Evolution at the base of Dorygnathus

The large pterosaur tree offers new insights into pterosaur evolution. Today we’ll look at the base of Dorygnathus (Middle Jurassic, Fig. 1), once thought an evolutionary oddity, but today is the key taxon at the base of all Cretaceous pterosaurs.

Figure 1. Evolution at the base of Dorygnathus to scale. Top: the BSP 1994 specimen assigned (erroneously) to Eudimorphodon. Top middle: Sordes. Bottom middle: Dorygnathus. Bottom: Jianchangnathus, a basal dorygnathid at the base of Wukongopteridae and Scaphognathia.

Sometimes a picture is worth a 1000 words. Here Sordes is derived from a sister to the BSP specimen, itself derived from the holotype Eudimorphodon, Dorygnathus is derived from a sister to Sordes. Jianchangnathus is derived from a sister to the basalmost Dorygnathus.

Other than the jianchangnathids, dorygnathus ultimately gave rise to ctenochasmatids and azhdarchids.

Jianchangnathids ultimately gave rise to wukongopteripterids and a more successful lineage, the scaphognathids, which ultimately gave rise to most other Cretaceous pterosaurs.

So, the generalized “plain brown sparrow” look of Sordes belies its genetic potential to create giant pterosaurs, crested pterosaurs and even flightless pterosaurs. And the large variation demonstrated by these four “sister” taxa indicates that there are many more taxa waiting to be discovered that will be transitional forms nesting between them.

The Edinburgh Rhamphorhynchus: biting its own tail!

The Edinburgh Rhamphorhynchus (Figs 1, 2, museum number unknown) is fairly complete, but my goodness, what a roadkill! Is it right side up? Or upside down? With the tail and head on the same side, and actually biting it’s own tail, this specimen offers nothing but confusion at first glance.

Figure 1. Click to enlarge. The Edinburgh specimen of Rhamphorhynchus, as if a Jurassic truck had just run over it. This one is biting its own tail!

It’s rare to find wing membranes and a tail vane, but both are present here. Things are a little easier to see when the parts are colorized (Fig. 2).

Figure 2. Click to enlarge. Edinburgh Rhamphorhynchus parts colorized. In the right wing membrane note how the trailing edge of the wing membrane curves back to the knee (proximal tibia).

Someday I’ll do a reconstruction when I find higher resolution. There’s likely more to this specimen that just cannot be seen without better data (higher resolution). Even so, note that the wing membrane appears to curve back toward the knee. Unfortunately this specimen will not solve any arguments with regard to deep chord (to the ankle) vs. narrow chord (to the elbow) wing membranes.

The specimen is not listed in the Wellnhofer (1975) catalog that lists 108 other Rhamphorhynchus specimens. Thus, I’ll guess that it is a newer specimen. It appears to be a R. muensteri species. Scale is unknown, but likely was mid-sized if so.

Reference
Wellnhofer P 1975a-c. Teil I. Die Rhamphorhynchoidea (Pterosauria) der Oberjura-Plattenkalke Süddeutschlands. Allgemeine Skelettmorphologie. Paleontographica A 148: 1-33.Teil II. Systematische Beschreibung. Paleontographica A 148: 132-186. Teil III. Paläokolgie und Stammesgeschichte. Palaeontographica 149: 1-30.

wiki/Rhamphorhynchus

Therapsid phylogeny revisited – svp abstracts 2013

From the abstract
Kammerer et al 2013 wrote: “Therapsida is comprised of five well-characterized major subclades, all of which appear simultaneously in the middle Permian fossil record: Biarmosuchia, Dinocephalia, Anomodontia, Gorgonopsia, and Eutheriodontia (containing Therocephalia and Cynodontia). Although these subclades have generally been recovered as reciprocal monophyla, the relationships between them has been subject to little consensus. In the past few decades the best-supported topology for therapsids has been a pectinate tree composed of the higher-level clades Theriodontia (Gorgonopsia + Eutheriodontia), Neotherapsida (Anomodontia + Theriodontia), and Eutherapsida (Dinocephalia + Neotherapsida). Recently, numerous advances have been made in our knowledge of the diversity and anatomy of the earliest therapsids, including new discoveries (e.g., Raranimus, Tiarajudens) and redescriptions (e.g., of the earliest known anomodont, Biseridens, and the earliest known gorgonopsian, Eriphostoma). Utilizing this new information, we have produced the most comprehensive phylogenetic analysis of early therapsid relationships yet, including almost every biarmosuchian, dinocephalian, and basal anomodont as well as representative basal dicynodonts, gorgonopsians, therocephalians, and cynodonts. The results of this analysis indicate that therapsid phylogeny is split into two major subclades (Dinocephalia+Anomodontia and Biarmosuchia + Theriodontia), with only Raranimus falling outside of this dichotomy. “Biarmosuchia” is found to be paraphyletic with regards to Theriodontia, with the South African “ictidorhinids” more closely related to theriodonts than Biarmosuchus. Dinocephalian monophyly is poorly supported, although its component subclasses Anteosauria and Tapinocephalia are recovered with strong support. “Neotherapsida” is found to be an artifact of long branch attraction; with the exception of the freestanding dentary coronoid process, all the characters traditionally used to support this clade are absent in early anomodonts like Biseridens. Intriguingly, this topology conforms with prominent pre-cladistic classifications of Therapsida, albeit with different characters supporting these relationships. Characters related to simplification of the palate and expansion of the jaw muscles are reconstructed as particularly homoplastic, with parallel trends in multiple therapsid clades. New work in the middle Permian is of vital importance towards documenting character acquisition during the rapid initial radiation of therapsids.”

Notes
Well, this doesn’t match the results of the large reptile tree, (synapsid subdivision) in which the Therapsida is essentially diphyletic (anomodonts + the rest) and all derived from Ophiacodon + Archaeothyris, not sphenacodontids. The problem may be at the very base. Despite listing “almost every’ therapsid taxon known, every good phylogenetic tree has to go deeply into its ancestral taxa to be sure the correct ancestors are found and to set patterns for the evolution of the basal therapsids. I only hope this was done.

Figure 1. Click to enlarge and see the latest version. This small image of the family tree of the Therapsida including the Dicynodontia is not up to date.

I fear that they also included Tetraceratops, a taxon often associated with the base of the Therapsida, but actually is a limnoscelid close to Tseajaia. Kammerer considered the basal therapsid, Stenocybus, a juvenile dinocephalian. So that’s a problem in that this key taxon at the base of the Anomodontia might not have been included because of its purported juvenile status.

Biseridens was not recovered as an anomodont, but as a dinocephalian in the large reptile tree. We need to look at this together.

Perhaps not a problem here, but…
There’s a disturbing habit emerging of professional paleontologists employing highly derived taxa, like Mesosuchus (Nesbitt 2011) and Tetraceratops, as basal (plesiomorphic) taxa. Remember, it’s the plain brown sparrows of the world that ultimately give rise to the wild bizarre forms, not the other way around.In any case, I look forward to the publication of this work.

References
Kammerer C, Jansen M and Frobisch J 2013. Therapsid phylogeny revisited. Journal of Vertebrate Paleontology abstracts 2013.

Lower Cretaceous Pterosaur Trackways from China

To no one’s surprise the new Lower Cretaceous pterosaur tracks (Xing et  al. 2013) belong to a species close to Jidapterus (Fig. 1), with its big hand, narrow foot and short toes nearly all the same length except digit 1 is shorter. Interesting  that the claw marks did not impress.

Figure 1. Jidapterus compared to the new Lower Cretaceous pterosaur tracks. It’s a pretty close match. Little pedal digit 5 makes a small impression. The unguals do not. 

I’m not sure about the relative lengths of the phalanges in manual digit 3. Sister taxa appear to go both ways (long middle phalanx and short) and the poor data I have is fuzzy at best for the tiny bones. If anyone has better data for any Lower Cretaceous Chinese azhdarchids, please send!

References
Xing L, Lockley MG, Piñuela L, Zhang J-P, Klein H, Li D-Q, Wang F-P 2013. Pterosaur trackways from the Lower Cretaceous Jiaguan Formation (Barremian–Albian) of Qijiang, Southwest China. Palaeo 661. doi: 10.1016/j.palaeo.2013.09.003

What do sea turtles and pterosaurs have in common?

Sea turtles can swim for long periods without resting. 
Pterosaurs could probably fly for long periods without resting.

Figure 1. Sea turtle and the pterosaur Arthurdactylus, both in dorsal view, not to scale.

Sea turtles have forelimbs transformed into underwater wings.
Pterosaurs have forelimbs transformed into aerial wings.

Sea turtles have complex lungs (large surface area and volume). 
Pterosaurs probably had complex lungs, too, as air sacs penetrated the skeleton, as in birds.

Sea turtles have regional endothermy. Active tissues are warmer than surrounding waters.
Pterosaurs were probably endothermic overall, but it’s interesting to think about them being able to regionalize hotter and cooler areas by restricting bloodflow.

This post was inspired by the question:
Did pterosaur wings act like gills to supplement gas exchange and blood cooling? After all, the wings were quilted with blood vessels. They were thin and exposed to lots of oxygen.

In sea turtles cloacal sacs supplement gas exchange while underwater. So, it’s true, some turtles do breathe through their butts! The champion turtle in this regard is Rheodytes leukops, an Australian river turtle.

Now, I’m not saying the same for the pterosaur cloaca.

I’m just suggesting that a thin membrane festooned with blood vessels and exposed to a clear airstream, might have taken advantage of this for gas and heat exchange. Probably a question that can never be answered, or else is painfully obvious.

Nature always finds a way.

Certainly in pterosaurs the trachea, lungs and air sacs did most of the work, even in pterosaurs with no external nares. Oxygen was always surrounding them. Yes, in certain pterosaurs, the external nares did not merge with the antorbital fenestra, but became vestigial, as in plesiosaurs. Even so, simply opening the mouth even a tiny slit, would admit plenty of air to the pterosaur trachea.

Added August 16, 2013. Reprinted from earlier. 
Sam Johnson commented on July 28, 2012: “It seems more logical that the hollow bones transported oxygen from the membranes to the flight muscles than from the lungs to the membranes without muscles.

“Flying requires an enormous amount of energy. the bat Pallas has the fastest metabolic rate of any mammal (including shrews), requiring a lot of oxygen. Mammalian lungs simply cannot supply such large amounts of oxygen, being far less efficient than bird lungs. For example, the Mexican free-tailes bat can reach 60 mph for a few seconds and its lungs simply cannot provide the huge amount of oxygen required for that much power. Their low wing loading (relative to birds) may perhaps be due to respiration. During flight the underside is at a higher pressure, enhancing oxygen uptake and the top side is at a lower pressure, enhancing CO2 release.”

So maybe there’s something to this wing/gill idea. It works in bats (Makanya and Mortola 2007), who wrote: “We conclude that in [the fruit bat] Epomophorus wahlbergi, the wing web has structural modifications that permit a substantial contribution to the total gas exchange.”

References
Makanya AN and Mortola JP 2007. The structural design of the bat wing web and its possible role in gas exchange. J Anat. 2007 Dec;211(6):687-97. Epub 2007 Oct 26.

Pterosaur isometric growth occurs after hatching. Allometric growth occurs in the egg.

For those who missed it, Neil Brocklehurst was kind enough to remind me that, “It is IMPOSSIBLE for evolution to occur without changes in ontogeny. Not unlikely, not rare, impossible.”

My answer: The changes you refer to occur in the egg, not after hatching. As I’ve posted before, the fact that Pterodaustro came into sexual maturity at half of its final old age size (Chinsamy et al. 2008) is the mechanism by which pterosaurs could rapidly produce half-size progeny (embryo can’t be larger than the egg > egg can’t be larger than the pelvic opening). Evidently this can be applied across the pterosaur board, based on the many tiny transitional series that phylogenetically follow larger specimens and then beget larger specimens that live longer, get bigger, have bigger eggs, etc.

References
Chinsamy A, Codorniú L and Chiappe LM 2008. Developmental growth patterns of the filter-feeder pterosaur, Pterodaustro guinazui. Biology Letters, 4: 282-285.

A third look at Daemonosaurus

Updated February 28, 2015 with a new skull for Daemonosaurus.

Daemonosaurus is an odd sort of dinosaur that we looked at earlier here and here. Huge teeth. Seems hyper-carnivorous. Apparently not so according to its phylogenetic nesting.

Earlier the large reptile tree nested Daemonosaurus at the base of the Ornithischia (more primitive than Scelidosaurus), not far from Pampadromaeus, Thecodontosaurus and Sacisaurus, among the other phytodinosaur clades. After learning about the palate of Heterodontosaurus, gaining new insights into Pantydraco and revising the scoring of several dinosaur taxa (thanks M. Mortimer), I traced another DGS illustration and created a new reconstruction.

Figure 1. Daemonosaurus skull in 3 views. The new reconstruction is narrower than previously with a new descending pterygoid flange and very few other refinements. The jaw is shorter. The dentary fang(s) appear to slip into that pmx/mx notch as in Heterodontosaurus.

See, I’m always ready to change and ready to learn.
And there’s still more to learn, I’m sure. Here are the new in situ images.

Figure 2. Skulls of Daemonosaurus, Haya and Jeholosaurus to scale.

The intermediate length cervicals suggest Daemonosaurus had a neck, midway between that of a typical ornithischian and sauropodomorph.

Daemonosaurus has a high naris, like Herrerasaurus and Massospondylus. The narial depression drops first, then the naris follows on other taxa.

Daemonosaurus has a shorter rostrum than Herrerasaurus and shorter than Pampadromaeus, but not as short as the phytodinosaurs.

The descending quadrate and quadratojugal in Daemonosaurus are also found in Heterodontosaurus and others.

References
Sues H-D, Nesbitt SJ, Berman DS and Henrici AC 2011. A late-surviving basal theropod dinosaur from the latest Triassic of North America. Proceedings of the Royal Society B, published online 

wiki/Daemonosaurus

The foot of Tropidosuchus: Another Lagerpeton sister without a pedal digit 5

This post follows
an earlier one that found fault with Niedzwiedzki et al. (2013) and Brusatte et al. (2011), which attempted to match four-toed Lagerpeton (Fig. 2) to five-toed Prorotodactylus and Rotodactylus ichnites, claiming these tracks represented the earliest examples of dinosauromorphs in the fossil record. Beside the morphological mismatch, which they acknowledged yet based their papers on, the large reptile tree found Lagerpeton was not even a dinosaur ancestor, but nested far afield with another chanaresuchid, Tropidosuchus (Fig. 3). Here we’ll show another Lagerpeton/Tropidosuchus sister with a metatarsal 5 lacking a pedal digit 5 sealing the deal that neither Lagerpeton, nor any close sister, could have made Prorotodactylus or Rotodactylus tracks. Even a further distant sister, Chanaresuchus (Fig. 4), lacks pedal digit 5.

Figure 1. An unidentified skeleton attributed to Tropidosuchus, but shares traits with Lagerpeton. No pedal digit 5 here.

Above is a specimen and its reconstruction attributed to Tropidosuchus (Bonaparte 1994), but notice the difference in the pedal proportions and other proportions. The foot morphology is much closer to Lagerpeton. This specimen also has a smaller humerus than Tropidosuchus. The pelvis is distinct from both genera. Chevrons are missing from this specimen. Chevrons may be missing form this clade, which otherwise shares a relatively wide tail base according to the caudal transverse processes.

Figure 2. Lagerpeton reconstructed. No pedal digit 5 here.

Tropidosuchus romeri (Arcucci 1990) Late Triassic was originally considered a lagosuchid like Marasuchus but here derived from a sister to BPI 2871 and Chanaresuchus. The pes of Tropidosuchus was quite similar to that of Chanaresuchus emphasizing digit 2 with a slender metatarsal 4. The tarsals did not have a calcaneal heel.

Figure 3. The holotype of Tropidosuchus retains the narrower digit 4 of Chanaresuchus. No pedal digit 5 preserved here.

Chanaresuchus bonapartei (Romer 1971) Anisian, Early Middle Triassic is a sister to Tropidosuchus. The most robust metatarsal was mt 2 Digit 3 was the longest. Metatarsal IV was extremely gracile and digit V was absent.

Figure 4. Chanaresuchus a quadrupedal ancestor to Tropidosuchus and Lagerpeton and the third Tropidosuchus-like taxon. No pedal digit 5 here.

So, why are professors promoting such mismatches? Why are reviewers approving such mismatches? Better matches to Prorotodactylus and Rotodactylus can be found in several untested taxa, as we saw earlier here. There must be a syndicate operating here, friends helping friends. Sometimes Science needs critics, not friends.

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.

References
Arcucci A 1987. Un nuevo Lagosuchidae (Thecodontia- Pseudosuchia) de la fauna de Los Chañares (edad reptil Chañarense, Triásico Medio), La Rioja, Argentina. Ameghiniana 24, 89–94.
Bonaparte JF 1994. Dinosaurios de America del Sur. Impreso en Artes Gráficas Sagitario. Buenes Aires. 174pp. ISBN: 9504368581
Brusatte SL, Niedz´wiedzki G and Butler RJ 2011. Footprints pull origin and diversification of dinosaur stem lineage deep into Early Triassic. Proceedings of the Royal Society B, 278, 1107–1113.
Niedzwiedzki G, Brusatte SL and Butler RJ 2013. Prorotodactylus and Rotodactylus tracks: an ichnological record of dinosauromorphs from the Early–Middle Triassic of Poland. Geological Society, London, Special Publications, first published April 23, 2013. doi 10.1144/SP379.12.

wiki/Lagerpeton

This Science poster seems to resonate

 

I pledge, in the name of Science, to continue making and correcting mistakes.

David Peters

A new look for Arizonasaurus and a new nesting

As a rauisuchid,
Arizonasaurus stands out as an oddball. The dorsal fin, first of all. No other rauisuchid has anything like it. The cervical centra are longer than tall, not taller than long as in other rauisuchids. The skull was restored by Nesbitt (2011) as dinosaurian, but the deep maxilla with its sharp ventral angle and extensive pre-ascending process appears to be more like the skull of Qianosuchus.

Due to a paucity of elements, the skull of Arizonasaurus has to be largely restored and there are several ways to do that.

Xilosuchus (Fig. 1) was associated with Arizonasaurus, but Xilosuchus has a much longer neck, again, very odd for a rauisuchid. Here (Fig. 1) the cervicals match pretty well with Yarasuchus, a sister to Qianosuchus.

Figure 1. Xilosuchus and Yarasuchus. Red arrows point to similar spacing of cervicals.

There is another clade of archosauriforms with elongate neck centra and large neural spines. At present only two taxa occupy this clade, Yarasuchus and Qianosuchus. They happen to be sisters to Ticinosuchus + aetosaurs in the large reptile tree. Here’s my case for inclusion of Xilosuchus and Arizonasaurus within the Yarasuchus clade. I’m not the first to do so. Brusatte et al. (2012 their figure 5) have already nested them together. So did Nesbitt (2011, his figure 51.)

But they didn’t reconstruct the skull of Arizonasaurus as if they were sisters. Here (Fig. 2) is a new restoration of Arizonasaurus.

Figure 2. The skull of Arizonasaurus reconstructed with available data, including maxilla and dentary bones from Nesbitt (2003) and postcranial drawings from Nesbitt (2011). So much is missing from the Arizonasaurus skull that several interpretations can be created. The distinctly angled maxilla is closest to Qianosuchus and was interpreted differently by Nesbitt (2011) as a more dinosaurian-type skull. The Xilosuchus skull is also by Nesbitt. Would be nice to see the actual materials.

When we put it all together (Fig. 3), using all available data, Arizonsaurus has a new look and a new nesting. The long neural spines of the Yarasuchus clade are simply that much longer in Arizonasaurus, along with a shorter neck — but not so short that the centra are taller than wide. In Arizonasaurus the neural spines might be tall, but the centra, at least some of them, appear to be longer than tall.

Figure 3. Arizonasaurus with a new Qianosuchus-like skull. The ventral pubis is common to the Qianosuchus clade. The scapula is typically larger and dorsally broader than in Arizonasaurus.

As a yarasuchid or qianosuchid, Arizonasaurus still stands out as a oddball, but less of one. Arizonasaurus appears to have a smaller scapula than the others, but it is incomplete and could be larger. The depth of the pubis also foretells longer hind limbs than in the short-limbed yarasuchid/qianosuchids. The large reptile tree will reflect this new nesting the next time I update it. Still no relation to the other finback, Lotosaurus, though, which continues to nest with the plant-eating dinosaurs.

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.

References
Butler RJ, Brusatte SL, Reich M, Nesbitt SJ, Schoch RR, et al. 2011. The Sail-Backed Reptile Ctenosauriscus from the Latest Early Triassic of Germany and the Timing and Biogeography of the Early Archosaur Radiation. PLoS ONE 6(10): e25693. doi:10.1371/journal.pone.0025693 Plos One paper
Nesbitt SJ 2003. Arizonasaurus and its implications for archosaur divergence
Sterling J. Nesbitt Proceedings of the Royal Society, London B (Suppl.) 270, S234–S237. DOI 10.1098/rsbl.2003.0066
Welles SP 1947 Vertebrates from the Upper Moenkopi Formation of the Northern Arizona. Univ. California Publ. Geol. Sci. 27, 241–294.

wiki/Arizonasaurus
wiki/Ctenosauriscus