The SMNS flathead anurognathid skull: bone by bone

Two recent blog posts by J. Headden, illustrations by M. Witton and various other odd reconstructions of the skull of the SMNS anurognathid (Bennett 2007), the flathead pterosaur, prompted today’s post. I realize I have not provided sufficient clear imagery and so, here it is (in rollover images or see below).

You’re looking at what appears to be a complete skull crushed ventrally with all parts intact and articulated. That is not the case. If so, you’d end up with a monster bearing little resemblance to other anurognathids and other pterosaurs in general (aka Bennett’s reconstruction, fig. 6 left). There was some shifting during taphonomy as will be made clear in the following photos. Reconstructing the parts produces a very nice anurognathid sharing most traits with other anurognathids (fig. 6 right), but the skull has a flatter morphology.

Figure 1. Skull of the flathead pterosaur. Some of the bones are orange. Others are tan. Still others, perhaps just the shapes they left, are white, as in the apparent fenestra in the top of the skull, which is nevertheless, divided medially. 

I realize the bones I found in this specimen are difficult to see. So today I present some of them as well as I possibly can (see figs. below). You can (and I encourage you to) use a rollover on the reptile evolution page on the flathead anurognathid here.

Bennett (2007) described this specimen and created reconstructions (fig. 6) in which his giant scleral ring was in the anterior half of the skull, the antorbital fenestra was relegated to a tiny zone bounded by bones he claimed he never found and the skull roof was as wide as any turtle’s with widely separated upper temporal fenestrae. Such a reconstruction is at odds with all other pterosaur skulls. Recall that Bennett also invoked the idea that the very wide frontal bones of the SMNS specimen were decayed centrally, providing windows to whatever bones were beneath them. Not so. His frontal bones are composed laterally on the left by a dorsal lacrimal piece (fig. 5) and medially by the narrow nasals (fig. 3). That’s a large corner of an ectopalatine seen through the bones.

SMNS-anurognathus maxillae

Figure 2. SMNS-anurognathus maxillae and teeth. The slender ascending process is broken in both cases. Note Bennett’s proposed “sclerotic ring” continues far beyond the confines of the skull. 

SMNS anurognathid sclerotic rings. They are small in the back half of the skull. Those are the narrow nasals between them.

Figure 3. SMNS anurognathid sclerotic rings. They are small in the back half of the skull. Those are the narrow nasals between them. All of these traits match those of other anurognathids. 

The SMNS anurognathid ascending process only of the premaxillae and frontals.

Figure 4. The SMNS anurognathid ascending process only of the premaxillae and frontals. Compare the frontals to the sclerotic rings (fig. 3), which nest within them.

SMNS anurognathus occiput and lacrimals. Note the extreme width of the occiput. This alone tells you the general shape of the skull, flatter than in other pterosaurs.

Figure 5. SMNS anurognathus occiput and lacrimals. Note the extreme width of the occiput. This alone tells you the general shape of the skull, flatter than in other pterosaurs. The fragile left lacrimal is broken into three parts. The right one has the sinuous shape that provides room for the dorsally bulging eyeballs (see fig. 6).

If you still can’t see the imagery I indicate above, please visit these rollover images. That way you won’t have to shift back and forth to compare images. You can also see the palate and other bones on the same web page.

Using DGS (digital graphic segregation, I was able to find symmetrical pairs for every bone in the skull, despite the layering of cranial bones atop displaced facial and palatal bones. All the bones, even those of the palate, resemble those in other anurognathid palates. Moreover, in the case of the SMNS anurognathid the palatal elements criss-cross and reinforce each other, providing maximum strength with minimum weight. The same cannot be said of the Bennett reconstruction.


Figure 6. The SMNS anurognathus. On the left as Bennett (2007) reconstructed it identifying the left maxilla has a giant sclerotic ring (ignoring the teeth). Bennett’s palate does not resemble those of other anurognathids. On the right is the more accurate reconstruction based on DGS. Click for more information.

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.

Bennett SC 2007. A second specimen of the pterosaur Anurognathus ammoni. Paläontologische Zeitschrift 81(4):376-398.

Book Review – Earth Before the Dinosaurs – Steyer

 Earth Before the Dinosaurs by Sébastien Steyer

Figure 1. Earth Before the Dinosaurs by Sébastien Steyer

Quite good – until you get to the reptiles.
Then mostly bogus in that it follows Carroll (1988) in putting Diadectomorphs as the outgroup to the Amniota (when its an ingroup) and includes mesosaurs with parareptiles, which we discussed earlier, among several other misfit nestings. There is no mention of Gephyrostegus and Cephalerpeton, which nest at the base of the Reptilia in the large reptile tree. And Longisquama is pictured gliding with paired plumes outstretched horizontally.

You get the picture. This is old school data.

Steyer S 2012. Earth Before the Dinosaurs. Indiana University Press. 182pp.

‘I’m a Dino’ cartoons on YouTube – Sordes and Pteranodon

I’m probably the last to know about this.

YouTube has a few “I’m a Dinosaur” cartoons. These two feature Sordes and Pteranodon talking to the camera and to each other. Click to play.

Click to play. I'm a Dinosaur cartoon - Sordes.

Click to play. I’m a Dinosaur cartoon – Sordes.

Cute and cheeky. Just two minutes long.

Click to play. I'm a dinosaur - Pteranodon.

Click to play. I’m a dinosaur – Pteranodon.

The myth of the Parareptilia

The large reptile tree is not the first attempt at classifying Reptiles. It is only the most recent and the most comprehensive.

Earlier attempts invented the clade “Parareptilia,” a name coined by Olson in
1947 to refer to a group of pre-Triassic reptiles leaving no living descendants, as opposed to the Eureptilia, which included all living reptiles and their last common ancestors. This group included turtles.

Gauthier et al. (1988) attempting to understand Reptile relationships using cladistic analysis, and were among the first to do so. They divided the Amniota into Synapsida  and Sauroposida, then divided the Sauropsida into the Reptilia and Parareptilia.

Unfortunately testing in the large reptile tree (using more taxa) does not support these divisions and several new basal reptiles have been described since 1988.

No longer do synapsids split off first from the rest of the Reptilia. Now the new Archosauromorpha (chiefly insect-eaters) splits from the new Lepidosauromorpha (chiefly plant-eaters). The new Archosauromorpha includes the Synapsida, members of which evolve to become the Diapsida, which includes the Enaliosauria (Mesosauriadae + Sauropterygia + Ichthyopterygia and kin). The new Lepidosaurormorpha includes Captorhinidae, Diadectidae, Millerettidae, Lepidosauriformes and kin.

No longer are mesosaurs nested with procolophonids, but nest far from them with several other marine reptiles.

Laurin and Reisz (1995) revised the concept of the Parareptilia. In their cladogram the Synapaside split off first, followed by the Mesosauridae. The remaining taxa were considered Reptilia. Parareptilia included Millerettidae, Pareiasauria, Procolophonidae and Testudines (turtles). Their Eureptilia included Captorhininidae and Romeriida (Protorothyrididae and Diapsida).

Others (always including O. Rieppel) have moved turtles to the Sauropterygia.

When you stop including suprageneric taxa (and the dangers that follow that practice) and start including hundreds more generic taxa, new nesting patterns emerge, as demonstrated by the large reptile tree and all of its subset clades. All the groupings for Parareptilia proposed by earlier workers get split up and recombined in new patterns and clades. These new clades actually demonstrate the gradual accumulation of character traits for any and all derived taxa without introducing any “strange bedfellows.”

So the Parareptilia and its membership has been falsified in a larger, more comprehensive study. The utility of this term in paleontological work has been invalidated.

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.

Gauthier J, Kluge AG and Rowe T 1988. The early evolution of the Amniota. In M. J. Benton (ed.). The phylogeny and classification of the tetrapods, Volume 1: amphibians, reptiles, birds. 103-155. Oxford: Clarendon Press.
Laurin M, Reisz RR 1995. A reevaluation of early amniote phylogeny. Zoological Journal of the Linnean Society 113 (2): 165–223.
Olson EC 1947. The family Diadectidae and its bearing on the classification of reptiles. Fieldiana Geology 11: 1–53.

Rethinking the “fused ribs” of Triassic gliders

Living from the Permian through the Cretaceous, the so-called “rib” gliders are an interesting lot. And I’m still trying to figure out what’s going on with the ribs and transverse processes. All their closest kin have ribs and none have transverse processes. Here’s the latest (a very minor change of thinking):


Figure 1. Coelurosauravus. It had ribs, but no transverse processes. The extradermal rods were more numerous than the ribs. Click to learn more.

Coelurosauravus (Fig. 1) is the earliest one (Late Permian). It had no transverse processes. It had ribs. It had extradermal rods likely supporting membranes, and many more rods than ribs anteriorly, but that became a one-to-one relationship posteriorly.

The other rib gliders were distinctly different. Icarosaurus (Fig. 2), Kuehneosaurus (Fig. 3), Mecistotrachelos and Xianglong all had long transverse processes, few to no ribs and the extradermal rods matched one to one with the transverse processes.

Traditional thinking (everyone else) considers the gliding membrane rods to be the ribs, following the pattern of Draco (Fig. 4), the living and genuine rib glider, which likewise has no transverse processes.

Since no other close taxa to the extinct “rib” gliders had transverse processes this led to the heretical idea that the long transverse processes WERE the ribs now fused to the centra and that the gliding spars continued to be extradermal in origin, as in Coelurosauravus (Fig. 1).


Figure 2. Icarosaurus. Note the tiny ribs near the shoulders. Or are those unfused ribs?

Icarosaurus and Kuehneosaurus display mid-change solutions. They have short transverse processes anteriorly and long ribs. They also have long transverse processes starting at the shoulder and tiny to no ribs. Look closely. You’ll see them.


Figure 3. Kuehneosaurus. Note the elongation of the transverse processes replacing the gradually shortening ribs posteriorly.

This is different from the situation in the convergent and largely unrelated living glider, Draco (Fig. 4) in which there are no transverse processes and those membrane spars definitely are the dorsal ribs. The spars on Icarosaurus (Fig. 2) are waaaay too long to be even considered as ribs.

Draco volans

Figure 4. Draco volans in dorsal view based on an X-ray. Click for more info.

Ultimately, I now see the evolution of increasingly longer transverse processes (restricted only to members of this clade) and the reduction of the ribs, not their fusion as I thought before. So, dermal extensions attached directly to transverse processes and the ribs are missing posteriorly.

Not sure how the rest of the gut was supported, or how the lungs expanded without traditional ribs. These oddballs figured out some other way to respire.

The trend was for a shorter body and fewer membrane spars taken to extremes in the most derived of these gliders, Mecistotrachelos and Xianglong, which was originally considered a lizard related to Draco. Here’s the family to scale.

The Triassic gliders and their non-gliding precursors.

Figure 5. Click to enlarge. The Triassic gliders and their non-gliding precursors.

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.

This marks the 500th post. 

Diandongosuchus and its Phytosaur Synapomorphies

For anyone following this line of thought on Diandongosuchus, I offer data from my matrix. The following synapomorphies were recovered with Diandongosuchus and phytosaurs (I make no claim that these traits are not found elsewhere as convergences within the tree):

1. Naris not larger than antorbital fenestra

2. Squamosal creates a temporal ledge

3. Squamosal descends at a right angle

4. Postorbital/parietal contact is long

5. Postfrontal present (plesiomorphic)

6. Postfrontal has no contact with upper temporal fenestra

7. Quadrate lean: vertical

8. Mandible tip rises

9. Angular lateral exposure: less than a third of jaw depth

10. Mandible ventral shape: straight

11. Cervical centra: height = length

12. Cervical ribs with free anterior processes

13. Scapulocoracoid fenestration present

14. Radius + ulna not longer than 3x width

15. Tarsus has double bend shape

16.  Pedal 3.1 > p2.1

17. Pedal 4 length subequal to metatarsal 4

With phytosaurs and Proterochampsa:

1. Ventral aspect of premaxilla vs. rostrum: a third or greater.

2. Squamosal and quadratojugal indentation: V-shaped

3. Choana orientation: deflected medially (unknown in Diandongosuchus)

4. Vomer teeth absent (unknown in Diandongosuchus)

The real sister taxa and close relatives of Diandongosuchus

Figure 1. The real sister taxa and close relatives of Diandongosuchus beginning with the Youngina with the longest, lowest rostrum, BPI 2871, and moving forward toward the parasuchians.

Of course lots more characters were recovered in larger clades surrounding the phytosaurs. If anyone disputes these or has others, please bring them to my attention.

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.

Li C, Wu X-C, Zhao L-J, Sato T and Wang LT 2012. A new archosaur (Diapsida, Archosauriformes) from the marine Triassic of China, Journal of Vertebrate Paleontology, 32:5, 1064-1081.

Dr. Ellenberger and his Petite Cosesaurus – part 1: Cranial Traits

Dr. Paul Ellenberger (pronounced “El-len-ber-zhay”) spent a large part of his life attempting to link a tiny Mid-Triassic fossil reptile, Cosesaurus aviceps, to birds. He considered it a precursor to Archaeopteryx in the years just following the publication of Ostrom (1969) on Deinonychus. Ellenberger published two small papers (Ellenberger and de Villalta 1974, Ellenberger 1978) and a very large (664 pp.) unpublished tome (Ellenberger 1993) on this little reptile perpetually entwined with an amorphous jellyfish. No one has spent more time studying Cosesaurus than Ellenberger. No one has put more effort into describing it and photographing it from every angle in the most precise detail.

Cosesaurus aviceps at close to actual size.

Figure 1. Cosesaurus aviceps at close to actual size. The blob next to it is a jelly fish. No actual bones are preserved. Cosesaurus is nothing but a deep impression faithfully preserving every aspect of its skeleton down to the finest soft tissue details. The tail is especially deep, which created the impression, when transferred to 2-D, of emanating feathers. Tomorrow the same image will be presented but flipped 180 degrees.

Even so…
Ellenberger (1993) got many things wrong. He had a mistaken preconception and that biased his observations. It can happen. I’ve seen it happen to the best paleontologists out there. Following tradition is easy, but it leads to problems. Testing tradition is good science. Distrusting the validity of autapomorphies is key. Phylogenetic analysis trumps all.

The Power of Pet Ideas
Ellenberger’s (1993) bird hypotheses were never taken seriously or supported by other writers in the literature. Nevertheless, Ellenberger created a body of data leading to an interest in the taxon that launched Cosesaurus in a new direction for me. It never occurred to Ellenberger to link Cosesaurus to Sharovipteryx, Longisquama and pterosaurs. I raised the subject with him after seeing Cosesaurus in Barcelona and while staying with Paul for a day or two at his home in Montpellier, France. Ellenberger didn’t like the idea (because it didn’t support his bird hypothesis), and he didn’t want to discuss it.

Giving Credit
Well, we’re going to explore Dr. Ellenberger’s view of this little predecessor taxon. The point of this report is to give credit where credit is due and to shine a light on any mistakes.

Overall view of Cosesaurus aviceps in standing pose.

Figure 1. From Ellenberger 1993. Overall view of Cosesaurus aviceps in standing pose, lateral and dorsal views. Note the bird-like restoration, a little too erect with hands flipped to more closely match the hands of birds. Other problems in the pectoral and pelvic regions will be discussed in part 2 of this blog tomorrow.

Ellenberger’s Reconstruction of Cosesaurus
Ellenberger saw Cosesaurus as a bird precursor, therefore he saw it as a digitigrade narrow-gauge biped. These are all true. Matching footprints (Rotodactylus) are evidence (Peters 2000). Despite being a footprint expert, Ellenberger (1993) did not consider a match of Rotodactylus to Cosesaurus. He did not produce an illustration with a bent-back pedal digit 5, which would have completed the match (Fig. 5).

The brain of Cosesaurus

Figure 2. The brain of Cosesaurus and the binocular vision reported by Ellenberger (1993). Note the elongated antorbital fenestra above the maxilla and below the prefrontal/nasal. The naris is indicated by two short lines here, better viewed in figure 3 (lower of the two skulls).

The Brain of Cosesaurus
No one but Ellenberger (1993) bothered to document the cranial capacity of Cosesaurus. Ellenberger applied reverse geometry to re-inflate the crushed skull of Cosesaurus to determine its likely dimensions in 3-D. Of course, he hoped to show that the brain of Cosesaurus had enlarged to bird-like proportions. It had also enlarged to pterosaur-like proportions. This was no ordinary reptile.

The skull of Cosesaurus with the antorbital fenestra and dorsal fibes/frill.

Figure 3. The skull of Cosesaurus in two lighting conditions with the antorbital fenestra and dorsal/cranial fibers/frill visible in the upper photo. From Ellenberger (1993). Also note the premaxilla crest in front of the orbit.

The Antorbital Fenestra
Ellenberger (1993) reported an antorbital fenestra in Cosesaurus and his images (Fig. 3) confirm that. I also confirm that, having seen the fossil in Barcelona.

By contrast, Sanz and Lopez-Martinez (1984) said there was no antorbital fenestra and considered Cosesaurus a juvenile Macrocnemus (Fig. 4). They also missed dozens of other traits that distinguish Cosesaurus from Macrocnemus (Fig. 5). They illustrated Cosesaurus in an inaccurate cartoonish fashion virtually identical to a cartoon Macrocnemus without any distinguishing traits other than a shortened rostrum, not realizing that in this clade hatchlings are virtually identical to adults. Altogether the Sanz and Lopez-Martinez (1984) report can be considered dated, biased and bogus because they didn’t put the effort in that was needed to trump earlier data by Ellenberger.

The same can be said of the Senter (2003) dissertation that reported no antorbital fenestra, even though he illustrated one, again in cartoonish fashion. I don’t understand how scientists can be so blinded by paradigm and bias that they cannot report the presence of an antorbital fenestra in Cosesaurus (Fig. 3). Unfortunately others (Evans 1988, Hone and Benton 2008) used the bogus data in phylogenetic analysis, preferring those simplified drawings to the precision of Elleberger (1978, 1993) and Peters (2000) or their own examinations(!)

 Cosesaurus illustrated as a juvenile Macrocnemus by Sanz and Lopez-Martinez

Figure 4. Cosesaurus illustrated as a juvenile Macrocnemus by Sanz and Lopez-Martinez (1984).

Binocular Vision
Ellenberger determined that the large eyes of Cosesaurus poised over the small rostrum probably delivered 50 degrees of overlapping vision. That seems reasonable and sets Cosesaurus apart from Macrocnemus.

The Teeth 
Ellenberger reported the upper and lower three posterior teeth in the jaws of Cosesaurus were different that the others: broader and less pointed. These were precursors to the multicusped teeth found in derived fenestrasaurs.

The Naris
Ellenberger reported a slit-like naris in Cosesaurus, displaced from the snout tip. Such a naris is also found in all descendants of a sister to Huehuecuetzpalli, including pterosaurs and tanystropheids.

The Jaw Tip
Ellenberger considered the extended jaw tips to be beak precursors. The skull was also longer than the tooth row in the more primitive lizard, Huehuecuetzpalli. In more derived fenestrasaurs teeth protruded from the anterior jaws.

Current interpretation of Cosesaurus.

Figure 5. Current interpretation of Cosesaurus. Click to enlarge. Post-crania will be presented tomorrow.

Ellenberger (1993) correctly illustrated a jugal with a new quadratojugal process in Cosesaurus.

Ellenberger (1993) reported the occiput leaned posteriorly, which would have been appropriate for a reptile standing erect on hind limbs, whether bird ancestor or pterosaur ancestor.

Ellenberger illustrated as much of the palate as was visible and it was essentially correct and similar to that of pterosaurs, as reported earlier.

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.
Evans SE 1988. The early history and relationships of the Diapsida. Pp. 221–260 in: Benton, M. J. (ed.) The Phylogeny and Classification of the Tetrapods, Volume 1: Amphibians, Reptiles, Birds. Syst Assoc Sp Vol No. 35A. Clarendon Press, Oxford.
Gauthier JA 1986. Saurischian monophyly and the origin of birds. Memoirs of the California Academy of Science 8: 1–55.
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
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.

The Aerodynamics of Sharovipteryx – the Hind Wing Glider

It’s obvious that Sharovipteryx was built to be a glider. But it’s the elongated, stick-like hind legs – rather than the barely visible fore legs – that acted like wings – and that’s enough to throw most of us for a loop. What gives here?? Everyone imagined a lack of pitch control in such a creature.

To counter this pitch problem, Sharov (1971, Fig. 1) invented a membrane stretching from the back of the largely imagined arms to the front of the legs in a rather cartoonish presentation.

Various reconstructions of Sharovipteryx

Figure 1. Various reconstructions of Sharovipteryx

Gans et al. (1987) observed no membrane between the front and hind legs. Even so, experiments with paper models (Fig. 1) showed that a canard membrane would have stabilized the animal in the air and extended its glide.

Wellnhofer (1991) simply re-illustrated what had been previously published, likewise imagining front limbs supporting a canard, but not connected to the hind limbs.

Dyke et al. (2006) could find no membranes anterior to the hind legs, but imagined some that created a delta wing like certain aircraft. These authors also imagined a small delta canard wing anterior to the forelimbs for pitch control.

As you can see, there has been no loss of imagination in trying to figure out the extent of the wing membranes in Sharovipteryx. Unfortunately, there has been an embarrassing lack of attention to detail.

Figure 2. The wing membranes of Sharovipteryx.

Figure 2. The wing membranes of Sharovipteryx.

The Search for Anterior Control Membranes
While others searched in vain for anterior membranes capable of controlling pitch in Sharovipteryx, I found them (Fig. 2), either by direct observation or phylogenetic bracketing.

Neck Strakes
Two large hyoids (tongue bones) could have spread that loose preserved neck skin up to 6x their original width (3x per side, in pink, Fig. 2). This action would have created aerodynamic strakes, also found in modern jets like the F-18. Lifting and depressing the head, via the neck, would have elevated and depressed the strake, giving Sharovipteryx great pitch control.

Forelimb Canards
Both forelimbs are preserved virtually in their entirety.  When recontructed they look stunted, but they were robust with traits also found in Longisquama and Cosesaurus like an expanded distal end and a large deltopectoral crest and parallel, appressed ulna and radius. These traits also include a tiny pteroid and preaxial carpal (Fig. 2.), bones traditionally found only in pterosaurs, but also found in Cosesaurus (Peters 2009). Digit 4 was the longest (reaching to the pelvis in Sharov’s (1971) original illustration). Digit 5 was a vestige. Since sister taxa, Cosesaurus and Longsiquama had trailing edge membranes, phylogenetic bracketing permits us to add them to Sharovipteryx (Fig. 2). These would have acted like canards, small forewings capable of independent movement, but most typically extended straight out, supported by a robust scapula and coracoid. The sternal complex remains buried in Sharovipteryx, but phylogenetic bracketing indicates it had this fenestrasaur trait as well.

Pancake-like Ribs
A trait that has gone overlooked in prior reconstructions (Fig. 1) is how broad and flat the rib cage was. In dorsal view it formed a circular disc. This shape filled the area between the fore and hind limbs without the invention of a connecting membrane.

Small Anterior Leg Membranes
A small stiff membrane anterior to the femur helped fill the gap otherwise present when the straight femur met the circular dorsal ribs. Another small stiff membrane appeared at the distal tibia.

Interdigital Membranes
Between the toes membranes extended only as far as the proximal interphalangeal joints medially, further laterally, extending to ungual 4 lateral to digit 4.

Aft of the hind legs, the patagia are well represented in the fossil. These fiber-embeded membranes extended to the second joint of digit 5 and to the second chevron on the tail.

Long, Flat Thighs
The elongated ilia, both fore and aft, would have supported wide thighs, at least proximally.

Far from lacking anterior pitch-control membranes, Sharovipteryx had both strakes and canards. Since the coracoids were elongated and stem-like, Sharovipteryx probably flapped its small forelimbs, generating a small amount of thrust. Primarily Sharovipteryx was a glider, and a well-controlled one! This obvious obligatory biped had to have been agile enough to land using its feet alone, as in birds. When approaching a landing spot, the anterior control surfaces would have been raised to stall the main flight membrane just prior to touch down.

More on Sharovipteryx in the next two blogs: the hands and the skull.

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.

Dyke GJ, Nudds RL and Rayner JMV 2006. Flight of Sharovipteryx mirabilis: the world’s first delta-winged glider. Journal of Evolutionary Biology.
Gans C, Darevski I and Tatarinov LP 1987. Sharovipteryx, a reptilian glider?Paleobiology, October 1987, v. 13, p. 415-426.
Peters D 2000. 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
Sharov AG 1971. New flying reptiles from the Mesozoic of Kazakhstan and Kirghizia. – Transactions of the Paleontological Institute, Akademia Nauk, USSR, Moscow, 130: 104–113 [in Russian].
Wellnhofer P 1991. The Illustrated Encyclopedia of Pterosaurs. Salamander, London


A New Skull for Silvanerpeton

Silvanerpeton is not a Reptile
But it is a sister to Gephyrostegus, the proximal outgroup to the Reptilia. So it has a bearing on the base of the Reptilia.

A new skull for Silvanerpeton

Figure 1. A new skull for Silvanerpeton (in yellow) is somewhat different from the original reconstruction provided by Ruta and Clack (2006). This is based on tracing a photo of the specimen provided in Ruta and Clack (2006).

A New Skull Based on Tracings
Silvanerpeton miripedes (Clack 1994) Visean, Early Carboniferous ~335 mya, ~40 cm in length was originally illustrated with the skull shown in figure 1 (in white) and on this data Silvanerpeton was entered into the large reptile family tree. The addition of a contemporary, Eldeceeon (Smithson 1994) introduced certain problems (autapomorphies) into the dataset. Those problems sparked a review of the specimen that included tracing the skull from photographs provided in Ruta and Clack (2006). A new reconstruction based on specimen NMS G.1998.51.2 (in yellow, Fig. 1) had distinctly different traits. There are several specimens known. Perhaps the original was based largely on another specimen.

A new skull for specimen 1 attributed to Gephyrostegus.

Figure 1. A new skull for specimen 1 attributed to Gephyrostegus. Click to enlarge.

A New Reconstrucion of Gephyrostegus (specimen 1)
As part of the process of weeding out invalid autapomorphies, I also retraced and reconstructed a sister of Silvanerpeton, specimen 1 of Gephyrostegus (Brough and Brough 1967) (CGH IIIB 21 c. 587) and counterpart (MP451) to produce a revised skull for that specimen (Fig. 2). Note the virtual loss of the otic notch, the development of the transverse pterygoid flange and those long reptilian toes, more robust than in the larger Silvanerpeton.

A New Nesting
Taxa shifted slightly from the prior tree due to the addition of Eldeceeon and the reinterpretation of traced traits. Both Silvanerpeton and specimen 1 moved away from Utegenia and toward Gephyrostegus. This clarifies the lineage of pre-reptiles. Needless to say at this point, diadectids and limnoscelids do not nest as pre-reptiles, but well within the new Lepidosauromorpha.

A Serial Size Reduction
Silvanerpeton was a 40 cm tetrapod. Specimen 1 of Gephyrostegus and G. watsoni were a quarter of its size. Silvanerpeton appeared just 25 million years after Ichthyostega, and the first amniotes may have been contemporaries, just getting started. These data fall in line with Carroll’s (1970) hypothesis on tiny tetrapods without a tadpole stage were the first to lay tiny amniotic eggs, as blogged just a few days ago.

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.

Carroll RL 1970. The Ancestry of Reptiles. Philosophical Transactions of the Royal Society London B 257:267–308. online pdf.
Carroll RL 1991. The Origin of Reptiles in Origins of the Higher Groups of Tetrapods: Controversy and Consensus.  Schultze H-P and Trueb L (eds). Cornell University Press.
Carroll RL 2008. 
Problems of the Origin of Reptiles. Biological Reviews 44(3):393-431.
Carroll RL 2009. 
The Rise of the Amphibians 365 Million Years of Evolution. The Johns Hopkins University Press. 360 pp.
Clack JA 1994.
 Silvanerpeton miripedes, a new anthracosauroid from the Visean of East Kirkton, West Lothian, Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences 84 (for 1993), 369–76.
Ruta M and Clack, JA 2006 A review of Silvanerpeton miripedes, a stem amniote from the Lower Carboniferous of East Kirkton, West Lothian, Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences, 97, 31-63.
Smithson TR 1994. Eldeceeon rolfei, a new reptiliomorph from the Viséan of East Kirkton, West Lothian, Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences 84 (3-4): 377–382.


What is Dalinghosaurus? It’s THE Most Primitive Lizard.

Will you look at the feet on that thing!!


Figure 1. Dalinghosaurus, a beautiful and complete specimen. IVPP V13281, Barremian, Hejiaxin, China. 2cm skull length

Dalinghosaurus longidigitus (Ji 1998) Early Cretaceous was reported (Evans and Wang 2005) to be related to the modern-day lizards in the family Xenosauridae (Shinisaurus in China), Xenosaurus in Mexico and two extinct lizards, Exostinus and Carusia. Unfortunately they didn’t expand their inclusion set (see below). One fossil skeleton of Dalinghosaurus included ten juveniles. Despite the disparity in limb length, Evans and Wang (2005) considered Dalinghosaurus a climber, not a biped.

Reconstruction of Dalinghosaurus

Figure 2. Reconstruction of Dalinghosaurus

A Key Taxon at the Base of the Squamata + Tritosauria
In the large study Dalinghosaurus nested outside the Squamata (Iguania + Scleroglossa) and the Tritosauria, as a sister to Homoeosaurus and Gephyrosaurus on the primitive side and Lacertulus + the Daohugou lizard and Iguana on the more derived side. Shinisaurus, Exostinus and Xenosaurus were not included in the large study, but it is doubtful that they were closer sisters because Carusia and a sister to Shinisaurus, Bahndwivici, nested at the base of the Anguimorpha, well within the Scleroglossa.

The Lepidosauria.

Figure 2. The Lepidosauria. Click to see entire reptile tree.

Until a better candidate comes along a sister to Dalinghosaurus was THE basal squamate + tritosaur from which all other squamates evolved. A new clade is needed to include the Squamata, the Tritosauria and DalinghosaurusThe Dalinghosauria is defined as Dalinghosaurus, Iguana, their last common ancestor and all of its descendants.

The Homoeosaurus Connection
Distinct from Homoeosaurus, the skull of Dalinghosaurus had a larger orbit and a deeper preorbital bone with a reduced lacrimal. The jugal and postorbital bones were relatively more gracile. The mandible was deeper with a longer, deeper retroarticular process.

The cervicals were longer. The dorsal ribs were much wider producing a flattened cross-section. The transverse processes of the anterior caudals were longer. The tail was longer and more robust.

The scapulocoracoid was fused and fenestrated anteriorly. The interclavicle is cruciform. The humerus, radius and ulna were more gracile. The metacarpals were longer.

The pubis was longer and the ischium was more robust. The feet were much longer and longer than the femur + tibia. The astragalus and calcaneum were co-ossified.


Figure 3. Homeosaurus, a more primitive sister to Dalinghosaurus

Adding Taxa to the Large Study
Along with Dalinghosaurus, I recently added several taxa to the Lepidosauria. Each one found a nesting site without rearranging any other portion of the tree. This is a sign of tree strength.

Evans SE and Wang Y 2005. The Early Cretaceous lizard Dalinghosaurus from China. Acta Palaeontologica Polonica 50 (4): 725–742. online pdf
Ji SA 1998. A new long−tailed lizard from the Upper Jurassic of Liaoning, China. In: Department of Geology, Peking University (ed.), Collected Works of International Symposium on Geological Science, Peking University, Beijing, China, pp 496–505. Seismological Press, Beijing.