Abstract Granitic rocks represent a ubiquitous component of upper continental crust but their origin remains highly controversial. This controversy stems from the fact that the granites may result from fractionation of mantle-derived basaltic magmas or partial melting of different crustal protoliths at contrasting pressure–temperature conditions, either water-fluxed or fluid-absent. Consequently, many different mechanisms have been proposed to explain the compositional variability of granites ranging from whole igneous suites down to mineral scale. This Special Publication presents an overview of the state of the art and envisages future avenues towards a better understanding of granite petrogenesis.

Abstract Granites ( sensu lato ) come in many types and flavours, defining distinct magmatic series/suites/types. A good classification not only gives generally accepted and understandable names to similar rocks but also links the bulk chemical composition to the stoichiometry of the constituent minerals and, potentially, also to the likely source, magmatic evolution and tectonic setting. The ‘ideal’ granitoid classification should be based on chemical criteria amenable to an objective treatment. Statistical analysis helps to identify the most discriminant variables. The key properties are (1) acidity/maficity, (2) alkalinity (balance of Na + K v. Ca), (3) aluminosity (balance of Al v. Ca, Na and K), (4) Fe/Mg balance and (5) Na/K balance and K contents at the given SiO 2 level. These are used by successful classifications, e.g. the I/S dichotomy is based mainly on aluminosity, while the Frost et al. (2001 ; ‘A geochemical classification for granitic rocks', Journal of Petrology , 42 , 2033–2048, https://doi.org/10.1093/petrology/42.11.2033 ) classification includes all but Na/K. Even though it is commonplace to use weight percentages of oxides, we suggest that a better strategy is to employ simple atomic parameters (e.g. millications-based) that can be directly linked to modal proportions and compositions/crystal structure of individual rock-forming minerals. This facilitates a petrological interpretation, which, in turn, can be related to petrogenesis and, ultimately, to likely tectonic setting(s).

Abstract The classical S–I–A-type granites from the Lachlan Orogen, SE Australia, formed as a tectonic end-member of the accretionary orogenic spectrum, the Paleozoic Tasmanides. The sequence of S- to I- to A-type granite is repeated at least three times. All the granites are syn-extensional, formed in a dominantly back-arc setting behind a single, stepwise-retreating arc system between 530 and 230 Ma. Peralkaline granites are rare. Systematic S–I–A progressions indicate the progressive dilution of an old crustal component as magmatism evolved from arc (S-type) to proximal back-arc (I-type) to distal back-arc (A-type) magmatism. The alkaline and peralkaline A-type Younger granites of Nigeria were generally hotter and drier than the Lachlan A-type granites and were emplaced into an anhydrous Precambrian basement during intermittent intracontinental rifting. This geodynamic environment contrasts with the distal back-arc setting of the Lachlan A-type granites, where magmatism migrated rapidly across the orogen. Tectonic discrimination diagrams are inappropriate for the Lachlan granites, placing them in the wrong settings. Only the peralkaline Narraburra suite of the Lachlan Orogen fits the genuine ‘within-plate’ setting of the Nigerian A-type granites. Such discrimination diagrams require re-evaluation in the light of an improved modern understanding of tectonic processes, particularly the role of extensional tectonism and its geodynamic drivers.

Abstract The origin of large I-type batholiths remains a disputed topic. One model states that I-type granites form by partial melting of older crustal lithologies (amphibolites or intermediate igneous rocks). In another view, granites are trapped rhyolitic liquids occurring at the end of fractionation trends defining a basalt–andesite–dacite–rhyolite series. This paper explores the thermal implications of both scenarios, using a heat balance model that abstracts the heat production and consumption during crustal melting. Heat is consumed by melting and by losses through the surface (conductive or advective, as a result of eruption). It is supplied as a basal conductive heat flux, as internal heat production or as advective heat carried by an influx of hot basalt into the crust. Using this abstract approach, it is possible to explore the role different parameters play in the balance of granites formed by differentiation of basalts or by crustal melting. Two end-member situations appear equally favourable to generating large volumes of granites: (1) short-lived environments dominated by high basaltic flux, where granites result mostly from basalt differentiation; and (2) long-lived systems with no or minimal basalt flux, with granites resulting chiefly from crustal melting.

Abstract New laboratory experiments using granulite xenoliths support a dual origin for I-type granites as primary and secondary. Primary I-type granites represent fractionated liquids from intermediate magma systems of broadly andesitic composition. Fluid-fluxed melting of igneous rocks that resided in the continental crust generates secondary I-type granites. The former are directly related to subduction, with Cordilleran batholiths as the most characteristic examples. Experiments with lower crust granulite sources, in the presence of water, show that amphibole is formed by a water-fluxed peritectic rehydration melting reaction . Entrainment of only 10% of restites composed of amphibole, pyroxene, plagioclase and magnetite, is sufficient to account for discrepancies in aluminium saturation index and maficity in secondary I-type granites. Lower crust granulite xenoliths, attached to a sanukitoid containing 6 wt% water, have been used in two-layer capsules to test fluid-fluxed melting reactions as the origin of secondary I-type granites. It is proposed that sanukitoid magmas act as water donors that trigger extensive melting of the lower crust, giving rise to granodioritic liquids. Because primary granites are related to coeval subduction, and secondary ones are crustal melts from older subduction-related rocks, the distinction between both I-types is essential in tectonic reconstructions of ancient orogenic belts.

Abstract Peraluminous granites and trondhjemites make up small plutonic bodies intruded into high-grade paragneisses in the Peloritani Mountains, marking the beginning of late Variscan granitoid magmatism in southernmost Italy. The granites range from low-Ca monzogranites to alkali feldspar granites, while the trondhjemites vary from trondhjemites s.s . to low-Ca trondhjemites. Relatively high radiogenic ( 87 Sr/ 86 Sr) i ratios (mostly from 0.7073 to 0.7125) and negative ε Nd values (mostly from −5.66 to −8.73) point to crustal sources for all the granitoids. Major and trace element compositions indicate an absence of genetic relationships between the trondhjemites s.s . and the low-Ca granitoids, but possible relationships between the low-Ca trondhjemites and the granites. All of the studied granitoids have near-pure melts compositions, consistent with H 2 O-fluxed and dehydration melting of metasediments for the trondhjemites and the granites, respectively. However, the unusual compositions of the low-Ca trondhjemites and microstructural evidence in these rocks for pervasive subsolidus replacement of magmatic feldspars by secondary sodic plagioclase indicate that they were derived instead from metasomatic alteration of the granites. Thus, water may be involved in the production of trondhjemites in two different ways, driving water-fluxed melting in the magma source and driving alkali metasomatism at the sites of granite emplacement in the upper crust.

Abstract Leucogranites are a characteristic feature of collisional orogens. Their generation is intimately related to crustal thickening and the active deformation and metamorphism of metapelites. Data from Proterozoic to present day orogenic belts show that collisional leucogranites (CLGs) are peraluminous, with muscovite, biotite and tourmaline as characteristic minerals. Isotopic ratios uniquely identify the metapelitic sequences in which CLGs occur as sources. Organic material in pelitic sources results in f O 2 in CLGs that is usually below the fayalite–magnetite–quartz buffer. Most CLGs form under vapour-poor conditions with melting involving a peritectic breakdown of muscovite. The low concentrations of Mg, Fe and Ti that characterize CLGs are largely related to biotite–melt equilibria in the source rocks. Concentrations of Zr, Th and rare earth elements are lower than expected from zircon and monazite saturation models because these minerals often remain enclosed in residual biotite during melting. Melting involving muscovite may limit the temperatures achieved in the source regions. A lack of nearby mantle heat sources in thick collisional orogens has led to thermal models for the generation of CLGs that involve flux melting, or large amounts of radiogenic heat generation, or decompression melting or shear heating, the last one emphasizing the link of leucogranites and their sources to crustal-scale shear zone systems.

Abstract Modern quantitative phase equilibria modelling allows the calculation of the stable phase assemblage of a rock system given its pressure, temperature and bulk composition. A new software tool (Rcrust) has been developed that allows the modelling of points in pressure–temperature–bulk composition space in which bulk compositional changes can be passed from point to point as the system evolves. This new methodology enables quantitative process-oriented investigation of the evolution of rocks. Procedures are outlined here for using this tool to model: (1) the control of the water content of a subsolidus system based on available pore space; (2) the triggering of melt loss events when a critical melt volume threshold is exceeded, while allowing a portion of melt retention; (3) the entrainment of crystals during segregation and ascent of granitic magmas from its source; (4) the modification of the composition of granite magmas owing to fractional crystallization; and (5) the progressive availability (through dissolution) of slow diffusing species and their control of the effective bulk composition of a system. These cases collectively illustrate thermodynamically constrained methods for modelling systems that involve mass transfer.

Abstract The ability of Rcrust software to conduct path-dependent phase equilibrium modelling with automated changing bulk compositions allows for a phase equilibrium approach to investigate an array of source controls for their effect on the bulk compositions of melts produced by sequential melting events. The following source controls of the rock system are considered: (1) initial magnesium and iron content; (2) initial sodium and calcium content; (3) pressure–temperature path followed by the system; and (4) threshold by which melt extractions in the system are triggered. These source controls are investigated in a water-restricted system and a water-in-excess system. The permutation of these cases resulted in 128 different modelled pressure–temperature bulk composition paths investigating the melting of an average pelite composition. The resultant melt compositions are compared to that of a natural granite dataset and provide a good fit for the incompatible elements Na 2 O and K 2 O with the allowance that granites most likely form as magmas consisting of melt and ferromagnesian-rich crystals. The fluid state of the system is shown to have the strongest control on melt compositions, with the pressure–temperature path having subordinate control on the volume and composition of melts produced.

Abstract The Buddusò Pluton in NE Sardinia (Italy) is a normally zoned intrusion composed of three units with chemical composition ranging from hornblende-bearing tonalites (SiO 2 ∼ 65 wt%) to leucocratic monzogranites (SiO 2 ∼ 76 wt%). Zircon crystals in the pluton are dated at 292.2 ± 0.7 Ma and have ε Hf values ranging from −4 to −8, with no systematic differences observed between the units. The pluton, which is isotopically homogeneous at the whole-rock scale in terms of Sr and Nd isotopes, shows textural evidence indicating local crystal–melt segregation. In this paper, we have implemented a novel approach based on path-dependent phase-equilibria modelling to test the hypothesis that the internal chemical variability of the pluton was generated by crystallization differentiation of a homogeneous parental magma. Our modelling indicates that this hypothesis is valid if the mechanism by which this occurs is compaction in a rheologically locked crystal-rich magma and if the separation occurs at 0.3 GPa from a tonalitic magma with water content >2 wt%. Finally, a subset of the magmatic enclaves in the pluton are considered to be autoliths, formed by the disruption of the compacted crystal mush and interaction between these cumulates and the felsic melt.

Abstract Whole-rock geochemistry represents a powerful tool in deciphering petrogenesis of magmatic suites, including granitoids, which can be used to formulate and test hypotheses qualitatively and often also quantitatively. Typically, it can rule out impossible/improbable scenarios and further constrain the process inferred on geological and petrological grounds. With the current explosion of high-precision data, both newly acquired and retrieved from extensive databases, the whole-rock geochemistry-based petrogenetic modelling of igneous rocks will gain further importance. Especially promising is its combination with thermodynamic modelling into a single, coherent and comprehensive software, using the R and Python languages.

Granites ( sensu lato ) represent the dominant rock-type forming the upper–middle continental crust but their origin remains a matter of long-standing controversy. The granites may result from fractionation of mantle-derived basaltic magmas, or partial melting of different crustal protoliths at contrasting P–T conditions, either water-fluxed or fluid-absent. Consequently, many different mechanisms have been proposed to explain the compositional variability of granites ranging from whole igneous suites down to mineral scale. This book presents an overview of the state of the art, and envisages future avenues towards a better understanding of granite petrogenesis. The volume focuses on the following topics: compositional variability of granitic rocks generated in contrasting geodynamic settings during the Proterozoic to Phanerozoic Periods; main permissible mechanisms producing subduction-related granites; crustal anatexis of different protoliths and the role of water in granite petrogenesis; and new theoretical and analytical tools available for modelling whole-rock geochemistry in order to decipher the sources and evolution of granitic suites.

Abstract The Kerguelen Archipelago is made up of a stack of thick piles of Tertiary flood basalts intruded by transitional to alkaline igneous centres at various times since 30 Ma ago. In the SW, the Rallier-du-Baty Peninsula is mostly occupied by two silicic ring complexes, each with an average diameter of 15 km, comprising dissected calderas cross-cut by subvolcanic cupolas. Previous radiometric determinations yield ages ranging from 15.4 to 7.4 Ma in the southern centre, and 6.2 to 4.9 Ma in the northern one. The felsic ring dykes were injected by coeval mafic magmas, forming, successively, swarms of early mafic enclaves, disrupted synplutonic cone sheets, and late cone-sheets. After the emplacement and subsequent unroofing of the plutonic ring complexes, abundant and thick trachytic pyroclastic flows and falls were emitted from the younger caldera volcanoes, while hawaiite and mugearite lava flows were erupted from marginal maars and cones. Huge trachyte ignimbritic flows filled the glacial valleys in the central Peninsula, and capped lacustrine deposits and older lava flows, while related pumice falls are widespread throughout the archipelago. This powerful plinian eruption took place after the network of glacial valleys was established, but before the Little Ice Age that occurred during the last centuries. In the south of the peninsula, even younger trachytic formations are exposed, and fumarolic vents are still active. The growth mechanisms of a caldera-related ring complex can be explained as a repetitive sequence of two eruptive episodes. The first episode of hydrofracturing, induced by volatile exsolution within the evolving magma chamber, creates a vertical circular fracture zone, along which highly vesiculated magmas are emitted during explosive eruptive events occurring at the surface in a caldera volcano. It is followed by a second episode of cauldron-subsidence of a crustal block down to the degassed magma chamber, induced by pressure release. Downward movement of the crustal block favours the emplacement at shallow depths within the older caldera-filling formations, of discrete magmatic sheets characterized by a 16-km mean diameter and a 1-km mean thickness, corresponding to an average unit volume of 200 km 3 . Actually, the estimated volumes of the different igneous episodes within the Rallier-du-Baty nested ring complex vary from 60 to 900 km 3 , and correspond to the production during 15 Ma of about 2800 ± 850 km 3 of new materials and a net crustal growth of about 100 ± 30 × 10 3 m 3 per year.