The Origin of Dinosaurs x2 (2010) revisited

Several years ago
the top vertebrate paleontologists in the world (Brusatte et al. 2010) reported on the origin of dinosaurs. Coincidentally Langer et al. (2010) wrote a similar report.  It is now 6 years later. Let’s see how well those report have held up as they compare to the current data (2016) in the large reptile tree.

Figure 2. The origin of dinosaurs to scale. Gray arrows show the direction of evolution. This image includes Decuriasuchus, Turfanosuchus, Gracilisuchus, Lewisuchus, Pseudhesperosuchus, Trialestes, Herrerasaurus, Tawa and Eoraptor.

From the Brusatte et al. introduction
“During the past 25 years, numerous new fossils, reinterpretations of long-forgotten specimens, and numerical analyses have significantly revised our understanding of this major macroevolutionary event, which is one of the most profound and important evolutionary radiations in the history of life.”

What has stood the test of time:

1. Dinosaurs are archosaurs: birds+crocs and last common ancestor
2. Archosaurs are diapsid reptiles = Eudibamis, Petrolacosaurus and all their descendants.
3. Dinosaurs are: “Triceratops horridus, Passer domesticus, and all descendants of their most recent common ancestor.” Or alternatively: ““the least inclusive clade containing Megalosaurus and Iguanodon.” Thus dinosaurs are monophyletic.
4. The suite of traits common to dinosaurs include: 1) upright and fully erect posture [shared with basal crocs and dinosauromorphs]; 2) an enlarged deltopectoral crest on the humerus [shared with Trialestes]; 3) a “specialized” hand; 4) a perforated acetabulum (hip socket) [which may close]; 5) a well-developed fourth trochanter on the femur; 6) a lesser trochanter on the femur; 7) and a simple hinge ankle joint with proximal tarsals fixed immovably to the tibia and fibula [shared with basal crocs and dinosauromorphs].
5. Dinosaurs likely originated during the Middle Triassic. They are diverse at the earliest Late Triassic.
6. Herrerasaurus and Eoraptor are some of the most complete specimens of any early dinosaur.
7. Langer: Herrerasaurs are basal to the Ornithischia-Saurischia dichotomy, but the actual dichotomy is Theropoda/Phytodinosauria
8. Langer: The oldest dinosaurs include Herrerasaurus, Eoraptor, Staurikosaurus, Saturnalia and Panphagia all from the Carnian (early Late Triassic). These are also among the most primitive dinosaurs. Missing from this list is Barberenasuchus, also Carnian, not commonly considered a dinosaur, but nests as a sister to Eodromaeus.

What has not stood the test of time:

1. Archosaurs (crocs + dinos alone) no longer include pterosaurs
2. Diapsids no longer include lizards, snakes, rhynchocephalians (including rhynchosaurs and trilophosaurs) and pterosaurs. Those have a diapsid skull by convergence.
3. Arizonasaurus is no longer an archosaur since crocs and birds had a more recent common ancestor, a sister to Gracilisuchus.
4. The clades Crurotarsi (= Pseudosuchia) and Avemetatarsalia (= Ornithodira, Ornithosuchia) are now junior synonyms for older nomenclature based on their inclusion sets (Archosauriformes and Reptilia respectively).
5. Pterosaurs no longer nest with archosaurs, but with lepidosaurs, in a new clade known as the Tritosauria nesting between basal rhynchocephalians and basal protosquamates.
6. Lagerpeton is not a dinosauromorph, but a sister to Tropidosuchus.
7. Marausuchus is does not nest outside the Dinosauria, but as a basal theropod.
8. Sacisaurus, Silesaurus and Asilisaurus are not the immediate sisters of dinosaurs. Rather they now nest with poposaurs, the proximal outgroup to the Archosauria (crocs + dinos only).
9. Overlooked by Brusatte et al., Lewisuchus, Zupaysaurus, Pseudhesperosuchus, Trialestes, and their kin are the now the immediate sisters of dinosaur, the true dinosauromorphs.
10. Some manner of feathers now diagnose the Dinosauria, which primitively had naked (not scaly) skin, like a plucked chicken.
11. Herrerasaurus and Eoraptor are no longer incerta sedis but the most basal dinosaur and one of the basal phytodinosaurs respectively.
12. Zupaysaurus no longer nests as a theropod, but a dinosauromorph
13. Berberosaurus no longer nests as a theropod, but as the basalmost phytodinosaur
14. Ornithischia no longer branch off first from Saurischia, but are derived from basal phytodinosaurs. Sauropodomorpha are sisters to basal Ornithisichia with Daemonosaurus and Chilesaurus at the base.
15. Langer: Eusaurischia (Sauropodomorpha + Theropoda) is a junior synonym for Dinosauria
16. Langer: Silsauridae (all taxa closer to Silesaurus than to Marasuchus + Heterodontosaurus) is a junior synonym for Poposauria, if kept monophyletic.
17. Langer: the basal-most dinosaurs were not probably omnivorous,
18. Langer: herrerasaurs were not theropods
19. Langer: there is no Onithischia-Saurischia dichotomy. Saurischia is a junior synonym  for Dinosauria.
20. Langer: Agnophitys is a dinosaur sister to Marasuchus.
21. Langer: Putative dinosaur Saltopus is a basal archosaur close to Gracilisuchus.

Figure 1. Click to enlarge. Subset of the large reptile tree focusing on the Archosauria (crocs + dinos). Sharp-eyed observers will find minor changes here.

Staurikosaurus
Langer et al. (2010) mentioned Staurikosaurus (Colbert 1970) as the first consensual early dinosaur to be collected. Here it nests as a basal theropod, basal to a clade of theropods that is often overlooked that includes Marasuchus, Procompsognathus and Segisaurus. Yes, Staurikosaurus has but two sacral vertebrae. So do other clade members.

Guaibasaurus
Langer et al. (2010) also mentioned Guaibasaurus (Bonaparte et al., 1999) who reported, “The mesotarsal condition and the outline of the distal section of tibia indicate the saurischian nature of this new form, but the almost unreduced medial wall of the acetabular portion of ilium shows an unrecorded primitive condition within the cited group. Several features suggesting affinities with both the Prosauropoda and Theropoda, imply that Guaibasaurus candelariensis may belong to the ancestral group for both of them.” The large reptile tree nests Guaibasaurus as a basal theropod and as the sister to Marasuchus + Procompsognathus, not far from Staurikosaurus. 

The Novas (1992) dinosaur definition
According to Langer et al., Novas (1992b) provided the first phylogenetic definition of Dinosauria as ‘‘the common ancestor of Herrerasauridae and Saurischia + Ornithischia, and all of its descendants’’. The addition of herrerasaurs does not change the current tree (Fig. 1). Padian & May (1993) explicitly restricted the use of Dinosauria to the clade composed of Saurischia and Ornithischia, exclusive of ‘‘Herrerasaurus and its allies’’. But Novas has priority. Moreover, the last common ancestor of Saurischia and Ornithischia is currently a herrerasaur. The diagnosis of the Dinosauria has seen some changes over the years. Many are traits that are not covered by the large reptile tree. Please check out the references below for lists and histories of those lists.

What does the large reptile tree diagnose dinosaurs?
The following suite of traits are found in basal dinosaurs and not their proximal outgroups, Trialestes, the Pseudhesperosuchus clade. However many of these traits are found elsewhere on the tree. And many traits are lost in more derived dinos.

1. Naris opening lateral
2. Parietal skull table weakly constructed
3. Mandible tip straight (neither upturned nor down)
4. Interclavicle poorly ossified or absent
5. Coracoid shape disc-like, even if fused (elongate or strap shape in outgroup)
6. Radiale and ulnare not elongated (as in outgroup)
7. Manus with long penultimate phalanxes and raptorial claws
8. Femoral head interned and sub rectangular (reversed in the Marasuchus clade).
9. Longest metatarsal: 3
10. Proximal metatarsals: 1 and 5 reduced

Bipedality
has long been touted as a key dinosaurian trait, but dinosaurs evolved from basal bipedal crocodylomorphs, like Gracilisuchus and Scleromochlus. Interesting that Scleromochlus has been often associated with unrelated pterosaurs. Pterosaur removal sets things a little straighter in the retelling of the dinosaur ancestry story. Scleromochlus has not often been touted as a dinosaur ancestor, but by virtue of its false association with pterosaurs in various cladograms, it has always been there.

The long coracoids and proximal carpals of basal bipedal crocs
have set them apart from consideration as possible dino ancestors. But if you just let the software do its job, then you’ll recover nestings that indicate the elongate coracoids and proximal carpals became reduced to shorter, more primitive conditions in basal dinos.

Traits found in dinosaurs exclusive of Herrerasaurus:

1. Feathers (not on the matrix, but worth mentioning)
2. Skull shorter than cervicals
3. Cranium convex
4. Naris opening
5. Maxilla ventral margin straight
6. Jugal qj process straight
7. Quadrate curls posterodorsally
8. Jaw joint aligned with ventral maxilla
9. Canine maxillary teeth not present
10. Nine or more cervical vertebrae
11. Some caudal vertebrae 3x longer than tall
12. Tibia not shorter than femur
13. Metatarsus not shorter than half the tibia
14. Phalanges on metatarsal 5: 0 (reversed in higher clades)

Then if wanted to
you could simply list all the traits of Herrerasaurus, the basalmost dinosaur, knowing full well that Herrerasaurus itself is derived from the first, as yet undiscovered, dinosaurs.

References
Bonaparte JF, Ferigolo J and Ribeiro M 1999. A new early Late Triassic saurischian dinosaur from Rio Grande do Sol state, Brazil” (PDF). Proceedings of the Second Gondwanan Dinosaur Symposium, National Science Museum Monographs 15: 89–109.
Brusatte SL, Nesbitt SJ, Irmis RB, Butler RG, Benton MJ and Norell MA 2010. The origin and early radiation of dinosaurs. Earth-Science Reviews 101 (2010) 68–100.
Colbert EH 1970. A Saurischian dinosaur from the Triassic of Brazil. American Museum Novitates 2405; 1-39
Langer MC. Ezcurra MD, BittencourtJS, Novas FE 2010. The origin and early evolution of dinosaurs. Biological Review 85, 55–110.