Yesterday we looked at Dr. Paul Ellenberger’s long-time interest in the tiny Mid-Triassic reptile, Cosesaurus aviceps, with a focus on the skull. He thought of his little Cosesaurus as an important transitional taxon, a bird precursor living several tens of millions of years earlier than Archaeopteryx.
New observations and phylogenetic analysis nest Cosesaurus within the Tritosauria at the base of the Fenestrasauria (Peters 2000b) leading toward the Pterosauria. Derived from a sister to the basal lizard, Huehuecuetzpalli, Cosesaurus is the Archaeopteryx of the Pterosauria. Cosesaurus could run on two legs (Peters 2000a, Fig. 12, based on a perfect match to Rotodactylus tracks) and flap its fiber-trailed forelimbs (Peters 2009), but it could not fly. That would come several million years later.
Having covered several cranial traits earlier, today we’ll look at the post-cranial traits observed by Ellenberger (1978, 1993, Fig. 1), many reinterpreted by Peters (2000, 2009, Fig. 12) both correctly and incorrectly and here corrected again.
The Cervicals, Dorsals and Sacrals
Ellenberger (1993) correctly interpreted the eight cervicals as individually longer than the dorsals and provided with elongated ribs (Fig. 1). The dorsals were procoelous and relatively short as a set. More than two sacrals were present. Unfortunately Ellenberger interpreted the largest sacral transverse process as an ischium (Fig. 7). There are no similarly-shaped ischia in candidate sisters. The actual sacral transverse processes were identified by Peters (2000, Fig. 9).
The tail was attenuated after the first ten caudals, which were provided with elongated transverse processes (Fig. 2). Thus the caudofemoral muscles were still important femoral abductors. The distal centra were three times longer than their depth. The chevrons did not descend as in most reptiles, but remained parallel to the centra, as in Archaeopteryx (and ignored by Ellenberger, as in Sharovipteryx and pterosaurs). Ellenberger (1993) noted a valid break in the tail of Cosesaurus, in which the tail was rotated 90 degrees on either side of the break, but that break is not reflected in the extended soft imprint.
The Tail – Soft Tissues
The tail of the Cosesaurus fossil was surrounded by a broad matrix area that sloped gently downward (Fig. 2). The area beyond the attenuated caudals was subdivided by a series of regularly spaced lines. Under his bird bias, Ellenberger (1993) considered such data an indication of elongated tail feathers (rectrices) with feather shafts. That would have preceded the development of theropod feathers by several tens of millions of years. Unfortunately, no Cosesaurus sister taxa have any sort of similar structure. If just the narrow, hair-like shafts are valid then the tail of Cosesaurus had keratinous hairs along its length. In certain pterosaurs these hairs at the tail tip coalesced to form a tail vane. This, of course, is an attempt to explain away the broader shape surrounding the tail. This could be the effect of water and granular matrix interacting with the tail hairs. Or this broad shape could be a valid structure because it appears to be continuous with the uropatagium coming off the left femur (Figs. 2, 11). Hard to figure.
The Pectoral Girdle
Ellenberger (1993) identified the strap-like scapulae more or less correctly (Fig. 3), but not quite long enough (Fig. 5). Birds and pterosaurs both have a strap-like scapula. Ellenberger considered the broad oval ventral plate a pair of giant coracoids fused medially (Fig. 3). Peters (2000) mistakenly followed this interpretation, but recently reinterpreted the plate as an anteriorly migrated sternum (Fig. 5). The prominent quadrant-shaped stem Ellenberger identified as a sternal keel (Fig. 3) turned out to be a very pterosaur-like coracoid stem (Fig. 5). The problem was: there are no other known taxa that have a pectoral girdle close to the Ellenberger (1993) restoration/reconstruction. However, there are three other taxa with the current Peters restoration/reconstruction: Sharovipteryx, Longsiquama and pterosaurs. Cosesaurus demonstrated the genesis of the sternal complex and the stem-like coracoid.
Ellenberger (1993) considered the sternum of Cosesaurus to be unossified (Figs. 3, 4), supporting that otherwise disconnected and ossified ventral keel. That was an invention created due to fulfill his bird-bias. What Ellenberger (1993) considered a giant ventral coracoid, is now identified as a sternum (Fig. 5) having migrated forward to a position it occupies in Longisquama and pterosaurs, up against the clavicle and the transverse processes of the interclavicle creating a sternal complex.
Ellenberger (1993, Fig. 3) correctly traced the clavicles of Cosesaurus as transversely oriented and straight overlapping medially and rimming the anterior interclavicle and coincident sternum. Then he restored the clavicles (Fig. 4) as disconnected from the other elements and V-shaped like a deep wishbone-shaped furcula (fused clavicles) following the restored V-shaped of his coracoids. Jurassic birds, like Archaeopteryx, did not attain such a shape in the clavicle. In Cosesaurus, Longisquama and pterosaurs the clavicles indeed rim the anterior interclavicle/sternum complex as shown in situ (Fig. 5) and reconstructed (Fig. 12). They are coplanar with the sternum.
Ellenberger (1993) correctly identified the scapulae. Peters (2000a) did not. While attempting to follow the example of a sister, Macrocnemus, I considered the scapulae to be disarticulated ribs and other elements to be the disc-like scapulae. Sanz and Lopez-Martinez (1984) made the same mistake. The day I could finally “see” my error was a good day for enlightenment. I could almost hear the Moody Blues.
Ellenberger (1993) illustrated the humerus (Figs. 3, 4) as essentially straight, and it was, with a slightly expanded proximal and distal end.
The Radius and Ulna
The radius and ulna in Cosesaurus were straight, parallel and closely appressed to each other (Fig. 1), unlike birds, just like pterosaurs (Fig. 6). The forearm was becoming increasingly restricted in pronation and supination due to the straighter shapes of the radius and ulna. Ellenberger (1993) reported fibers emerging from the posterior ulna (Fig. 1). These are clear in his photos and I confirmed them when I visited the fossil (Peters 2009). Due to his bird bias, Ellenberger (1993) considered these to be feather precursors. Finding closer connections with pterosaurs in phylogenetic analysis (Peters 2000), Peters (2009) considered these to be aktinofibril precursors, the fibers that support the wing and uropatagium in pterosaurs and their kin.
Ellenberger (1993) labeled the fingers correctly several times, but when restoring Cosesaurus he flipped the hands, making #2 the longest finger (Fig. 6). This created a more bird-like hand. Unflipped the hands of Cosesaurus resemble those of fenestrasaur and tritosaur sister taxa with shorter medial metacarpals and fingers, except digit 5, which is very short, on its way to becoming a vestige, as seen in Huehuecuetzpalli and the wings of Sharovipteryx, Longisquama and pterosaurs.
Ellenberger found two spots on the medial wrist that he ascribed to cartilage. Peters (2009) identified those as migrated centralia, now homologous with the pteroid and preaxial carpal. The other carpals were poorly ossified, as in the closest Cosesaurus sister taxa, including the basal lizard, Huehuecuetzpalli.
Attempting to find homologs for the retroverted pubis of birds, Ellenberger (1993) considered a displaced prepubis (Fig. 9) and a drifted gastralium to be the retroverted pubis of birds (Figs. 7-8). Never mind that the pubis of Archaeopteryx was not so retroverted or attenuated. Never mind that no other sisters had a similar pelvis. Ellenberger (1993) described elongated ilia, which is essentially correct (but perhaps over elongated). Ellenberger (1993) considered a displaced sacral transverse process to be an ischium, overlooking the actual coosified ischium + pubis (Fig. 9) and completely overlooking the pterosaur-like prepubis, probably because it was not expected in a bird ancestor. Once I recognized one prepubis in plain sight on the ilium, the other, tucked beneath the femur (Fig. 9), was easier. Unfortunately a manuscript describing these revelations and synapomorphies was rejected. Hence this website.
The Hind Limb
Ellenberger (1993) correctly interpreted the femur without any sort of head or neck, as in basal reptiles including lizards. Unfortunately he reconstructed the femur as a parasagittal element (Fig. 1), as in birds, rather than a sprawling one, as in the basal pterosaur, MPUM 6009.
The tibia and fibula were both straight and closely appressed to one another with the fibula less than half the diameter of the tibia, as in birds and pterosaurs. The length of the tibia/fibula becomes longer than the femur in derived theropods in the bird lineage, and in Sharovipteryx and higher taxa, in the pterosaur lineage.
The Tarsus and Pes
Ellenberger (1993) correctly interpreted the pes of Cosesaurus with an astragalus, calcaneum, centrale and distal tarsal 4 as the four largest tarsal elements. He also found tiny distal tarsals 1-3. Such a tarsus is a synapomorphy Cosesaurus shared with Tanystropheus, Sharovipteryx and pterosaurs. In addition, all these taxa had a very short metatarsal 5 and a hyper-elongated phalanx 5.1, which Ellenberger (1993) correctly identified, but did not make the pterosaur connection and no bird or bird ancestor has such a toe. One would have to go back to Proterosuchus to find a bird ancestor with digit 4 longer than 3. And certain armored aetosaurs develop a longer fourth toe. This is the trait that first drew me toward this taxon as a possible sister to pterosaurs. No gracile archosaur or archosauriform had such a toe.
Ellenberger photographed and noted soft tissues emanating from the left femur and tibia (Fig. 11). He considered these to be possible feather precursors. These impressions also greatly resembled the fiber-embedded uropatagia of sister taxa, Sharovipteryx and pterosaurs. Note the fibers anterior to the knee in figure 11. Cosesaurus was much decorated in such soft tissue, and this makes phylogenetic sense as a precursor to the party queen of the Triassic, Longisquama.
Despite intense study, Ellenberger (1993) invented some insightful and strange structures in Cosesaurus due to his strong bird-bias. These have been both credited and criticized. As a suite and from head to toe, Cosesaurus shares more traits with the tritosaurs, Huehuecuetzpalli through pterosaurs. The changes in the pectoral and pelvic region distinguish Cosesaurus from Macrocnemus and nest it with similarly endowed reptiles, including Sharovipteryx, Longsiquama and pterosaurs.
As always, I encourage readers to see specimens, make observations and come to your own conclusions. Test. Test. And test again.
Evidence and support in the form of nexus, pdf and jpeg files will be sent to all who request additional data.
Ellenberger P and de Villalta JF 1974. Sur la presence d’un ancêtre probable des oiseaux dans le Muschelkalk supérieure de Catalogne (Espagne). Note preliminaire. Acta Geologica Hispanica 9, 162-168.
Ellenberger P 1978. L’Origine des Oiseaux. Historique et méthodes nouvelles. Les problémes des Archaeornithes. La venue au jour de Cosesaurus aviceps (Muschelkalk supérieur) in Aspects Modernes des Recherches sur l’Evolution. In Bons, J. (ed.) Compt Ren. Coll. Montpellier12-16 Sept. 1977. Vol. 1. Montpellier, Mém. Trav. Ecole Prat. Hautes Etudes, De l’Institut de Montpellier 4: 89-117.
Ellenberger P 1993. Cosesaurus aviceps . Vertébré aviforme du Trias Moyen de Catalogne. Étude descriptive et comparative. Mémoire Avec le concours de l’École Pratique des Hautes Etudes. Laboratorie de Paléontologie des Vertébrés. Univ. Sci. Tech. Languedoc, Montpellier (France). Pp. 1-664.
Ostrom JH 1969. Osteology of Deinonychus antirrhopus, an unusual theropod from the Lower Cretaceous of Montana. Peabody Museum of Natural History Bulletin 30: 1–165.
Peabody FE 1948. Reptile and amphibian trackways from the Lower Triassic Moenkopi formation of Arizona and Utah. University of California Publications, Bulletin of the Department of Geological Sciences 27: 295-468.
Peters D 2000a. Description and Interpretation of Interphalangeal Lines in Tetrapods. Ichnos 7:11-41.
Peters D 2000b. A Redescription of Four Prolacertiform Genera and Implications for Pterosaur Phylogenesis. Rivista Italiana di Paleontologia e Stratigrafia 106 (3): 293–336.
Peters D 2009. A reinterpretation of pteroid articulation in pterosaurs. Journal of Vertebrate Paleontology 29: 1327-1330.
Peters D 2011. A Catalog of Pterosaur Pedes for Trackmaker Identification Ichnos 18(2):114-141.
Sanz JL and López-Martinez N 1984. The prolacertid lepidosaurian Cosesaurus aviceps Ellenberger & Villalta, a claimed ‘protoavian’ from the Middle Triassic of Spain. Géobios 17: 747-753.