Since 2002, a collaboration between The Yichang Institute of Geology and Mineral Resources, The Martin-Luther-Universität Halle-Wittenberg (Prof. G.H. Bachmann), the Muschelkalkmuseum of Ingelfingen (Dr. Hans Hangdorn), the University of Bonn (Prof. Martin Sander) and the Geological Museum in Copenhagen has focused on the study of the Middle-Late Triassic of the Guanling area (Guizhou Province, Southwest China). The most spectacular fossils include complete crinoid colonies and marine reptiles found in the Carnian Xiaowa Formation, but screen-washing of sediments from the Yangliujing (Anisian-Ladinian), Zhuganpo (Ladinian/Carnian) and Xiaowa (Carnian) Formations has also yielded various chondrichthyan ichthyoliths. It is these fossils that we present here. Teeth of Polyacrodus contrarius were found in the Ladinian/Carnian, as well as a tooth of ?Parvodus in the Anisian, and a tooth of an indeterminate elasmobranch, possibly with neoselachian affinity, in the Ladinian/Carnian. Dermal denticles similar to Arctacanthus are also relatively common in the Ladinian/Carnian interval. They are much smaller than the Permian Arctacanthus, and the structure of their root prevents them for being hybodont cephalic spines.

Despite over 180 years of study and a number of virtually complete specimens, major questions remain concerning aspects of plesiosaur anatomy. The functional morphology of the trunk region is one of these. Most reconstructions assume that plesiosaurs lacked a sternum, but the Nicholls and Russell model argues the presence of a sternal basket based upon ontogenetic, functional and comparative phylogenetic evidence. Taphonomic and morphological evidence from a number of plesiosaur families is consistent with this model. A specimen representing a new taxon from the Pliensbachian of England corroborates this view.

This specimen also informs the current debate on where the boundary between the cervical and dorsal regions of the vertebral column should be placed, and whether the concept of intermediate �pectoral� vertebrae is either useful or necessary. Pectoral vertebrae are traditionally defined as those in which the rib facet is divided by the centrum-neural arch suture. It is here proposed that the dorsal series in plesiosaurs starts with the first vertebra to bear ribs with cupped distal ends, signifying involvement in the sternal basket, and connection to the sternum. In this particular case, this vertebra is the last of the traditional �pectoral� series. This shows that the concept of the pectoral series as a distinct region of the plesiosaur vertebral column is invalid.

However, the usefulness of this definition of the starting point of the dorsal series remains somewhat unsatisfactory, as it as it requires that a specimen displays a high degree of preservation and articulation.

The Jehol Group of China is composed of three formations in ascending order: the Dabeigou Formation, the Yixian Formation and the Jiufotang Formation, spanning a period from the Late Hauterivian to Aptian (131-120 Ma). Rocks of the Jehol Group are best exposed in western Liaoning and northern Hebei provinces, but the Xinminpu Group, Gansu Province, and the Qingshan Group, Shandong Province, may be lateral equivalents. Together, the Yixian and Jiufotang formations, and their equivalents, have produced numerous lizard specimens (Dalinghosaurus , Mimobecklesisaurus , Pachygenys, Yabeinosaurus), but, as yet, none have been recovered from the older Dabeigou Formation, leaving a gap in the early record.

The Daohugou Bed of Inner Mongolia and Liaoning (well represented at Daohugou, Ningcheng County, southeastern Inner Mongolia) predates the Jehol Group, although its age is controversial. While some have argued for a Middle or Late Jurassic age (based mainly on invertebrates), others regard the bed as younger (latest Jurassic or Early Cretaceous). The Daohugou locality has produced hundreds of well-preserved salamander remains (including a new species of the Jehol salamander Liaoxitriton ) as well as a few dinosaurs and pterosaurs, and a mammal, but has only recently yielded lizards. These are rare but, like the salamanders, they show exquisite soft tissue preservation. The new Daohugou specimens are the oldest lizards currently known from China and represent a significant addition to the emerging picture of lizard evolution in eastern Asia.

The Chalicotheriidae are large, claw-bearing perissodactyls. They are divided into two subfamilies, the Chalicotheriinae and the Schizotheriinae, distinguished by the more derived stage of the cheek dentition but more primitive postcranial anatomy of the latter group. 58 upper cheek teeth of three species representing both subfamilies (Metaschizotherium fraasi , Chalicotherium grande and C. goldfussi ) from 27 m
Middle and Late Miocene vertebrate localities of Southern Germany were analyzed with respect to their mesowear signatures. The localities are situated within the Upper Freshwater Molasse (MN5 to MN9), the Dinotheriensande (MN9), and on the karst plateau of the Franconian Alb (MN6). Mesowear signatures are interpreted as indicating a specific dietary regime and as such to reflect a habitat-driven signal of food availability. In the present study the two cervids Axis axis (chital) and Cervus duvauceli (swamp deer) are the extant reference species classified closest to all Chalicotheriidae investigated. We thus contradict the common assumption of chalicotheres being solely browsers and conclude a mixed feeding dietary regime with substantial components of abrasive food items. This observation is independent from the stratigraphic level. Since grasses were not very widespread before the end of the Miocene, another abrasive food component must have applied for the chalicotheres, such as Equisetum (horsetail), or fruit containing lignine and phytoliths. The resemblance in the diet of representatives of both Schizotheriinae and Chalicotheriinae contrasts with the different morphologies of their terminal phalanges. Thus the often proposed function of the claws in obtaining food must be reconsidered.

The partial skeleton of a long-necked plesiosaur was donated to Lincoln Museum in 1906. Unusually for an historic specimen, it can be placed in a sound stratigraphical context, and be confidently dated as Pliensbachian. Although the specimen has been on display in Lincoln at various times, it is not recorded in the scientific literature. Its preparation and description have been undertaken as a PhD project.

Morphological evidence suggests a close relationship to the Toarcian genera Microcleidus and Occitanosaurus . Cladistic analyses of the plesiosauria place these, with �Plesiosaurus� brachypterygius , as part of a radiation of Lower Jurassic forms from within which the Upper Cretaceous Elasmosauridae are derived. However, character-based support for this relationship is rather weak, and it involves also an interval of about 90 million years from which no related specimens are recorded.

An unusual feature of the specimen is that the spinal column is broken in the middle of the dorsal series, with the posterior portion displaced forward by about 50cm to lie parallel to the anterior portion. Preparation has revealed possible tooth marks in the vertebral column at the base of the neck, and twisting of the of dorsal vertebrae anterior to the break. The presence of ribs which cannot be matched to specific centra suggests that two or three vertebrae are missing from the mid-dorsal region. It is suggested that this is evidence that the animal was attacked by a large and powerful predator, possibly a Rhomaleosaurid pliosaur.

A new dinosaur locality has been discovered in the Yuliangze Formation (late Maastrichtian) of Wulaga (Heilongjiang Province, China). More than 500 hadrosaurid bones, together with well-preserved skin impressions, have been unearthed from a continuous bonebed in a diamictite layer. This facies is interpreted as a sediment gravity flow deposit. The Wulaga quarry can be regarded as a monodominant bonebed: more than 80% of the bones belong to a new genus of lambeosaurine hadrosaurid. Phylogenetic analysis shows that this new taxon is a derived lambeosaurine that forms a monophyletic group with the corythosaur and parasauroloph clades. Besides lambeosaurine fossils, isolated bones display typical hadrosaurine morphology and are also referred to a new taxon. This taxon displays a series of plesiomorphic characters and phylogenetic analysis indicates that it would be the most basal hadrosaurine known to date. It suggests that hadrosaurines, and thus hadrosaurids, would have had an Asian origin, which implies a relatively long ghost lineage, of approximately 13 my, for basal hadrosaurines in Asia. The size-frequency distribution of different types of bones collected in the Wulaga bonebed suggests that a sudden event caused the non-selective mass mortality of the lambeosaurine population, regardless of the age, size and robustness of the individuals.

Excavations directed by the Jura-Museum Eichstätt during the late 1990s in the silicified Upper Jurassic limestones of Schamhaupten (Franconian Alb, Bavaria, Southern Germany) yielded the new, small-sized theropod dinosaur Juravenator starki . The almost complete, articulated skeleton of Juravenator , is the best preserved non-avian theropod in Europe, regarding its completeness as well as its soft tissue preservation.

Juravenator is the second dinosaur ever discovered in the famous Jurassic limestones of the Solnhofen archipelago in Southern Germany – the first discovery was the celebrated Compsognathus some 150 years ago. However, Compsognathus , as well as a number of skeletons of the most ancient bird Archaeopteryx , come from the Lithographic Plattenkalk (Solnhofen Limestones, Lower Tithonian, Malm zeta), a stratigraphic unit slightly younger than the 151-million-year-old Silicified Plattenkalk of Schamhaupten (uppermost Kimmeridgian to Tithonian, Malm epsilon – zeta 1).

Several osteological characters support the assignation of Juravenator to Coelurosauria and other anatomical features place it phylogenetically together with Compsognathus within compsognathids.

Particularly important is the exquisite preservation of portions of soft tissue, which are best observed along the tail and somewhat also along the lower legs. Integumentary impressions in the proximal portion of the tail show a scaly, tuberculated skin; additionally, UV-light illumination visualizes mineralized remains of soft parts along the tail and lower legs. Noteworthy, the soft tissue does not show any evidence of feathers or feather-like integumentary appendages, structures well-known among compsognathids (Sinosauropteryx ) and several other non-avian coelurosaurs (e.g. Caudipteryx , Microraptor, Sinornithosaurus ). The lack of feather-like integumentary coverings in Juravenator complicates current understanding of the evolution of feathers among the dinosaurian predecessors of birds.

Endothermy in non-avian dinosaurs cannot be answered by first level observations but could be inferred from oological observations. Evolutionary trends in eggs, eggshell microstructures, clutches, and nests shapes, unlike bone histology and presence of feathers, provide a powerful proxy that suggests that troodontids, a group of deinonychosaurian theropods found at high latitudes, covered with feathers in Chinese Lagerstätten, and fossilized in a death pose identical to an avian sleeping posture, might have been endothermic.

Re-interpretation of four dubious characters among the mosaic of oological and nesting characters displayed by D . antirrhopus yields interesting results by placing this dromaeosaurid closer to troodontids than oviraptorids, a topology that mirrors the current skeletal hypothesis. This topology is mostly supported by 1) the presence of an egg found outside of the body cavity and apposed to the outer surface of gastralia and 2) the aspect and size of two truncated polar sections in respect to the preserved length of its equatorial region. This observation not only revives the interesting debate of missing or partial data in analyses of fossils and the methodology to deal with such issues but also suggests that D.antirrhopus would have been endothermic. This hypothesis stands although there are missing data in the oological characters of D. antirrhopus and the observed oological features are a mosaic of characters shared by oviraptorosaurids and troodontids. In conclusion, endothermy would be synapomorphic for Deinonychosaurians and, in addition to other characters that were long-held to be exclusively avian, endothermy would have predated the rise of birds.

Reconstructions of the cranial musculature of the two plesiosauroid genera Hydrorion brachypterygius and Seeleyosaurus guilelmiimperatoris from the Liassic Posidonia shale show similarities in the general arrangement of muscles. However, differences in functional anatomy do exist and are probably related to the mode of predation. The musculature of Hydrorion was capable of exerting higher bite forces than that of Seeleyosaurus , whereas in the latter genus the muscles achieved higher speeds for closing the jaws. While the postcranial of the two taxa is remarkable alike, the limbs show distinct differences in length, which leads to divergent swimming abilities. The limb aspect ratio of both genera was calculated, and shows low to moderate values for Hydrorion , indicating a high manoeuvrability, and relatively high values for Seeleyosaurus , which denotes efficiency.

These differences in the feeding apparatus and the locomotion, in comparison with the excellent and abundant fossil content of the Posidonia shale, allows to assign possible prey to each taxon. Hydrorion probably fed on Leptolepis , a sprat-like swarm fish common in certain levels of the Posidonia shale, whereas Seeleyosaurus preyed upon coleoids.

The first well-preserved, partial associated skeleton of Anoplotherium latipes is described from the UK earliest Oligocene. Tibia, femur, ulna and cervical vertebrae are more completely known for the genus than previously. A. latipes and related species and genera are interpreted as facultatively bipedal, extended-limb, high browsers, based especially on the following: pelvis with flared ilia and long pubic symphysis; medially bowed tibiae shorter than femur; trunk vertebrae enlarging caudally; extensive attachment for supraspinous and deltoid muscles for raising the forelimbs; long muscular tail for balance; and large hind-foot processes for attachment of suspensory ligaments. The adaptations are closest to those of ground sloths, although are not exactly like any other bipedal browser living or extinct in combining long tail, unclawed unguals and relatively low intermembral index. This is probably because Anoplotherium had evolved hooves but retained a primitively long tail. Emphasis was therefore on maintaining an erect stance without forelimb support. Having only toe-number differences, A. commune and A. latipes may have been sexual dimorphs. The large Anoplotherium species would have been able to browse 2-3 metres above the ground without competition from other contemporaneous European terrestrial mammals. Bipedal high browsing represents an adaptation previously unrecognised in European Eocene communities.

The vertebral pachyostosis of Carentonosaurus mineaui , an “aigialosaurid” (Mosasauroidea, Squamata) from the Cenomanian (late Cretaceous) of Charente-maritime (S-W of France) corresponds to an osteosclerosis, caused by a slowing or a pause of the chondroclastic and osteoclastic activities. This characteristic, observed also in the other Cretaceous squamates Pachyvaranus crassispondylus and Simoliophis rochebrunei , cannot be observed in any extant squamate. On the contrary, their vertebrae present a very strong porosity linked to the intensity of the osseous reworking accompanying their growth. This particularity of the osseous structure may be regarded as a heterochronic phenomenon, more specifically neoteny. Its link with an adaptation to shallow marine environment appears to be confirmed by the supposed ecology of C. mineaui . Nevertheless, the polarity, plesiomorphic or derived, of this character state within squamates remains unknown. The histological analysis of the periostic bone tissue of the vertebrae of this taxon gives information (asymmetry, rate, cyclic growth) about its way of growth which may be compared to that of P. crassispondylus and S. rochebrunei . The pachyostosis strictly speaking, morphologically observed at the level of the lateral edges of the centrum and maybe at the base of the neural canal, could not be determined histologically on the performed cross sections and seems to be restricted to these vertebral zones.

The vertebrate faunas from Lower and Middle Devonian concretion-bearing beds of Bolivia essentially yield chondrichthyans and acanthodians, whereas placoderms and osteichthyans are virtually lacking. This unusual faunal composition is a thought to be linked to the circumpolar, cool water environments of the “Malvinokaffric Realm”. New braincase material now shows that these chondrichthyans fall into at least two major types, elasmobranch-like and Pucapampella -like, respectively. Pucapampella , a presumed basal chondrichthyan with a complete ventral fissure and a palatobasal articulation, now seems to have had a hitherto unsuspected specific diversity, and new material allows the reconstruction of its complete mandibular arch. In addition a complete, articulated gnathostome braincase recently discovered in the top part of the Emsian-Eifelian Icla Formation (Subandean zone, Sucre area) displays a peculiar character assemblage. Although its braincase morphology is rather chondrichthyan-like, it is not lined with prismatic calcified cartilage, and its mandibular arch elements are covered with denticle-bearing platelets. Its snout ends anteriorly with a pair of sigmoid, denticle-bearing elements that recall the tenacula of the Carboniferous presumed holocephalimorph Harpacanthus . Various interpretations are proposed for this enigmatic skull. It may actually be a Harpacanthus -like form, predating the earliest known evidence of the taxon by 65 Myr, but the lack of prismatic calcified cartilage raises the question of whether it is a holocephalimorph, or even a chondrichthyan. It also bears some resemblance to the equally enigmatic stensioellids, from the Emsian of Germany, long thought to be primitive placoderms, but now tentatively reinterpreted as possible basal holocephalimorphs.

As early as 1908, Launay mentioned that fossil birds had been found in the Palaeocene of central France. These fossils came from the Menat locality (Puy-de-Dôme). However, these birds were only briefly described and never re-described in detail. Of the three birds mentioned in the literature, one is currently held in the collection of the University Claude Bernard Lyon 1, and another in the collection of the Natural History Museum in Paris (MNHN). The third one appears to be lost. In addition, three new fossils have been recovered from the original site of Menat - one isolated foot, one isolated humerus, and a nearly complete forelimb – in addition to an old, never before described specimen that was �found� in the collection of the MNHN. These fossils are described - although their exact phylogenetic affinities are difficult to decipher due to poor preservation, they certainly represent some of the oldest records of �landbirds� known from the fossil record.

Allin�s theory that the reduced postdentary bones and quadrate of non-mammalian cynodonts were not only the morphological homologues, but also the functional equivalents of the mammalian tympanic bone and ear ossicles is tested on the basis of detailed new information of a specimen of Chiniquodon . The anatomy is shown to be a compromise between the respective functional requirements for a persistent, though reduced, stress transmission function of a jaw articulation, and an acoustic transformation function of a middle ear. There was a sound pressure level transformer ratio of about 30, but the mass and compliances of the elements restricted sensitivity to low frequencies, up to perhaps 2kHz. Neither an air-filled tympanic cavity, nor a specialised tympanic membrane were present; snakes and other modern reptiles lacking a tympanic cavity offer a better mechanical analogy than modern mammals for the ear function of a cynodont.

The sensitivity to high frequency sound characteristic of the mammalian acoustic transformer system, with tympanic cavity and tympanic membrane, could only have evolved after the origin of the dentary-squamosal jaw articulation, and was correlated with miniaturisation in the lineage leading to basal mammaliaforms.

New petrosal bones, assigned to Pucadelphys and Andinodelphys from the early Paleocene of Tiupampa (Bolivia) are here described, and provide supplementary information concerning the anatomy of the ear region of those taxa. The re-examination of characters from the petrosal and basicranium usable for a parsimony analysis, shed light on the phylogenetic relationships of the three Tiupampan genera known by complete cranial remains (i.e., Mayulestes , Pucadelphys , Andinodelphys ). The combination of dental, general cranial, and basicranial characters led to two alternative hypotheses. The first hypothesis states that borhyaenoids (including Mayulestes ) are nested within Notometatheria. Pucadelphyds (i.e., Pucadelphys and Andinodelphys ) are the sister group of a clade comprising MHNC 8369 (one isolated petrosal from Tiupampa) and Marsupialia. The second hypothesis favours the paraphyly of "borhyaenoids" (i.e., the exclusion of Mayulestes from borhyaenoids) and the polyphyly of "Notometatheria". In this case, Mayulestes and borhyaenids represent the stem group of a clade including Asiatic, American and Australian metatherians.

This analysis, conducted with combined datasets (i.e., dental, general cranial, and basicranial), highlighted contradictory information brought by dental and cranial characters. This emphasizes the importance of considering a large anatomical complex (here the entire skull) for systematic purposes, that is likely to reflect the mosaic evolution of the characters among metatherians.

Recent excavations of a Late Cretaceous dinosaur bonebed at Blagoveschenk (Far Eastern Russia) bring new information about its taphonomy and the age profile of its lambeosaurine population. The observed mixture of both fine and coarse sediments, without stratification, is typical for sediment gravity flow deposits. It can therefore be postulated that sediment gravity flows, originating from the uplifted areas at the border of the Zeya-Bureya Basin, concentrated the bones. The fossils form a monodominant bonebed: bones of the lambeosaurine Amurosaurus riabinini form 90% of the recovered fossils, suggesting a herding behaviour for this lambeosaurine. The small number of associated skeletal elements indicates that the carcasses would have been disarticulated well before the transportation of the bones. Less than 2% of the bones exhibit potential tooth marks: scavenging activity was therefore limited, or scavengers had not to actively seek out the bones for nutrient intake. The under-representation of small and light skeletal elements, the dislocation of the dental batteries and the numerous fractured long bones suggest that most of the fossils were transported along a relatively important distance. The random orientation of the bones might indicate a sudden end of transport before its stability can be reached. The size-frequency distributions of some bones suggest an attritional mortality profile for the Amurosaurus population. Predators preferentially killed younger specimens. The absence of fossils that may be attributed to nestling to early juvenile individuals indicate that they were segregated from the adults, and could join the herd only when they were half-grown.

The extinct family Lithornithidae constitutes the earliest, well-known clade of palaeognathous birds. They were an important and diverse constituent of the Northern Hemisphere avifauna in the Palaeocene – Lower Eocene. Three genera with a total of eight species are currently recognised from deposits in North America and the North Sea Basin.

The holotype of the type taxon for the family, Lithornis vulturinus Owen 1841, was destroyed during the Second World War. Subsequently, a neotype for this taxon and several new species of Lithornis were described by Houde (1988). However, recent re-analysis of lithornithid material from the North Sea Basin reveals that there is little or no osteological overlap between material preserved in the holotype and the neotype. The original differential diagnoses of the six species placed within Lithornis are founded on extremely weak criteria of relative size and not on osteological characters. For example, recent re-examination of the original holotype of the species L. nasi and new fossil material from the Danish Fur Formation, revealed a number of osteological characters, which are not present in other lithornithids from the North Sea Basin, but are present in the much larger lithornithid Paracathartes howardae from the Lower Eocene of North America. This suggests that the species is probably not referable to the genus Lithornis .

Finally, many specimens, some of them holotypes, previously referred to various species of Lithornis, consist of incomparable or non-diagnostic material. The genus Lithornis , and possibly the entire family is in need of revision and re-diagnosis.

The first mention of fossil footprints from Thailand was made in 1854 by Jean-Baptiste Pallegoix (1805-1862), a French priest, Titular Bishop of Mallus, who was the Vicar Apostolic of Siam between 1841 and 1862. Pallegoix, who was also the author of a useful Thai-French-Latin-English dictionary, mentioned footprints of antediluvian animals seen by him at Phrabat (or Phra Phuttabat, Saraburi Province) in Central Thailand. This discovery was evoked in 1868 by the French explorer Henri Mouhot (1826-1861), who travelled in Siam and Laos between 1858 and 1861. In 1883 the French geographer Elisée Reclus noticed that these footprints had not yet been studied by geologists. 122 years later we went to Phrabat and we could conclude that these Siamese footprints were erosional figures in marine Permian limestones mistakenly considered as fossil footprints.

In the meantime Triassic and Cretaceous vertebrate footprints have been discovered in Thailand. Triassic track makers from Nam Nao (Phetchabun province) were wide-gauge plantigradous animals (high pace angulation, large track width, large metatarsal impression) with tetradactyl mesaxonic pes and tetradactyl ectaxonic manus, using an asymmetrical gait. The systematic affinities of the track makers are not yet clear but they could be some primitive archosauromorphs such as phytosaurs. A new Cretaceous tracksite at Tha Uthen (Nakhon Phanom province) has recently been discovered and now belongs to the Department of Mineral Resources. It has revealed an impressive assemblage of small theropods, crocodiles and ornithopods.

New catfish remains from the Paleogene of Mali, North Africa are described and compared with known species of Nigerium from the Paleogene of Nigeria. The shape and interrelationships of skull bones preserved in the numerous crania have been studied and analysed using various morphometric techniques. The results from the fossils are compared with those from recent bagrid catfish species in order to investigate intra-and inter- specific variation and to see if morphometric analyses can help determine how many species might be present within the Mali and Nigeria catfish faunas.

Herpetotherium is a long-lived fossil marsupial genus from North America and Europe, and it is common in the fossil record anywhere from Early Eocene to Late Oligocene. Most of the specimens attributed to the genus are dental fragments. One particular specimen (AMNH 22304) of the genus is among the few well-preserved fossil marsupial crania known from the Cenozoic. Gabbett (1994) described the external basicranial anatomy of the genus with emphasis on the auditory region of this particular specimen.

Here, I re-examine the same specimen by utilizing a high-resolution microCT technology. MicroCT scan allows me to digitally remove the matrix and study the otherwise inaccessible cranial regions, and complements the previous work. This study investigates the anatomy of the endocranium and the nasal regions, and compares them to various extant didelphids and fossil marsupials.

Stegosauria is a clade of ornithischian dinosaurs characterised by a bizarre array of dorsally projecting dermal plates or spines extending from the cervical region to the distal end of the tail. Although the first stegosaur was discovered in Swindon, U.K., in 1875, and numerous remains have since been recovered from North America, Tanzania and China, as well as locations throughout Europe, taxonomy and relationships within the group are largely unstudied. The first cladistic analysis of Stegosauria based upon direct observation of specimens is presented. In contrast to previous analyses, the resulting cladogram is well resolved, and although relationships are not strongly supported by statistical tests, groupings accord well with the current understanding of palaeogeography during the Upper Jurassic and Lower Cretaceous: North American and British genera are closely related, and the African genus Kentrosaurus is more distantly related. Chinese genera are basal and imply a relict primitive fauna in China during the Middle and Upper Jurassic.

At the beginning of the Oligocene in Europe, following the global climatic change known as Stehlin’s “Grande Coupure”, mammalian communities present a deep change due to the arrival of new exotic taxa. Among small mammals the eomyids (Mammalia, Rodentia), that are already known from the American and Asian Eocene fossil record, appear in the European fossil record with the single species Eomys antiquus . Previous studies (Comte & Vianey-Liaud 1989; Fahlbusch 1975) have proposed that the diversification known for this family from the Late Oligocene and the Miocene onward is the result of an evolution from this Early Oligocene species.

New material from Southern Germany and Southeastern France, especially two new taxa from the Early Oligocene, evidences that the diversity of this family is more deeply rooted and that the principal forms of eomyids known in the Late Oligocene were already present in Europe at the beginning of the Oligocene, as hinted by Engesser (1990) based on some lineage recognized within Swiss localities. A preliminary quantitative study based on rodent communities from Bavaria, including this new material, also emphasizes this drastic faunal change at the beginning of the Oligocene.

Recently discovered Cretaceous bird fossils, particularly from Liaoning in China, have provided important insights into the theropod-avialan transition and the evolution of flight. Nonetheless, little is known about how the avian brain evolved in response to the development of flight, largely because very few avian fossil endocranial casts are known. CT and 3D-based analysis of two Lower Eocene neornithines from the London Clay of England, Odontopteryx toliapica and Prophaethon shrubsolei, demonstrate that they possessed brains comparable in size and shape to those of living seabirds, indicating that the bird brain had reached an evolutionary level close to that of Recent species by that time. However, poor development of the eminentia saggitalis (which is responsible for many advanced functions including binocular vision) in both species, especially in Odontopteryx , shows that important telencephalic features characteristic of extant clades had not developed fully by the Eocene. These data nevertheless represent the earliest evidence of the avian eminentia saggitalis in the fossil record and have important implications for the evolution of avian cognitive ability. Our results support earlier hypotheses of a general increase in avian brain size over time, but indicate a trend toward diversification in telencephalic architecture that culminated in the extreme cerebrotypes today seen in, for example, Psittaciformes, Strigiformes and some Passeriformes.

>Synodontis (Mochokidae, Siluriformes) is a diverse freshwater catfish genus living in Africa. It is represented in all the ichtyological provinces except Maghreb and South Africa. Like other Siluriformes Synodontis fish are characterised by their pectoral and dorsal spines and by the presence of maxillary and mandibular barbells. The Synodontis fossil record starts in the Middle Miocene (Tunisia) and continues in the Mio-Pliocene mainly in East Africa (e.g. Kenya, Uganda). These fossils are rarely identified at a specific level. In fact, criteria of identification of the species are based on characters which are not preserved in the fossil (e.g. morphology of the barbells).

Here, we present the Synodontis fishes associated with the oldest known hominid, Sahelanthropus tchadensis (Brunet et al ., 2002). They were collected by the Mission Paléoanthropologique Franco-Tchadienne in the Late Miocene of Toros-Menalla, Chad (Vignaud et al ., 2002). This first Synodontis assemblage from Central Africa consists of abundant and well-preserved specimens. To identify this material, we have constructed the first identification key at a specific level. It is based on osteological characters and addressed to the well-preserved bones of the fossil record i.e. cleithrum, pectoral and dorsal spine, nucal plate, supraoccipital and frontal. The Synodontis fishes from Chad show a high taxonomic diversity including both living and fossil species. Moreover, this study reveals specific associations depending on the localities; this could testify to environmental particularity of these localities. Such diversity has never been revealed in the Synodontis fossil record.

According to the generally accepted explanation, alligatoroids first appeared in North America during the Late Cretaceous and soon after migrated to Europe from the direction of the Atlantic Ocean. It is supported by the occurrence of two phylogenetically basal species, Leidyosuchus canadensis and Deinosuchus rugosus in the Campanian and Maastrichtian sediments of North America.

In Europe at least two Late Cretaceous alligatoroid taxa have been described: Acynodon and Musturzabalsuchus , discovered in the area of the former Mediterranean. Recently the Hungarian dinosaur bearing locality in the Bakony Mountains has yielded several isolated alligatoroid mandible remains. Some of the autapomorphic characters of the group are present on these remains (laterally shifted foramen aërum on the articular) and comparison with other Late Cretaceous taxa suggests that the Hungarian alligatoroid is probably a new taxon. The biogeographical importance of the discovery is given by the age of the material which is Santonian, so far the oldest alligatoroid remains ever reported. Some of the characters on the dentary (such as the separated 3^rd and 4^th alveoli) indicate that the Hungarian alligatoroid is more closely related to the derived, blunt-snouted alligatorids like Brachychampsa, Stangerochampsa and Acynodon than to the basal Alligatoroidea group (including Leidyosuchus and Musturzabalsuchus) . These results question the North American origin of Alligatoroidea as by this new discovery the oldest member of the group is presently known from Europe, nevertheless this taxon morphologically seems to be a derived form.

Snakes are elongate and very flexible animals. However, in erycine Boidae (‘sand boas’), the posterior part of the tail is not flexible. The vertebrae of this portion of the vertebral column of erycines are markedly shortened and they bear additional apophyses that form extra intervertebral articulations, in fact non-flexible joints. A snake from the Eocene, Cadurceryx filholi , has posterior caudal vertebrae similar to those of extant Erycinae; but in this snake, additional apophyses are present also on more anterior vertebrae: they are present in the whole caudal region and on vertebrae from the trunk. In the original description of Cadurceryx , it was presumed that, aside from the caudal region, only the posteriormost trunk vertebrae were provided with additional apophyses. But, new material has shown that the vertebral column of Cadurceryx was entirely composed of vertebrae with additional apophyses, except perhaps the anteriormost vertebrae (‘cervical’ region).

A living snake, the African Mehelya (Colubridae), has rather similar trunk vertebrae. It displays usual habits and movements of snakes, but its additional apophyses do not form additional intervertebral articulations. On the contrary, in Cadurceryx , the apophyses are comparatively longer and provided extra intervertebral articulations in at least a part of the trunk region. Therefore, the movements of Cadurceryx were likely limited to a degree that cannot be estimated. Cadurceryx was first assigned to the erycine Boidae, but its trunk vertebrae cast doubt on this referral. No skull bone may be confidently attributed to this snake and its affinities remain enigmatic.

How well do our Finite Element model results reflect reality? We will never be able to definitively answer this question in extinct animals, but we can phylogenetically and functionally bracket extinct animal FE-models with data from living animals detailing (a) how well FE-models replicate experimentally recorded in-vivo/vitro bone strain; and (b) which parameters matter the most for accuracy. FE-model validation against bone strain data is currently underway for macaques and crocodiles. To close the phylogenetic bracket around dinosaurs and pterosaurs, extant birds are an obvious candidate for FE-validation.

Here I present the first FE-model of a bird skull and its biomechanical behaviour during pecking and biting. Ostriches are chosen as a model system as they are large and retain a paleognathous palate and open cranial sutures in-keeping with their theropod ancestors. Loadings are applied to the beak based on (i) preliminary pecking forces recorded in-vivo, and (ii) biting forces calculated from musculoskeletal architecture. Results show that the behaviour of the skull is fundamentally different during pecking versus biting. The jugal bar is excessively stressed, particularly during biting, but is alleviated by introducing soft tissues and passive kinetic movement at sutures and joints. Results are useful on two counts in that they (1) suggest that a deviation from biting towards pecking behaviour in early avian evolution can be detected from characters of bony morphology, and (2) provide an idea where to attach strain gauges in the next step of the research – in-vitro strain measurement and comparison to FE-model derived strain.

The Norian Calcare di Zorzino (Zorzino Limestone) yield an interesting vertebrate fauna, comprising mainly fishes, but also reptiles were well represented. Although it was deposited amid a carbonate platform, terrestrial reptiles are more abundant and diverse than aquatic ones, with archosauriforms, protorosaurs, drepanosaurids and sphenodontids, in addition to pterosaurs. Several lines of evidence suggest that these reptiles lived on small islands that surrounded the basins in which the Calcare di Zorzino was deposited. Preservation is usually good and most specimens are fairly complete and articulated. In most cases however, no traces of soft parts are preserved, but there are some remarkable exceptions. In fact, skin patches are present on the holotypes of Drepanosaurus unguicaudatus , of Vallesaurus cenensis , and of Langobardisaurus pandolfii, as well as on one specimen of Megalancosaurus preonensis . Preliminary investigation shows that the skin is not preserved as impression but scales are actually fossilized; that all specimens with fossilized skin belong to terrestrial taxa and the preserved portion in most cases corresponds to the tail or pelvic region (but soft tissues are present also in the shoulder region of Drepanosaurus ); and, finally, that specimens with fossilized skin were found only in the more marly and thinly laminated beds, while the skin is not preserved in those specimens found in thicker and more calcareous beds.

The detailed morphology of the lower teeth of ringtail possums (Marsupialia: Pseudocheiridae), particularly m1, is highly significant in determining species. Several new fossil pseudocheirid species have been recognised on the basis of lower teeth from the Oligo-Miocene deposits of Riversleigh, Australia. However, there is an equally large number of upper dental specimens, including two skulls, that are more difficult to identify to species, let alone positively match with an occluding group of lower teeth. Upper teeth are known for some previously identified fossil pseudocheirids, but pairings have been aided by low species diversity in their localities and distinct size correlations - little has been discussed of their diagnostic features. Species-level resolution of the upper teeth holds potentially valuable information regarding phylogeny and the palaeobiology of pseudocheirids.

A morphometric prediction method has been developed to help identify the upper teeth for fossil species. Using paired measurements of upper and lower molars of extant pseudocheirid species, regression analyses were performed and prediction equations calculated. As expected, a high correlation was found between matched occluding molars, and %SEE and %PE were also calculated to test prediction reliability. Initial application of the prediction equations to lower teeth of fossil species has satisfactorily narrowed the possible selection of almost 300 matching upper dental specimens, but only functions as a first step. Ultimately, matching of occluding teeth, and species identification of upper teeth must be based on morphology and more specific morphometric analyses.

The Osteostraci (Cephalaspids) have a wide range both spatially and temporally. They share a last common ancestor with the jawed vertebrates which represents one of the most crucial stages of vertebrate evolution – the transition from jawless to jawed vertebrates. Osteostracan phylogeny therefore impacts not only upon our understanding of the acquisition of many key vertebrate characters but also ideas of palaeobiogeographic evolution across Laurussia and Siberia.

An in-depth phylogeny for the group is lacking and little consensus has been reached so far. Previous studies have been limited by out-dated methodology, taxonomic scope, or accuracy of findings. Here, novel observations are used to construct a computer generated phylogeny including recent new finds. The phylogeny impacts upon palaeobiogeography, osteostracan evolution and development of paired appendages.

The ancestral osteostracan state has broad implications for not only the group but also general early vertebrate evolution and gnathostome origins. The new phylogeny lends support to the long disputed claim that the tremataspids are derived rather than ancestral and thus supports the placement of Osteostraci as sister to the jawed vertebrates. However, the phylogeny indicates that the evolution and development of pectoral fins is more complicated than previously envisaged, with losses and potential reversions occuring within the group.

Osteostraci taxa generally demonstrate strong endemism and as such have a strong palaeogeographical signal. This new phylogeny enables cladistic palaeobiogeographic reconstruction of dispersal and vicariance events occuring across the arctic. The Osteostraci are found to have originated and diversified in South Norway and Britain.

Plesiosaurian remains from non-marine sediments are not very common, but such occurrences have been reported from several different horizons in various countries. In Canada, two Upper Cretaceous non-marine

units, the Dinosaur Park Formation and Horseshoe Canyon Formation in Alberta, have yielded plesiosaurian fossils. A previous study of the plesiosaurian fauna of the Dinosaur Park Formation (Campanian) confirmed their non-marine occurrences, and the dominance of elasmosaurids relative to polycotylids was demonstrated. The taxonomic status of the specimens from the non-marine Horseshoe Canyon Formation (Campanian-Maastrichtian, Upper Cretaceous), Alberta, Canada was recently reassessed. The holotype of Leurospondylus ultimus and a previously undescribed partial skeleton from Red Deer River Valley represent indeterminate elasmosaurids, whereas the “plesiosaurian” vertebra from Edmonton is actually an ornithopod caudal. The two plesiosaurian specimens came from different horizons for which different environments are inferred. The larger individual may be closely related to Elasmosaurus . These studies support the hypothesized decline of polycotylid plesiosaurs in the North American during the Late Campanian. Well-preserved plesiosaurian specimens from non-marine sediments are usually of juveniles and small adults, and it is suggested that small adults and juveniles existed in both marine and non-marine environments whereas large adults were limited to the former.

Stereoscopic microwear and mesowear analyses have proven very useful as independent tools for paleodietary reconstruction to test paleodietary inferences drawn from the study of craniodental morphology in extinct species. These techniques lend themselves particularly well to dietary studies involving large samples and spanning vast tracks of evolutionary time. Here, we compare dental microwear (i.e., immediate wear) and mesowear (i.e., cumulative wear) results for both Tertiary and Quaternary representatives of the families Antilocapridae, Camelidae, Dromomerycidae, Equidae, and Oreodontidae. Results offer insights into the origin of hypsodonty in certain lineages and also show that the classic story in paleontology regarding the timing of changes in mammalian communities in North America due to the spread of savanna grasslands replacing more closed habitats is greatly oversimplified. For example, peak shifts toward more abrasive diets apparently occurred in the earliest Miocene in Equids but not until the late Miocene in Antilocaprids and Dromomerycids. Results also indicate that the widely ascribed ideas of brachyodonty being indicative of browsing, mesodonty being indicative of mixed feeding, and hypsodonty being indicative of grazing are in need of serious revision and more applicable to extant versus extinct ungulates.

The first undisputed record of mammals from Wealden Group (Early Cretaceous) strata of Britain dates from 1911. Subsequently, no Wealden mammal fossils were found until a sustained search was commenced in 1960. This resulted in the recovery of a number of new specimens from the Valanginian of mainland Britain. During the 1970s, in a further attempt to isolate Wealden mammal remains, a number of workers also undertook bulk screening of Barremian strata exposed on the Isle of Wight. Only two multituberculate teeth were found and work to recover mammal fossils from the Wealden of the Isle of Wight ceased. Despite this unpromising start, recent comprehensive bulk screening of Isle of Wight Wessex Formation strata has resulted in the recovery of a diverse microvertebrate fauna including at least six mammals. Among these are a new spalacolestine spalacotheriid representing the first European record of the Spalacolestinae. Hitherto, five species (one unnamed) of spalacotheriid mammal were known from the Lower Cretaceous of Britain. All are referred to the genus Spalacotherium but it is now evident that these referrals should be treated with caution. Furthermore, the new Wessex Formation spalacotheriid and recently described spalacotheriids from the ?Barremian of Japan, and the Barremian and Aptian of China exhibit combinations of characters that suggest that these tiny mammals were more diverse and that their evolution was more complex than previously recognized. The discovery of a spalacolestine in the Barremian Wessex Formation supports the concept of faunal interchange between Europe, Asia and North America during the Early Cretaceous and possible derivation of North American spalacotheriids from a European or Eurasian ancestor.

The largest phosphate basin of Morocco, the Ouled Abdoun Basin, has yielded a very rich vertebrate fauna, ranging in age from the latest Cretaceous (Maastrichtian) to the Early Eocene (Ypresian). Turtle remains are abundant there; they include both articulated skeletons and isolated elements. This turtle assemblage shows considerable diversity. More than a dozen taxa, mostly based on skulls, have been collected, belonging to both cryptodires and pleurodires. The cryptodiran turtles are represented by cheloniid sea turtles (Osteopygis emarginatus from the Danian beds, Tasbacka ouledabdounensis from the Thanetian beds), and a dermochelyid from the Maastrichtian beds. The pleurodires consist of several taxa of bothremydids, including Phosphatochelys tedfordi from the Ypresian beds. All turtles hitherto recorded from this near-shore marine phosphate sequence are marine forms. The remarkable diversity in skull morphology, especially among bothremydids, varying from a very long, narrow snout to a wide, telescoped short face, and from a narrow triturating surface to a very large secondary palate with deep pits, reveals sensory modifications and adaptations in feeding mechanisms.

If pterodactyloids had never existed pterosaur evolutionary history would have been a truly modest affair that faded out in the Early Cretaceous. As it is, following their debut in the Late Jurassic, pterodactyloids became remarkably diverse, achieved giant size, and seem to have been the dominant group of flying vertebrates throughout the Cretaceous. To understand this evolutionary success story we need to gain insights into the origin of pterodactyloids, and to do this we must first establish how pterodactyloids are related to more basal forms ('rhamphorhynchoids'). Traditional schemes are vague and reveal nothing of consequence regarding this relationship. Phylogenetic analyses, by contrast, largely concur in grouping pterodactyloids with Rhamphorhynchus (or Rhamphorhynchidae), with all other 'rhamphorhynchoids' lying basal to this clade. The problem with this arrangement is that it requires several character state reversals (rhamphorhynchids and more basal clades such as campylognathoidids share various apomorphies not found in pterodactyloids) and reveals very little regarding the origins of typical features of pterodactyloids. An alternative arrangement, supported here, unites pterodactyloids with anurognathids, an unusual group of 'frog-headed' insectivorous pterosaurs that lie at or near the base of the pterosaur tree. If correct, this hypothesis means that classic pterodactyloid apomorphies (e.g. loss of cervical ribs and short tail) occurred early in pterosaur evolution and are not restricted to pterodactyloids. Moreover, the lineage leading to anurognathids + pterodactyloids must have appeared before the Jurassic and most or (if anurognathids are the most basal clade) possibly all long-tailed pterosaurs belong within a distinct clade.

The Late Jurassic faunas of the Morrison Formation (USA) and Tendaguru (Tanzania) apparently share several dinosaurian genera (e.g. Elaphrosaurus , Barosaurus , Brachiosaurus and Dryosaurus ). This faunal similarity has been interpreted as evidence for a terrestrial connection between North America and Africa during the Late Jurassic, which allowed regular dispersal/ migration between the two areas. Faunal dissimilarities have been explained as either sampling error or palaeoecological differences. These interpretations can be challenged on a number of grounds. First, although palaeogeographic reconstructions are not unanimous, the majority depict marine barriers between North America and Africa from ca. the Oxfordian onwards. Second, taxonomic revision has undermined the evidence for generic-level similarities between the two faunas (e.g. “Barosaurus “ africanus from Tendaguru has recently been re-interpreted as Tornieria africana , and there are doubts about the congeneric status of the Morrison and Tendaguru Brachiosaurus species). Finally, cladistic biogeography sheds light on the biotic relationships between North America and Africa. Area cladograms for the Late Jurassic have placed North America and Africa as sister areas, but this is compatible with either a land connection (allowing dispersal) or separation (promoting vicariance). Furthermore, analysis using an updated data-set supports a closer relationship between Europe and North America compared to Africa in the Late Jurassic. Thus, there is little biogeographic or palaeogeographic evidence for faunal exchange between North America and Africa in the Late Jurassic. Instead, a more plausible explanation for the observed distribution patterns requires the formation of a widespread biota during the Middle Jurassic, followed by separation and vicariance in the Late Jurassic.

A quantitative and qualitative study has been undertaken of surface modifications on enamel, dentine and bone of micromammals from a single bed in the Osborne Member, Headon Hill Formation. The theridomyids, glirids, marsupials and nyctitheres were living in close association with a small floodplain pond in which the fossiliferous sediments accumulated. The primates, bats and apatothere were living further away from this pond, probably in forested patches on the floodplain. The pantolestid shows evidence of longer distance waterborne transport. The majority of theridomyids, many marsupials, fewer glirids and even fewer nyctitheres were locally predated/scavenged. The amphicyonid Cynodictis cf. lacustris is identified as the predator/scavenger of the theridomyids and marsupials, based on bite marks and fragmentation patterns, whereas the possibility of the same predator/scavenger for the glirids and the nyctitheres is not excluded. Lesser predation/scavenging of the nyctitheres and glirids may be due to their scansorial locomotion. The primates were eaten by an animal that caused little fragmentation or corrosion of their skeletal elements: probably a carnivorous bird. The bats, apatothere and pantolestid also show indications of predation/scavenging, but the identity of the predator/scavenger is not clear. A crocodile or other reptile predator is excluded as all cause complete demineralization of their prey. Large creodonts are possible predators/scavengers for some mammals, especially the larger pantolestid, but there is no evidence of this interaction. This detailed taphonomic analysis of individual species in a single bed has resulted in a comprehensive understanding of the accumulation of the micromammalian assemblage.

From 1986 to 1998, no new lissamphibian fossils were reported from China, and all known material was collected from Cenozoic deposits. The past eight years have witnessed an acceleration in the study of Mesozoic frogs and salamanders in China. This has been based on good new material (including a large number of articulated skeletons and exquisite soft tissue impressions) from the tuff-interbedded lacustrine deposits of northeastern China (Liaoning, Hebei and Inner Mongolia). Five frogs (Callobatrachus sanyanensis , Dalianbatrachus mengi, Liaobatrachus grabaui , Mesophryne beipiaoensis , Yizhoubatrachus macilentus ) and seven salamanders (Chunerpeton tianyiensis , Jeholotriton paradoxus , Laccotriton subsolanus , Liaoxitriton daohugouensis, L. zhongjiani, Pangerpeton sinensis, Sinerpeton fengshanensis) have been named. Most of them are from the Lower Cretaceous Jehol Group (i.e., Dabeigou, Yixian and Jiufotang formations in ascending sequence), whereas four salamanders (Chunerpeton, Jeholotriton, Liaoxitriton daohugouensis, Pangerpeton ) come from a lower horizon, the Daohugou fossil bed, argued as Middle Jurassic to Lower Cretaceous by different workers. Among the recent amphibian discoveries have been frog skeletons with three-dimensional preservation and salamander specimens representing different stages of ontogenetic development (larva to adult). Phylogenetic analyses suggest that the frogs are advanced over typical Jurassic frogs worldwide, and the salamanders advanced over Middle–Late Jurassic early caudates from England and Central Asia. Together with the findings from the Lower Cretaceous Tetori Group (Japan), these may represent one or two early adaptive radiations of East Asian lissamphibians, which were characterized by high taxonomic diversity and the first records of living amphibian families (e.g., Discoglossidae, Hynobiidae and Cryptobranchidae).

In recent years the Cedar Mountain Formation of Utah has become famous for vertebrate remains. Only a few tracksites have been reported, probably due, in large part, to the fact that the unit is made up largely of mudstones. Most tracks have mainly been the usual iguanodontid-dominated assemblage typical of the early Cretaceous. However, in recent years, a more diverse track assemblage was reported from Arches National Park, which included the tracks of sauropods, and possibly didactyl theropods, e.g. dromaeosaurids. These tracks are very deeply impressed in what was evidently a wet muddy cohesive substrate. In addition to these, at a lower stratigraphic level, some enigmatic traces are also preserved, on a ripple-marked sandstone layer. These traces consist of sets of elongate scratch marks radiating out from a central area and forming ovals or paired crescents 8-10cm in maximum dimension. These have been variously interpreted as tool marks, invertebrate traces or pterosaur feeding traces.

Outside the park, bird tracks were discovered in the Poison Strip Sandstone in summer 2005. The tracks are preserved on three fallen blocks, although distinctive lithology allowed identification of the track horizon. All the tracks are preserved in concave epirelief on a thin fine-grained sandstone layer at the top of a metre thick erosional based coarser sandstone with a rippled top surface, which may be of crevasse splay origin. The tracks are approximately 30-40mm in length with no traces of interdigital webbing. These are the earliest known bird tracks in North America.

These tracksites provide a more complete view of the early Cretaceous vertebrate assemblages in western North America during the Early Cretaceous. This part of the Cedar Mountain Formation is thought to have been deposited on the margins of a large lake as channels, crevasse splays and lakeshore/floodplain sediments.