Petrogenesis of a basalt-comendite-pantellerite rock suite: the Boseti Volcanic Complex (Main Ethiopian Rift)

Petrological and geochemical data for basic (alkali basalts and hawaiites) and silicic peralkaline rocks, plus rare intermediate products (mugearites and benmoreites) from the Pleistocene Boseti volcanic complex (Main Ethiopian Rift, East Africa) are reported in this work. The basalts are slightly alkaline or transitional, have peaks at Ba and Nb in the mantle-normalized diagrams and relatively low ⁸⁷Sr/⁸⁶Sr (0.7039–0.7044). The silicic rocks (pantellerites and comendites) are rich in sanidine and anorthoclase, with mafic phases being represented by fayalite-rich olivine, opaque oxides, aenigmatite and slightly Na-rich ferroaugite (ferrohedenbergite). These rocks were generated after prolonged fractional crystallization process (up to 90–95 %) starting from basaltic parent magmas at shallow depths and fO₂ conditions near the QFM buffer. The apparent Daly Gap between mafic and evolved Boseti rocks is explained with a model involving the silicic products filling upper crustal magma chambers and erupted preferentially with respect to basic and intermediate products. Evolved liquids could have been the only magmas which filled the uppermost magma reservoirs in the crust, thus giving time to evolve towards Rb-, Zr- and Nb-rich peralkaline rhyolites in broadly closed systems.     

A pantelleritic welded ash-flow tuff from the Ethiopian Rift Valley

Fantale (Lat. 8° 58 N., Long. 39° 54 E) is a typical Quaternary silicic strato-volcano situated on the floor of the northern part of the Main Ethiopian Rift. Following the growth of the main central cone, a voluminous ash-flow tuff was erupted in association with the development of the 4 km summit caldera. On the upper parts of the volcano the tuff is virtually restricted to three major eruptive pathways descending from the caldera. The tuff is more extensive on the lower flanks and forms a continuous sheet surrounding the volcano.The tuff is welded throughout, even when it is less than 1 m thick. It shows typical vitroclastic texture and is markedly eutaxitic in the intensely welded sections. The degree of welding appears to he largely independent of the total thickness of the tuff. Microscopic observations provide confirmatory evidence of postdepositional vesiculation of the tuff and also suggest that recrystallisation has not induced the development of cavities in the tuff.The detailed analytical data presented suggest that the chemical composition of the tuff was modified by recrystallisation which produced relative depletion in SiO₂, total Fe, Cl, and Na₂O and enrichment in Al₂O₃, K₂O and CaO. The trace elements Ba, La, Nb, Rb, and Zr were not affected by this process. Study of serial samples from three sections through the tuff has led to the identification of five of the constituent flow units of the tuff. The units are approximately homogeneous in composition but differ from each other. The earliest unit is more silicic and pantelleritic, whereas the later flow units are more trachytic in composition. This data is interpreted in terms of the progressive emptying of a small but strongly zoned high-level magma chamber.The presence of primary laminar flowage structures, postdepositional vesiculation effects, intense welding in thin units and high initial dips suggests that the Fantale tuff was deposited from a series of dense, perhaps partially fluidized, magma pulses.     

攀西古裂谷内龙帚山玄武岩—碱玄响岩建造

在川、滇、黔三省晚二叠世峨眉山玄武岩省的中部，产于攀西古裂谷内的龙帚山火山岩建造自下而上有4个喷发旋回：拉斑玄武岩、碱玄响岩、钾粗面玄武岩以及拉斑玄武岩。上下两个玄武岩旋回的岩性与该裂谷带以外大面积展布的高钛玄武岩岩性相同；而其碱玄响岩旋回主要由响碱玄岩—碱玄响岩—响岩岩系组成，夹粗安岩和安粗岩层，此旋回厚达千米，类似于非洲肯尼亚裂谷带的高原溢流响岩岩系。龙帚山晚二叠世火山岩建造的层序和岩石组合不同于古裂谷带外正常的大陆板内暗色岩喷出相的层序和岩浆演化规律，可能是地幔热柱和裂谷背景下岩石圈拉张复合作用的结果。     

Structure of the Ethiopian lithosphere: Xenolith evidence in the Main Ethiopian Rift

The lithospheric and sublithospheric processes associated with the transition from continental to oceanic magmatism during continental rifting are poorly understood, but may be investigated in the central Main Ethiopian Rift (MER) using Quaternary xenolith-bearing basalts. Explosive eruptions in the Debre Zeyit (Bishoftu) and Butajira regions, offset 20 km to the west of the contemporaneous main rift axis, host Al-augite, norite and lherzolite xenoliths, xenocrysts and megacrysts. Al-augite xenoliths and megacrysts derived from pressures up to 10 kb are the dominant inclusion in these recent basalts, which were generated as small degree partial melts of fertile peridotite between 15 and 25 kb. Neither the xenoliths nor the host basalts exhibit signs of carbonatitic or hydrous (amphibole + phlogopite) metasomatism, suggesting that infiltration of silicate melts resulting in pervasive Al-augite dyking/veining dominates the regional lithospheric mantle. Recent geophysical evidence has indicated that such veining/dyking is pervasive and segmented, supporting the connection of these Al-augite dykes/veins to the formation of a proto ridge axis. Al-augite xenoliths and megacrysts have been reported in other continental rift settings, suggesting that silicate melt metasomatism resulting in Al-augite dykes/veins is a fundamental processes attendant to continental rift development.     

Chemistry of basalts from the Icelandic Rift Zone

The chemistry of basalts from the Icelandic Rift Zone is discussed on the basis of 34 analysis, both new and previously published. It is concluded that a distinct chemical grouping is present between alkalibasalts and tholeiitic basalts.The tholeiitic basalts can furthermore be grouped into batches with slightly different primitive chemistry but identical differentiation trends.The chemical variation within a group of basalts from the same volcanic center can be related to a time scale in such a way, that the youngest (historic) lavas all show advanched differentiation, but the lavas known to be oldest have the most primitive chemistry.The possibility is suggested, that the volcanic activity along the Icelandic Rift Zone represents an independent cycle of chemical differentiation, distinct from the Tertiary volcanic activity along the Wyville Thompson Ridge.Attention is drawn to a fairly well defined grouping of basalts from different parts of the Midatlantic Ridge using the plot combined iron against magnesia.     

Landslides in the Ethiopian highlands and the Rift margins

Landslide hazard is one of the crucial environmental constraints for the development of Ethiopia, representing a limiting factor for urbanization and infrastructures. The high relief and the rugged topography induced by a strong Plio-Quaternary uplift, the occurrence of clayey horizons within the sedimentary sequences, the dense network of tectonic fractures and faults, the thick eluvial mantles on volcanic outcrops, and the thick colluvial–alluvial deposits at the foot of steep slopes are the predisposing factors for a large variety of mass movements. Heavy summer rainfall is the main triggering factor of most landslides, some of which undergo a step-like evolution with long-lasting quiescence intervals. First generation movements are commonly restricted to shallow phenomena, such as soil slips or mud flows in eluvial–colluvial material. Fast moving slope failures, such as rock slides, topplings and falls, are also triggered by earthquakes. To mitigate the landslide risk, any first priority measure should include adequate drainage of slopes in order to reduce water infiltration. On the other hand, appropriate site selection for buildings, transferring risky settlements, accurate geological control of works, and education campaigns are all strongly recommended.     

Obsidian from the northern sector of the Main Ethiopian Rift: implications for archeology

Obsidian is abundant in the Main Ethiopian Rift (MER). Petrological and geochemical features of obsidian from four volcanic centers in the MER, namely Birenti, Dofen, Fentale and Kone, are presented. Compositional and petrological variability is noted among the Dofen and Fentale obsidian, but not in those from Kone and Birenti where each have separate but uniform elemental composition. The Fentale and Kone obsidian were source materials for the artifacts of a number of Middle Stone Age and Later Stone Age/Neolithic sites in the region. We have yet to determine whether Dofen and Birenti were sources for archeological artifacts. The study also shows that volcanic episodes from a single center do not necessarily result in compositional variability.     

Peralkaline magma evolution and the tephra record in the Ethiopian Rift

The 3.119 ± 0.010 Ma Chefe Donsa phreatomagmatic deposits on the shoulder of the Ethiopian Rift mark the northern termination of the Silti-Debre Zeyit Fault Zone, a linear zone of focused extension within the modern Ethiopian Rift. These peralkaline pumice fragments and glass shards span a wide range of glass compositions but have a restricted phenocryst assemblage dominated by unzoned sanidine. Glass shards found within the ash occupy a far more limited compositional range (75–76 wt% SiO₂) in comparison with the pumice (64–75 wt% SiO₂), which is rarely mingled. Thermodynamic modeling shows that liquids broadly similar to the least evolved glass composition can be achieved with 50–60 % fractionation of moderately crustally contaminated basalt. Inconsistencies between modeled solutions and the observed values of CaO and P₂O₅ highlight the important role of fluorine in stabilizing fluor-apatite and the limitations of current thermodynamic models largely resulting from the scarce experimental data available for the role of fluorine in igneous phase stability. On the basis of limited feldspar heterogeneity and crystal content of pumice at Chefe Donsa, and the difficulties of extracting small volumes of Si-rich melt in classical fractional crystallization models, we suggest a two-step polybaric process: (1) basaltic magma ponds at mid-upper-crustal depths and fractionates to form a crystal/magma mush. Once this mush has reached 50–60 % crystallinity, the interstitial liquid may be extracted from the rigid crystal framework. The trachytic magma extracted at this step is equivalent to the most primitive pumice analyzed at Chefe Donsa. (2) The extracted trachytic liquid will rise and continue to crystallize, generating a second mush zone from which rhyolite liquids may be extracted. Some of the compositional range observed in the Chefe Donsa deposits may result from the fresh intrusion of trachyte magma, which may also provide an eruption trigger. This model may have wider application in understanding the origin of the Daly Gap in Ethiopian magmas—intermediate liquids may not be extracted from crystal-liquid mushes due to insufficient crystallization to yield a rigid framework. The wide range of glass compositions characteristic of the proximal Chefe Donsa deposits is not recorded in temporally equivalent tephra deposits located in regional depocenters. Our results show that glass shards, which represent the material most likely transported to distal depocenters, occupy a limited compositional range at high SiO₂ values and overlap some distal tephra deposits. These results suggest that distal tephra deposits may not faithfully record the potentially wide range in magma compositions present in a magmatic system just prior to eruption and that robust distal–proximal tephra correlations must include a careful analysis of the full range of materials in the proximal deposit.     

The origin of high bicarbonate and fluoride concentrations in waters of the Main Ethiopian Rift Valley, East African Rift system

Thermal waters in the Main Ethiopian Rift Valley are characterized by high Na, bicarbonate and fluoride concentrations, and near-neutral to alkaline pH. Sodium, bicarbonate and fluoride are positively correlated in the waters. The principal reason for the bicarbonate in the area is the high rate of carbon dioxide outgassing. This, combined with acid volcanics, geothermal heating, low Ca and low salinity, is also one of the causes of high fluoride in this part of the active volcanic zone of the East African Rift. Evaporative concentration is responsible for the high salinity, alkalinity and fluoride in the closed-basin lakes of the region. The waters are undersaturated with respect to fluoride and anhydrite. Calcium tends to be fixed in Ca bearing minerals such as calcite and epidote, which are abundant in the system. Hence, it appears that fluoride is a mobile component in acid volcanic geothermal systems.     

Local ecological knowledge and wetland management in the Ethiopian Rift Valley

GeoJournal - Freshwater systems are productive environments and vital for the livelihood of local people. The provision of goods and services is, however, related to their functionality and...     

Ground cracks in Ethiopian Rift Valley: facts and uncertainties

No accurate relationship has been obtained in this study between ground cracks in Ethiopian Rift Valley and the most common causes of earth fissures such as aquifer-system compaction and increased horizontal seepage stresses. This is due to the fact that the level of groundwater withdrawal responsible for these processes is still negligible in Ethiopia. If aquifer-system compaction and increased horizontal seepage stresses have a certain role, then it should be through the long-term effect of groundwater flow from basins to neighboring lakes. The ground cracks appeared also not to have a direct link with active faulting or distant earthquakes. Structurally, the Ethiopian Rift Valley is dominated by NE–SW-trending tensional faults, but no evidence is obtained in this study to associate the process of surface cracking with major tectonics. However, an aseismic elastic strain, which originates at depth and propagates upward through sediments without the formation of bedrock faults, could result in conditions conducive to the development of cracks. Then, fissures might ultimately be created after heavy rainfalls by near-surface processes such as piping and hydrocompaction along water-line sources.     

Geochemistry of a Rapakivi Granite Suite in a Proterozoic Rift Trough in Beijing and Its Vicinity

Controlled by E-W-trending faults, a Proterozoic (1.4-1.8 Ga old) rapakivi granite suite was intruded inBeijing and the area to its east (within Hebei Province), forming three parallel belts of igneous rocks. Theisotopic, trace element and rare earth element geochemical data of a bimodal rock association made up ofanorthosite, gabbro and alkali basalt and olivine-bearing quartz-syenite, rapakivi granite and trachyte as wellas potassic A-type granites and anorogenic granites—— all suggest that there exists an incipient rift in thestudy area. Fractional crystallization of a mixed magma formed by the magma derived from the upper mantleand the magma derived by small degrees of fusion of the lower crust produced anorthosite cumulates. Thewater-deficient granitic magma was differentiated into a subalkaline series. When the fractional crystallizationwas incomplete, rhythmic eruptions took place.     

Petrogenesis of the anorthosite-chromitite association: crystal-chemical and petrological insights from the Rum Layered Suite, NW Scotland

Thirteen Cr-bearing spinels from major horizons of magma replenishment in the open-system Rum Layered Suite have been analysed by X-ray single crystal diffraction and electron microprobe analyses. On the basis of the structural parameters and the chemistry of these spinels the so-called Rum trend, in which Al-content increases at the expense of Cr and Fe³⁺, has been easily recognised. In addition, natural spinels with Fe³⁺ content similar to synthetic spinels on the MgCr₂O₄?CMgFe₂O₄ join have been analysed for the first time. Layers of chromitite, anorthosite and peridotite situated within several cm of one another have yielded different intracrystalline exchange temperatures using an intercrystalline spinel-olivine thermometer. The Rum anorthosite Cr-spinels are interpreted as having crystallised within the cumulus pile following rejuvenation of the crystal mush. Their low Al-content is a function of simultaneous plagioclase crystallisation, reducing the amount of Al³⁺ present for the Cr-spinel. By contrast, Cr-spinels in well-known Archean anorthosites (e.g. Ujaragssuit nunat and Fisken?sset, western Greenland) and Sittampundi (southern India) are very aluminous in composition, attributed to crystallisation of Cr-spinel from high-alumina basalts in lower crustal magma chambers and linked to the control exerted by plagioclase crystallisation on Al content of the melt, in the absence of clinopyroxene crystallisation. The compositional differences between the Rum anorthosite Cr-spinels and the Fisken?sset and Sittampundi Cr-spinels suggest that postcumulus reaction of Cr-spinel and melt to low (800?C900°C) temperatures, as invoked for the Rum crystals, may not have been as important a process in the Archean anorthosites.     

Holocene extension direction along the Main Ethiopian Rift, East Africa

Assessment of the extension direction and its spatial and temporal variations is critical for evaluating a rifting process. The extension direction in the East African Rift System is a matter of debate and the NE–SW-trending Ethiopian portion is one of the most controversial areas, for which several extension directions have been proposed. Field analysis was performed along the axis of the Main Ethiopian Rift (MER), aiming at recognizing the opening direction of the NNE–SSW-trending Holocene extension fractures. The matching of pairs of asperities along the sides of these fractures allowed evaluation of the horizontal displacement and, thus, the extension direction. The collected data reveal a consistent Holocene extension direction, with a mean value of N52°W ± 20°. This NW–SE direction is constant along the MER, despite the ∼15° variations observed in its trend and the NNE–SSW trend of the extension fractures, oblique to the NE–SW trend of the MER.     

A Dynamic Model of Rift Zone Petrogenesis and the Regional Petrology of Iceland

The article describes the petrochemical evolution of oceanicrocks in terms of plate tectonics with special reference toIceland. The compositional variation along the rift zone isrelated to different production rates of mantle-derived olivinetholeiite of invariant composition which is added to the crustfrom below and modified by mixing with anatectic melts in thecrust and concomitant crystal fractionation. The kinematic processes of crustal accretion cause rocks depositedin the rift zone to subside towards higher temperatures wherethey suffer hydration and progressive metamorphism before becominga part of the stable crustal plate. Rocks deposited near therift-centre assume the highest metamorphic grade (greatest depth)while rocks deposited at the rift-margins follow a shallow pathbefore being carried towards lower temperatures in the stableplate. The material transport through stationary metamorphiczones produces the layering of the oceanic crust. As the hydrated rocks cross their solidus isotherm, silicicmagma is formed by incongruent partial melting. The meltingcontinues until rocks crossing the boundary between the amphiboliteand granulite fades are finally dehydrated by the break-downof amphibole. This reaction boundary defines the surface ofthe upper mantle. The segregation and retention of crust-derived magmas withinthe rift zone results in chemical fractionation in the oceaniccrust, for its lower sections are depleted in elements enteringthe early melt fractions, which are silicic and enriched inthe dispersed elements. The last melt-increments from the samesubsiding pile are ne-normative basalts. The rift-zone rocks are shown to be mantle-derived olivine tholeiitemodified by minor amounts of crustal rhyolite and nephelinebasalt, while volcanism outside the rift zone is dominated bythe crust-derived magmas themselves. All mixtures undergo furthermineralogical evolution towards invariant compositions in thebasalt system, resembling the olivine tholeiite, quartz tholeiite,and nepheline basalt of synthetic systems. The dispersed-elementgeochemistry of the oceanic rocks is but slightly modified bycrystal fractionation, and reflects the mixing ratios of theolivine tholeiite and the different crustal magmas. The geochemistry of radiogenic isotopes is controlled by continuousprocesses of crustal fractionation separating mother and daughterelements. The oxygen-isotope geochemistry can be referred tomagma mixing, for rhyolites formed by anatexis in the hydratedcrust are enriched in light oxygen relative to the mantle-derivedmelts.     

Petrogenesis of the Kinsman Intrusive Suite: Peraluminous Granitoids of Western New Hampshire

The Kinsman Intrusive Suite occurs in six major plutons of westernNew Hampshire, covering a total area of 2240 km². It is an Acadian-agesyntectonic gneissic S-type peraluminous granitoid, rangingin composition from quartz diorite to granite. Much of the Kinsmanis characterized by very large (up to 120 mm in maximum dimension)megacrysts of alkali feldspar, but the bulk chemistry of therocks indicates that these cannot be phenocrysts crystallizedfrom initially homogeneous melts. Locally, there is abundant(20 per cent) almandine-rich garnet, and graphite is a commonaccessory.In contrast to the unannealed orthoclase in surroundingmetapelites, the alkali feldspar of the Kinsman has, for themost part, inverted to maximum microcline. The garnets havecore temperatures in the range of 800 to 900 ?C, and are pseudomorphedby, or show reaction rims to, biotite. Plagioclase commonlyshows zoning, some of it oscillatory. These features are magmaticin nature, and argue against the conclusions of previous investigatorsthat the mineralogy and textures of the rock are due to regionalmetamorphism of a previously-crystallized two-mica granitoidwhich has undergone prograde reactions such as:muse + bio +3 qtz 2 Kfs + gar + 2H₂O.The intrusives have also producedrecognizable contact-metamorphic features in the wallrocks andare probably coeval with the dominant M2 Acadian metamorphism.Majorelement analytical data for the Kinsman suite has been examinedby least squares mixing-model and extended Q-mode factor analysis.These calculations, supported by consideration of REE data,suggest that the most likely origin for the Kinsman magmas isby deep-crustal anatexis of slightly calcareous metapelites,and involves a reaction such as:bio + Al₂SiO₅ + qtz + feldspars gar + cord + Kfs + plag + melt.In non-calcareous pelites thisreaction produces a water-undersaturated peraluminous melt attemperatures above 700 ?C, and allows for the early crystallizationor recrystallization of K-feldspar, plagioclase, and garnetin a crystal-liquid mush or migma. Geochemically, garnet + plagioclaseare treated as restite, and a minimum-melt granite as the magmain the Q-mode and mixing-model calculations. The variabilityin chemistry of the Kinsman Intrusive Suite is best explainedon the basis of mixing of leucogranitic anatectic melts withgarnet-plagioclase restitic material and a quartz-feldspar-sillimanite-biotiterock, but only very slightly affected by crystal settling.     

The Role of Pre-existing Structures in the Origin, Propagation and Architecture of Faults in the Main Ethiopian Rift

Four major fault systems oriented N–S to NNE–SSW, NE–SW, E–W and NW–SE are identified from Landsat Thematic Mapper (TM) images and a high resolution digital elevation model (DEM) over the Ethiopian Rift Valley and the surrounding plateaus. Most of these faults are the result of Cenozoic - extensional reactivation of pre-existing basement structures. These faults interacted with each other at different geological times under different geodynamic conditions. The Cenozoic interaction under an extensional tectonic regime is the major cause of the actual volcano-tectonic landscape in Ethiopia. The Wonji Fault Belt (WFB), which comprises the N–S to NNE–SSW striking rift floor faults, displays peculiar propagation patterns mainly due to interaction with the other fault systems and the influence of underlying basement structures. The commonly observed patterns are: curvilinear oblique-slip faults forming lip-horsts, sinusoidal faults, intersecting faults and locally splaying faults at their ends. Fault-related open structures such as tail-cracks, releasing bends and extensional relay zones and fault intersections have served as principal eruption sites for monogenetic Plio-Quaternary volcanoes in the Main Ethiopian Rift (MER).     

Petrogenesis and Tectonic Significance of An Early Paleoproterozoic High-Mg Boninite-Norite-Diorite Suite of Rocks from the Bastar Craton,Central India

正Mafic/ultramafic rocks derived from high siliceous(high-Si)high magnesium(high-Mg)mantle melts are an important feature observed during early Precambrian.They provide much important geological and petrological     

Relationships between Mafic and Peralkaline Silicic Magmatism in Continental Rift Settings: a Petrological, Geochemical and Isotopic Study of the Gedemsa Volcano, Central Ethiopian Rift

Petrological and geochemical data are reported for basalts andsilicic peralkaline rocks from the Quaternary Gedemsa volcano,northern Ethiopian rift, with the aim of discussing the petrogenesisof peralkaline magmas and the significance of the Daly Gap occurringat local and regional scales. Incompatible element vs incompatibleelement diagrams display smooth positive trends; the isotoperatios of the silicic rocks (⁸⁷Sr/⁸⁶Sr = 0·70406–0·70719;¹⁴³Nd/¹⁴⁴Nd = 0·51274–0·51279) encompassthose of the mafic rocks. These data suggest a genetic linkbetween rhyolites and basalts, but are not definitive in establishingwhether silicic rocks are related to basalts through fractionalcrystallization or partial melting. Geochemical modelling ofincompatible vs compatible elements excludes the possibilitythat peralkaline rhyolites are generated by melting of basalticrocks, and indicates a derivation by fractional crystallizationplus moderate assimilation of wall rocks (AFC) starting fromtrachytes; the latter have exceedingly low contents of compatibleelements, which precludes a derivation by basalt melting. ContinuousAFC from basalt to rhyolite, with small rates of crustal assimilation,best explains the geochemical data. This process generated azoned magma chamber whose silicic upper part acted as a densityfilter for mafic magmas and was preferentially tapped; maficmagmas, ponding at the bottom, were erupted only during post-calderastages, intensively mingled with silicic melts. The large numberof caldera depressions found in the northern Ethiopian riftand their coincidence with zones of positive gravity anomaliessuggest the occurrence of numerous magma chambers where evolutionaryprocesses generated silicic peralkaline melts starting frommafic parental magmas. This suggests that the petrological andvolcanological model proposed for Gedemsa may have regionalsignificance, thus furnishing an explanation for the large-volumeperalkaline ignimbrites in the Ethiopian rift. KEY WORDS: peralkaline rhyolites; geochemistry; Daly Gap; Gedemsa volcano; Ethiopian rift     

A high-pressure experimental study on the evolution of the silicic magmatism of the Main Ethiopian Rift

In this contribution we report the results of an experimental study that investigated equilibrium and fractional crystallization of hydrous, transitional alkaline basalt at low oxygen fugacity, under lower to middle crustal conditions to constrain the generation of subaluminous and peralkaline differentiation products that typically occur in rift systems. The experiments reveal that liquids produced by equilibrium crystallization in the range 0.7–1 GPa cannot cross the subaluminous/peralkaline compositional divide. In contrast, fractional crystallization experiments under isobaric and polybaric conditions approach closer the naturally observed trend from subaluminous to evolved peralkaline products suggesting that polybaric differentiation starting at elevated pressures can indeed lead to the transition from subaluminous to peralkaline derivative liquids. The presence of water in the parental magmas of silicic derivative products is of prime importance for the fractionation equilibria as well as for the mobility of such magmas toward shallow crustal levels.

We suggest that peralkaline magmas in rift environments are indicative for differentiation under relatively low oxygen fugacity conditions in an extensional environment characterized by a high degree of crustal fracturing that allows rapid upward migration of mafic parental magmas and formation of shallow magma reservoirs. Crystallization–differentiation of parental, hydrous transitional alkaline basalt in such reservoirs is controlled by low pressure phase equilibria that typically evolve through early saturation of anorthite-rich plagioclase and suppressed amphibole crystallization resulting in ‘low-alumina’, peralkaline derivative liquids.