The assembly of the avian body plan (Cau 2018) pt. 3 of 3

Yesterday and the day before we looked at parts 1 and 2 of the Cau 2018 tree topology of theropods leading to birds (Fig. 3). Today: part 3 of 3.

Node 19. Maniraptora: (Alvarezsauroidea + Pennaraptora) Here’s where the second Compsognathus (CNJ79) would affect the Cau topology. The basal member of Maniraptora in the Cau tree is Jianchangosaurus, which has a toothless premaxillla and nests with therizinosaurs traditionally and in the LRT (Fig. 5). You don’t want a toothless premaxilla at this basal node because most succeeding taxa have premaxillary teeth. In the LRT, Jianchangosaurus is a derived therizinosaur, close to a surprise tiny therizinosaur with long forelimbs and a trenchant pedal digit 2, Rahonavis. That may change or be confirmed with more complete specimens.

The LRT agrees with Cau in nesting Shuvuuia with Haplocheirus.

Figure 1. Jianchangosaurus nests at the base of the Maniraptora in Cau 2018, but with therizinosaurs in the LRT.

Figure 1. Jianchangosaurus nests at the base of the Maniraptora in Cau 2018, but with therizinosaurs in the LRT, where it nests with Rahonavis.

Figure 2. Rahonavis nests in the LRT as a tiny derived therizinosaur based on the few bones currently known.

Figure 2. Rahonavis nests in the LRT as a tiny derived therizinosaur based on the few bones currently known.

Node 20. Pennaraptora (Oviraptorosauria + Paraves) The Cau study and the LRT agree that Caudipteryx and Khaan nest together. Lacking from the Cau study, Limusaurus (Fig. 1) nests as a basal oviraptorid in the LRT. In turn the Cau study includes taxa not listed in the LRT.

Node 21. Paraves: Distinct from the Cau tree, the LRT nests Microraptor with Ornitholestes and apart from Deinonychus and Velociraptor. The LRT nests Fukuiraptor with Zhenyuanlong with tyrannosaurs. The Cau study does not include Zhenyuanlong. 

Figure 1. The origin of birds cladogram according to Cau 2018. Taxon exclusion forces a mixup of basal taxa.

Figure 3. The origin of birds cladogram according to Cau 2018. Taxon exclusion forces a mixup of basal taxa.

Node 22. Averaptora: In the Cau study Sinovenator nests with Jinfengopteryx and Mei. In the LRT, Jinfengopteryx (Fig. 4) nests as a basal troodontid, derived from a sister to Velociraptor and Haplocheirus. Sinovenator nests closer to birds. Mei nests within birds (Scansoriopterygidae). Yi and Epidexpteryx are also scansoriopterygids. Cau nests them basal to Archaeopteryx.

 Jinfengopteryx, a basal troodontid in both studies.

FIgure 4. Jinfengopteryx, a basal troodontid in both studies. Think of this taxon like a neotonous velociraptor, leading to all troodontids including (with further neotony) birds. Note the resemblance to Solnhofen birds.

Employing only one Archaeopteryx in the Cau study
overlooks the variety in Solnhofen birds recovered by the LRT. When this is repaired with more taxa, let’s see what happens when more Solnhofen birds are added (Fig. 5):

Figure 1. Cladogram subset of the LRT focusing on Theropoda.

Figure 4. Cladogram subset of the LRT focusing on Theropoda, including extant birds.

As we learned
earlier, no two Solnhofen birds are identical. In the LRT they are distinct enough to nest in several basal bird clades. This was completely missed by Cau and most other bird workers.

Missing from the Cau taxon list
are any living birds. In the LRT, the toothed Cretaceous birds nest between paleognaths and neognaths, so that branch was missed.

Sometimes
taxon exclusion adversely affects tree topologies. Start with a wide gamut analysis (Fig. 5) that sets limits on the more focused study that you want to look at.

References
Cau A 2018. The assembly of the avian body plan: a 160-million-year long process. Invited Paper, Bollettino della Societa Paleontologica Italiana 57(1):1–25.

Advertisements

The assembly of the avian body plan (Cau 2018) pt. 2 of 3

Yesterday we looked at part 1 of the Cau 2018 cladogram of theropods (including birds). Certain taxa within this study were new to me, so I added several to the the large reptile tree (LRT, 1213 taxa). Today we’ll continue with Node 9, still within the Huxley stage of theropod evolution.

Figure 2. Zuolong skull revised with a backward tilting lacrimal and other minor modifications.

Figure 1. Zuolong skull revised with a backward tilting lacrimal and other minor modifications.

Node 9. Tetanurae: (Zuolong + Chilesaurus + Neotetanurae). Adding Zuolong to the LRT nests it as a basal theropod, basal to the rarely tested taxa in the Segisaurus + Marasuchus + Procompsognathus clade (Fig. 2). Zuolong is the first of these with a relatively complete skull. Cau 2018 nest Zuolong and the phytodinosaur, Chilesaurus, with Neotetanurae apparently by excluding certain relevant taxa.

Figure 1. Chilesaurus and kin, including Damonosaurus and basal phytodinosauria.

Figure 2. Chilesaurus and kin, including Damonosaurus and basal phytodinosauria. No close relatives of theropods here!

Node 10. Chilesaurus + Neotetanurae: (See node 9). Cau also reports, “The parsimony analysis confirms the basal tetanuran affinities of the enigmatic Chilesaurus and dismisses ornithischian relationships suggested by Baron & Barrett (2017).” This is only true base on taxon exclusion. Add back the missing taxa, like Daemonosaurus , Jeholosaurus and Haya and the tree topology will change.

Figure 1. The origin of birds cladogram according to Cau 2018. Taxon exclusion forces a mixup of basal taxa.

Figure 1. The origin of birds cladogram according to Cau 2018. Taxon exclusion forces a mixup of basal taxa.

Node 11. Neotetanurae (Carnosauria + Coelurosauria): The Cau tree and LRT share many taxa here, including Allosaurus with Sinraptor and Acrocanthosaurus. The Cau tree has only one Compsognathus. The LRT has two, each at the base of its own clade. The Cau tree nests Megalosaurus between the phytodinosaur, Chilesaurus, and the small compsognathid, Aorun, which is odd on the face of it (= no gradual accumulation of traits).

Node 12. Coelurosauria: The LRT indicates that Sinocalliopteryx does not belong in this clade, as Cau recovers it, but Sinocalliopteryx nests much more primitively, basal to Coelophysis and kin.

Nodes 13. Compsognathid grade + Tyrannoraptora: The spinosaurs and kin are not present in the Cau taxon list. When present these long rostrum taxa attract Guanllong and Megaraptor to more primitive theropods, away from tyrannosaurs, despite convergent traits.

Node 14. Sinocalliopteryx + Tyrannoraptora: See Node 13.

Node 15. Tyrannoraptora: (Tyrannosauroids and maniraptoromorphs) See Node 13.

Node 16. Maniraptoromorpha: (includes Vultur, excludes Tyrannosaurus). This definition is a little vague. Wish it had at least one included basal taxon. In the Cau tree Coelurus is a basal taxon. Unfortunately, too little of it is known to add it to the LRT.

Node 17. Ornitholestes + Maniraptoriformes: Distinct from the Cau tree, Ornitholestes is basal to microraptorids and tyrannosaurs, as well as dromaeosaurs, troodontids and birds.

Node 18. Maniraptoriformes: (Ornithomimosauria + Maniraptora). Distinct from the Cau tree, ornithomimosaurs in the LRT are derived directly from the holotype of Compsognathus, separate from oviraptorids and dromaeosaurs, closer to tyrannosaurs and kin. The LRT nests Ornitholestes and kin on the bird side of therizinosaurus + oviraptorids. The Cau tree does the opposite.

Figure 1. Cladogram subset of the LRT focusing on Theropoda.

Figure 2. Cladogram subset of the LRT focusing on Theropoda.

Part 3 tomorrow.

References
Cau A 2018. The assembly of the avian body plan: a 160-million-year long process. Invited Paper, Bollettino della Societa Paleontologica Italiana 57(1):1–25.

 

The assembly of the avian body plan (Cau 2018) pt. 1 of 3

Dr. Andrea Cau 2018 summarizes traditional knowledge
on the origin of birds breaking the process into three stages:

  1. Huxleyian stage: Early Triassic to Middle Jurassic the earliest ancestors of birds acquired postcranial pneumatisation, an obligate bipedal and digitigrade posture, the tridactyl hand and feather-like integument
  2. Ostromian stage: Middle to Late Jurassic is characterised by a higher evolutionary rate, the loss of hypercarnivory, the enlargement of the braincase, the dramatic reduction of the caudofemoral module, and the development of true pennaceous feathers.
  3. Marshian stage: Cretaceous. The transition to powered fl ight with the re-organisation of both forelimb and tail as fl ight-adapted organs and the full
    acquisition of the modern bauplan

This is a pretty good plan overall.
Unfortunately Dr. Cau uses an antiquated cladogram (Fig. 1) riddled with taxon exclusion (especially among the outgroups), so the details tend to get a little messed up. Let’s review the pluses and minuses.

Figure 1. The origin of birds cladogram according to Cau 2018. Taxon exclusion forces a mixup of basal taxa.

Figure 1. The origin of birds cladogram according to Cau 2018. Taxon exclusion forces a mixup of basal taxa.

The Cau cladogram and LRT
both feature many of the same basal theropods at the beginning, birds at derived nodes and a variety of carnivores in between, with dromaeosaurs then troodontids leading to birds.

Dr. Cau opens his paper
with several paragraphs devoted to nomenclature. He finds the term ‘non-avian’ particularly irksome. Cau employed 132 taxa and 1781 (1431 informative) characters. He reports that he decided not to include pterosaurs as outgroup taxa. That shows wisdom.

Unfortunately
Cau was not wise to largely ignore basal bipedal crocodylomorphs, including such favorites as Scleromochlus and Gracilisuchus. Thankfully Lewisuchus made his list.

So Cau starts off with four very distant outgroup taxa (Euparkeria, Teleocrator, Dormomeron and Lagerpeton), and that is never good (relevant taxon exclusion, again). It also shows a lack of understanding that could have been had with a quick glance at the large reptile tree (LRT, 1213 taxa). That’s what it’s there for.

Cau 2018 Results
3072 shortest trees (vs. LRT has one, fully resolved tree, last time I tested the whole tree).

Here are Cau’s nodes:

  1. Teleocrater + Dinosauromorpha: Unfortunately this clade does not include the Crocodylomorpha, so it is invalid. ‘Dinosauromorpha’, at best, is a junior synonym of Archosauria in the LRT.
  2. Dinosaurormorpha: (Lagerpetids + Dinosauriformes). Unfortunately this clade does not include the Crocodylomorpha, so it is invalid. When more taxa are added, lagerpetids nest with Tropidosuchus among the chanaresuchidae. Thus,  ‘Dinosauriformes’, at best, is a junior synonym of Archosauria in the LRT.
  3. Dinosauriformes: (Marasuchus + Dracohors). More taxa move Lewisuchus into the Crocodylomorpha, Silesaurus into the Poposauria and Pisanosaurus deep into the Phytodinosauria.
  4. Dracohors: (includes Megalosaurus, but not Marasuchus). More taxa (e.g. Segisaurus, Procompsognathus) move Marasuchus into the Theropoda and other taxa as listed above in the LRT.
  5. Dinosauria: (Eodromaeus, Herrerasauridae, Sauropodomorpha and Ornithoscelida). This is too many taxa and shows a lack of understanding. No basal dichotomy can be made. The LRT defines Dinosauria as Theropoda + Phytodinosauria, their last common ancestor (Herrerasaurus) and all descendants.
  6. Ornithoscelida: (Ornithischia + Theropoda) Adding more taxa will split up and invalidate this clade, based on LRT results.
  7. Theropoda: (Coelophysoidea + Averostra) In the LRT several theropods are listed as outgroups in the Cau analysis and it includes the phytodinosaur, Chilesaurus. (Daemonosaurus is curiously absent from this paper). Almost toothless Limusaurus should nest with oviraptorids. Elaphrosaurus has not been tested in the LRT. The basalmost coelophysoid (with feathers!), Sinocalliopteryx, nests as a derived compsognathid in the Cau taxon list.
  8. Averostra: (Ceratosauria + Tetaneurae) The LRT recovers a clade of large carnivores between Sinocalliopteryx and Compsognathus. This clade includes ProceratosaurusDeinocheirus, Xiongguanlong, Suchomimus and Spinosaurus, taxa not employed by Cau. These taxa attract Guanlong and Dilong to this basal feathered clade, away from tyrannosaurs. Otherwise, the LRT and Cau both place the same long list of medium to large basal theropods in clades at the base of this clade/grade.
Figure 1. Cladogram subset of the LRT focusing on Theropoda.

Figure 2. Cladogram subset of the LRT focusing on Theropoda.

More tomorrow.

References
Cau A 2018. The assembly of the avian body plan: a 160-million-year long process. Invited Paper, Bollettino della Societa Paleontologica Italiana 57(1):1–25.

 

Megaraptor: closer to spinosaurs than to tyrannosaurs

First of all,
what we know of Megaraptor is a chimaera. There is no complete skeleton. What we know comes from a little bit here (Fig. 1a) and a little bit there (Figs. 2, 3).

Wikipedia reports:
“Initially considered a giant dromaeosaur-like coelurosaur, then an eovenatorid allosauroid, then a basal tyrannosauroid coelurosaur…”

Here Megaraptor nests
with Xiongguanlong and other pre-spinosaurs, like Suchomimus (Fig. 7).

Porfiri et al. report,
“Megaraptorids are characterized by the formidable development of their manual claws on digits I and II and the transversely compressed and ventrally sharp ungual of the first manual digit. Phylogenetic relationships of megaraptorans have been the focus of recent debate. Megaraptorans have been alternatively interpreted as basal coelurosaurians (Novas,1998), basal tetanurans (Calvo et al., 2004; Smith et al., 2008), and allosauroids closely related with carcharodontosaurids (Smith et al., 2007; Benson et al., 2010; Carrano et al., 2012). However, recent evidence has been presented in favour of their inclusion within Coelurosauria, and possibly as basal members of Tyrannosauroidea (Novas et al., 2013). The [juvenile] skull material conforms a primary source of information for both coelurosaurian and tyrannosauroid features of Megaraptoridae, unavailable in previous studies.”

Figure 1. Megaraptor manus and ulna from xxx.

Figure 1a. Megaraptor manus and ulna from xxx.

Figure 1b. Suchomimus manus.

Figure 1b. Suchomimus manus.

Figure x. Suchomimus restoration.

Figure 1c. Suchomimus restoration. Note the large hands.

Novas et al. (2013) found Megaraptor and related taxa as deeply nested within Coelurosauria, notably as the sister group of Xiongguanlong + Tyrannosauridae. In the LRT Megaraptor also nests close to Xiongguanlong, but far from Tyrannosauridae.

Porifi et al. compared the juvenile skull
of Megaraptor to that of Dilong, which they nest (Fig. 4) as a basal tyranosauroid. In the LRT (Fig. 5) Dilong does indeed nest close to Megaraptor, but both nest far from Tyrannosaurus. Those long snouts are a spinosaur trait.

Figure 2. Scale bars seem to be amiss here, but these are the bones and scale bars from the MUCPv 595 specimen of the skull of Megaraptor.

Figure 2. Scale bars seem to be amiss here, but these are the bones and scale bars from the MUCPv 595 specimen of the skull of Megaraptor.

A new Megaraptor skull restoration
is presented below (Fig. 3), based on Xiongguanlong, a sister in the LRT.

Figure 1. Megaraptor skull restoration revised.

Figure 3. Megaraptor skull restoration revised (in blue). As in Xiongguanlong.

Several traditional long-snouted ‘tyrannosauroids’
nest with spinosauroids in the LRT. The two clades converge in many traits and paleontologists have, so far, been willing to accept that tyrannosaur ancestors had long snouts and elaborate rostral crests. Perhaps, someday, the consensus will swing the other way.

Figure x. The skull of Xiongguanlong is long and low, like that of spinosaurs and kin, not like that of tyrannosaurs and kin.

Figure 4. The skull of Xiongguanlong is long and low, like that of spinosaurs and kin, not like that of tyrannosaurs and kin.

Evidently
the Theropoda is a wicked clade to pylogenetically analyze. Workers have been shuffling the nodes trying to figure them out. Taxon exclusion may be the cause of this lack of consensus. The LRT lacks certain taxa. So do other studies.

Figure 2. Cladogram nesting Megaraptor from xxx.

Figure 6. Cladogram nesting Megaraptor from xxx.

Many branches are similar in many studies. Others are not.

Figure 5. Basal theropods with the addition of Zuolong and Megaraptor.

Figure 7. Basal theropods with the addition of Zuolong and Megaraptor. Scipionyx was added later (see figure 8).

Tyrannosauroids never had a long, low skull
in the LRT. They were derived from the the CNJ79 specimen of Compsognathus (yes, it needs a new generic name), which also includes ornithomimids, Fukuivenator, Tianyuraptor, Huaxiagnathus, Zhenyuanlong and traditional tyrannosaurids, like Alioramus.

Figure 3. The Scipionyx clade includes Allosaurus, Deinocheirus, Spinosaurus and other larger theropods.

Figure 8. The Scipionyx clade includes Allosaurus, Deinocheirus, Spinosaurus and other larger theropods.

References
Novas FE 1998. Megaraptor namunhuaiquii, gen. et sp. nov., a large-clawed, Late Cretaceous theropod from Patagonia. Journal of Vertebrate Paleontology. 18: 4–9.
Novas FE, Agnolin FL, Ezcurra MD.Porfiri J, Canale JI 2013.
Evolution of the carnivorous dinosaurs during the Cretaceous: the evidence from Patagonia. Cretaceous Research 45, 174e215.
Porfiri JD et al. (5 co-authors) 2014. Juvenile specimen of Megaraptor (Dinosauria, Theropoda) sheds light about tyrannosauroid radiation. Cretaceous Research 51:35–55.

wiki/Megaraptor

What is Scipionyx?

Scipionyx samniticus (Dal Sasso and Signore 1998; Dal Sasso and Maganuco 2011; SBA-SA 163760; 46.1cm; Early Cretaceous; Figs. 1,2) was originally considered a hatchling compsognathid, like Compsognathus. Here, in the large reptile tree (LRT, 1211 taxa; Fig. 3) it is a late-surviving member of an Early Jurassic radiation that produced SpinosaurusDeinocheirus and their kin, with basal taxa close to Proceratosaurus (Fig. 2). The short rostrum and large orbit indicate a juvenile status, perhaps only a few days old. Internal organs were preserved along with the bones, but no outer integument was preserved. The feet and long tail are conjectural. The vertebrae of the back are not pneumatised. This taxon provides some clues to the post-crania of Proceratosaurus.

Figure 1. Scipionyx skull and overall. The tail and feet are restored.

Figure 1. Scipionyx skull and overall. The tail and feet are restored.

Long considered a compsognathid
Scipionyx, nests at the base of its clade, as several other compsognathids nest in the LRT. Thus, various compsognathids gave rise to most theropods, including giant killers and tiny flyers. The Scipionyx clade includes a series of long-snouted taxa culminating with Deinocheirus and Spinosaurus. Proceratosaurus (Fig. 2) is a close relative from the Middle Jurassic known from a partial skull.

Figure 2. The hatchling Early Cretaceous Scipionyx compared to the adult Middle Jurassic Proceratosaurus. Scipionyx provides clues to the post-crania of Proceratosaurus, while Proceratosaurus provides clues to the adult skull shape of Scipionyx.

Figure 2. The hatchling Early Cretaceous Scipionyx compared to the adult Middle Jurassic Proceratosaurus. Scipionyx provides clues to the post-crania of Proceratosaurus, while Proceratosaurus provides clues to the adult skull shape of Scipionyx.

The Scipionyx clade
(Fig. 3) originated in the Early Jurassic. Several members continued into the Late Cretaceous. Others did not. One trait, among several, this trade seems to hold in common, and originating with Scipionyx, is a nuchal crest created by a rise in the posterior rim of the parietal. This deepens the bone surrounding the upper temporal fenestra for greater anchors of jaw adductors.

Figure 3. The Scipionyx clade includes Allosaurus, Deinocheirus, Spinosaurus and other larger theropods.

Figure 3. The Scipionyx clade includes Allosaurus, Deinocheirus, Spinosaurus and other larger theropods.

References
Dal Sasso C and Signore M 1998. Exceptional soft tissue preservation in a theropod dinosaur from Italy. Nature, 392: 383–387.
Dal Sasso C and Maganuco S 2011. Scipionyx samniticus (Theropoda: Compsognathidae) from the Lower Cretaceous of Italy — Osteology, ontogenetic assessment, phylogeny, soft tissue anatomy, taphonomy and palaeobiology, Memorie della Società Italiana de Scienze Naturali e del Museo Civico di Storia Naturale di Milano XXXVII(I): 1-281

Tanycolagreus compared to Yutyrannus

Tanycolagreus topwilsoni (Carpenter et al., 2005, Late Jurassic, TPII 2000-09-29, 3.3m, restored skeleton based on several specimens) was originally considered a coelurid, then a basal tyrannosauroid by several authors. The large reptile tree nests it with the much larger Yutyrannus, but with a relatively longer neck and longer tail.

Figure 1. Tanycolagreus compared to Yutyrannus.

Figure 1. Tanycolagreus compared to Yutyrannus.

Yutyrannus huali (Xu et al. 2012 ZCDM V5000 Zhucheng Dinosaur Museum, Shandong, Lower Cretaceous Yixian Formation) was originally considered a tyrannosauroid theropod likeT-rex. Here it nests as a sister to Tanycolagreus derived from a sister to Sinraptor. Note the lacrimal horns, large three-fingered hand and long torso. Yutyrannus is famous for being a giant feathered theropod, several times larger than the next largest contender.

Figure 2. Yutyrannus skull compared to Tanycolagreus.

Figure 2. Yutyrannus skull compared to Tanycolagreus.

References
Carpenter K, Miles C and Cloward K 2005. New small theropod from the Upper Jurassic Morrison Formation of Wyoming. in Carpenter, K. 2005. The Carnivorous Dinosaurs, Indiana University Press: 23-48
Xu X, Wang K, Zhang K, Ma Q, Xing L, Sullivan C, Hu D, Cheng S, Wang S et al. 2012. A gigantic feathered dinosaur from the Lower Cretaceous of China. Nature 484 (7392): 92–95. doi:10.1038/nature10906. PDF here.

wiki/Yutyrannus
wiki/Tanycolagreus

The hands of Juravenator

There’s a tiny disarticulated fourth finger hidden in there…
that was originally overlooked (Fig. 1, in pink).

Figure 1. The hands of Juravenator in situ and reconstructed. Note on the left hand, digit 2 is lateral. Somehow digit 3 has moved below it.

Figure 1. The hands of Juravenator in situ and reconstructed. Note on the left hand, digit 2 is lateral. Somehow digit 3 has moved below it. The grayscale drawing is from the original paper.

Oddly, in the left hand
(Fig. 1) digit 3 ended up between digits 1 and 2.

See how DGS (digital graphic segregation) can be helpful?
Try it yourself. It just clarifies things and make reconstructions easy without changing the traced data, even by one pixel.

Careful readers will note
that few PILs (parallel interphalangeal lines) can be drawn through the joints here. Probably they come into alignment only when the fingers are curled (flexed) as in the human hand (try it yourself!). I say ‘probably’ because these need to be 3D modeled to test the possibility. Metacarpals do align. So do m1.2, m2.2 and m3.3.

Figure 4. Juravenator reconstructed. Note the many similarities with Compsognathus (Fig. 3).

Figure 2. Juravenator reconstructed. Note the many similarities with Compsognathus (Fig. 3).

References
Chiappe LM and Göhlich UB 2010. Anatomy of Juravenator starki (Theropoda: Coelurosauria) from the Late Jurassic of Germany.Neues Jahrbuch für Geologie und Paläontologie – Abhandlungen, 258(3): 257-296. doi:10.1127/0077-7749/2010/0125
Göhlich UB and Chiappe LM 2006. A new carnivorous dinosaur from the Late Jurassic Solnhofen archipelago. Nature 440: 329-332.
Göhlich UB, Tischlinger H and Chiappe LM 2006. Juravenator starki (Reptilia, Theropoda) ein nuer Raubdinosaurier aus dem Oberjura der Suedlichen Frankenalb (Sueddeutschland): Skelettanatomie und Wiechteilbefunde. Archaeopteryx, 24: 1-26.

wiki/Compsognathus
wiki/Juravenator