mesozoic history of Aar and Gotthard massifs (Central Alps) by isotopic dating - a review. Schweiz Mineral Petrogr Mitt 74:41-51 Schaltegger U (1995) High-resolution chronometry of late Variscan extensional magmatism and basin formation: examples from the Vosges, Black Forest and the Alpine basement. Terra Nostra 7/95:109-111 Schaltegger U, Corfu F (1992) The age and source for late Hercynian magmatism in the Central Alps: evidence from precise U-Pb ages and initial Hf isotopes. Contrib Mineral Petrol 111: 329-344 Schaltegger U, Corfu F (1995) Late Variscan "Basin and Range" magmatism and tectonics in the Central Alps: evidence from U-Pb geochronology. Geodinam Acta 8:82-98 Schaltegger U, Gnos E, Kiipfer T, Labhart TP (1991) Geochemistry and tectonic significance of late Hercynian potassic and ultrapotassic magmatism in the Aar Massif (Central Alps). Schweiz Mineral Petrogr Mitt 71:391-403 Variscan granitoids of central Europe 95 Scharbert S (1987) Rb-Sr Untersuchungen granitoider Gesteine des Moldanubikums in Osterreich. Mitt Osterr Mineral Gesell 132:21-37 Schermaier A (1991) Geologisch-petrographische Untersuchungen zur pr~ialpidischen Entwicklung des Tauernfensters am Ostrand des Venedigermassivs (Hohe Tauern). Jb Geol Bundesanst 135:746-750 Schermaier A (1993) Gliederung der Zentralgneise im mittleren und westlichen Tauernfenster. Geologie - Petrographie - Zirkontypologie - Geochemie. Thesis, University of Salzburg, pp 174 Schermaier A, Haunschmid B, Finger F (1997) Distribution of I- and S-type granites in the eastern Alps: a possible clue to unravel pre-Alpine basement structures. Tectonophysics 272:315-333 Scotese CR (1984) Paleozoic paleomagnetism and the assembly of Pangea. In: van der Voo R, Scotese CR, Bonhommet N (eds) Plate reconstruction from Paleozoic paleomagnetism. Am Geophys Union Geodynam Srs 12:1-10 Siebel W (1994) Geochemical and isotopic (Sr Nd) investigations of NE Bavarian granitoids: a comparison between the Fichtelgebirge and the northern Oberpfalz. Rundgespr~ich 10:22-24 Siebel W (1995a) Constraints on Variscan granite emplacement in north-eastern Bavaria, Germany: further clues from a petrogenetic study of the Mitterteich granite. Geol Rundsch 84:384-398 Siebel W (1995b) Anticorrelated Rb-Sr and K-Ar age discordances: Leuchtenberg granite, NE Bavaria, Germany. Contrib Mineral Petrol 120:197-211 Silver LT, Chappell BW (1988) The Peninsular Ranges Batholith: an insight into the evolution of the Cordilleran batholiths of southwestern North America. Trans Roy Soc Edinburgh Earth Sci 79:105-121 Spillmann P, Biichi H (1993) The pre-Alpine basement of the Lower Austroalpine nappes in the Bernina Massif (Grisons Switzerland, Valtellina Italy). In: yon Raumer JE Neubauer F (eds) The Pre-Mesozoic geology in the Alps. Springer, Berlin Heidelberg New York Tokyo, pp 457-468 Stemprok M (1994) The Krusne Hory (Erzgebirge) granite batholith. Mitt Osterr Mineral Gesell 139:374-376 Thiele 0 (1962) Neue geologische Ergebnisse aus dem Sauwald (OC)). Verhandl Geol Bundesanst 1962:117-129 Thompson RN, Morrison MA, Hendry GL, Parry SJ (1984) An assessment of the relative roles of crust and mantle magma genesis: an elemental approach. Phil Trans Roy Soc Lond A310:549-590 Th6ni M (1991) Neue Rb/Sr-Daten an Mineralien und Gesteinen des Leopold von BuchDenkmales und am Tonalit von Schaitten. Mitt Gesell Geol Bergbaust Osterr 37: 157162 Todt W (1976) Zirkon-U/Pb-Alter des Malsburg-Granits, Stid-Schwarzwald. N Jb Mineral Monat 12:532-544 Troll G (1964) Geologische 0bersichtskarte des Bayerischen Waldes 1: 100000. Geologica Bavarica 58 (enclosure) Uher P, Pushkarev Y (1994) Granitic pebbles of the Cretaceous flysch of the Pieniny klippen belt, Western Carpathians. Geol Carpath 45:375-378 Vavra G, Hansen BT (1991) Cathodoluminescence studies and U/Pb dating of zircons in pre-Mesozoic gneisses of the Tauern Window: implications for the Penninic basement evolution. Geol Rundsch 80:703-715 Vellmer C (1992) Stoffbestand und Petrogenese von Granuliten und granitischen Gesteinen der siJdlichen B6hmischen Masse in Nieder6sterreich. Thesis, University of G6ttingen, pp 111 96 F. Finger et al.: Variscan granitoids of central Europe Vielzeuf D, Holloway JR (1988) Experimental determination of the fluid-absent melting relations in the pelitic system. Contrib Mineral Petrol 98:257-276 Vielzeuf D, Montel JM (1994) Partial melting of metagreywackes, part 1. Fluid-absent experiments and phase relationships. Contrib Mineral Petrol 117:375-393 Volker F, Altherr R (1987) Lower Carboniferous calc-alkaline volcanics in the northern Vosges: evidence for a destructive continental margin. Terra Cognita 7:174 yon Blanckenburg F, Davies JH (1995) Slab breakoff: a model for syncollisional magmatism and tectonics in the Alps. Tectonics 14:120-131 von Quadt A, Finger F (1991) Geochronologische Untersuchungen im 6sterreichischen Teil des Sfidbohmischen Batholiths. U-Pb Datierungen an Zirkonen, Monaziten und Xenotimen des Weinsberger Granits. Eur J Mineral 3 (Beiheft 1): 281 von Quadt A, Griinenfelder M, Biichi H (1994) U-Pb zircon ages from igneous rocks of the Bernina nappe system (Grisons, Switzerland). Schweiz Mineral Petrogr Mitt 74: 373382 Wenzel T, Mercolli I, Oberhiinsli R (1991) The plutonic rocks of the Meif3en Massif (Germany): evidence for open and closed system fractionation processes. Schweiz Mineral Petrogr Mitt 71 : 371-390 White A JR, Chappell BW (1983) Granitoid types and their distribution in the Lachlan Fold Belt, southeastern Australia. In: Roddick JA (ed) Circum Pacific plutonic terranes. Geol Soc Am Mem 159:21-34 Authors' addresses: F. Finger, M. P. Roberts, B. Haunschmid, and A. Schermaier, Institut ffir Mineralogie, Universit~it Salzburg, Hellbrunnerstrasse 34, A-5020 Salzburg, Austria; H. P. Steyrer, Institut ffir Geologie und Pal~iontologie, Universit~it Salzburg, Hellbrunnerstrasse 34, A-5020 Salzburg, Austria