Carnosaur - Lower Cretaceous - Japan

A new carnosaur (Dinosauria: Theropoda) from the Lower Cretaceous of Japan[1]

Yoichi Azuma and Philip J. Currie

Abstract: An associated skeleton from the Kitadani quarry of the Lower Cretaceous (Albian) strata in Fukui Prefecture,Japan, represents a new genus and species of carnosaur (Dinosauria: Theropoda). The immature specimen, which is about 4.2 m long, is well preserved and includes elements of the skull, vertebral column, front limbs, and hind limbs. The hands are relatively large and armed with strongly curved, sharp claws, suggesting that it was an active predator. Phylogenetic analysis indicates it is a basal allosauroid. An isolated astragalus from Australia may represent a closely related species.

Résumé : Un squelette provenant de strates du Crétacé inférieur (Albien) dans la carrière Kitadani, de la préfecture de Fukui, au Japon, représente un nouveau genre et une nouvelle espèce de carnosaure (Dinosauria : Theropoda). Le spécimen immature, d’une longueur d’environ 4,2 mètres est bien conservé; il comporte des éléments du crâne, de la colonne vertébrale ainsi que des membres antérieurs et postérieurs. Les mains sont relativement larges et munies de griffes acérées, fortement recourbées, suggérant qu’il était un prédateur actif. Une analyse phylogénétique indique qu’il s’agit d’un allosauroïde basal. Un astragale isolé de l’Australie pourrait représenter une espèce étroitement liée.

[Traduit par la Rédaction] Azuma and Currie 1753

Introduction

      Over the last decade, many terrestrial vertebrate fossils (Hirayama 1996) have been recovered from the Tetori Group of the Hokuriku region of central Japan (Azuma et al. 1995;  Azuma and Tomida 1997). The Tetori Group is divided into three subgroups (Kuzuryu, Itoshiro, and Akaiwa), which range in time from Bajocian to Albian (Maeda 1961) and in sedimentary environments from marine to fluvial. The most productive Japanese dinosaur sites are found in Fukui and neighboring prefectures in the Kitadani Formation (Albian), which is the uppermost level of the Akaiwa subgroup. Rivers under a moist, temperate climate laid down the alternating sandstones, mudstones, and tuffs. The best Kitadani locality is on the Sugiyama River, within the city limits of Katsuyama in Fukui Prefecture (Fig. 1). The quarry was worked by the Fukui Prefectural Museum between 1988 and 1993 and from 1996 to 1998. Enough bones and teeth were recovered from the bottom of a narrow ravine to justify the expense of heavy equipment to take down the overburden. The back wall of the quarry is as much as 45 m high, making it one of the largest dinosaur quarries in the world. Fossils were recovered from several layers on the way down to the bonebed. Dinosaurs identified so far from the Kitadani Formation include at least one sauropod, an iguanodontian, and a possible psittacosaurid. Footprints include those of theropods, birds, sauropods and iguanodontians (Azuma and Takeyama 1991).

      Indeterminate amphibian and small reptile tracks provide some indication of other types of animals that were living in the region with the dinosaurs. Fossil plants, freshwater mollusks, fish, turtles, and crocodiles have been found in the same beds, giving the most complete record of a dinosaurian ecosystem presently known in Japan. The recovery of a nearly complete crocodile skeleton and a partial skeleton of a theropod shows that articulated and associated skeletons can be collected at this site and gives the promise of more exciting discoveries in the future. Enough material has been recovered from the quarry to produce a composite skeleton of the iguanodontid. This and other major specimens can be seen on exhibit in the Fukui Prefectural Museum. A variety of theropod teeth have been collected in the quarry. The larger ones are flattened and bladelike, and are reminiscent of the central Asian tyrannosaur Alectrosaurus. Smaller teeth are similar in size and shape to those of dromaeosaurids, but do not have the peculiarities of denticulation found in either Dromaeosaurus or known velociraptorines. Nevertheless, teeth from the Katsuyama quarry are identified as dromaeosaurid on the basis of a characteristic twist in the anterior carina (Currie et al. 1990), and at least one dorsal vertebra (FPM 96082619) is suggestive of dromaeosaurid affinities. In 1993, a right manual ungual (I-2), a right astragalus, and a right metatarsal III were found in a small area of the Kitadani quarry, and were recognized as part of an associated skeleton. Their discovery led to the reopening of that section of the quarry in 1996 and 1997, and many additional theropod bones were recovered. 

      To ascertain whether or not the elements were likely to belong to the same individual, a series of measurements Can. J. Earth. Sci. 37: 1735–1753 (2000) © 2000 NRC Canada 1735 Received September 3, 1999. Accepted June 7, 2000. Published on the NRC Research Press Web site on December 1, 2000. Paper handled by Associate Editor H.-D. Sues. Y. Azuma. Fukui Prefectural Museum, Omiya 2-19-15, Fukui City 910, Japan. P.J. Currie.1 Royal Tyrrell Museum of Palaeontology, Box 7500, Drumheller, AB T0J 0Y0, Canada. 1Corresponding author (e-mail: pcurrie@mcd.gov.ab.ca). © 2000 NRC Canada 1736 Can. J. Earth. Sci. Vol. 37, 2000 from this specimen were compared with the same measurements in a wide range of theropods. The femur (507 mm long) was used as the standard against which all other measurements were compared, because this bone is one of the least variable amongst theropod dinosaurs (Christiansen 1998). The length of the coracoid (which was found away from the main concentration of theropod bones) falls within the 95% confidence limits of a theropod with a 507 mm long femur, although it is at the lower extreme of the limits. Comparison of the humerus with femur length produced a figure that falls within the expected range of this measurement for all theropods. Although there is a considerable amount of variation in interspecific comparisons of theropod front limbs, the humerus compares closely in relative length with theropods like Allosaurus, Eustreptospondylus, and Gasosaurus. Ulnar length falls on the regression line for this measurement in Allosaurus, Ceratosaurus, compsognathids, coelophysoids, Herrerasaurus, and oviraptorosaurs. The relative length of this element is greater in Deinonychus and ornithomimids, but is significantly shorter in Acrocanthosaurus, Carnotaurus, and tyrannosaurids. Like other bones of the forelimbs, manual phalanges show a wide range of variability. Whereas the length of the second phalanx of the second finger is close to what would be expected in an average theropod, the manual unguals (I-2, II-3) are larger than most and cluster with those of Deinonychus, Allosaurus, and Alxasaurus. Hind limb elements have more interspecific stability. The maximum width of the proximal head of the tibia, the shaft and distal widths of the fibula, the lengths of metatarsals II and III, the width of the astragalus, and the lengths of pedal phalanges III-1 and III-2 all fall within the 95% confidence limits for these measurements in theropods. Given the fact that the lengths of all elements fall within the expected range of size variability for a theropod 4.2 m in total length, that they were recovered from the same region of the quarry, and that no elements are duplicated, it is highly probable that they represent a single individual (Fig. 2). Bivariate analyses of theropod measurements are in the process of being refined and published by the second author. The term Carnosauria was first used by von Huene (1920, 1926) to depict all large theropods excluding Ceratosaurus and tyrannosaurids. It has been used consistently since that time, although the taxa included have fluctuated. In this paper we follow Hutchinson and Padian (1997) in defining “Carnosauria” as all theropods closer to Allosaurus than to birds.

      Fig. 1. Quarry map showing positions of theropod bones recovered. The cluster of bones centered on H-4 includes the material assigned to the holotype of Fukuiraptor kitadaniensis. Bones clustered around C-4 include juvenile Fukuiraptor material that is under preparation. The arrow in the first inset points to the location of the quarry west of Fukui City. The second inset shows the position of the second inset on the West Coast of the island of Honshu, Japan. Systematic palaeontology Dinosauria Owen, 1842 Theropoda Marsh, 1881 Carnosauria von Huene, 1920 Fukuiraptor gen.nov. Etymology "Fukui" refers to the prefecture in the central part of Honshu, Japan, where the specimen was found; "raptor" is Latin for robber. Genoholotype FPMN 97122 (associated skull and limb bones) plus FPMN 96082443 (left humerus). Generic Diagnosis Monospecific genus. See Species Diagnosis. Fukuiraptor kitadaniensis sp.nov. Etymology The species is named after the Lower Cretaceous Kitadani Formation. Holotype FPMN 97122 is an associated skeleton found in one small area of the Kitadani quarry. The lengths of all bones belong to an individual that was an estimated 4.2 m long when it was alive. The associated skeletal parts include (with their element numbers) a left maxillary fragment (FPMN 9712201), right dentary fragment (FPMN 9712202), four isolated teeth (FPMN 9712203, 9712204, 9712205, 9712206), dorsal centrum (FPMN 9712207), distal caudal (FPMN 9712208), both humeri (FPMN 96082443, 9712209), right ulna (FPMN 9712210), right manual ungual I-2 (FPMN 9712211), right manual phalanx II-2 (FPMN 9712212), left manual ungual II-3 (FPMN 9712213), fragments of the right ilium (FPMN 9712214), one pubis (FPMN 9712215 shaft, and possibly 9712216) and both ischia (FPMN 9712217, 9712218), the right femur (FPMN 9712219), the proximal half of the right tibia (FPMN 9712220),the distal half of the right fibula (FPMN 97080206), right astragalus (FPMN 9712221), right metatarsal I (FPMN 9712222), right metatarsal II (FPMN 9712223), right metatarsal III (FPMN 9712224) and pedal phalanges left I-1 (FPMN 9712225), right III-1 (FPMN 9712226), right III-2 (FPMN 9712227), and right IV-2 (FPMN 9712228). Referred specimens FPMN 9712229 is a fragment of the left maxilla with two complete and one partial alveoli, whereas FPMN 9712230 is part of a dentary with three alveoli. FPMN 9712231, 9712232, 9712233, 9712234, 9712235, 9712236, 9712237, 9712238, and possibly 9712239 are isolated, shed teeth. FPMN 9712240 is a cervical centrum tentatively referred to Fukuiraptor. A partial cervical neural arch (FPMN 9712241) fits onto the centrum, and even though it was found in a different part of the quarry, it might be from the same individual. An almost complete dorsal neural arch (FPMN 9712242) may also belong to this genus. A theropod coracoid (FPMN 9712243) is the right size to belong to the holotype and was recovered from the same level as the main concentration of theropod bones, but was found five metres away in a different part of the quarry. Locality and age The specimen was recovered from the Lower Cretaceous (Albian) Kitadani Formation (Akaiwa subgroup, Tetori Group). The Kitadani locality is on the Sugiyama River in the northern part of the city of Katsuyama, Fukui Prefecture (Latitude 36°7¢N, Longitude 136°33¢E). Diagnosis Medium-sized theropod (holotype has an estimated length of 4.2 m, Fig. 2). Carnosaur with the following autapomorphies: fused interdental plates in maxilla and dentary (convergent with dromaeosaurids); narrow dentary; narrow bladelike teeth with oblique blood grooves (convergent with tyrannosaurids); ratio of ulna to humerus is 0.92, which is significantly higher than any other carnosaur, except © 2000 NRC Canada

      A new carnosaur (Dinosauria: Theropoda) from the Lower Cretaceous of Japan[2] Yoichi Azuma and Philip J. Currie

      Azuma and Currie 1737

      A partial centrum (FPMN 9712207) of a mid-cervical vertebra, possibly the fifth, is 58 mm long. This length does not include the convex anterior intervertebral articulation, which was not preserved. The posterior intervertebral articulation is deeply concave, which shows that the cervical centra were opisthocoelous. The parapophysis is preserved on the lower margin of the left side, and extends ventral to the lower edge of the anterior intervertebral articulation on the midline. The ventral surface of the centrum is concave between the parapophyses. The lateral surface of the centrum is concave dorsal to the parapophysis and is penetrated by one (possibly two) pneumatopores as in most theropods. The pair of sutures with the neural arch was still open at the time of death. A neural arch fragment (FPM 9712241) fits almost perfectly on the left side of the centrum. The suture on the arch fragment for the centrum is 52 mm long and has the same curvature and contours. Only the base of the transverse process is preserved, and its hollow interior clearly communicated with pneumatopores in the deep infradiapophysial fossa. Laminae anterior and posterior to the fossa supported the transverse process from below. The infraprezygapophysial fossa is present, but poorly preserved. Dorsal neural arches can be assigned tentatively to Fukuiraptor. The almost complete arch (FPM 9712242) is from one of the last vertebrae of the presacral series. The transverse process bears both parapophysis and diapophysis, separated by 58 mm on the right side. It extends laterally, posteriorly, and dorsally about 60 mm from the longitudinal midline of the body. Each process is triangular in crosssection distally, with an anteroposterior length of 19 mm and a dorsoventral height of 13.5 mm. The base of the transverse process is excavated posteriorly by a deep infrapostzygapophysial fossa, in the floor of which several pneumatopores invade the interior of the neural arch. The width across the anterior zygapophyses is 35 mm, but the posterior zygapophyses are damaged enough to make such a measurement unreliable. The zygapophyses are inclined at an angle of about 45 degrees. The neural canal is large (minimum diameter is 12 mm). The neural spine is tall, laterally compressed (the distal end is only 7.5 mm thick), and platelike, and stood at least 10 cm above the roof of the neural canal. The anteroposterior length of the neural spine is 37 mm near the base, and more than 40 mm distally. One dorsal centrum (FMPN 9712207, Figs. 5A–F) is long (77.5 mm), but the intervertebral articulation is 60 mm high and 56 mm wide. The centrum is narrow-waisted with a deep lateral depression ventral to the suture for the neural arch. Below the pair of lateral depressions, the width of the centrum is 23.5 mm. The neural arch was not fused to the centrum, which is another reason to believe the animal was immature at the time of death. The suture extends along the entire length of the centrum and is elevated 15 mm above the floor of the neural canal, which is 11 mm across at midlength. There is no pleurocoel. A distal caudal vertebra (FPMN 9712208, Figs. 5G–K) is 26.7 mm long and has a posterior intervertebral articulation that is 7.1 mm high and 8.3 mm across. The centrum has two longitudinal ventral ridges, and one on each side. Both the paired prezygapophyses and the medial postzygapophyses are broken, making it impossible to determine their extent.