Geochemistry and petrogenesis of basalts from DSDP Leg 92 holes

Basalts recovered on DSDP Leg 92 include all the major basalt types so far recovered from the ocean crust of the eastern Pacific. Basalts from Holes 597, 597A, 597B, 597C, and 599B are tholeiites exhibiting all the mineralogical and geochemical characteristics of N-type mid-ocean ridge basalts (MORB). Fragments of ferrobasalts and alkali basalts were also obtained, however, from Holes 60IB and 602B, respectively. Hole 597C, which penetrated 91 m into basement and is the deepest hole so far drilled in fast-spreading crust, yielded basalts that can be divided into three major lithologic units. The lowest unit, Unit III, contains modal olivine and comprises basalts which, at about 8 to 10% MgO, are as basic as any sampled from fast-spreading crust. The middle unit, Unit II, is the most evolved; its basalts are olivine free and contain between 6 and 7.5% MgO. The upper unit, Unit I, is intermediate in composition between Units II and III; it is characterized by both modal olivine and glomerocrysts made up of plagioclase and rare olivine. Unit I is probably a massive flow, whereas Units II and III may be massive flows or sills. The basalts appear to have undergone three stages of alteration ("deuteric," "relatively reducing," and "oxidizing"), the intensity of alteration decreasing markedly downcore. Hole 597B, at 26.4 m of basement penetration the only other "deep" hole, contains just one lithologic unit, which closely resembles Unit I of Hole 597C. Petrogenetic modeling reveals that the three lithologic units in Hole 597C are cogenetic and that they were derived from a depleted mantle source similar to the source of the tholeiites and ferrobasalts sampled in other holes; the alkali basalts are the only rocks derived from enriched mantle. Lavas of Unit III probably lay on the olivine-plagioclase cotectic, whereas the other lavas lay on an olivine-plagioclase-clinopyroxene peritectic. Some 60% of closed-system crystallization is needed to generate the most-evolved from the last-fractionated tholeiite, and a further 50% crystallization (80% overall) is needed to generate the ferrobasalts. Xenocrysts of calcic plagioclase and pseudomorphosed olivine in tholeiites from Hole 597B and Unit I of Hole 597C, and in the ferrobasalts from Hole 601B, provide evidence, however, that some magma mixing may have taken place.

Trace element abundance, and Sr and Nd isotope ratios of dust samples in the Pacific Ocean (Table 2)

Eolian dust preserved in deep-sea sediment cores provides a valuable indicator of past atmospheric circulation and continental paleoclimate. In order to identify the provenance of eolian dust, Nd and Sr isotopic compositions and Rb, Sr and rare earth element (REE) concentrations have been determined for the silicate fractions of deep-sea sediments from the north and central Pacific Ocean. Different regions of the Pacific Ocean are characterized by distinct air-borne inputs, producing a large range in epsolin-Nd (-10 to +1), 87Sr/86Sr (0.705-0.721), La/Yb (5-15), EuN/EuN* (0.6-1.0) and Sr/Nd (4-33). The average Nd isotopic composition of Pacific deep-sea sediments (epsilon-Nd = -6), is more radiogenic than the average from the Atlantic (epsilon-Nd = -8). In contrast, the average147Sm/144Nd ratio for Pacific sediments (0.114) is identical to that of Atlantic sediments and to that of global average riverine suspended material. The values of epsilon-Nd and147Sm/144Nd are positively correlated for the Pacific samples but negatively correlated for Atlantic samples, reflecting a fundamental difference between the dominant components in the end members with radiogenic Nd (island-arc components in the Pacific and LREE-enriched intraplate ocean island components in the Atlantic). Samples from the north central Pacific have distinctive unradiogenic epsilon-Nd values of -10, 87Sr/86Sr > 0.715, high La/Yb (> 12), and low EuN/EuN* (0.6) and Sr/Nd (3-6). These data are virtually identical to the values for loess from Asia and endorse the use of these sediments as indicators of Asian paleoclimate and paleowind directions. Island-arc contributions appear to dominate in the northwest Pacific, resulting in higher epsilon Nd (-1 to +1) and lower 87Sr/86Sr (~ 0.705) and La/Yb (~ 5). Sediments from the eastern Pacific tend to have intermediate Sr and Nd isotopic compositions but regionally variable Sr/Nd and REE patterns; they appear to be derived from the west margin of the North and South American continents, rather than from Asia. Our results confirm that dust provenance can be constrained by isotopic and geochemical analyses, which will facilitate reconstructions of past atmospheric circulation and continental paleoclimate.

Geochemistry of ODP Hole 111-504B basalts (Table 2)

Fifty samples of basalt recovered during ODP Leg 111 from the dikes (Layer 2C) of Hole 504B (1350.0-1562.3 m below seafloor) were analyzed by X-ray-fluorescence techniques. All of the samples are highly depleted in magmaphile elements relative to other mid-ocean ridge basalts, with TiO2 = 0.75-1.24 wt%, Na2O = 1.59-2.22 wt%, Zr = 38-64 ppm, Nb = 0.3-1.5 ppm, and Y = 20-30 ppm (for samples containing 0%-2% phenocrysts), but have ratios of highly incompatible elements similar to normal Type I mid-ocean ridge basalts (e.g., Zr/Nb > 30). Abundances of compatible elements are similar to those of typical mid-ocean ridge basalts, with MgO = 7.2-9.2 wt%, Fe2O3* = 9.3-12.5 wt%, Ni = 55-164 ppm, and Cr = 26-388 ppm. Approximately 2% of the samples recovered from the top part of Hole 504B are similar to normal Type I or Type II ocean floor basalts. However, all of the analyzed Leg 111 samples from Hole 504B are depleted basalts. Aphyric dike rocks from Leg 111 are virtually identical to the depleted aphyric samples recovered from the pillow lavas and dikes in the upper 1075 m of Hole 504B during DSDP Legs 69, 70, and 83, with the exception of elements readily altered by seawater (Sr, Rb, and K). These elements reach a maximum in both abundance and variability in the pillow lavas of the upper 571.5 m of Hole 504B and decline to more constant values in the dike system sampled on Legs 83 and 111, apparently as a result of a decrease in porosity and increase in alteration temperatures relative to the pillow lavas. Based on compositional similarities to the vast majority of the pillows and flows, the dikes sampled on Leg 111 appear to be the feeder system for the pillow lavas in the upper part of Hole 504B. The incompatible-element-depleted compositions of the Costa Rica Rift Zone basalts are consistent with multistage melting of a normal mid-ocean ridge source.

Geochemistry of basalts at DSDP Site 89-462

At Site 462 in the Nauru Basin, western Pacific Ocean, 56 lithologic units have been recovered from an extensive flood basalt province. Fossil evidence suggests that the lavas were emplaced during the interval 100-115 Ma, some 30 m.y. after formation of the underlying Jurassic ocean crust. The lithologic units can be broadly divided into three chemical units, the lowermost two of which are chemically monotonous, suggesting rapid eruption of basalt from a compositionally homogeneous magma chamber. All the basalts are hypersthene- (hy-) rich tholeiites, with approximately chondritic La/Sm, La/Yb, Zr/Nb, La/Ta, and Th/Hf ratios. Chemically they resemble, in part, "transitional" mid-ocean ridge basalts (MORB) from areas such as the Reykjanes Ridge, although Rb, Ba, and K contents are very low and similar to those of "normal" MORB. Their 87Sr/86Sr ratios are higher than in N-type MORB (Fujii et al., 1981). The chemistry of the Nauru basalts differs from that of continental flood basalts, which tend to be strongly enriched in large-ion lithophile (LIL) elements, although the extent to which the differences result from sialic contamination or source variability is not clear.

Lead isotope ratios of basalts from DSDP Leg 37 (Table 1)

Analyses of Pb from mid-ocean ridge basalts obtained from DSDP Leg 37 lie on a line of near-zero-age slope. The data seem consistent with an evolutionary model of three stages, the beginning of the second stage being at about 600 m.y. - the model age obtained for the least radiogenic Pb. The beginning of the third stage of evolution at the time of formation of the rocks apparently did not change the average Th/U ratio since measured values are consistent with the ratio deduced from Pb isotope ratios. Ratios of U/Pb are not consistent with Pb isotope ratios, however, thus ruling out a simple two-stage evolutionary model.

Lamproites from Gaussberg, Antarctica: Possible transition zone melts of Archaean subducted sediments

Petrogenetic models for the origin of lamproites are evaluated using new major element, trace element, and Sr, Nd, and Pb isotope data for Holocene lamproites from the Gaussberg volcano in the East Antarctic Shield. Gaussberg lamproites exhibit very unusual Pb isotope compositions (Pb-206/Pb-204 = 17.44-17.55 and Pb-207/Pb-204 = 15.56-15.63), which in common Pb isotope space plot above mantle evolution lines and to the left of the meteorite isochron. Combined with very unradiogenic Nd, such compositions are shown to be inconsistent with an origin by melting of sub-continental lithospheric mantle. Instead, a model is proposed in which late Archaean continent-derived sediment is subducted as K-hollandite and other ultra-high-pressure phases and sequestered in the Transition Zone (or lower mantle) where it is effectively isolated for 2-3 Gyr. The high Pb-207/Pb-204 ratio is thus inherited from ancient continent-derived sediment, and the relatively low Pb-206/Pb-204 ratio is the result of a single stage of U/Pb fractionation by subduction-related U loss during slab dehydration. Sr and Nd isotope ratios, and trace element characteristics (e.g. Nb/Ta ratios) are consistent with sediment subduction and dehydration-related fractionation. Similar models that use variable time of isolation of subducted sediment can be derived for all lamproites. Our interpretation of lamproite sources has important implications for ocean island basalt petrogenesis as well as the preservation of geochemically anomalous reservoirs in the mantle.

Study of the Saldanha Massif (MAR, 36 degrees 34 ' N): Constrains from rock magnetic and geophysical data

We present a study of the magnetic properties of a group of basalt samples from the Saldanha Massif (Mid-Atlantic Ridge - MAR - 36degrees 33' 54" N, 33degrees 26' W), and we set out to interpret these properties in the tectono-magmatic framework of this sector of the MAR. Most samples have low magnetic anisotropy and magnetic minerals of single domain grain size, typical of rapid cooling. The thermomagnetic study mostly shows two different susceptibility peaks. The high temperature peak is related to mineralogical alteration due to heating. The low temperature peak shows a distinction between three different stages of low temperature oxidation: the presence of titanomagnetite, titanomagnetite and titanomaghemite, and exclusively of titanomaghemite. Based on established empirical relationships between Curie temperature and degree of oxidation, the latter is tentatively deduced for all samples. Finally, swath bathymetry and sidescan sonar data combined with dive observations show that the Saldanha Massif is located over an exposed section of upper mantle rocks interpreted to be the result of detachment tectonics. Basalt samples inside the detachment zone often have higher than expected oxidation rates; this effect can be explained by the higher permeability caused by the detachment fault activity.

The insular shelves of the Faial-Pico Ridge (Azores archipelago): a morphological record of its evolution

Shelves surrounding reefless volcanic ocean islands are formed by surf erosion of their slopes during changing sea levels. Posterosional lava flows, if abundant, can cross the coastal cliffs and fill partially or completely the accommodation space left by erosion. In this study, multibeam bathymetry, high-resolution seismic reflection profiles, and sediment samples are used to characterize the morphology of the insular shelves adjacent to Pico Island. The data show offshore fresh lava flow morphologies, as well as an irregular basement beneath shelf sedimentary bodies and reduced shelf width adjacent to older volcanic edifices in Pico. These observations suggest that these shelves have been significantly filled by volcanic progradation and can thus be classified as rejuvenated. Despite the general volcanic infilling of the shelves around Pico, most of their edges are below the depth of the Last Glacial Maximum, revealing that at least parts of the island have subsided after the shelves formed by surf erosion. Prograding lava deltas reached the shelf edge in some areas triggering small slope failures, locally decreasing the shelf width and depth of their edges. These areas can represent a significant risk for the local population; hence, their identification can be useful for hazard assessment and contribute to wiser land use planning. Shelf and subaerial geomorphology, magnetic anomalies and crustal structure data of the two islands were also interpreted to reconstruct the long-term combined onshore and offshore evolution of the Faial-Pico ridge. The subaerial emergence of this ridge is apparently older than previously thought, i.e., before approximate to 850 ka.

Major, minor, trace element, Sm-Nd and Sr isotope compositions of mafic rocks from the earliest oceanic crust of the Alpine Tethys

Recent isotopic and biochronologic dating has demonstrated that the Gets nappe contains remnants of the oldest part of the oceanic crust of the Alpine Tethys. The ophiolites are associated with deep sea sediments, platform carbonates and continental crustal elements suggesting a transitional environment between continental and oceanic crust. Therefore, the ophiolites from the Gets nappe provide the opportunity to assess the nature of mantle source and the magma evolution during the final rifting stage of the European lithosphere. Trace clement analyses of mafic rocks can he divided into two sets: (1) P, Zr and Y contents are consistent with those of mid-ocean ridge basalts and REE patterns have a P-MORB affinity. (2) P,Zr Ti and Y contents are compatible with within-plate basalts and are characterized by REE spectra similar to that of T-MORB. Both have Nd isotopic compositions similar to those of synrift magma of the Red Sea and to the Rhine Graben. The model ages are in agreement with an LREE-enriched subcontinental mantle source derived from depleted mantle 800 to 900 Ma ago. Minor, trace element and Sm-Nd compositions suggest that these rocks are basaltic relies of an earliest stage of oceanic spreading i.e. an embryonic ocean. Comparison between REE patterns, Nd and Sr isotope compositions, isotopic and biochronologic ages from different Alpine Tethys ophiolites shows that samples with enriched LREE are from the older ophiolitic suites and are relies of the embryonic ocean floor. Later phases of ocean spreading are characterized by basalts that are depleted in LREE.

New structural and petrologic data on Mesozoic schists in the Rhodope (Bulgaria): geodynamic implications

Structural analysis of low-grade rocks highlights the allochthonous character of Mesozoic schists in southeastern Rhodope, Bulgaria. The deformation can be related to the Late Jurassic-Early Cretaceous thrusting and Tertiary detachment faulting. Petrologic and geochemical data show a volcanic arc origin of the greenschists and basaltic rocks. These results are interpreted as representing an island arc-accretionary complex related to the southward subduction of the Meliata-Maliac Ocean under the supra-subduction back-arc Vardar ocean/island arc system. This arc-trench system collided with the Rhodope in Late Jurassic times. (C) 2003 Academie des sciences. Published by Editions scientifiques et medicales Elsevier SAS. All rights reserved.

The magmatic to hydrothermal evolution of the intrusive Mont Saint-Hilaire Complex: Insights into the late-stage evolution of peralkaline rocks

The Cretaceous Mont Saint-Hilaire complex (Quebec, Canada) comprises three major rock units that were emplaced in the following sequence: (I) gabbros; (II) diorites; (III) diverse partly agpaitic foid syenites. The major element compositions of the rock-forming minerals, age-corrected Nd and oxygen isotope data for mineral separates and trace element data of Fe-Mg silicates from the various lithologies imply a common source for all units. The distribution of the rare earth elements in clinopyroxene from the gabbros indicates an ocean island basalt type composition for the parental magma. Gabbros record temperatures of 1200 to 800 degrees C, variable silica activities between 0 center dot 7 and 0 center dot 3, and f(O2) values between -0 center dot 5 and +0 center dot 7 (log delta FMQ, where FMQ is fayalite-magnetite-quartz). The diorites crystallized under uniform a(SiO2) (a(SiO2) = 0 center dot 4-0 center dot 5) and more reduced f(O2) conditions (log delta FMQ similar to-1) between similar to 1100 and similar to 800 degrees C. Phase equilibria in various foid syenites indicate that silica activities decrease from 0 center dot 6-0 center dot 3 at similar to 1000 degrees C to < 0 center dot 3 at similar to 550 degrees C. Release of an aqueous fluid during the transition to the hydrothermal stage caused a(SiO2) to drop to very low values, which results from reduced SiO(2) solubilities in aqueous fluids compared with silicate melts. During the hydrothermal stage, high water activities stabilized zeolite-group minerals. Fluid inclusions record a complex post-magmatic history, which includes trapping of an aqueous fluid that unmixed from the restitic foid syenitic magma. Cogenetic aqueous and carbonic fluid inclusions reflect heterogeneous trapping of coexisting immiscible external fluids in the latest evolutionary stage. The O and C isotope characteristics of fluid-inclusion hosted CO(2) and late-stage carbonates imply that the surrounding limestones were the source of the external fluids. The mineral-rich syenitic rocks at Mont Saint-Hilaire evolved as follows: first, alkalis, high field strength and large ion lithophile elements were pre-enriched in the (late) magmatic and subsequent hydrothermal stages; second, percolation of external fluids in equilibrium with the carbonate host-rocks and mixing processes with internal fluids as well as fluid-rock interaction governed dissolution of pre-existing minerals, element transport and precipitation of mineral assemblages determined by locally variable parameters. It is this hydrothermal interplay between internal and external fluids that is responsible for the mineral wealth found at Mont Saint-Hilaire.

Partial melting of basic dykes in the contact aureoles of two shallow-level, gabbro-pyroxenite intrusions, Fuerteventura

THESIS ABSTRACT Low-pressure anatexis of basic dykes gave rise to unusual, zebra-like migmatites, in the contact metamorphic aureoles of two layered gabbro-pyroxenite intrusions, PXl and PX2, in the root zone of an ocean island, Fuerteventura Basal Complex (Canary Islands). This thesis focuses on the understanding of processes attributing to the partial melting and formation of these migmatites, characterised by a dense network of closely spaced, millimetre-wide leucocratic segregations with perfectly preserved igneous textures. The presence of fluids are required to decrease the solidus of basic igneous lithologies, to allow partial melting in such aloes-pressure (1-2 kb) environment. An oxygen isotope study was thus carried out on dykes inside and beyond the PX2 aureole, in order to decipher the nature and origin of such fluids. Low or negative δ18O values were obtained for whole rocks and mineral-separates, decreasing towards the contact, with the intrusion itself retaining fairly high values. This trend has been attributed to the advection of meteoric water during magma emplacement, with increasing fluid/rock ratios (higher dyke intensities towards the intrusion acting as fluid-pathways) and higher temperatures promoting increasing exchange during recrystallisation. A comparison of whole rock and mineral major- and trace- element data allowed the redistribution of elements .between different mineral phases and generations, during contact metamorphism and partial melting to be assessed. Certain trace-elements, e.g. Zr, Hf, Y, and REEs, were internally redistributed during contact metamorphic recrystallisation, causing- the enrichment of neocrystallised diopsides compared to relict phenocrysts. This has been assigned to the liberation of trace elements on the breakdown of primary minerals, kaersutite and sphene, on entering the thermal aureole. Major and trace element compositions of minerals in migmatite melanosomes and leucosomes are almost identical, pointing to a syn- or post- solidus reequilibration on cooling of the migmatite terrain. The mineralogical, textural and geochemical evolution of dykes in a contact metamorphic aureole, is recorded around an apophysis of the PX1 intrusion, where there is evidence of incipient partial melting. Hydrothermal mineral pseudomorphs in the outer parts of the aureole are progressively replaced by dry mineral assemblages, with increasingly recrystallised diopside and evidence of partial melting -the extent of which varies from one lithology to another. The appearance of more mafic lithologies towards the intrusion, with lower whole rock SiO2 and mobile element abundances, e.g. Rb, Cs, K, has been explained by the migration and accumulation of feldspathic material into leucosomes outside the samples. A micro-structural study of leucosomes and leucocratic pods, with the aid of high-resolution X-ray computed micro-tomography (HRXµCT), allowing the visualization and quantification of shapes and orientations, was carried out in order to better understand the processes of melt segregation in the PX1 aureole. Leucocratic pods, representing former amygdales, are considered as natural strain ellipsoids. Their short axes are oriented perpendicular to leucosome planes, which sub-parallel the intrusive contact. Leucosomes thus effectively represent foliation planes. This implies that the direction of maximum shortening, during migmatisation, was perpendicular to the orientation of leucosomes, contradicting earlier models that suggest leucosomes represent tension veins. RESUME DE LA THESE Un phénomène rare de fusion partielle de filons basiques à basse pression a été étudié dans les auréoles de contact de deux intrusions litées de gabbro-pyroxénite, PX1 et PX2, localisées dans le soubassement de l'île volcanique de Fuerteventura aux Canaries. Cette anatexie a engendré des migmatites finement zébrées d'aspect très inhabituel, dont les processus de formation ont été étudiés dans le présent travail. Ces roches sont caractérisées par un réseau dense de veinules leucocrates d'épaisseur millimétrique, dont les textures ignées sont parfaitement préservées. La fusion partielle de roches basiques à basse pression (1-2 kbar) requiert la présence d'eau afin d'abaisser le solidus du système à des températures géologiquement réalistes. Une étude comparative des isotopes de l'oxygène a ainsi été menée sur des filons respectivement affectés et non affectés par le métamorphisme de contact, afin de confirmer la présence de ces fluides, de déterminer l'importance de leur interaction avec les roches et leur origine. Des valeurs de δ180 basses ou négatives ont été mesurées sur roche totale et minéraux séparés, décroissantes en direction du contact, alors que l'intrusion elle-même a conservé des valeurs élevées. Ce gradient a été attribué à l'advection d'eau météorique durant la mise en place du magma, les températures les plus élevées favorisant d'autant plus la circulation des fluides et les échanges isotopiques durant la recristallisation des roches. Cette recristallisation engendré une redistribution chimique complète des éléments entre les différentes générations de minéraux résultant du métamorphisme de contact et de l'anatexie, mise en évidence par microanalyse. Certains éléments traces comme Zr, Hf, Y et les REE ont été concentrés dans le diopside néoformé consécutivement à la déstabilisation de minéraux primaires riches en ces éléments comme la kaersutite ou le sphène. Les compositions en éléments majeurs et traces des minéraux des mélanosomes et leucosomes des migmatites sont pratiquement identiques, indiquant une rééquilibration syn- à postsolidus lors du refroidissement de l'auréole de contact. La transformation progressive des filons basiques au niveau de leur minéralogie, textures et composition chimique a pu être observée en détail à l'approche du contact d'une apophyse de l'intrusion PX1. La paragenèse magmatique initiale n'est jamais préservée, les faciès les plus distants du contact étant constitués d'un assemblage pseudomorphique hydrothermal. Ce dernier est progressivement remplacé par des assemblages anhydres incluant du diopside néoformé, puis apparaissent les premiers signes de fusion partielle, dont l'importance varie fortement d'une lithologie à l'autre. L'apparition de faciès plus basiques en direction du contact, avec des teneurs réduites en SiO2 et en éléments incompatibles tels Rb, Cs, K, a été attribuée à l'échappement de leucosomes feldspathiques hors du système. Une étude microstructurale de la distribution spatiale du matériel leucocrate au sein des migmatites par microtomographie X de haute résolution (HRXµCT) a été menée pour mieux comprendre les processus de ségrégation des liquides dans l'auréole de PX1. De petites entités ovoïdes, représentant d'anciennes structures amygdalaires au sein des filons, ont été considérées comme des ellipsoïdes marqueurs de la déformation finie. Leur petit axe est orienté perpendiculairement aux plans définis par les leucosomes, eux-mêmes subparallèles au contact intrusif. Les leucosomes matérialisent donc des plans de clivage. Ainsi, la direction de raccourcissement maximum durant la fusion partielle était perpendiculaire à l'orientation des leucosomes, contrairement à ce qui a été dit dans de précédentes publications, qui suggéraient que les leucosomes représentaient des veines de tension. RESUME DE LA THESE (POUR LE GRAND PUBLIC) L'observation directe du soubassement d'une île volcanique est une occasion rare, accessible dans le «complexe de base » de l'île canarienne de Fuerteventura. Ce dernier a enregistré divers phénomènes magmatiques, métamorphiques et de fusion partielle induits par l'intrusion répétée de magmas alimentant des appareils volcaniques sus jacents, sous forme de petits plutons, essaims de filons et complexes annulaires de gabbros alcalins, pyroxénites, syénites et carbonatites. Dans ce contexte de flux de chaleur élevé, des filons basiques ont subi une fusion partielle au contact de deux intrusions de gabbro-pyroxénite, un phénomène extrêmement rare à une profondeur aussi réduite, estimée à quelque 3-6 km. Les produits de cette fusion partielle sont des liquides très riches en feldspath, concentrés en un réseau dense de veinules blanches (leucosomes) au sein du matériau résiduel sombre non fondu (mélanosome) pour former ce qu'on appelle des migmatites. Outre les aspects pétrologiques liés à la formation de ces migmatites, l'intérêt majeur du phénomène réside dans le fait qu'il puisse représenter la source des magmas évolués parfois observés sur les îles océaniques. A des pressions aussi faibles que dans le soubassement de Fuerteventura (1-2 Kbar), la présence de fluides abondants est nécessaire pour abaisser la température de début de fusion des roches (solidus) à des valeurs géologiquement réalistes. Des expériences ont montré que même en présence de plusieurs %-poids d'eau, une température de 1000°C était encore nécessaire pour obtenir une proportion de liquide équivalente à celle observée sur le terrain, soit 25%. Or les magmas alcalins des îles océaniques, bien qu'hydratés, n'en contiennent de loin pas autant, ce qui implique une source d'eau externe. Une étude isotopique de l'oxygène a été entreprise afin de tester cette hypothèse. Les valeurs obtenues en 5180 sont basses ou négatives et indiquent l'influence d'eau d'origine météorique. Cette eau de pluie se serait infiltrée le long des filons depuis la surface du volcan et les aurait complètement hydrothermalisés en profondeur (situation encore visible à l'extérieur de l'auréole de contact), leur permettant ainsi de stocker l'eau nécessaire à leur fusion partielle ultérieure. L'interaction entre eau de pluie et filons a été d'autant plus importante que ces derniers étaient proches du contact avec l'intrusion, ce qui suggère que la circulation de ces eaux et leur interaction avec les roches a été favorisée par la chaleur fournie par l'intrusion elle-même. Un autre aspect de ce travail s'est focalisé sur la redistribution des éléments traces au sein des minéraux des filons basiques durant le métamorphisme de contact et la fusion partielle. Ainsi, le pyroxène de seconde génération est-il sensiblement enrichi en traces telles Zr, Hf, Y et les terres rares, par rapport au pyroxène magmatique originel, en relation avec la déstabilisation de minéraux primaires riches en ces éléments tels le sphène et la kaersutite. Cependant, les compositions en éléments majeurs et traces des minéraux recristallisés des migmatites sont pratiquement les mêmes dans les leucosomes et les mélanosomes, suggérant une rééquilibration chimique complète durant le refroidissement de ces lithologies. Si certaines migmatites se sont comportées en système chimiquement fermé (hormis l'eau météorique), d'autres filons ont manifestement perdu une partie de leurs leucosomes, ainsi qu'en témoigne leur composition progressivement appauvrie en silice et autres éléments incompatibles mobiles, tels K et Rb à l'approche du contact de l'intrusion. Parallèlement à cette évolution chimique, les paragenèses hydrothermales distantes du contact sont progressivement remplacées par des paragenèses anhydres, puis par l'apparition des premiers leucosomes, tandis que les textures magmatiques initiales sont complètement effacées au profit d'une combinaison de textures magmatique dans les leucosomes et en mosaïque dans les mélanosomes. Enfin, la distribution spatiale des liquides de fusion partielle a été étudiée par microtomographie X de haute résolution, dans des filons contenant des entités ovoïdes leucocrates, sans doute d'anciennes amygdales à zéolites. Ces dernières ont été considérées comme des ellipsoïdes de la déformation finie. L'orientation de leur petit axe, perpendiculaire au plan défini par les veinules de leucosomes, indique que ces derniers représentent des plans de clivage perpendiculaires à la direction de raccourcissement maximum. Ainsi, la ségrégation des liquides de fusion partielle se serait faite dans les plans de compression et non dans des plans de dilatation, contrairement à ce que laisserait penser le sens commun.