Search

The application of carbonate clumped isotope (Δ47) thermometry in deep-time is often limited by modification of the original temperature signal by thermal resetting. New modeling approaches to estimate the initial isotopic composition of partially reset calcites and maximal burial temperatures, however, open promising avenues in temperature reconstruction. Such approaches strongly depend on laboratory-derived kinetic parameters of calcite materials, which may differ in their microstructure, water content and distribution, and minor and trace element composition, and thus may have different resetting kinetics. The rostra of belemnites, an extinct group of mollusks with a wide temporal and spatial occurrence in the Mesozoic, have been extensively used for deep-time paleoclimate reconstructions using oxygen isotope geochemistry. Belemnites are also important targets for clumped isotope-based temperature reconstructions, but often are found to have reset Δ47 compositions. Here, we present results from heating experiments on belemnite rostral calcite and optical calcite and provide belemnite-specific kinetic parameters for clumped isotope resetting. We show that belemnite calcite is altered faster and at lower temperatures than optical calcite and all other calcites reported in previous studies. We suggest that fast initial resetting results from oxygen isotope exchange of belemnite calcite with internal skeletal water present as fluid inclusions or organic-derived water, a process completed within 2–4 min at the experimental temperatures used here. Extrapolation to geological timescales using different solid-state bond reordering models shows that belemnite calcite resetting starts at lower burial temperatures than brachiopod, spar, and optical calcites. This susceptibility to thermal resetting results in a measurable (+3 ◦C) increase of the apparent Δ47 temperature even under shallow to moderate burial conditions (i.e., 40–50 ◦C for 106–107 years timescales). Following the overprint to higher apparent Δ47 temperatures during burial, the belemnite Δ47 may further reequilibrate during exhumation resulting in a decrease of apparent Δ47 temperatures. Such “retrograde resetting” is similar to what is observed for carbonatites and marbles during cooling, and may cause underestimation of the thermal resetting a sample experienced during its geological history. Overall, our results demonstrate the importance of material-specific kinetic parameters and we urge caution when interpreting Δ47-derived temperatures of biogenic carbonates from deep-time archives., Swiss National Science Foundation project number 200021_169849, AF from Juan de la Cierva Fellowship (IJC2019040065-I), Spanish Ministry of Science and Innovation, European Development Fund and the European Social Fund, The European Commission, Horizon 2020 (ICECAP; grant no. 101024218), The Research Council of Norway Centre of Excellence funding scheme, Project number 223272. EBSD data for the WA-CB-11 brachiopod provided by the authors of Henkes et al. (2014), US National Science Foundation (EAR-1226832)

Crustal geochemical signatures in carbonatites may arise from carbon recycling through the mantle or from fluid-mediated interaction with the continental crust. To distinguish igne-ous from fluid-mediated processes, we experimentally determined rare earth element (REE) partitioning between calcite/melt and apatite/melt at subvolcanic emplacement conditions (1-2 kbar, 750-1000 degrees C). Our data allow modeling of calcite-apatite (Cc/Ap) partition coef-ficients (D), representing a new tool to bypass the previously required but largely unknown carbonatite melt composition. Experimentally determined magmatic calcite/apatite REE patterns are flat, as DLa D Cc /Ap/ Cc Ap Lu / is similar to 0.75, and they show a slight U-shape that becomes more pronounced with temperature decreasing from 1000 to 750 degrees C. Application to texturally well-equilibrated natural Ca-carbonatites and calcite-bearing nephelinites shows that some calcite-apatite pairs follow this pattern and, hence, confirm the magmatic nature of the car-bonates. DLaD Cc/Ap / values of other mineral pairs range from 10-2 to 10-3, which, together Cc/Ap Lu with a substantial light REE depletion in the calcite, is interpreted as fluid-mediated light REE removal during secondary calcite recrystallization. Calcite/apatite REE distributions are well suited to evaluate whether a carbonatite mineralogy is primary and magmatic or has been affected by secondary recrystallization. In this sense, our tool provides information about the sample's primary or secondary nature, which is essential when assigning isotopic crustal signatures (in Ca, C, or Sr) or REE patterns to related geologic processes., Geology, 51 (1), ISSN:0091-7613, ISSN:1943-2682

Magmas readily react with their wall-rocks forming metamorphic contact aureoles. Sulphur and possibly metal mobilization within these contact aureoles is essential in the formation of economic magmatic sulphide deposits. We performed heating and partial melting experiments on a black shale sample from the Paleoproterozoic Virginia Formation, which is the main source of sulphur for the world-class Cu-Ni sulphide deposits of the 1.1 Ga Duluth Complex, Minnesota. These experiments show that an autochthonous devolatilization fluid effectively mobilizes carbon, sulphur, and copper in the black shale within subsolidus conditions (≤ 700 °C). Further mobilization occurs when the black shale melts and droplets of Cu-rich sulphide melt and pyrrhotite form at ∼1000 °C. The sulphide droplets attach to bubbles of devolatilization fluid, which promotes buoyancy-driven transportation in silicate melt. Our study shows that devolatilization fluids can supply large proportions of sulphur and copper in mafic–ultramafic layered intrusion-hosted Cu-Ni sulphide deposits., Nature Communications, 12 (1), ISSN:2041-1723

Carbon isotope exchange between carbon-bearing high temperature phases records the carbon (re-) processing in the Earth's interior, where the vast majority of global carbon is stored. Redox reactions between carbonate phases and elemental carbon govern the mobility of carbon, which then can be traced by its isotopes. We determined the carbon isotope fractionation factor between graphite and a Na2CO3-CaCO3 melt at 900–1500 °C and 1 GPa; The failure to isotopically equilibrate preexisting graphite led us to synthesize graphite anew from organic material during the melting of the carbonate mixture. Graphite growth proceeds by (1) decomposition of organic material into globular amorphous carbon, (2) restructuring into nano-crystalline graphite, and (3) recrystallization into hexagonal graphite flakes. Each transition is accompanied by carbon isotope exchange with the carbonate melt. High-temperature (1200–1500 °C) equilibrium isotope fractionation with type (3) graphite can be described by (temperature T in K). As the experiments do not yield equilibrated bulk graphite at lower temperatures, we combined the ≥1200 °C experimental data with those derived from upper amphibolite and lower granulite facies carbonate-graphite pairs (Kitchen and Valley, 1995; Valley and O'Neil, 1981). This yields the general fractionation function usable as a geothermometer for solid or liquid carbonate at ≥600 °C. Similar to previous observations, lower-temperature experiments (≤1100 °C) deviate from equilibrium. By comparing our results to diffusion and growth rates in graphite, we show that at ≤1100 °C carbon diffusion is slower than graphite growth, hence equilibrium surface isotope effects govern isotope fractionation between graphite and carbonate melt and determine the isotopic composition of newly formed graphite. The competition between diffusive isotope exchange and growth rates requires a more careful interpretation of isotope zoning in graphite and diamond. Based on graphite crystallization rates and bulk isotope equilibration, a minimum diffusivity of Dgraphite = 2 × 10−17 m2s−1 for T > 1150 °C is required. This value is significantly higher than calculated from experimental carbon self-diffusion constants (∼1.6 × 10−29 m2 s−1) but in good agreement with the value calculated for mono-vacancy migration (∼2.8 × 10−16 m2 s−1)., Geochimica et Cosmochimica Acta, 253, ISSN:0016-7037, ISSN:1872-9533

Carbon isotope fractionation in the CO2–CO–CH4–C system was investigated at 300–1200 °C at near-atmospheric pressures by thermally decomposing a variety of organic materials in sealed quartz tubes. Measured gas speciations correspond well to the expected range from thermodynamic calculations. We show that chemical and isotopic equilibrium among gas species is obtained when applying a nickel catalyst for CO2/CH4, CH4/CO, and CO2/CO at ≤600 °C or without a catalyzing agent for CO2/CO at ≥800 °C. The experiments define carbon isotope fractionation factors for the CO2/CH4, CO2/CO and CH4/CO pairs as (i) 103lnαCO2/CH4 = 8.9 (±0.6) ⋅ 105⋅ (1/T2)0.825(±0.005) (200–1200 °C) (ii) 103lnαCO2/CO = 1.07 (±0.05) ⋅ 106⋅ (1/T2)0.830(±0.003) (300–1200 °C) (iii) 103lnαCH4/CO = 1.1 (±0.2) ⋅ 103⋅ (1/T2)0.462(±0.001) (300–1200 °C), which reproduce the experimental values within 0.2‰ for CO2/CH4 and CO2/CO and within 0.12‰ for CH4/CO (T in K, 1σ fit uncertainties in brackets, CO2/CH4 includes the ≤600 °C experimental data of Horita, 2001). Carbon isotope fractionation factors at 1000 °C are still large for CO2/CH4 and CO2/CO (6.6 and 7.5‰ respectively) but only 1.5‰ for CH4/CO. Elemental carbon precipitated through thermal decomposition of the organic starting materials yields δ13C values that depend on the X(O) = O/(O + H) of the organic starting material, i.e. the initial oxidation state of carbon in the organics. We further observe a catalytic effect of the quartz walls on chemical and isotopic exchange in the CO2/CO system, probably due to the activation of the silicate surface by H+ and OH− ions at >650 °C. Our experimental results yield improved calibrations of the CO2/CH4 equilibria and the first experimental calibration of CO2/CO and CH4/CO carbon isotope fractionation. Applications are in the tracing of magmatic hydrothermal gas emissions, in carbon-precipitating COH-fluids, and in monitoring of coal-seam fires, but our results may also be applied for quality control during steel-making processes., Earth and Planetary Science Letters, 506, ISSN:0012-821X, ISSN:1385-013X

Compositional variations of peridotites from the Cretaceous ophiolites in southern Turkey and Northern Cyprus are presented to document the nature of partial melting and possible effects of reactive dissolution of primary mantle phases during fore-arc spreading. The peridotites overall exhibit a range of 187Os/188Os ratios from 0.1171 to 0.1266 and appear to represent a mantle region that preserves a record of ancient melt depletion. The samples are depleted in 187Os/188Os compared to the ambient oceanic upper mantle (187Os/188Os ~0.127), suggesting that they are representatives of a shallow fore-arc mantle where transport of radiogenic 187Os during slab dehydration was limited. Chemical variations of primary mantle minerals indicate that the peridotites are the residues of moderate to high degrees (>16%) of partial melting and have experienced significant modal and chemical compositional modification through interaction with oxidizing hydrous basaltic melts. Interacting melts, which appear to be similar in composition to primitive arc tholeiite, are likely to have originated from sub-lithospheric lower part of the mantle wedge during early stages of fore-arc spreading and migrated upward to react with variably depleted harzburgites to induce further melting in the overlying lithospheric mantle through open-system reactive flow. This second stage melting resulted in (1) common occurrence of reactive harzburgites and dunites by incongruent melting of orthopyroxene and crystallization of olivine through interaction with olivine saturated melt; and (2) local development of refertilized peridotites by shallower melt impregnation that involves interaction with olivine + clinopyroxene saturated melt. The dissolution of orthopyroxene caused the reacting melt to be enriched in silica and diluted in incompatible elements which led to the production of the final melts similar in composition to fore arc basalt and boninite. Involvement of compositionally variable mantle and melt components with different rates of melt influx therefore appears to explain the generation of fore-arc crust with a range of diverse rock suites including temporally and spatially associated arc tholeiites and boninites with significant depletion in incompatible elements.

Progress in Earth and Planetary Science, 1 (1)

Many studies have documented hydrous fractionation of calc-alkaline basalts producing tonalitic, granodioritic, and granitic melts, but the origin of more alkaline arc sequences dominated by high-K monzonitic suites has not been thoroughly investigated. This study presents results from a combined field, petrologic, and whole-rock geochemical study of a paleo-arc alkaline fractionation sequence from the Dariv Range of the Mongolian Altaids. The Dariv Igneous Complex of Western Mongolia is composed of a complete, moderately hydrous, alkaline fractionation sequence ranging from phlogopite-bearing ultramafic and mafic cumulates to quartz–monzonites to late-stage felsic (63–75 wt% SiO[subscript 2]) dikes. A volumetrically subordinate more hydrous, amphibole-dominated fractionation sequence is also present and comprises amphibole (±phlogopite) clinopyroxenites, gabbros, and diorites. We present 168 whole-rock analyses for the biotite- and amphibole-dominated series. First, we constrain the liquid line of descent (LLD) of a primitive, alkaline arc melt characterized by biotite as the dominant hydrous phase through a fractionation model that incorporates the stepwise subtraction of cumulates of a fixed composition. The modeled LLD reproduces the geochemical trends observed in the “liquid-like” intrusives of the biotite series (quartz–monzonites and felsic dikes) and follows the water-undersaturated albite–orthoclase cotectic (at 0.2–0.5 GPa). Second, as distinct biotite- and amphibole-dominated fractionation series are observed, we investigate the controls on high-temperature biotite versus amphibole crystallization from hydrous arc melts. Analysis of a compilation of hydrous experimental starting materials and high-Mg basalts saturated in biotite and/or amphibole suggests that the degree of K enrichment controls whether biotite will crystallize as an early high-T phase, whereas the degree of water saturation is the dominant control of amphibole crystallization. Therefore, if a melt has the appropriate major-element composition for early biotite and amphibole crystallization, as is true of the high-Mg basalts from the Dariv Igneous Complex, the relative proximity of these two phases to the liquidus depends on the H[subscript 2]O concentration in the melt. Third, we compare the modeled high-K LLD and whole-rock geochemistry of the Dariv Igneous Complex to the more common calc-alkaline trend. Biotite and K-feldspar fractionation in the alkaline arc series results in the moderation of K[subscript 2]O/Na[subscript 2]O values and LILE concentrations with increasing SiO[subscript 2] as compared to the more common calc-alkaline series characterized by amphibole and plagioclase crystallization and strong increases in K[subscript 2]O/Na[subscript 2]O values. Lastly, we suggest that common calc-alkaline parental melts involve addition of a moderate pressure, sodic, fluid-dominated slab component while more alkaline primitive melts characterized by early biotite saturation involve the addition of a high-pressure potassic sediment melt.

Journal of Petrology, 47 (3), ISSN:0022-3530, ISSN:1460-2415

Unlike in many other stable isotope systems, equilibrium isotope fractionation in the carbon system is predicted by statistical mechanics’ theory to be in the order of several permil at high temperatures corresponding to Earth’s deep and extreme environments. Experimental studies confirming such theoretical predictions are rare, and lacking for most mineral-mineral, mineral-fluid, and fluid-fluid pairs. This thesis investigates carbon isotope exchange experiments between 300 – 1500 °C in the systems (i) CH4-CO2-CO, (ii) graphite-Na2CO3-CaCO3 (melt) and (iii) CH4-CO2-CO-graphitic carbon. Experiments have been conducted with box- and gas-mixing furnaces, externally-heated pressure vessels (“cold seals”), and piston cylinder apparatuses. Sample characterization (before and after experiment) involved gas-chromatography (GC), elemental analyzer (EA) and gas bench (GB) systems connected to an isotope ratio mass spectrometer (IRMS). Project (i) examined the carbon isotope exchange in a gas phase by using a variety of organic starting materials that were, sealed under vacuum in quartz tubes, thermally decomposed (300 -1200 °C) to create a CH4-CO2-CO gas mix plus an elemental carbon residual. While the decomposition and gas generation is nearly instantaneous, chemical and isotopic equilibration is protracted: vials with an additional piece of Ni-foil catalyzed the equilibration reactions, whereas experiments without Ni yield only minor chemical and isotopic reactions, if any. Measured gas speciation of catalyzed gases correspond well to the expected range from thermodynamic calculations. The experiments define carbon isotope partition functions for the CO2/CH4, CO2/CO and CH4/CO pairs as (T in Kelvin) 10^3ln a (CO2/CH4) = 8.9(±0.6)*10^5 (1/T^2)^0.825(±0.005) 10^3ln a (CO2/CO) = 1.07(±0.05)*10^6 (1/T^2)^0.830(±0.003) 10^3ln a (CH4/CO) = 1.1(±0.2)*10^3 (1/T^2)^0.462(±0.001) Project (ii) explored the carbon isotope exchange during graphite crystallization in a Na2CO3-CaCO3 melt at 900-1500 °C, 1 GPa using a piston-cylinder device. Graphite was grown anew from organic material during the melting of the carbonate mixture. Graphite growth proceeds by (1) decomposition of organic material into globular amorphous carbon, (2) restructuring into nano-crystalline graphite, and (3) recrystallization into hexagonal micron-sized graphite flakes. Each transition is accompanied by carbon isotope exchange with the carbonate melt. As the experiments did not yield bulk-equilibrated graphite at lower temperatures, the >1200 °C data was combined with empirical fractionation factors for carbonate-graphite obtained from upper amphibolite and lower granulite facies carbonate-graphite pairs reported by Kitchen & Valley (1995; J. metamorphic Geol., 13: 577-594) and Valley & O’Neil (1981; Geochim. Cosmochim. Acta 45: 411-419) which results in the general fractionation function (T in Kelvin) ∆13C(carbonate-graphite) = (3.37(4)*10^6)/T^2 This function is usable as a geothermometer for solid or liquid carbonate at >600 °C. Similar to previous observations, lower-temperature experiments (

The northern Tanzanian sector of the Gregory Rift plays a key role in the understanding of different scientific problems that span the fields of petrology, volcanology and anthropology. There are only few areas on Earth, where continental rifting is currently active. The East African Rift System is one of these areas, and the Gregory Rift is one of the most multifaceted parts of it, where early and recent episodes of the continental rift development are recorded in exposed igneous and sedimentary rocks. One of the challenges related to continental rifting is to explain the broad compositional and isotopic diversity of generated magmas. The genesis of alkali-rich carbonatites at the Oldoinyo Lengai volcano is one of the most frequently studied aspects of this chemical diversity. However, the carbonatites constitute a volumetrically insignificant part of the volcano’s total volume. In order to fully understand the role of carbonatite genesis in relation to continental rifting processes, it is therefore also important to investigate the associated, and considerably less-studied, silicate rocks. In addition to this, several anthropologically important sites occur in the Gregory Rift, which record the premodern human evolution at the Pliocene-Holocene transition. In this thesis, the silicate rocks from the northern Tanzanian sector of the Gregory Rift are used to investigate the temporal and compositional evolution of the area, with special emphasis on the evolution of silicate magmas at the currently active Oldoinyo Lengai volcano. For a better understanding of magma generation in a continental rift environment, isotope characteristics of a representative suite of rocks belonging to the northern sector of the Gregory Rift were investigated. The oldest of these (3.8-1.4 Ma) are exposed in the Rift Escarpment, and the youngest rocks are represented by different rocks from Oldoinyo Lengai and from the Lake Natron Engaruka-Monogenetic Volcanic Field (LNE-MVF). The results show that Rift Escarpment magmas were produced from a heterogeneous mantle source, where the heterogeneity is the result of carbonatite-related metasomatism. The eruptive centres of the considerably younger LNE-MVF produced the most primitive (melilititic) magmas in the area. These magmas originate from a homogeneous mantle source, while the more evolved (nephelinitic and basanitic) magmas from the same area are derived from a slightly more heterogeneous source. The silicate lavas from Oldoinyo Lengai define two separate geochemical trends, which reflect two distinct stages of volcanism at Oldoinyo Lengai. During the early stage, the volcano erupted various nephelinitic and phonolitic magmas with enriched isotopic compositions. The evolution of these magmas was controlled by crystal fractionation. A total fractionation of approximately 90% crystals allows to successfully reproduce the observed geochemical trend. The second stage of volcanism, which is currently ongoing, is characterized by eruptions of nephelinitic magmas with phenocrysts of wollastonite, combeite and melilite. These magmas are also intimately associated with the alkali-rich carbonatites. The evolution of magmas belonging to Stage II is controlled by crystal fractionation up until the point of carbonatite exsolution, after which the silicate magmas evolve by a combination of processes such as fractionation, remelting and assimilation of cumulates left over from the first stage of volcanism and/or magma mixing with crystal mushes. The pronounced differences in geochemistry as well as in the isotope characteristicsof the volcanic rocks belonging to the first and the second stage of evolution indicate different primary magmas for the two series. A combination of isotope mixing calculations and fractional crystallization modelling indicates that the primary magma(s) of the second stage can be explained by interaction (partial remelting and assimilation) of cumulates left over in the crust from the first stage of volcanism with magma(s) that are geochemically and isotopically similar to the primitive magmas erupted in the neighbouring LNE-MVF. The second part of the thesis provides the first steps aimed at establishing a tephrostratigraphic framework for this part of the Gregory Rift. The recently discovered Engaro Sero Footprint Site hosts one of the best-preserved sets of fossilized hominid footprints in the world, which are preserved in a sequence of volcaniclastic sedimentary rocks. In this thesis, it is shown that the footprint-bearing horizon consists of a primary volcanic ash-fall that has been slightly reworked by water, and that it was produced during a voluminous eruption of the Oldoinyo Lengai volcano. The sedimentary unit, which covered the footprints and helped to preserve them, consists of the wind-blown material derived from the same eruption intermixed with locally derived detrital material. A combination of field observations, regional stratigraphy and a possible correlation with ash layers preserved in the lacustrine sequence of Lake Emakat, located 20 km SW of the volcano, indicates an early Holocene age of the footprints. In a sediment core from Lake Emakat, seven distinct ash layers of late Pleistocene – early Holocene age were described interbedded with a sequence of lacustrine sediments. These ash horizons have nearly identical mineralogy to that found in most recent and pre-modern pyroclastic rocks of Oldoinyo Lengai. Based on information from this sediment core, an average recurrence rate 810 years between voluminous explosive eruptions can be calculated for Oldoinyo Lengai during this time span. This data fill an important informational gap regarding the eruptive history of the volcano. The main reason for this is that the written accounts for the area only cover the last ~150 years, and reliable age determinations are sparse at best.

In this thesis a series of piston cylinder experiments has been conducted on partially molten systems, to understand the textural and mechanical relationship between the solid and liquid phases. Three different experimental apparati have been used for these the experiments. A standard end-loaded piston cylinder to perform annealing experiments; an end-loaded piston cylinder equipped for in-situ measurement of electrical conductivity of samples has been exploited to determine the conductivity of silicate-metallic samples and a newly developed centrifuging piston cylinder, which applies high temperature, pressure and acceleration to partially molten samples. These apparati and the assemblages to run experiments with them are described in detail. Before the description and analysis of the four systems investigated in detail, all nonsuccessful experiments are briefly reviewed. Most of the latter experiments have been performed in an olivine plus Fe-S system and a natural gabbro system, with negative results. A series of annealing experiments and conductivity measurements has been performed to investigate the relationship between solid olivine and molten Fe-S. A percolation threshold larger than 15 vol% has been found for this system as bulk conductivity similar to that of pure olivine has been measured for mixtures containing less than 20 vol% of Fe-S. The measurements of dihedral angles at different oxygen contents of the metallic alloy has revealed that the dihedral angle itself is always larger than 60° for oxygen contents up to 16.6 atom%. A grain growth law for olivine surrounded by molten ironsulfide has been computed. From these data the final conclusion that in an accreting planetary body , in which the silicate fraction remain crystalline, only the metal melt in excess of 15 vol% could segregate to form a core, the rest being trapped in the silicate mantle. In the system “peridotite plus Fe-S” gravitational separation of Fe-S melt from solid or partially molten peridotites has been studied, together with determination of a percolation threshold for the system. Segregation rates of small Fe-S droplets in a peridotite matrix were determined to 2.4·10-4 to 1.1·10-3 mm/h in presence of silicate melt, dropping to just 8.4-8.6·10-5 mm/h in solid peridotite. This corresponds to permeabilities of 10-16-10-17 m2 for the silicate melt present system, and to a permeability of 10-19 m2 for silicate melt absent system. These data yield a calculated core formation time for the Earth of ca. 300 My, for Mars of about 240 My and for Vesta of about 220 My. These values are at least one order of magnitude larger than Hf/W ages for core formation in these planetary bodies. Conductivity experiments show results similar on the “olivine plus Fe-S” and the “peridotite plus Fe-S” systems. Bulk conductivities were measured and are similar for mixtures containing up to 15 vol% of Fe-S. From these results the following conclusions arise: Fe-S melt percolation in solid or partially molten peridotites is too slow, to account for core formation in earth size bodies and the percolation threshold for peridotite plus Fe-S melt is larger than 15 vol%. Melt extraction experiments have been performed, with the use of the centrifuging piston cylinder, in a system constituted of olivine aggregates with small amounts of basaltic melt added (system olivine plus MORB). A permeability between 1.57·10-10 and 4.48·10- 8 10 m2 has been found for aggregates containing 5 to 13 vol% of melt, which correspond to extraction rates of 3.2 to 16.7 m/yr. These values do not agree to those computed with the models of McKenzie (1984) or supposed by Faul (1997), however they are close to those of Faul’s (1997) model. A comparison with the segregation rates calculated on the base the isotopic ratios of Th, U and Ra in natural MORB melt yields good agreement with the permeabilities of this study. A last series of experiments has also been done with the centrifuging piston cylinder dealing with a system constituted by basaltic melt containing 30 vol% of olivine (system MORB plus olivine). The settling velocity for a suspension of olivine grains has been measured to be 1/10 of the settling velocity of a single isolated grain (settling with a Stokes’ velocitiy), confirming experimentally the observations of Schwindinger (1999). A formation time of ca. 0.32 m/yr has been computed for olivine orthocumulates, which is in agreement with evidences from a natural olivine orthocumulate present in the succession of the Murotomisaki gabbroic Complex (Japan). After mechanical settling is completed, gravity induced dissolution-repricipitation processes lead to the formation of adcumulates where the previous porous space is filled with the cumulus phase. A possible formation time for olivine adcumulates of ca. 1 cm/yr, after maximum close packing of grains is reached, has been estimated on the base of experimental data. This last value is valid only for adcumulate layers which are a few centimeters thick.