A newly investigated travertine deposit in southern Tibet provides a window into Holocene hydrological, geomorphic and climatic processes near the boundary of the Tibetan Plateau and High Himalaya. Travertine, deposited as a result of the degassing of CO2-rich groundwater as it emerges on the Earth's surface, is in many cases formed along the trace of major crustal-scale faults in primarily extensional tectonic regimes. A travertine platform measuring roughly 1 km by 0.5 km exists near the town of Nyalam in southern Tibet along a major Himalayan down-to-the-N normal fault, the South Tibetan Fault System. A wide variety of travertine depositional textures and features are recorded in the platform on a series of terraces. Active travertine deposition was observed from spring mounds and seeps along the base of the platform at the modern river level. Palaeotemperatures of spring water, calculated from δ18O of the travertine, range from 9 to 25 °C, which closely matches the temperatures recorded from modern springs in the area. A complex geomorphological landscape records interaction between growing alluvial fans, travertine accumulation, and a rapidly down-cutting river with associated fluvial terraces. River incision was contemporaneous with travertine deposition, as indicated by cemented fluvial river gravel layers interbedded with travertine. High 87Sr/86Sr ratios in the travertine (mean of 0.7168) indicate subsurface fluid interaction with radiogenic crystalline rocks of the underlying Greater Himalaya. Uranium-series ages of the travertine platform range from 5400 a (±950 a) to 11600 a (±1000 a), and indicate a younging progression from higher terraces near the valley wall to lower terraces at present-day river level. Travertine that overlies a river gravel terrace 18 m above river level formed at 11600 a. This date yields a local incision rate of 1.6 mm a−1, consistent with estimated fluvial incision rates in the High Himalaya. The range of our U-series ages coincides with an interval of higher precipitation associated with greater intensity of the Indian monsoon, which led to elevated spring discharge and carbonate precipitation in this part of the High Himalayas.

This paper presents new U–Pb zircon analyses from garnet–sillimanite paragneisses from the Gweta borehole in northeast Botswana. Concordant to near-concordant analyses of zircon from these rocks reveal a billion year history from 3015 ± 21 Ma for the oldest detrital grain measured, to the age of high-grade metamorphism, 2027 ± 8 Ma. The maximum age of sedimentation in the Magondi belt is constrained by the age of the youngest concordant detrital zircon at 2125 ± 6 Ma. This contrasts with the age of sedimentation in the Central Zone of the Limpopo belt which is Archaean. The comparison of our results with U–Pb zircon data from the Magondi belt in Zimbabwe suggests that the granulite-facies metamorphism in this belt extended between c. 2027–1960 Ma. Granulite-facies rocks with U–Pb zircon ages in this interval are also known in the Ubendian belt and lend support to the correlation of these two segments of Palaeoproterozoic belts in southern and central–eastern Africa. The granulite facies metamorphism in the Magondi belt is coeval with the high-grade metamorphism and granitoids documented further south in the Central Zone of the Limpopo Belt.

During the past few months some papers on early amphibia and reptiles have reached me from America, Russia, and China. These have afforded me much pleasure as a change from my present unaccustomed work, and I therefore feel that I can usefully express my thanks to their authors by writing a criticai summary of their conclusions. And such a review is justified because collectively these papers add very greatly to our knowledge and understanding of the evolution of early Tetrapods. Reference will be made to a list of these papers at the end of this review. It is most convenient and logical to discuss firstly the individual animais and then their distribution in time.

A formation composed of unbedded scaly bentonitic clay containing abundant scattered exotic blocks, boulders, pebbles, and smaller fragments is described from eastern Timor. It is proposed to call this formation the Bobonaro Scaly Clay. This formation is compared closely with the Argille scagliosa of the northern Apennines. Its origin is ascribed to submarine sliding of an unstable clay mass from the area north of Timor under the influence of gravity, closely associated with the emplacement of large overthrust slices of strata that marked a climax of the mid-Tertiary orogeny in Timor. The discovery of this formation has helped to bring about a radical revision of the published views of the stratigraphy and structure of Timor.

New data implying crustal activation of Eastern Avalonia along the Anglo-Brabant fold belt are presented. Late Ordovician subduction-related magmatism in East Anglia and the Brabant Massif, coupled with accelerated subsidence in the Anglia Basin and in the Brabant Massif during Silurian time, indicate a foreland basin development. Final collision resulted in folding, cleavage development and thrusting during the mid-Lochkovian to mid-Eifelian. In the southeast of the Anglo-Brabant fold belt, Acadian deformation produced basin inversion and the regional antiformal structure of the Brabant Massif. The uplift, inferred from the sedimentology, petrography and reworked palynomorphs in the Lower Devonian of the Dinant Synclinorium is confirmed by illite crystallinity studies. The tectonic model discussed implies the presence of two subduction zones in the eastern part of Eastern Avalonia, one along the Anglo-Brabant fold belt and another under the North Sea in the prolongation of the North German–Polish Caledonides.

Arrested-type charnockite formation occurs in an assemblage of high-grade gneisses at several localities of the Chilka Lake area that belongs to the Proterozoic Eastern Ghats Belt of India. The isolated ellipsoidal domains are found exclusively in leucogranite (leptynite) bands that intruded lit-par- lit interbanded granulite-grade supracrustal and intermediate igneous rocks (khondalite–enderbite). Macrostructures and microfabrics document a multiple deformation of the rock assemblage under high-grade conditions. The intrusion of the leucogranitic melts separates a first episode of deformation, D1, from a younger progressive deformation, D2–D4. A transpressive regime and inhomogeneous deformation is indicated for D2–D4 by the associated structures and fabrics. But quartz c-axis patterns show that pure shear prevailed during the closing stages of deformation. The spatial distribution and orientation of the ellipsoidal charnockite domains within the host leptynite and the orientation pattern of orthopyroxene c-axes inside the domains provide evidence for a synkinematic in situ formation of the domains during D3, through partial breakdown of the leptynite assemblage (Bt+Grt+Qtz+Fl1[rlhar ]Opx+Fsp+Ilm+Fl2/L). Local fluid migration along steep foliation planes associated with large-scale D3 folds triggered the reaction. Orthopyroxene blastesis was confined to the centre of the domains, and an envelope formed in which the residing fluid caused secondary intergranular formation of chlorite, ore and carbonate, imparting the domains' typical greenish-brown charnockite colour. The shape of the envelope, which varies from prolate in limbs to oblate in hinges of D3 folds, is responsive to the local stress field. Comparison of chemical rock compositions supports the in situ formation of charnockite in leptynite. Subtle compositional differences are controlled by the changing mineralogy. Compared to the host leptynite, the charnockite domains are enriched in K2O, Ba, Rb and Sr, but depleted in FeO*, MnO, Y and Zr. The data obtained in this study provide conclusive evidence that the ellipsoidal charnockite domains do not represent remnants of stretched enderbite layers as proposed by Bhattacharya, Sen & Acharyya, but formed in situ in the leptynite as a result of localized synkinematic fluid migration late in the deformation history.

There is little consensus on the systematic position of the colossal theropod dinosaur Chilantaisaurus tashuikouensis from the Cretaceous (Aptian–?Albian or Upper Cretaceous) Ulansuhai Formation of Inner Mongolia, which has been recovered as a derived member of both Allosauroidea and Spinosauroidea by numerical phylogenetic analyses. Redescription of the type material of C. tashuikouensis reveals an unusual combination of morphological features that render determination of its systematic position problematic. It possesses anatomical features that have been proposed as synapomorphies of Neotetanurae: a preacetabular fossa on the ilium, and a wedge-shaped cross-section of the shaft of the third metatarsal. It also shares some features with specific allosauroid taxa: a pronounced ulnar epicondyle on the humerus, and a prominent medial shelf bounding the preacetabular fossa on the ilium (also present in tyrannosauroids). However, it lacks some features that are present in all other allosauroids: a marked depression on the anterior surface of the distal humerus adjacent to the ulnar condyle, and a humerus that is less than 0.4 times the length of the femur; it furthermore possesses a tibial astragalar facet that is approximately 10% of the tibial length, which suggests a more basal position within Tetanurae. Chilantaisaurus shares certain features with some spinosauroids: an enlarged and elongated first manual ungual, and a suprastragalar buttress that has been modified to a vertical ridge, but these characters are not unique to spinosauroids. A highly reduced fourth trochanter may be an autapomorphy of Chilantaisaurus, as has previously been suggested, or unite the taxon with Coelurosauria in an entirely novel grouping. On the basis of these observations it is likely that Chilantaisaurus is a neotetanuran, but unlikely that it is an allosauroid. Chilantaisaurus may belong to an alternative lineage of very large theropods that continued into the Cretaceous from the diversification of basal neotetanurans during the Middle Jurassic.

The Neoproterozoic–Cambrian succession in the Zavkhan Basin of western Mongolia preserves early Cambrian bioconstructions of Nemakit-Daldynian to Botomian age. As elsewhere (Siberia, Morocco), the Nemakit-Daldynian bioconstructions in the upper Tsagaan Oloom Formation to lower Bayan Gol Formation interval were purely calcimicrobial. Spectacular calcimicrobial ‘patch reefs’ of presumed Tommotian age are present in the lower Bayan Gol Formation. In contrast, late Atdabanian–early Botomian bioconstructions are varied and well developed in the upper Salaany Gol Formation including, in addition to calcimicrobial bioherms, Gordonophyton–Razumovskia crusts, radiocyathan—archaeocyathan bioherms and thickets of the ramose archaeocyath Cambrocyathellus.

These latter Zavkhan Basin buildups show some ecological and sedimentological features in common with coeval bioconstructions in South Australia, which also developed under a tectonically active regime.

We review historical approaches to the systematics of Enantiornithes, the dominant birds of the second half of the Mesozoic, and describe the forelimb remains of a new Cretaceous euenantiornithine. This taxon is known on the basis of fossil specimens collected from southern France, Argentina and the United States; such a wide geographical distribution is uncharacteristic for Enantiornithes as most taxa are known from single localities. Fossils from the Massecaps locality close to the village of Cruzy (Hérault, southern France), in combination with elements from New Mexico (USA) and from the Argentine locality of El Brete (Salta Province) testify to the global distribution of large flighted euenantiornithine birds in the Late Cretaceous. We discuss the systematics and taxonomy of additional isolated bones of Enantiornithes that were collected from the Argentine El Brete locality in the 1970s; the presence of these flying birds in Cretaceous rocks on both sides of the equator, in both northern and southern hemispheres, further demonstrates the ubiquity of this avian lineage by the latter stages of the Mesozoic.

The earliest nematophorid (planktic) dendroid graptolites (Family Anisograptidae) are described and revised from several heterofacial facies sections of the Broom Point Formation, Cow Head Group, western Newfoundland (Broom Point North and St Paul's represent medial to upper continental slope, Martin Point and Green Point distal slope environments). Noting potential conodont recycling in the palaeoslope environments of the proximal Cow Head Group sections, graptolites are the only group of fossils that could not have been subject to post-burial transport. The graptolites belong to an informal group of quadriradiate form-species principally representing two originally related but then diverging phyletic groups: pendent Rhabdinopora Eichwald, 1855 and horizontal to sub-horizontal Staurograptus Emmons, 1855; both being stratigraphically within Assemblage Zone I (emended after Cooper, 1979) of earliest Tremadoc or early Ibexian age (top of Cordylodus intermedius to base of C. lindstromi conodont zones). Assemblage I graptolites offer a higher resolution than coeval conodonts and are a better tool for inter-facies correlations both within the Broom Point Formation and with other graptolitic sequences on a global scale. Form-species of quadriradiate Rhabdinopora rustica (such as ?R. praeparabola and R. parabola) and of Staurograptus dichotomus (e.g. ‘Radiograptus’ or ‘Heterograptus’ flexibilis or ‘Aletograptus’) are interpreted as astogenetic morphs, or as phylogenetically descendant variants of these forms. Upon evidence of late astogenetic insertion of dissepiments in Staurograptus and Anisograptus the names ‘Heterograptus’ and ‘Radiograptus’ are placed into junior synonymy, and ‘Aletograptus’ is regarded a junior synonym of Staurograptus because its post-primary dichotomies are merely delayed distally (or omitted) to be observed exclusively in very mature specimens. The Cow Head Group fauna is compared with earliest nematophorids from Oslo, Norway (Naersnes) and Dayangcha, Jilin, China (Xiaoyangqiao), and a new comprehensive ecostratigraphic zonation for early Tremadoc graptolites is proposed.

In March of the present year (1909) the Rev. R. Ashington Bullen, who was then travelling in Majorca, wrote to inform me of the existence of a supposed ossiferous breccia situated on the east coast of that island. This caused a long-deferred project of exploring the Balearic Islands in search of Pleistocene cave-deposits to be carried out, and a month was spent in this manner, although a much longer period would be required for an exhaustive examination of Majorca alone, which was the only one of the group visited. A considerable tract of limestone country was investigated, and the result was the discovery of three ossiferous cave-deposits, including the one mentioned above. Similar remains were procured from each of these, and on examination prove to be those of an ungulate which appears to be without parallel. No great quantity of material was obtained, but this fortunately includes a well-preserved skull of an old individual with the associated mandible and atlas vertebra—the typespecimen, see Fig. 1—as well as a number of perfect examples of various limb-bones.

The Global Stratotype Section and Point (GSSP) for Stage 2 of the Ordovician System, now the Floian Stage and approximately equivalent to the lower and middle Arenig of England and Wales, is defined by the first appearance datum (FAD) of the graptolite Tetragraptus approximatus in the Diabasbrottet Quarry section at Mount Hunneberg, Sweden. One of the issues this raises is how to correlate the base of Stage 2 at the GSSP with areas and successions that do not contain a correlative graptolite fauna. The distinctive Cymatiogalea messaoudensis–Stelliferidium trifidum acritarch assemblage is present in the upper Tremadocian Araneograptus murrayi Graptolite Biozone of NW England and ranges across the Tremadocian–Stage 2 boundary there (the Tremadoc–Arenig boundary of Anglo-Welsh nomenclature). It also occurs widely at other high southern Ordovician palaeolatitudes around the margin of Gondwana, being reported from Ireland, Wales, the Isle of Man, Belgium, Germany, Spain and Turkey, and may also be present in Bohemia and Argentina (Eastern Cordillera). It therefore has the potential to contribute towards the recognition and correlation of the base of Stage 2 in those areas. Of particular interest are the First Appearance Datums of various taxa within the stratigraphical range of the messaoudensis–trifidum assemblage, notably that of Aureotesta clathrata simplex, which is considered to be close to the base of Stage 2 in NW England.

Elements of the messaoudensis–trifidum assemblage also occur in Baltica, the palaeoplate on which the GSSP for the base of Stage 2 is located. However, many of the taxa used to subdivide the messaoudensis–trifidum assemblage around Gondwana have not been recorded from Baltica and may be restricted palaeobiogeographically to the Gondwanan margin. Furthermore, acritarch microfloras have not been reported from the Diabasbrottet Quarry section itself, and there are hiatuses across the base of Stage 2 in the two sections from Baltica considered in this paper. Hence, direct correlation of the base of Stage 2 between the GSSP and other sections using acritarchs is not yet possible. Nevertheless, some taxa, for example the genera Peteinosphaeridium and Rhopaliophora, are shown to have FADs at similar stratigraphical levels in the late Tremadocian Stage of both Baltica and Gondwana, and therefore have the potential to correlate time slices in the late Tremadocian Stage between palaeoplates.

Geochemical and Sr–Nd–Pb isotopic data are presented for volcanic rocks from Zougouyouchaco (30.5 Ma) and Dogai Coring (39.7 Ma) of the southern and middle Qiangtang block in northern Tibet. The volcanic rocks are high-K calc-alkaline trachyandesites and dacites, with SiO2 contents ranging from 58.5 to 67.1 wt % The rocks are enriched in light REE (LREE) and contain high Sr (649 to 986 ppm) and relatively low Yb (0.8 to 1.2 ppm) and Y (9.5 to 16.6 ppm) contents, resulting in high La/Yb (29–58) and Sr/Y (43–92) ratios, as well as relatively high MgO contents and Mg no., similar to the compositions of adakites formed by slab melting in subduction zones. However, the adakitic rocks in the Qiangtang block are characterized by relatively low εNd(t) values (−3.8 to −5.0) and highly radiogenic Sr ((87Sr/86Sr)i=0.706–0.708), which are inconsistent with an origin by slab melting. The geochemistry and tectonics indicate that the adakitic volcanic rocks were most likely derived from partial melting of delaminated lower continental crust. As the pristine adakitic melts rose, they interacted with the surrounding mantle peridotite, elevating their MgO values and Mg numbers.

Parringtonia gracilis Huene, 1939 is represented by both cranial and postcranial material collected from the lower Middle Triassic (Anisian) Lifua Member of the Manda beds in southwestern Tanzania. This aberrant taxon was previously proposed to have affinities with pseudosuchian archosaurs, and specifically with the enigmatic Erpetosuchus granti from the Upper Triassic of Scotland. Here, we confirm the close affinities of Parringtonia gracilis and Erpetosuchus granti based on the following unambiguous synapomorphies: mediolaterally expanded posterior portion of the maxilla, alveoli present only in the anterior half of the maxilla, and absence of tooth serrations. Furthermore, the two taxa share osteoderms with deep sculpturing, a deep fossa on the dorsal margin of the neural spines and a heavily waisted shaft of the scapula. We added both Parringtonia gracilis and Erpetosuchus granti into a comprehensive phylogenetic analysis of early archosaurs and found that these taxa are clearly referable to Archosauria but that relationships are poorly resolved at the base of this clade. However, our analysis demonstrates that Erpetosuchus granti is not closely related to Crocodylomorpha, as has been hypothesized previously. The Erpetosuchidae are a clade of small-bodied archosaurs that have a poor fossil record but have members from both northern and southern Pangaea, ranging temporally from the Middle to Late Triassic. Thus, Erpetosuchidae is part of the early archosaurian radiation.

A dinoflagellate cyst record with strong Mediterranean/Lusitanian affinities is described from marine deposits of Eemian age (Last Interglacial; Late Pleistocene) at Ristinge Klint, Denmark, revealing new information about the hydrographic evolution of the southwestern Baltic Sea. A revised correlation of the pollen record at Ristinge Klint with that of the annually laminated site at Bispingen in northern Germany provides temporal control. Approximately the first quarter of Eemian time is represented. A marine ingression into a lake took place during the Quercus rise, about 300 years into the interglacial, and is marked by low (< c. 3 psu) salinities at the base of the Cyprina Clay that increased progressively. An abrupt and significant rise in the inflow of warm, saline waters from the North Sea occurred at about 750 years into the interglacial (the Corylus rise), and at about 1900 years into the interglacial, strongly stratified waters developed. Higher in the Cyprina Clay and continuing to its top, at nearly 3000 years into the interglacial, more open-marine waters are indicated, although fully marine conditions were not reached. The dinoflagellate record throughout the Cyprina Clay at Ristinge Klint is therefore one of increasing marine influence. Summer sea-surface temperatures approached, and may have exceeded, 26–28 °C during early Eemian time, indicating temperatures at least 5 °C warmer than at present. These warm conditions persisted to the top of the record at Ristinge Klint. No evidence exists at Ristinge Klint for the influence of Arctic watermasses, and the paucity of cold-water species throughout the section reflects mild winter temperatures in the southwestern Baltic Sea.

The new species Spiniferites ristingensis is formally described, and the name Operculodinium centrocarpum var. cezare de Vernal, Goyette & Rodrigues, 1989 is validated.

Whitcliffian and Downtonian macroplants of rhyniophyte morphology are described from Capel Horeb Quarry, Powys, Wales. Both floras contain Hostinetla Stur, Cooksonia Lang and Steganotheca Edwards. They are briefly compared with others of similar age, and the contribution, particularly of the Whitcliffian assemblage, to our knowledge on the early colonisation of the land by vascular plants is discussed.

Details of the Infra-Cambrian Polarisbreen Series are given, arising partly out of observations made since they were defined in outline by the authors in 1956. The Polarisbreen Series includes the distinctive Polarisbreen Tillite Formation, which correlated with the neighbouring Sveanor Formation of Nordaustlandet and with other tillites in Vestspitsbergen. They are regarded as a product of the great infra-Cambrian ice age and are correlated with world-wide tillite horizons of this age.

A diverse assemblage of Carboniferous plant fossils occurs in the upper Benxi Formation at the Kaihuagou Section near Taiyuan, Shanxi, China. It consists of impressions and fusain fragments, the latter revealing anatomical details. Unlike previously published assemblages from the Benxi Formation, there are no pteridosperms, but a predominance of noeggerathioids and fern fragments. There are three new species: Achlamydocarpon taiyuanensis, Conchophyllum suboblongifolium, and the first reported example of a Selaginella from the Carboniferous System of China, S. benxiensis. The arborescent lycophytes do not to belong to Lepidodendron, as previously claimed, but to Synchysidendron, and three new combinations are therefore proposed for species from the Benxi Formation: S. galeatum, S. subrhombicum and S. tripunctatum. The Benxi Formation flora represents a transition between the tropical swamp vegetation represented in the Westphalian floras of Europe and North America, and the Stephanian and Permian Cathaysian floras of China. It is evidence of an essential continuity between the Late Palaeozoic vegetation of the western and eastern tropics, which should be united as a single phytochorion, the Amerosinian Realm.

New carbon isotopic data from the Devonian of Ardennes (Belgium) and partly from the Holy Cross Mountains (Poland) highlight an abrupt and high-amplitude negative excursion in the punctata conodont Zone. Published information from Moravia and China suggests that this Middle Frasnian negative excursion, jointly with the preceding large-scale positive shift, should be used as a global chemostratigraphic marker. Causation scenarios for this negative ‘punctata Event’ are correlated neither with major biota turnover nor major sea-level changes, but may be related to: (1) the Alamo Impact Event, that led to (2) the massive dissociation of methane hydrates and (3) the rapid onset of global warming.