Gondwana Research

Nd isotopic data and U/Pb zircon ages suggest that only 7–13% of continental crust was formed between 1.35 to 0.9 Ga. Calculated crustal production rates during this time fall within the 1.1 km3/y average production rate of continental crust. This distribution of juvenile continental crust supports the existence of only two major superplume events at 2.7 and 1.9 Ga, and one or two minor events in the Phanerozoic at about 300 and 110 Ma. The absence of a 1.35-0.9 Ga juvenile crust peak may be related to supercontinent history. Results of this study confirm that although both the supercontinent and superplume cycles are episodic, each cycle can operate independently of the other.

Several mafic dyke swarms of similar composition and age (tholeiite- ca.1.0 Ga) occur on both sides of the Atlantic Ocean in eastern South America and western Africa. When assembled to their pre-drift position in the Mesozoic, the Brazilian coastal dyke province of Bahia, and the African dykes in Cameroun (Ebolowa suite) and Congo (Comba and Sembe-Ouesso provinces) define a giant radiating pattern (1200 km × 800 km) similar to other dyke swarms elsewhere associated with large-scale continental rifting. Magma flow indicators of the Brazilian dykes and branching propagation styles of their African counterparts indicate that the dyke conduits were fed with magmas diverging from a source beneath the long axis of the Meso-Neoproterozoic West-Congolian Basin in Africa. There, MORB-like metabasalts have been described in the La Bikossi Group of the Mayombian Supergroup. Whether the rifting event and intrusion of dyke swarms were triggered or not by a mantle plume beneath part of the Rodinia subcontinental lithosphere remain to be confirmed.

The 1.3– 1.2 Ga fan- shaped Mackenzie dyke swarm and other similar- aged dyke swarms in the Canadian Shield constitute the subswarms of a Late Mesoproterozoic giant radiating swarm. The Late Mesoproterozoic mafic dyke swarms in Australia and East Antarctica might constitute additional subswarms of the giant radiating dyke swarm. A possible Late Mesoproterozoic mantle plume is placed at the focal area of the giant radiating dyke swarm between North America and the landmass (West Australia– East Antarctica). This mantle plume triggered the continuous extension at ca. 1.3–1.2 Ga, which extended into much of Columbia and led to the final fragmentation of the Columbia supercontinent.

New data obtained over the past few years indicate that, by the end of the Archaean, the North China Craton consisted of two major crustal entities, identified as the eastern and western blocks. The formation of island and magmatic arcs along the western margin of the eastern block record the latest Archaean events so far identified. The location of the western block at this time is unknown, however, the occurrence of Palaeoproterozoic passive margin sediments, similar in age to those occurring in the eastern block, indicates essentially similar conditions and possible separation by an ocean of unknown dimensions. Re-initiation of east-dipping subduction beneath the western margin of the eastern block resulted in closure of this Palaeoproterozoic ocean, bringing the western block into close proximity. Continued subduction resulted in a major continental-continental collision and the western margin of the eastern block was again reactivated, leading to extensive thrusting, high-pressure metamorphism and the generation of crustal melts. This occurred between 1.88 and 1.79 Ga ago, resulting in the formation of the Trans-North China Orogen and final amalgamation of the North China Craton. Although information is sparse to non-existent, we have reviewed the possible location of the North China Craton at the time of collision and note a similarity in terms of age, lithology and distribution of major crustal components between the North China Craton and Baltica. Although speculative, we present this model in the hope that it will stimulate debate and lead to further research in these areas.

A newly identified northwest–southeast oriented, deeply-rooted, steep to vertical, large-scale structural system within the Proterozoic Curnamona Province, Australia, which we term the “Benagerie Shear Zone”, is imaged in regional magnetic and gravity datasets. In this study, we use a combination of field analysis and quantitative geophysical methods, to establish a 1100 Myr history of activity along the Benagerie Shear Zone during which the location of younger geological structures are influenced by the pre-existing shear zone. This deformational system is interpreted to have 1) aided ascent and emplacement of the ca. 1600 Ma Ninnerie (magmatic) Supersuite; 2) controlled the loci of nucleation of normal faults during rifting and continental breakup at ca. 800 Ma; and 3) influenced the development of fold structures as well as acting as a plane co-linear to the rotation axis of pre-existing normal faults such that they were steepened and reactivated as strike slip structures during the ca. 500 Ma Delamerian Orogeny. We interpret that the Benagerie Shear Zone has not undergone uni-directional propagation during its evolution but rather through reactivation was a primary influence on controlling the nucleation of Neoproterozoic rift faults, thereby playing a major role in accommodating strain over a significant period of the evolution of the Curnamona Province. This study demonstrates that crustal-scale shear zones can evolve over hundreds of millions of years, have strike-lengths of hundreds to thousands of kilometers, and have vastly different surface expressions along strike.

The dolomitic carbonates from the Jhamarkotra Formation of the Palaeoproterozoic Aravalli Supergroup are characterized by widely variable carbon isotope ratios (δ13C) ranging from near zero to as high as + 12‰ V-PDB. The estimated maximum age (ca. 2150 Ma) of the Aravalli carbonates help bracketing these with the coeval carbonate bodies of the world that show high positive carbon isotope values. The intriguing existence of normal marine δ13C values in some pockets suggest influence of local scale depositional conditions prevailing in different sub-basins. Amongst these two sub-basins which showed high δ13C values, a hypersaline evaporative condition is considered responsible for the necessary enrichment in one, while methanogenesis (possibly in conjunction with sulphate reduction processes) might have caused such enrichment in the other. From the empirical association of the sub-basins with profuse stromatolitic phosphorite we infer that the depositional setting that favoured cyanobacterial growth (leading to formation of stromatolites) prevented growth of methanogenetic archaea in such anoxic environments. Our study therefore highlights the fact that the early Palaeoproterozoic 13C excursion in the Aravalli Supergroup is not essentially a time-specific event but is greatly dependent on the variation in the depositional palaeoenvironment prevailing in different sub-basins.

A global review of the stratigraphical and geographical distribution of Tyloplecta reveals that the genus ranges in age from Kungurian to Changhsingian (Middle to Late Permian). Tyloplecta first evolved in South China in the Kungurian (late Early Permian). The genus went through its first diversification in the Guadalupian, suffered a major extinction at the end of the Guadalupian, and re-diversified in the Wuchiapingian. T. yangtzeensis persisted into the Changhsingian as the only survivor of the genus involved in the end-Permian mass extinction. Palaeogeographically, South China is not only the centre of origin for the genus but also an area of diversification and evolution. In addition to South China, Tyloplecta has also been recorded from the Far East Russia, Japan, central Thailand, Laos, Cambodia, Qiangtang Terrane of Tibet, Salt Range, Iran, Armenia, Hungary, Yugoslavia, and Slovenia. This geographic spread suggests that Tyloplecta was primarily restricted to the Palaeotethys and is indicative of warm-water palaeoequatorial conditions. Its presence in some of the northeast Asian terranes (e.g., parts of Japan and Far East Russia) and in the Salt Range (Pakistan) and central and north Iran (part of the Cimmerian microcontinents) demonstrate that the genus invaded the middle palaeolatitudinal regions in both hemispheres during the late Middle Permian in response to increased shallow marine biotic communications between Cathaysia in the eastern Palaeotethys and southern Angaraland, and between Cathaysia and Peri-Gondwanaland. The invasion of Tyloplecta (and some other taxa) into the southern shore waters of Angaraland may be explained by assuming ocean surface current connections and close palaeogeographical proximities between the South China, Sino-Korea and Bureya blocks. In comparison, the invasion of Tyloplecta into the Peri-Gondwanaland region is more likely a result of reduced palaeogeographical distance between South China and Peri-Gondwanaland and the appearance of the Cimmerian microcontinents as migratory stepping stones.

Recent and new faunal data from the Cambrian to Silurian rocks of the Precordillera, Famatina and Northwest Argentina basins are used to discriminate between different paleogeographic models, and especially to establish to what extent they are compatible with a previous conclusion that the Precordillera is a Laurentian-derived microcontinent. There is no paleontological evidence to support a para-autochthonous Gondwanan origin of the Precordillera. The strong differences in the Cambrian trilobite faunas and lithologic successions preclude a common origin of the Precordillera terrane, eastern Antarctica and South Africa. Recent discoveries of brachiopods and organisms of the Phylum Agmata strengthened Laurentian affinities during the Cambrian. The latest Cambrian-early Ordovician faunas that inhabited the autochthonous Northwest Argentina basin, including the western Puna volcaniclastic successions, are mostly peri-Gondwanan. The early Ordovician brachiopods, ostracods and trilobites display mixed Laurentian, Baltic and Avalonian biogeographical links supporting a drifting of the Precordillera across the Iapetus Ocean. Increasing Gondwanan elements during the Llanvirn, along with varied geological evidence, indicate that the first stages of collision may have begun at that time, involving a major change in the plate kinematics. The distribution of facies and faunas, basin development, and timing of deformation are interpreted as resulting from a north to south diachronous closing of the remnant basin during the last phases of convergence and oblique collision of the Precordillera terrane with the Gondwana margin. The high level of endemism of Caradoc faunas may be a consequence of the rearrangement and partial isolation of sedimentary areas during the strike-slip movement of the colliding Precordillera plate with respect to the Gondwana margin. Suggested relationships between facies distribution, geographic barriers and faunal migrations before and during the collision are depicted in a series of schematic reconstructions at five time slices from late Cambrian to Silurian.

The tsunami run-up, inundation and damage pattern observed along the coast of Tamilnadu (India) during the deadliest Indian Ocean tsunami of December 26, 2004 is documented in this paper. The tsunami caused severe damage and claimed many victims in the coastal areas of eleven countries, bordering the Indian Ocean. Along the coast of Indian mainland, the damage was caused by the tsunami only. Largest tsunami run-up and inundation was observed along the coast of Nagapattinam district and was about 10–12 m and 3.0 km, respectively. The measured inundation data were strongly scattered in direct relationship to the morphology of the seashore and the tsunami run-up. Lowest tsunami run-up and inundation was measured along the coast of Thanjavur, Puddukkotai and Ramnathpuram districts of Tamilnadu in the Palk Strait. The presence of shadow of Sri Lanka, the interferences of direct/receded waves with the reflected waves from Sri Lanka and Maldive Islands and variation in the width of continental shelf were the main cause of large variation in tsunami run-up along the coast of Tamilnadu.

The Mesoarchean (ca. 3075 Ma) Ivisaartoq greenstone belt in southern West Greenland includes variably deformed and metamorphosed pillow basalts, ultramafic flows (picrites), serpentinized ultramafic rocks, gabbros, sulphide-rich siliceous layers, and minor siliciclastic sedimentary rocks. Primary magmatic features such as concentric cooling-cracks and drainage cavities in pillows, volcanic breccia, ocelli interpreted as liquid immiscibility textures in pillows and gabbros, magmatic layering in gabbros, and clinopyroxene cumulates in ultramafic flows are well preserved in low-strain domains. The belt underwent at least two stages of calc-silicate metasomatic alteration and polyphase deformation between 2963 and 3075 Ma. The stage I metasomatic assemblage is composed predominantly of epidote (now mostly diopside) + quartz + plagioclase ± hornblende ± scapolite, and occurs mainly in pillow cores, pillow interstitials, and along pillow basalt-gabbro contacts. The origin of this metasomatic assemblage is attributed to seafloor hydrothermal alteration. On the basis of the common presence of epidote inclusions in diopside and the local occurrence of epidote-rich aggregates, the stage I metasomatic assemblage is interpreted as relict epidosite. The stage II metasomatic assemblage occurs as concordant discontinuous layered calc-silicate bodies to discordant calc-silicate veins commonly associated with shear zones. The stage II metasomatic assemblage consists mainly of diopside + garnet + amphibole + plagioclase + quartz ± vesuvianite ± scapolite ± epidote ± titanite ± calcite ± scheelite. Given that the second stage of metasomatism is closely associated with shear zones and replaced rocks with an early metamorphic fabric, its origin is attributed to regional dynamothermal metamorphism. The least altered pillow basalts, picrites, gabbros, and diorites are characterized by LREE-enriched, near-flat HREE, and HFSE (especially Nb)-depleted trace element patterns, indicating a subduction zone geochemical signature. Ultramafic pillows and cumulates display large positive initial εNd values of + 1.3 to + 5.0, consistent with a strongly depleted mantle source. Given the geological similarities between the Ivisaartoq greenstone belt and Phanerozoic forearc ophiolites, we suggest that the Ivisaartoq greenstone belt represents Mesoarchean supra-subduction zone oceanic crust.

The chemical Th–U total Pb isochron method (CHIME) of dating was carried out on accessory minerals in samples from the Okcheon metamorphic belt in Korea. Dated minerals include xenotime and monazite with overgrown mantles in a granitic gneiss clast from the Hwanggangri Formation, metamorphic allanite in garnet-bearing muscovite–chlorite schist of the Munjuri Formation, and polycrase and monazite in post-tectonic granite from the Hwanggangri area. Overgrowth of mantles took place at 369 ± 10 Ma on c. 1750 Ma cores of xenotime and monazite in the granitic gneiss. Allanite, occurring in textural equilibrium with peak metamorphic minerals, yields a CHIME age of 246 ± 15 Ma that is discriminably older than the polycrase (170 ± 6 Ma) and monazite (170 ± 3 Ma) ages of the post-tectonic granite. These chronological data suggest that some of the metasedimentary rocks in the belt formed through a single stage of metamorphism at c. 250 Ma from post-370 Ma sediments. Late Permian age signatures have also been reported from the Precambrian Gyeonggi and Yeongnam massifs that border the Okcheon metamorphic belt, and indicate that parts of the basement massifs and the metamorphic belt were affected by the same regional metamorphic event.

The time frame of the three main geological events in the Neoproterozoic Cambaí Complex, juvenile São Gabriel belt in the southern Brazilian Shield is established by integrating field mapping, back-scattered electron imaging and sensitive high-resolution ion microprobe (SHRIMP II) U–Pb dating of 96 zircon crystals from nine granitic and metasedimentary rock samples. The three events are: (1) voluminous flat-lying paragneisses (Cambaizinho Complex) and orthogneisses (Vila Nova gneisses) between 735 and 718 Ma, (2) tonalite–trondhjemite association (Lagoa da Meia-Lua Suite) between 710 and 690 Ma, and (3) late granodiorite intrusions (Sanga do Jobim Suite) at 680 Ma. An additional older volcanic event (Campestre Formation) was dated at 753 Ma. These results are most significant for the reconstruction of West Gondwana.

During the late stages of the Brasiliano orogenic cycle (Lower Cambrian), the Camaquã Basin was gradually filled by the alkaline-trending, bimodal volcanic rocks of the Acampamento Velho Alloformation. This volcanic package consists of two facies associations: the lower one composed of andesites and basaltic andesites, and the upper one of rhyolitic rocks. The rhyolitic association comprises alternating pyroclastic rocks (lapilli, tuffs and welded tuffs) in the middle section and flows at the top. Geochemical evidence, especially trace elements and REE, confirmed the stratigraphic succession proposed herein for the volcanic rocks, as well as their co-genetic relationships and the fractional crystallization of the felsic sequence. The Acampamento Velho Formation seems to have been generated in an extensional regime preceding the collision of the Rio de la Plata and Kalahari continental plates. This extensional regime probably occurred during subduction of the Adamastor oceanic plate beneath the Rio de la Plata plate in a retroarc setting.

U-Pb isotopic analyses of zircon from the lowest structural units of the Acatlán Complex of southern Mexico indicate that Paleozoic tectonothermal events are overprinted by mid-Jurassic (175±3 to 171±1 Ma), low pressure migmatization (5–6 kb), polyphase deformation, and intrusion of felsic and mafic magmas. Ensuing rapid cooling recorded by 40Ar/39Ar muscovite, biotite and K-feldspar ages is estimated to have taken place at 21±3°C/my at exhumation rates of 0.6 mm/yr. Such rapid exhumation requires a combination of erosion and tectonic unroofing that is recorded by top-to-the-west kinematic data. Synchronous tectonic unroofing is also recorded 100 km to the east in the adjacent Oaxaca terrane, where top-to-the-north, extensional shear zones occur in Paleozoic strata.This pattern of extension suggests tectonic unroofing in response to domal uplift (radius >100 km) like that associated with core complexes, slab windows, and hotspots. Most tectonic analyses for the Jurassic place the Acatlán Complex in the forearc region of an arc in Colombia lying 600–800 km inboard of the subduction zone, presumably in response to flat-slab subduction. Modern analogues suggest that flat-slab subduction reflects subduction of young buoyant oceanic lithosphere adjacent to either a mid-oceanic ridge, or a plume. Since core complexes are typical of arc-backarc regions, and slab windows generally produce metamorphic belts, the forearc setting and associated domal uplift suggest a plume to be the most likely cause of this Jurassic tectonothermal pulse in southern Mexico. This plume activity is synchronous with the opening of the Gulf of Mexico during the breakup of Pangea, to which it may have contributed.

Field relationships, geochemistry and U–Pb zircon and Sm–Nd whole-rock geochronology are used to constrain the genesis of Palaeoproterozoic massive amphibolites and orthogneisses of the Algodões Sequence, Central Ceará Domain, Northeastern Brazil. The 2236 ± 55 Ma-old Algodões amphibolites show trace element geochemical signature and positive εNd(t) values similar to Phanaerozoic oceanic plateau basalts and less often to back-arc basalts. The amphibolites are intruded by tonalitic to quartz-dioritic gneisses of the Cipó unit which were dated at 2160–2170 Ma and 2130 Ma. Neodimium isotope data and whole-rock geochemistry for these gneisses show they are akin to juvenile arc plutonics of the adakitic suite. The likely occurrence of Palaeoproterozoic plateau basalts and arc plutonics in this part of northeastearn Brazil render comparisons with similar terranes elsewhere in South America and Africa, which in turn indicate a significant contribution of accretionary orogens in the early assembly of the Columbia Supercontinent.

The K-bentonite, black shale and flysch successions at the Ordovician–Silurian transition in South China have been the subject of comprehensive investigations relative to the probable accretion of the Yangtze Block and the questionable Cathaysia Block. First, the geochemical analyses of K-bentonites show that the parent magma originated in syn-collisional, volcanic-arc and within-plate tectonic settings, which produced mainly intermediate-to-felsic series magmas, associated with continuous collision and subduction of paleo-continental blocks/arcs. Further, the regional distribution of K-bentonite thickness indicates that voluminous explosive volcanism was located in the present southeastern shoreline provinces of China. Secondly, northwestwardly migrating, Ordovician–Silurian, transitional flysch successions, and the accompanying diachronous K-bentonite-bearing black-shale interval, as well as the related, overlying, shallowing-upward succession at the interior of the Yangtze Block, developed as an unconformity-bound sequence that mirrors foreland-basin tectophase cycles in the Appalachian basin. The above features suggest that the sequence accumulated in a similar foreland basin, which formed in response to adjacent deformational loading in a northwesterly migrating orogen located to the southeast. Geochemical and paleocurrent data from the turbiditic flyschoid sandstones also support these depositional settings. Accordingly, it seems that all criteria strongly support the presence of an Ordovician–Silurian, subduction-related orogen resulting from collision with a block to the southeast that must have been the original “Cathaysia Block” of Grabau and later workers. The K-bentonite, black-shale and flysch successions can be regarded as distal, foreland responses to the continuous northwestward collision and accretion of the Cathaysia Block to the Yangtze Block. Hence, we prefer to suggest that the suture zone with the sensu stricto Cathaysia Block probably developed along previously identified late Early Paleozoic suture relicts in the shoreline provinces of southeast China. On the other hand, although accretion of fragments with Cathaysian affinities to the Yangtze Block may have begun as early as Middle to Late Proterozoic time, the Ordovician–Silurian orogeny described above probably reflects the final phase of accretion between the two blocks. Moreover, when combined with similar peri-Iapetan orogenic events in other areas during the same period, this accretion event may have been part of a major stage of global tectonic reconstruction in the evolution of Gondwana.

New age, petrochemical and structural data indicate that the Banda Terrane is a remnant of a Jurassic to Eocene arc–trench system that formed the eastern part of the Great Indonesian arc. The arc system rifted apart during Eocene to Miocene supra-subduction zone sea floor spreading, which dispersed ridges of Banda Terrane embedded in young oceanic crust as far south as Sumba and Timor. In Timor the Banda Terrane is well exposed as high-level thrust sheets that were detached from the edge of the Banda Sea upper plate and uplifted by collision with the passive margin of NW Australia. The thrust sheets contain a distinctive assemblage of medium grade metamorphic rocks overlain by Cretaceous to Miocene forearc basin deposits. New U/Pb age data presented here indicate igneous zircons are less than 162 Ma with a cluster of ages at 83 Ma and 35 Ma. 40Ar/39Ar plateau ages of various mineral phases from metamorphic units all cluster at between 32–38 Ma. These data yield a cooling curve that shows exhumation from around 550 °C to the surface between 36–28 Ma. After this time there is no evidence of metamorphism of the Banda Terrane, including its accretion to the edge of the Australian continental margin during the Pliocene. These data link the Banda Terrane to similar rocks and events documented throughout the eastern edge of the Sunda Shelf and the Banda Sea floor.

Linear belts of Gondwana basins developed in the Indian continent since Late Palaeozoic along favoured sites of Precambrian weak zones like cratonic sutures and reactivated mobile belts. The Tibetan and Sibumasu - West Yunnan continental blocks, that were located adjacent to proto-Himalayan part of the Indian continent, rifted and drifted from the northern margin of the East Gondwanic Indo-Australian continent, during Late Palaeozoic, when the said northern margin was under glacial or cool climatic condition and rift-drift tectonic setting. The Indo-Burma-Andaman (IBA), Sikule, Lolotoi blocks were also rifted and drifted from the same northern margin during Late Jurassic. This was followed by the break-up of the Australia-India-Madagascar continental block during the Cretaceous. The activity was associated with hot spot related volcanism and opening up of the Indian Ocean. The Late Cretaceous and Tertiary phases of opening of the Arabian Sea succeeded the Early Cretaceous phase of opening of the Bay of Bengal, part of the Indian Ocean. The Palaeo- and Neo-Tethyan sutures in Tibet, Yunnan, Laos, Thailand and Vietnam reveal the complex opening and closing history of the Tethys. The IBA block rotated clockwise from its initial E-W orientation because of 90°E and adjacent dextral transcurrent fault movements caused due to faster northward movement of the Indian plate relative to that of Australia. The India-Tibet terminal collision during Early-Middle Eocene initiated Himalayan orogenesis and contemporaneously there was foreland basin development that was accompanied with sporadic but laterally extensive continental-flood-basalt (CFB) type and related volcanism. The Paleogene rocks of the Himalayan foreland basin are involved in tectonism and are mostly concealed under older rocks.

Structural, metamorphic and isotopic data obtained from the Nogoli Metamorphic Complex of western Sierra de San Luis indicate that the Early Paleozoic Famatinian Orogeny overprinted an already structured and metamorphosed older basement. The older geological features are relict NW trending fabric associated with high-grade (amphibolite facies) regional metamorphism preserved within thin strips of schists and paragneisses and in the core of mafic to ultramafic lenses. Arc magmatism, medium P (Barrovian type)/high T (amphibolite to granulite facies) regional metamorphism and penetrative NNE to NE trending foliation are related to the building of the Famatinian orogenic belt. The P-T conditions of the Famatinian prograde metamorphism reached a pressure peak of ca. 8 kb, with a thermal peak from -750°C up to -820°C. U-Pb conventional and chemical dating and Ar-Ar plateau ages constrain the peak of the main orogenic phase related to the Famatinian belt to 470–457 Ma (Early to Mid-Ordovician). Greenschist facies retrograde metamorphism closely associated with shear zones and secondary Ar-Ar plateau and Sm-Nd ages suggest that a late to post-orogenic phase of the Famatinian belt was active at least since -445 Ma. This phase continued during the Silurian to Late Devonian times through multiple reactivation of early shear zones. The Famatinian Orogeny reset a previous thermal history and therefore, the timing of the relict fabric could not be constrained conclusively with radiometric dates. Despite this difficulty, a range of 520 to 490 Ma suggests some inheritance from Pampean events registered by the older NW-SE fabric. The Early to Mid-Ordovician regional metamorphism and ductile deformation of the western Sierra de San Luis is interpreted as the orogenic effects of the collision of the allochthonous Cuyania terrane with the autochthonous proto-Pacific margin of Gondwana during the Famatinian Orogeny.

Numerous bedded manganese and manganese–iron ore deposits occur in Shikoku, Southwest Japan. The mineralogy, wall rock characteristics and fossil age data of the deposits are synthesized in this paper. Their mineral assemblages correlate well with the progressive metamorphic grade of the accretionary complexes in Shikoku. Based on the plate tectonic framework of subduction–accretion process, the deposits are considered to have formed by metamorphism of Mn–Fe-nodule-bearing siliceous sediments on the ocean floor.

We have reinvestigated the mid-Cretaceous plume pulse in relation to paleo-oceanic plateaus from accretionary prisms in the circum-Pacific region, and we have correlated the Pacific superplume activity with catastrophic environmental changes since the Neoproterozoic. The Paleo-oceanic plateaus are dated at 75–150 Ma; they were generated in the Pacific superplume region and are preserved in accretionary prisms. The volcanic edifice composed of both modern and paleo-oceanic plateaus is up to 10.7 × 106 km2 in area and 19.1 × 107 km3 in volume. The degassing rate of CO2 (0.82 − 1.1 × 1018 mol/m.y.) suggests a significant impact on Cretaceous global warming. The synchronous occurrence of paleo-oceanic plateaus in accretionary complexes indicates that Pacific superplume pulse activities roughly coincided at the Permo-Triassic boundary and the Vendian–Cambrian boundary interval. The CO2 expelled by the Pacific superplume probably contributed to environmental catastrophes. The initiation of the Pacific superplume contributed to the snowball Earth event near the Vendian–Cambrian boundary; this was one of the most dramatic events in Earth's history. The scale of the Pacific superplume activity roughly corresponds to the scale of drastic environmental change.

The kinematics of the crustal-scale Achankovil Shear Zone in the Southern Granulite Terrane, India is often debated. This shear zone is considered to mark the terrane boundary between Madurai Granulite Block towards the north and Kerala Khondalite Belt towards the south. Closely spaced 4500 new gravity measurements in the region reveal the detail crustal fabric across the Achankovil Shear Zone. A subdued Bouguer gravity anomaly along a N–S profile that trends approximately orthogonal to the Achankovil Shear Zone signifies that this part of the crust is devoid of any crustal-scale density discontinuity. Furthermore, horizontal-gradient, analytical signal and second vertical derivative analyses of the gravity data suggest that the density inhomogeneity across the Achankovil Shear Zone is relatively shallow in origin thereby refuting it as a terrane boundary. A prolonged zone of positive gravity gradient, as much as 125 km wide towards south, is due to the effect of the continental margin. 2 1/2D gravity modelling along the profile, constrained from seismic results, reveals a three layer crustal configuration with the depth to Moho varying from 41 km beneath the Vattalkundu to about 34 km beneath the Kanyakumari and then attaining to about 32 km beneath the continental shelf region. The 22 km thick quasi-continental crust in the adjoining Indian Ocean indicates a transitional crust in the Gulf of Mannar region. Analysis of new gravity data thus supports the idea that the Achankovil Shear Zone is an intracratonic litho-tectonic feature and the two provinces across it are related by a continual progression in single metamorphic terrain rather than an ancient geo-suture.

The rocks of Marwar Supergroup in the trans-Aravalli sector in western India are presumed to span the time interval between Neoproterozoic and early Cambrian. This, predominantly unfossiliferous, marine sedimentary sequence is characterized by a lower arenaceous facies (Jodhpur Group), middle carbonate facies (Bilara Group) and upper argillaceous— arenaceous facies (Nagaur Group) rocks. The sedimentation has been essentially in a shallow basin, described either as the fore-land slope of the rising Aravalli mountains or a sag-basin which developed and evolved due to subsidence of the updomed crust during Neoproterozoic Malani magmatism that failed to open rifts. The carbon isotopic profile for the Bilara Group carbonate rocks in the lower part shows marked oscillations and broadly negative δ13C character with negative anomalies as low as <−4.3‰PDB, observed near the base of Dhanapa Formation (lower unit) and <−6.5‰PDB in the overlying Gotan Formation (middle unit). The upper part of the profile shows a gradual positive shift. The carbon isotopic signatures of the Bilara Group rocks can be correlated with the end-Neoproterozoic — early Cambrian (Vendian — Tommotian) carbon isotopic evolution curve. Extremely low δ13C values indicate the glaciation related cold climatic postulates of the end-Neoproterozoic, followed by the warmer climatic conditions as indicated by the positive shift. The carbon isotopic data for Gotan Formation carbonates, at variance with the globally observed δ13C trends for early Tertiary, do not support the recently proposed Tertiary age for the Bilara Group.

The passive continental margins of India and conjugate Antarctica (30°–80° E) have been reconstructed quantitatively by eliminating the intervening ocean floor. Recently acquired seismic and gravity data from the margins define the boundary between continental and oceanic crust (COB) and indicate the thickness of the extended (rifted) continental crust. The COBs are restored to their pre-rift position by eliminating pre-drift extension and are fitted together. This fully palinspastic reconstruction reveals the Lambert and Mahanadi Rifts aligned in a Permian–Triassic rift system, the Napier salient draped by the bight of India, and strong connections within each of four sectors: (1) Southern Granulite Terrain–Sri Lanka–Lützow–Holm Terrane–Rayner Complex; (2) Napier–southern Eastern Ghats Mobile Belt–Dharwar Craton; (3) Kemp Land–MacRobertson Land–northern Eastern Ghats Mobile Belt–Bastar Craton; and (4) Prydz–Rauer–Vestfold Hills–Singhbhum Province. Major events registered are (a) a 0.9–1.3 Ga (Grenville) convergence that formed the Eastern Ghats–Rayner Mobile Belt in Rodinia, and (b) 0.50–0.60 Ga (Pan-Gondwanaland) events that accreted the Southern Granulite Terrain and Sri Lanka to the Antarctic–Indian region. Gondwanaland broke up at 0.14 Ga along the grain of the Eastern Ghats–Rayner Mobile Belt and across the composite Archean Dharwar–Napier Craton and the long axis of the Permian–Triassic rift system.

Adakitic intrusive rocks of ∼ 430–450 Ma were discovered in the North Qilian orogenic belt, the western section of the Central Orogenic System (COS) in China. These adakitic rocks were lower crust melts rather than slab melts as indicated by their crustal Ce/Pb, Nb/U, Ti/Eu, and Nd/Sm ratios and radiogenically enriched (87Sr/86Sr)i of 0.7053–0.7066 and εNd(t) of − 0.9 to − 1.7. While they are all characterized by low Yb (< 1.1 ppm) and Y (< 11.5 ppm) abundances with high Sr/Y (> 65) and (La/Yb)N (> 13.7) ratios, these adakitic rocks are classified into the low-MgO–Ni–Cr and high-MgO–Ni–Cr groups. The low-MgO samples were derived from partial melting of thickened lower crust, whereas the high-MgO samples were melts from delaminated lower crust, which subsequently interacted with mantle peridotite upon ascent. Adakitic rocks from the adjacent North Qinling orogenic belt also originated from thickened lower crust at ∼ 430 Ma. In addition, the North Qilian and North Qinling orogenic belts both consist of lithological assemblages varying from subduction-accretionary complexes at south to central arc assemblages, which include adakitic rocks, then to backarc phases at north. Such a sequence reflects northward subduction of the Qilian and Qinling oceans. In these two orogenic belts, the occurrence of adakitic rocks of common origin and ages together with the similarities in tectonic configurations and lithological assemblages are considered to be the evidence for the continuity between eastern Qilian and western Qinling, forming a > 1000 km Early Paleozoic orogenic belt. In such a tectonic configuration, the Qilian and Qinling oceans that subducted from south possibly represent parts of the large “Proto-Tethyan Ocean”. This inference is supported by the coexistence of Early Paleozoic coral and trilobite specimens from Asia, America and Australia in the North Qilian orogenic belt. Post-400 Ma volcanic rocks occur in the North Qinling orogenic belt but are absent in the North Qilian orogenic belt, indicating that these two orogenic belts underwent distinct evolution history after the closure of the Proto-Tethyan Ocean (∼ 420 Ma).

The Gangdese batholith emplaced during the time span of Cretaceous to Neogene in the southern Lhasa terrane of Tibet has been considered as a major constituent of an Andean-type convergent margin derived from the northward subduction of the Neo-Tethyan oceanic lithosphere under Asia. Whereas previous studies assigned the Gangdese granitoids to be comprised predominantly of calc-alkaline rocks, here we report a suite of charnockites from the eastern part of the belt and characterize their petrology, geochemistry and age. These rocks possess an assemblage of andesine, enstatite, diopside, calcic amphibole, Ti-rich biotite, quartz and minor K-feldspar. Geochemically, they are characterized by intermediate SiO2 (54–63 wt.%), relatively high Al2O3 (15.9–18.9 wt.%), REE (55.7–89.4 ppm) and Sr (419.6–619.4 ppm), and low Y (11.3–17.2 ppm) and Yb (1.2–1.8 ppm) concentrations. The rocks display geochemical affinities similar to those of adakites derived from the partial melting of a subducted slab, and also can be compared to magnesian charnockites formed within a continental magmatic arc. The crystallization conditions of the charnockites were estimated at 900 °C and 1.0 GPa. LA-ICP-MS zircon U–Pb analyses of eleven samples yield consistent 206Pb/238U weighted mean ages of 86 to 90 Ma, indicating that the charnockites were emplaced in the Late Cretaceous. Considering the coeval calc-alkaline magmatism and high-temperature granulite-facies metamorphism, we propose that such high-temperature and low-H2O activity charnockites were derived through Neo-Tethyan mid-ocean ridge subduction before the collision of India with the Asian continent.

A new Late Cretaceous Mesoeucrocodylian from the Adamantina Formation (Bauru Basin), São Paulo State, Brazil is described. The main features of this new species are the short, high oreinirostral rostrum, the large laterally positioned orbital notches and external nares in the anteriormost portion of the rostrum. The mandible is robust and concave-shaped in relation to the skull. The dentition is highly specialized, with two prominent incisiform teeth, a hypertrophied caniniform, and seven molariform teeth. The molariform teeth are ornamented with denticles in their lingual surface and are smooth on their labial surface. The molariforms are elliptical in cross-section, presenting the largest axis in the labial-lingual direction. Such dental characteristics are unique among the terrestrial crocodylomorphs of the Gondwana.

A determination of the seismic structure of the crust and uppermost mantle of East Antarctica, in the region of Casey station, Wilkes Land and Dumont DUrville station, Terre Adelie is presented. High-fidelity waveforms from teleseismic earthquakes recorded at stations CASY and DRV (1996-2001) are used to calculate the seismic receiver function, the signal produced as energy passes through layers in the seismic velocity structure under the receiving station. The receiver functions are stacked to improve the signal-to-noise ratio and then modelled using an inverse algorithm to find the structure that best fits the observed waveform at each station. Inferences are made regarding the tectonic structure, in particular, the crustal thickness and character of the seismic Moho.The crustal thickness under Casey Station is found to be 30 km (+/- 2 km) with a fairly sharp Moho, considerably less than Dumont D'Urville Station, where the crustal thickness is 42 km, and there IS a significant low velocity region the deep crust. The structure of the Wilkes Land lithosphere is comparable to that of the Albany-Fraser Orogen, Western Australia, part of its conjugate margin. This places a new constraint on the relative position of East Antarctica and Australia in the reconstruction of Gondwana, and earlier, supercontinents. A recent reinterpretation of Antarctic geology proposes tectonic province boundaries trending perpendicular to the coast with counterparts in southern Australia. Seismic techniques, determining structure beneath regions with no surface exposure, are vital tools in testing such tectonic hypotheses, towards the reconstruction of Gondwana to full lithospheric depth.

The Lega Dembi Primary Gold Deposit in southern Ethiopia is related to the shear zone-hosted vein in the Neoproterozoic metamorphosed volcano-sedimentary succession of greenschist- to amphibolite-facies metamorphism. The rocks consists of a sequence of biotite-feldspar-quartz schists, carbonaceous mica-schists, amphibolites and basic to ultrabasic rocks. This unit is separated from a foot wall biotite gneiss by a major shear zone. The ore bodies are hosted in the volcano-sedimentary sequence and consist of swarms of quartz veins, lenses, and stockworks that propagated along mesoscale ductile to brittle-ductile shear zones.The mineralization is defined by a complex paragenesis of gold in association with Cu-Pb-Zn-Fe sulphides, tellurides and sulphosalts. The presence of Ni-bearing minerals in amphibolites of the host sequence, together with the ore mineral association, suggests an origin related to mafic volcanism.

Questions about the taxonomic status, diversity, and pace of evolution of basal ornithischian dinosaurs persist in part because some historically important taxa have been based on incomplete material of uncertain ontogenetic status. We examined the morphology of critical “fabrosaurid” specimens and analyzed the bone tissues of small and large individuals. We conclude that the case for the existence of a non-heterodontosaurid ornithischian distinct from Lesothosaurus diagnosticus in the upper Elliot Formation of southern Africa is not conclusive and we suggest that this species and Stormbergia dangershoeki may actually represent ontogenetic stages of one taxon that reached maturity in approximately four years.

Geochemical, Sr-isotopic and geochronological data on the major granitic activity in the south Indian shield suggest a significant event of intracrustal melting and differentiation processes, particularly at middle- and lower crustal levels, resulting in irreversible geochemical differentiation of the crustal column into a large ion lithophile (LIL) element - depleted anhydrous lower crust and an upper crust characterised by copious granitic activity, marking the Archean-Proterozoic boundary. A perceptible link between the formation of granulites in the lower crust and the granitic activity in the upper crust is recognised in this region.

A portion of the aeromagnetic anomaly map of India, from 170 to 200 N and 78o to 84o E has been analysed to understand the tectonics of the region. The distribution of magnetic sources in the study region are clearly brought out in the analytic signal map and found to be associated with charnockitic rocks, iron formation and trap flows. The Godavari Graben is devoid of any magnetic sources. High-grade charnockitic rocks on surface and sub-surface, flank the shoulders of the Godavari Graben on either side. From the analysis of magnetic data, Sileru Shear Zone (SSZ) is identified as the contact of the Bastar craton and the Eastern Ghat Mobile Belt (EGMB). The Eastern Ghat is divided into two blocks: Block-N north of Srikakulam is devoid of magnetic sources while the charnockitic rocks are the main magnetic carriers in Block-S. The difference in magnetic characteristics of the two blocks has been attributed to the difference in metamorphic history. Block-N has an over print of amphibolite facies metamorphism while Block-S to the south depicts granulite facies metamorphism. The Euler solutions within the EGMB shows that the magnetic sources along SSZ is shallower than the south east implying that the exhumation process in the EGMB has a differential rate.

Drillhole sampling of the rocks beneath the post-upper JUrdSSiC sedimentary cover of the Southeastern Coastal Plain of the United States has revealed that tholeiitic and alkaline basalts and diabases, in addition to high-K rhyolites and granites, are present in the exotic Suwannee terrane. The most reliable published ages for basaltic rocks range from 183r5 to 199r8 Ma, overlapping the accepted age range for magmatism of the Central Atlantic Magmatic Province (CAMP). Ages for rhyolites range from 16526 to 189's Ma, while the only dated granite has a well-constrained U-Pb zircon age of 159?3 Ma. The geochemistry of all lithologies is consistent with their eruption or emplacement in an extensional tectonic environment. The tholeiitic basalts are derived from continental lithospheric mantle and crust that is approximately 1 billion years old, suggesting an affinity with the South American portions of Gondwana. Alkali basalts found in southwestern Florida may contain a component of sub-lithospheric mantle and have compositions similar to some basalt s from Liberia and Brazil. Associated silicic magmas may have been generated by intracrustal melting induced by the heat of underplated or injected mafic magmas during the waning stages of igneous activity.

Permian–Triassic drainage radiates from the Gamburtsev Subglacial Mountains (GSM) in central Antarctica. Proximal to the GSM are Permian–Triassic fluvial sandstones in the Prince Charles Mountains (PCM), and neighbouring ?Triassic red beds in Prydz Bay (PB) ODP740A. We analysed detrital zircons for U–Pb ages, Hf-isotope compositions, and trace elements to determine the age, rock-type and source of the host magma, and “crustal” model age (TDMC).Populations of detrital zircons are (1) 700 to 500 Ma, host magmas granitoid and alkaline rock, TDMC ranges from 2.5 to 1.1 Ga, and (2) 1200–800 Ma, host magmas mafic granitoid and alkaline rock, TDMC 2.1 to 1.5 Ga. The bedrock of the PCM-PB region is a potential provenance of the detrital zircons, but the same populations in Permian siltstone south of the PCM and in sediment inclusions in ice at Lake Vostok indicate that the GSM–Vostok Subglacial Highlands (VSH) are the main provenance. Similar detrital zircons in other sandstones in Gondwanaland downslope from a wider central Antarctic reflect an upslope provenance including the GSM–VSH as a complex of 1200–800 Ma (Grenville) and older cratons with mafic granitoids embedded in 700–500 Ma fold belts with granitoids and alkaline rocks. During the past 1000 Ma, the GSM has undergone intermittent uplift on a scale resembling that of the present uplands of Central Asia.

Aeromagnetic surveys help reveal the geometry of Precambrian terranes through extending the mapping of structures and lithologies from well-exposed areas into areas of younger cover. Continent-wide aeromagnetic compilations therefore help extend geological mapping beyond the scale of a single country and, in turn, help link regional geology with processes of global tectonics. In Africa, India and related smaller fragments of Gondwana, the margins of Precambrian crustal blocks that have escaped (or successfully resisted) fracture or extension in Phanerozoic time can often be identified from their aeromagnetic expression. We differentiate between these rigid pieces of Precambrian crust and the intervening lithosphere that has been subjected to deformation (usually a combination of extension and strike-slip) in one or more of three rifting episodes affecting Africa during the Phanerozoic: Karoo, Early Cretaceous and (post-) Miocene. Modest relative movements between adjacent fragments in the African mosaic, commensurate with the observed rifting and transcurrent faulting, lead to small adjustments in the position of sub-Saharan Africa with respect to North Africa and Arabia. The tight reassembly of Precambrian sub-Saharan Africa with Madagascar, India, Sri Lanka and Antarctica (see animation in http://kartoweb.itc.nl/gondwana) can then be extended north between NW India and Somalia once the Early Cretaceous movements in North Africa have been undone. The Seychelles and smaller continental fragments that stayed with India may be accommodated north of Madagascar. The reassembly includes an attempt to undo strike-slip on the Southern Trans-Africa Shear System. This cryptic tectonic transcontinental corridor, which first formed as a Pan-African shear belt 700–500 Ma, also displays demonstrable dextral and sinistral movement between 300 and 200 Ma, not only evident in the alignment of the unsuccessful Karoo rifts now mapped from Tanzania to Namibia but also having an effect on many of the eventually successful rifts between Africa-Arabia and East Gondwana. We postulate its continuation into the Tethys Ocean as a major transform or megashear, allowing minor independence of movements between West Gondwana (partnered across the Tethys Ocean with Europe) and East Gondwana (partnered with Asia), Europe and Asia being independent before the ∼250 Ma consolidation of the Urals suture. The relative importance of primary driving forces, such as subduction ‘pull’, and ‘jostling’ forces experienced between adjacent rigid fragments could be related to plate size, the larger plates being relatively closely-coupled to the convecting mantle in the global scheme while the smaller ones may experience a preponderance of ‘jostling’ forces from their rigid neighbours.

A geological study of the hitherto poorly described Neoproterozoic Gifberg Group, with emphasis on lithogeochemistry and O, C and Sr isotopic composition of the carbonate-dominated Widouw Formation (Vredendal Outlier, westernmost South Africa) revealed that the entire group is an equivalent of the relatively well constrained Port Nolloth Group in the external, paraautochthonous part of the Pan-African Gariep Belt further north. Thus, the Vredendal Outlier can be regarded as the southern extension of the Port Nolloth Zone. Two diamictite units are recognised in the Vredendal Outlier, which can be correlated respectively with the c. 750 Ma Kaigas Formation diamictite and the 583 Ma, syn-Gaskiers Numees Formation diamictite in the Gariep Belt proper. The dominating carbonate unit in the studied area is post-glacial with respect to the older of the two diamictite units. The combined textural, structural and geochemical evidence suggests that parts of the variably dolomitised limestone succession represent former evaporite beds. Sedimentation in a restricted, very shallow and proximal basin led to a wide range in C isotope ratios (δ13CPDB from − 4.2 to + 4.8‰), very high Sr concentrations (derived from original anhydrite) and initial 87Sr/86Sr ratios that are significantly higher (0.70785) than those of coeval seawater. As C and Sr isotopes are commonly used for chemostratigraphic correlation, and high Sr concentrations in Neoproterozic carbonates are often interpreted as evidence of former aragonite, the findings of this study should be used as warning against uncritical use of geochemical and isotopic parameters for describing ancient seawater composition. Thus C and Sr isotope ratios alone in Neoproterozoic carbonates may be less powerful proxies of ancient seawater composition, and high Sr contents are not necessarily indicative of an “aragonite sea”, as previously inferred.

The whole-rock geochemistry of metamorphosed greywackes, arenites and arkoses within the Mesoproterozoic Namaqua-Natal-Maudheim Province is interpreted with the aim of establishing geochemical correlations and defining common sediment source terrains. Metasediments of the Mfongosi Group of the Natal Sector of the Namaqua-Natal Metamorphic Province were sampled from their type area in the Mfongosi Valley. Metagreywackes from the northern limits of the Mfongosi Valley, directly adjacent to the Kaapvaal Craton, show ocean island arc signatures while metagreywackes from the southern limits of the Mfongosi Valley, near the contact with the Madidima Thrust of the Natal nappe zone, show mainly active continental margin signatures. Interleaved, geochemically distinct low-Ca+Na, high-K metamorphosed arkoses to lithic arkoses indicate a minor passive margin sediment component. Geochemical classification of low-grade Ahlmannryggen Group greywackes, arenites and arkoses of the Grunehogna Province, Antarctica, indicates both active and passive continental margin sediment sources. An oceanic island arc signature is not evident in Ahlmannryggen Group data. The active continental margin signature in both Natal Sector and Grunehogna Province metasediments potentially provides for a common link between these terranes. Discriminant Function Analysis, using three pre-defined provenance sub-sets within the Mfongosi Group and two pre-defined provenance sub-sets within the Ahlmannryggen Group, indicate that metasediments with active continental margin signatures from both groups are geochemically identical, implying that the active continental margin of the Grunehogna Province shed immature sediments westwards (African azimuths) into the developing, narrow or restricted Mesoproterozoic ‘Mfongosi Basin.’ This was accompanied by minor sediment influx from a stable continental platform, potentially the Kaapvaal Craton. Oblique and diachronous collision, initiated in the southwestern portions of the combined Natal Sector/Grunehogna Province system produced a laterally variable Mfongosi Group, which formed in the ‘Mfongosi Basin’. Coarse-grained sediments dominated in its eastern portions while basalts with thin sapropelite units dominated in its western portions.

The Permo-Triassic Beaufort Group (Karoo Basin) of South Africa is biostratigraphically subdivided into eight, temporally successive assemblage zones based on therapsids (‘mammal-like reptiles’). The Temnospondyli, fossil tetrapods usually regarded as extinct amphibians, are second only to therapsids in terms of diversity and abundance in these strata, with nine higher-level taxa (‘families’) known. Temnospondyls are also playing an increasingly important role in biostratigraphy and correlation of the Beaufort strata. The lower Beaufort Group (Late Permian) contains six of the eight biozones, but only one temnospondyl ‘family’, the Rhinesuchidae, whose record in the Karoo is the richest in the world. However, rhinesuchid taxonomy remains in flux and the group is thus of limited biostratigraphic utility. The Early Triassic Lystrosaurus Assemblage Zone (middle Beaufort Group) contains the Rhinesuchidae, Amphibamidae, Lydekkerinidae, Tupilakosauridae, Rhytidosteidae, Mastodonsauridae and Trematosauridae, although the biostratigraphy of temnospondyls within this biozone is poorly constrained. The uppermost reaches of the Lystrosaurus biozone contain a paucity of fossils but includes ‘Kestrosaurus’ (Mastodonsauridae) and ?Trematosuchus (Trematosauridae), taxa previously thought to pertain to the lower part of the overlying Cynognathus biozone. The late Early to Middle Triassic Cynognathus Assemblage Zone (upper Beaufort Group) hosts the Mastodonsauridae, Trematosauridae, Brachyopidae, Laidleriidae and, possibly, the Rhytidosteidae. Based largely on the spatial and temporal distribution of mastodonsaurids, this biozone has been biostratigraphically subdivided into a lower A, middle B and upper C subzones, characterised by differing ages and faunas.

The closure of the Palaeozoic witnessed the greatest biotic crisis in earth history. Surprisingly little is known about the effects and timing of the terrestrial counterpart of the well-described End-Permian mass extinction from known marine successions worldwide. In the present study, reliable paleomagnetic results were obtained from a PT boundary section in the terrestrial Karoo Basin of South Africa. Permo-Triassic aged mudstones from a locality in the Eastern Cape Province yielded two magnetic chrons, reverse followed by normal (with the boundary possibly close to the reversal). This extends to results from a previous study: thereby jointly identifying a R/N/R polarity pattern for this boundary interval. The PTB interval is constrained below the red mudstones of the Beaufort Group at the present locality and within reverse-magnetised green mudstone, implying a diachronic relation between the marine and terrestrial End-Permian mass extinction events.

Rb-Sr and Sm-Nd isotopic studies were carried out for metamorphic rocks in the Namaqualand Metamorphic Complex, South Africa. The metamorphic rocks give the Rb-Sr mineral isochron ages (whole-rock - biotite - felsic fractions) of 844±85 Ma and 811.6±6.6 Ma for the lower granulite zone and of 776.5±5.4 Ma for the upper granulite zone. The rocks yield the Sm-Nd mineral isochron ages of 1071±18 Ma (whole-rock - garnet - felsic fractions) and 1067±158 Ma (whole-rock - hornblende - biotite rich fraction - felsic fractions) for the lower granulite zone and of 1052.0±3.6 Ma and 1002.5±1.4 Ma (whole-rock - garnet - felsic fractions) for the upper granulite zone. These age data suggest that the granulite facies metamorphism took place at 1060-1000 Ma, and that the rocks cooled down at 850-780 Ma. The Sr and Nd isotopic compositions of metamorphic rocks are different between the lower and upper granulite zones.

Sedimentological, palaeontological and geological data from the glacial to postglacial transition in the late Paleozoic successions of the Paganzo-Calingasta Basin (PC) in southern South America and the Great Karoo-Kalahari Basin (GKK) in southern Africa are analysed, revised and reinterpreted. A brackish depositional setting is inferred for main areas previously considered to be nonmarine based upon ichnological interpretations. Three stratigraphic intervals have been defined based on changes in sedimentary facies and trace fossils association: The glacial interval (GI), early postglacial interval (EPI) and late postglacial interval (LPI). The GI and EPI contain a dominance of arthropod trackways, fish trails with and subordinate grazing and feeding traces. The EPI in the PC Basin comprises both nonmarine and brackish-marine ichnocoenoses without significant differences in ichnological composition. Trace fossils are preserved in underflow and turbidite beds of deltaic deposits. Opportunistic grazing traces constitute a post-event ichnocoenosis, while a pre-event ichnocoenosis is preserved at the base of turbidite beds. In the GKK Basin ichnofossils were documented in turbidite fans. The LPI in the GKK Basin contains the first evidence of shallow water deltaic infauna and subordinate grazing traces. Conversely, in the PC Basin the infauna is lacking.

The eastern part of the Cape Fold Belt, near Steytlerville, South Africa, reveals a typical pattern of numerous, north-verging thrust faults and associated folds, interpreted as part of a large duplex structure that formed along the southern margin of Gondwana during the Late Palaeozoic. Steeply-dipping fore- and backthrusts occur in the Bokkeveld Group (middle Cape Supergroup), where strata are composed of predominantly argillaceous rocks, whereas in the more arenaceous Witteberg Group (upper Cape Supergroup) there are fewer recognizable and less closely-spaced thrusts. Open style folds characterize areas in which the Bokkeveld Group crops out, but in areas of Witteberg outcrop, folds, especially those adjacent to thrusts, are often overturned.In spite of a general absence of marker horizons, a displacement of at least 500 metres can be inferred for one prominent thrust, the Jackalsbos thrust. This fault, the northernmost in the area investigated, is probably the sole thrust in the duplex structure, linked through southward-dipping imbricates to a projected roof thrust (the Baviaanskloof thrust) cropping out immediately south of the study area.Displacements on imbricates within the duplex are difficult if not impossible to measure, but the net effect is certainly accumulative and incremental. Truncation by a roof thrust and subsequent erosional processes may explain why so few of the many thrusts so far identified in the eastern part of the fold belt can be successfully mapped, and their displacements measured. Normal and strike-slip faults, less common than thrust faults, formed during extensional tectonism related to the breakup of Gondwana, during the Mesozoic.

The Mozambique belt of eastern and southern Africa is polyorogenic and marks the sites for the assembly (collision and suturing) and dispersion (rifting and drifting) of the Proterozoic supercontinents. Subduction zones and collisional sutures in this belt are of variable ages. Reliable isotope and geological data from the Mozambique belt of Holmes (1951) suggest that there existed three major Proterozoic oceans within this belt: the Palaeoproterozoic, Mesoproterozoic and Neoproterozoic “Mozambique Oceans”. However, the accretion and collisional tectonic history of this orogenically coalescent belt are complex and thus still enigmatic.

The northeastern part of Madagascar is characterized by Archaean to early Proterozoic rocks composed principally of Archaean granite and greenstone/amphibolite as well as reworked migmatite with subordinate Proterozoic paragneisses. The southern part is mostly occupied by Proterozoic rocks, composed mostly of Meso to Neo-Proterozoic and less metamorphic metasediments (Itremo Group) in the northwest, para- and ortho-gneisses in most other areas, with minor granitic gneisses with some Archaean components in the southeast. The north-northwest trending Central Granite-Gneiss-Migmatite Belt (CGGMB) is situated at the western margin of the Archaean-early Proterozoic terrain. The CGGMB is composed of granite, gneiss and migmatite with distinct lithologies and structures. They are: i) many types of granites including alkaline to mildly alkaline granites, and calc-alkaline granites; ii) batholitic granites, migmatitic granites and granite dyke swarm, iii) eclogite, and iv) the Ankazobe-Antananarivo-Fianarantsoa Virgation.The CGGMB was formed by the collision of the palaeo-Dharwar Craton to the east and the East African Orogen to the west at ca. 820-720 Ma and suffered indentation by a part of the western part of the East African Orogen at ca. 530 Ma that produced the Ankazobe-Antananarivo-Fianarantsoa Virgation at the centre of the CGGMB. Thus, the CGGMB is proposed to be the continuation of the eastern suture between the palaeo Dharwar Craton and the East African Orogen, and carries the main feature of the Pan-African collisional event in Madagascar.

The Pan-African Kid-Malhak Dokhan volcanic suite (609 ± 12 Ma) is exposed in the northernmost part of the Arabian–Nubian Shield. The suite consists of non-metamorphosed varicolored alternating succession of porphyritic lava flows of commonly felsic composition (rhyolite–dacite) interlayered with compositionally equivalent pyroclastic beds (dominantly ignimbrites). These Dokhan volcanics are quite evolved (SiO2 ≈ 65–77 wt.%), with strong high-K calc-alkaline affinity and are characterized by relative enrichment in total alkalis, Ba, Y, Zr and total REEs, depletion in Sr, and a LREE-enriched REE patterns with significant negative Eu anomalies. The Kid-Malhak Dokhan lavas display geochemical characteristics of both orogenic arc-type and anorogenic within-plate environments, suggesting eruption in a transitional “post-collisional tectonic setting. The ages of emplacement of the Dokhan volcanics in Egypt including that of Kid-Malhak region (580–620 Ma) coincide with end of the documented collision between the juvenile Arabian–Nubian crust and Saharan Metacraton and the subsequent extensional collapse event. This post-collision transition from compression to extension is explained by the extensional collapse following continental collision, which was controlled mainly by lithospheric delamination and slab breakoff (passive rifting). Various trace element characteristics discussed herein have indicated that the studied Dokhan magma was highly likely generated from crustal sources and that assimilation–fractional crystallization (AFC) and crustal contamination have played a major role and are most probably superimposed on fractional crystallization during the magmatic evolution of Kid-Malhak Dokhan volcanic suite. The eruption of the high-K calc-alkaline post-collisional Dokhan volcanics in Egypt defines a tectono-magmatic transition between the older calc-alkaline arc-related and the subsequent alkaline magmatism in the northern part of the Arabian–Nubian Shield.

Graphite is present in nature in several forms. Genetically they may be broadly classified as biogenic and abiogenic. The biogenic forms are those that are clearly derived from an organic precursor while the abiogenic or inorganic forms are more complex from the point of view of their origin, nature and geological relations.As a geomarker, biogenic graphite has certain advantages. It is easily recognized and shows different degrees of crystallinity depending on the relative grades of metamorphism it had undergone. Once it attains a certain degree of crystalline order, it does not revert to a lower state even under changing metamorphic conditions, thereby making it a good mineral geothermometer. It is also found in specific, restricted geological environments and is therefore useful as a boundary marker of ancient sedimentary terrains.These special characteristics of the biogenic type of graphite can be effectively used to trace sites of sedimentary basins and subsequent ocean closures that may have resulted in geosutures. Studies of the Pan-African terrains of the Gondwana crustal fragments as exemplified by the sutures of the Mozambique Belt running through East Africa, Madagascar, Sri Lanka and Antarctica illustrate this point. A further example comes from the Mashan Group of East China, one of the most productive graphite - bearing regions of the world.

Three main tectonic events related to Pan-African collision and post-collision evolution have been identified in Cameroon: i) crustal thickening; ii) left lateral wrench movements; and iii) right lateral wrench movements, successively. Correlation of these events throughout the Pan-African belt, and an integrated synthesis that includes the domains of Central Africa, Trans-Sahara and NE Brazil, enable us to define an indent-type structure in northwestern Cameroon. This one suggests the existence of a rigid prong in this area which collided with the São Francisco–Congo Craton (SFCC) active margin at ca 640 Ma prior to subsequent widespread remobilization and granitization that partly affected both landmasses between 640 and 580 Ma. Major indent-related shear zones were further reactivated during regional clockwise rotation of the newly stabilized domain of northwestern Cameroon, suggesting that collision with the West African Craton (WAC) was still active after 580 Ma. Geodynamic implications of this model are discussed in the context of the Pan-African global evolution in western Gondwana.

The pre-Early Cambrian Sandikli Basement Complex in western Central Anatolia comprises a low-grade meta-sedimentary succession (Güvercinoluk Formation) and meta-rhyolites intruded by meta-quartz porphyry rocks (Kestel Cayi Porphyroid Suite). The Güvercinoluk Formation consists of alternation of meta-siltstones and meta-sandstones with olistostromal conglomerates, rare black chert and cherty meta-dolomite lenses. The Kestel Cayi Porphyroid Suite is a deformed, highly sheared dome-shaped rhyolitic body with quartz porphyry rocks. Quartz porphyry dykes intrude both the volcanic carapace and the meta-sedimentary rocks of the Güvercinoluk Formation. Both the meta-quartz porphyry rocks and meta-rhyolites are typically mylonitic with relict igneous textures. Geochemical data indicate that the felsic rocks of the Kestel Cayi Porphyroid Suite are subalkaline and display characteristic features of post-collisional, I-type granitoids. The basement complex is unconformably overlain by variegated conglomerates, mudstones and arkosic sandstones with andesitic lavas, followed by siliciclastic rocks and carbonates that yielded Early Middle Cambrian fossils.Based on the geochemical characteristics of the felsic rocks of Kestel Cayi Porphyroid Suite and the depositional features of the associated sediments it is suggested that the Sandikli Basement Complex is related to a post-collisional extension event in NW Gondwanaland. Similar occurrences elsewhere have been related to a transition from continental plate convergence to continental plate divergence along the Pan-African Belt.

The Sidi Flah and Ougnat inliers are located in the eastern Anti-Atlas antiform between the Anti-Atlas Major Fault (AAMF) and South Atlas Fault (SAF). They consist of many granitoid intrusions emplaced into Neoproterozoic metasedimentary rocks and surmounted by upper Neoproterozoic A-type granites. The Sidi Flah (Saghro) and Ougnat granitoids are part of the Neoproterozoic magmatic activity related to northwards subduction of an oceanic plate beneath the Saghro continental margin. They are post-orogenic I- and S-type granitoids related to the ending of the compressional deformation in this Pan-African belt. A petrographic, geochemical and zircon typology study leads us to subdivide these rocks into three magmatic groups: (1) a medium- to high-K calc-alkaline group formed by quartz diorites and amphibole granodiorites is found in both Sidi Flah and Ougnat inliers; (2) a high-K calc-alkaline group is present in Sidi Flah. These two groups have a (deeper and) hybrid mantle-crust origin; (3) a peraluminous group in Ougnat is linked to the post-collisional setting and has a shallow crustal source. On a primitive mantle-normalized trace-element diagram, almost all of these rocks show a significant Nb depletion relative to K and La, which is typical of the calc-alkaline magmatism from the subduction-zone environment. Absence of structural marks of thrusting upon the West African craton (WAC) of this arc system and the ophiolitic suite in Bou-Azzer, and the presence of Imiter muscovite-bearing granite as part of Pan-African belt do not support the localization of northern limit of WAC at the level of SAF.

The Transcaucasian Massif (TCM) in the Republic of Georgia includes Neoproterozoic–Early Cambrian ophiolites and magmatic arc assemblages that are reminiscent of the coeval island arc terranes in the Arabian–Nubian Shield (ANS) and provides essential evidence for Pan-African crustal evolution in Western Gondwana. The metabasite–plagiogneiss–migmatite association in the Oldest Basement Unit (OBU) of TCM represents a Neoproterozoic oceanic lithosphere intruded by gabbro–diorite–quartz diorite plutons of the Gray Granite Basement Complex (GGBC) that constitute the plutonic foundation of an island arc terrane. The Tectonic Mélange Zone (TMZ) within the Middle-Late Carboniferous Microcline Granite Basement Complex includes thrust sheets composed of various lithologies derived from this arc-ophiolite assemblage. The serpentinized peridotites in the OBU and the TMZ have geochemical features and primary spinel composition (0.35) typical of mid-ocean ridge (MOR)-type, cpx-bearing spinel harzburgites. The metabasic rocks from these two tectonic units are characterized by low-K, moderate-to high-Ti, olivine-hypersthene-normative, tholeiitic basalts representing N-MORB to transitional to E-MORB series. The analyzed peridotites and volcanic rocks display a typical melt-residua genetic relationship of MOR-type oceanic lithosphere. The whole-rock Sm–Nd isotopic data from these metabasic rocks define a regression line corresponding to a maximum age limit of 804 ± 100 Ma and εNdint = 7.37 ± 0.55. Mafic to intermediate plutonic rocks of GGBC show tholeiitic to calc-alkaline evolutionary trends with LILE and LREE enrichment patterns, Y and HREE depletion, and moderately negative anomalies of Ta, Nb, and Ti, characteristic of suprasubduction zone originated magmas. U–Pb zircon dates, Rb–Sr whole-rock isochron, and Sm–Nd mineral isochron ages of these plutonic rocks range between ∼ 750 Ma and 540 Ma, constraining the timing of island arc construction as the Neoproterozoic–Early Cambrian. The Nd and Sr isotopic ratios and the model and emplacement ages of massive quartz diorites in GGBC suggest that pre-Pan African continental crust was involved in the evolution of the island arc terrane. This in turn indicates that the ANS may not be made entirely of juvenile continental crust of Neoproterozoic age. Following its separation from ANS in the Early Paleozoic, TCM underwent a period of extensive crustal growth during 330–280 Ma through the emplacement of microcline granite plutons as part of a magmatic arc system above a Paleo-Tethyan subduction zone dipping beneath the southern margin of Eurasia. TCM and other peri-Gondwanan terranes exposed in a series of basement culminations within the Alpine orogenic belt provide essential information on the Pan-African history of Gondwana and the rift-drift stages of the tectonic evolution of Paleo-Tethys as a back-arc basin between Gondwana and Eurasia.