Clavicles, the furcula, and what’s going on in basal archosaurs

Today’s post
comes from a paper on clavicles and furculae by Bryant and Russell (1993). The question is: since most theropods, and most dinosaurs, do not have clavicles, is the furcula of birds a neomorph (new structure)? Or is this the reappearance of topologically identical clavicles? The question becomes more complex withe the occasional appearance of clavicles in the Dinosauria.

Nesbitt et al. 2009
updated the furcula issue. They write, “Given this absence of clavicles and interclavicle in dinosaurian outgroups, the homology of the furcula to other components of the shoulder girdle has been contentious (Bryant and Russell, 1993). This debate is based largely on absence of evidence and will be explored more fully latter in this article.” They continued, “Additionally, topographic connectivity of the furcula to the other pectoral girdle elements in avian and other theropod dinosaurs is entirely consistent with and supportive of homology of the avian clavicle with the ancestral reptilian and tetrapod clavicle.” 

Of course, the Nesbitt et al. outgroups
are not the same outgroups in the LRT.

Before we start
Bryant and Russell followed the invalidated tradition of including pterosaurs with dinosaurs in the outmoded clade, Ornithodira. Worse than that, they and a number of high-profile paleontologists before them believed that pterosaurs had neither clavicles nor an interclavicle. Wild 1993 demonstrated that the pterosaur sternal complex is comprised of fused clavicles, interclavicle and sternum. These are separate in the pterosaur ancestor, Cosesaurus. Nesbitt et al. 2009 acknowledged that observation by Wild 1993 and agreed on the fusion of the sternal and clavicle elements in pterosaurs, However they followed in the wake of the invalid Ornithodira hypothesis.

In the large reptile tree
(LRT, 1012) clavicles that are mediallly broad are present in fish and basal tetrapods. That’s where we start. The situation changes in:

1. Frogs and kin – clavicles are medially narrow
2. The Tuditanus clade – clavicles are medially narrow
3. Pantylus – clavicles are medially narrow
4. The Microbrachis clade – clavicles are medially narrow
5. Within the Reptilia/Lepidosauromorpha: the Milleretta clade (remaining Lepidosauromorpha – clavicles are medially narrow,
6. Except the Caseasauria and except Turtles, where clavicles become part of the plastron,
7. Except the Coelurosauravus clade where the clavicles are absent
8. Except Longisquama + Pterosauria where the clavicles are fused to the sternal complex
9. Except Tetrapodophis + snakes.
10. Within the Reptilia/Archosauromorpha: the Diplovertebron/Romeriscus clade – clavicles are medially narrow
11. The Anomodontia – clavicles are medially narrow
12. Titanophoneus + Cynodontia (including mammals) – clavicles are medially narrow
13. Except in higher Carnivora – clavicles are absent
14. Except in Odontoceti – clavicles are absent
15. Except in Phenacodontidae (includes all hoofed mammals)
16. Prodiapsdia – clavicles are medially narrow
17. Except Araeoscelida  – clavicles are absent
18. Except Nothosaurus – clavicles are medially broad
19. Except the Simosaurus clade (plesiosaurs) – clavicles are medially broad
20. Except higher mesosaurs – clavicles are medially broad
21. Except the Wumengosaurus (pro-ichthyosaurs) clade – clavicles are medially broad, until Qianichthyosaurus and higher ichthyosaurs – clavicles are medially narrow
22. Within Prodiapsida, basal Younginiformes (including Archosauriformes) – clavicles are medially narrow
23. Except in certain Rauisuchids, Decuriasuchus, Turfanosuchus, Silesaurus, Shuvosaurus + Effigia (derived poposaurs) – clavicles are absent.
24. Clavicles are present in the poposaurs, Poposaurus and Lotosaurus.
25. Except Crocodylomorpha (crocs) – clavicles are absent
26. In basal Dinosauria and Prodinosauria the pectoral region is not well preserved
27. Except possibly in Junggarsuchus (Fig. 1) – tiny clavicles may be present
28. Within Orinithischia: Psittacosaurus – clavicles are medially narrow (neomorph)
29. Within Sauropodomorpha: Massospondylus – clavicles are medially narrow (neomorph?)
30. Segisaurus, Coelophysis and higher theropods – furcula is present, but lost in ornithomimosaurs and several other derived theropod clades 

The furcula and the absence of clavicles
Then furcula goes back to basal theropods, but is lost in certain theropod clades, like Ornithomimosauria. In basal theropods it appears to be  a neomorph without direct antecedent. At present, as Newbitt et al. noted, in basal Archosauria (all crocs and basal dinos) the clavicles have not been found. In general the clavicle appears to be lost in taxa that are preserved incomplete and scattered.

But all is not lost…
If we could only find a clavicle in Lewisuchus,. Gracilisuchus, Junggarsuchus, Herrrerasaurus and Tawa we would have a more or less continuous clavicle presence from fish to birds in the LRT. These taxa need to either have a set of clavicles, or some excuse for not preserving them. The latter appears to be the case often enough, as demonstrated here:

1. Lewisuchus – incomplete and jumbled specimen in which the clavicles could have been washed away.
2. Gracilisuchus – pertinent area lost during excavation
3. Junggarsuchus – appears to be minimally and tentatively present (Fig. 1), but really,  who knows what that little green bone is?
4. Herrerasaurus and Sanjuansaurus – pertinent area lost during excavation
5. Tawa – represented by “two nearly complete skeletons and several other partial specimens collected in a tightly associated small grouping at a single locality.” but no clavicle was reported and no in situ images were published suggesting that the skeletons were disassociated. Moreover, any out-of-place clavicles could be mistaken for ribs.
6. Eodromaeus – forelimbs and pectoral girdle missing from holotype.
7. Eoraptor – clavicles were not found and the pectoral girdle has taphonomically shifted.

In summary,
the data is largely missing from the transitional taxa at and near the base of the Archosauria. So there’s still hope that the clavicles were present in these taxa and will someday be discovered among more complete fossils.

Finding overlooked traits
For decades it was thought that no dinosaurs had a furcula or even clavicles. As it turns out, at least in Theropoda, they were largely overlooked. Nesbitt et al. 2009 write: “Furculae occur in nearly all major clades of theropods, as shown by new theropod specimens from the Early Cretaceous of China and a close inspection of previously collected specimens.” Finding overlooked traits is something we should all be doing. Some of these turn out to be important.

On a similar note…
earlier we looked at the homology of the central bones of the wrist and their migration to the medial rim in pterosaurs and pandas, and the previously overlooked evidence for that.

References
Bryant HN and Russell  AP 1993. The occurrence of of clavicles within Dinosauria: Implications for the homology of the avian furcula and the utility of negative evidence. Journal of Vertebrate Paleontology 13(2):171–184.
Nesbitt S, Turner AH, Spaulding M and Norell MA 2009. The Theropod Furcula. Journal of Morphology 270(7):856–879.
Wild R 1993. A juvenile specimen of Eudimorphodon ranzii Zambelli (Reptilia, Pterosauria) from the upper Triassic (Norian) of Bergamo. Rivisita Museo Civico di Scienze Naturali “E. Caffi” Bergamo 16: 95–120.