Figure 1. A series of fenestrasaur and pterosaur sternal complexes scaled to a common humerus length in phylogenetic order.
When you have a robust phylogenetic tree, as we do for the pterosaurs, and you have a catalog of reconstructions, as we do here, many interesting things can happen.
Here (Fig. 1) we’ll take a look at the development of the sternal complex, the anchor for the large pectoral flying adductors in one clade of pterosaurs that happened to reduce the sternal complex for reasons unknown. The wings, in this case, were not reduced.
This (Fig.1) is close to an evolutionary sequence based on a cladogram, represented by specimens not on, but close to a direct lineage.
Earlier we looked at the migration of the sternum anteriorly above the interclavicle to meet the short, transverse clavicles in Cosesaurus and Langobardisaurus.
In Longisquama, a taxon more committed to flapping, the clavicles wrapped posteriorly around the sternum (what some (Jones et al. 2000, Martin 2004) have erroneously considered homologous to a bird wishbone, aka – fused clavicles). The interclavicle developed a deep keel. The humerus developed a deltopectoral crest and expanded distal condyles, like those of a pterosaur.
The most basal pterosaur MPUM 6009 was little different, but had a smaller keel and a deeper sternal portion.
Eudimorphodon ranzii had more robust elements, and a shape like a triangle added to a cardioid. In this lineage, the size and shape of the sternal elements make Eudimorphodon appear to be an outlier, trending more toward the large sternal complex clade of Campylognathoides + Rhamphorhynchus.
Eudimorphodon cromptonellus had a smaller sternum but retained a semicircular shape.
Eudimorphodon Bsp 1994 had a smaller sternum, triangular in shape, with anterior processes at the coracoid joint.
Sordes was similar without the anterior processes.
The sternal complex of the Donau specimen of Dorygnathus was tear-drop shaped, semi-circular posteriorly.
All other Dorygnathus specimens had a tiny sternal complex (yet individually distinct in shape), a character that readily distinguishes them from Rhamphorhynchus. This one, SMNS 50164 nests at the base of the following pterosaurs.
TM 10341 was a tiny dorygnathid, yet its sternal complex was relatively larger with lateral processes. Unrelated Scaphognathus had a very similar shape by convergence.
“Rhamphodactylus” expanded laterally and posteriorly, creating a circular outline.
Beipiaopterus had sharp anterior corners creating a shield shape.
Among later taxa in this clade, the flightless JME SoS 2428 had a relatively larger sternal complex than the flighted azhdarchids, like Quetzalcoatlus and Zhejiangopterus.
Tiny sternal complex for Dorygnathus
Padian (2008) reported that, “The sternum [sternal complex] in Campylognathoides (Wellnhofer 1974) is frequently preserved and is much larger than in Dorygnathus; however, the former genus has relatively longer wings and a smaller skull, which suggests functional differences that may be connected with sternal morphology. Additionally, Dorygnathus may have had a larger posterior sternal extent that was not calcified, but merely cartilaginous.”
This is what we call wishful thinking or following a paradigm.
I’ve done it. Padian did it. In all Dorygnathus specimens gastralia are few and appear only in the posterior region. So there is a large gap between ossified parts ventrally (Fig. 2), distinct from most other pterosaurs with a larger sternal complex. If the region was a cartilaginous extension it would have been odd and improbable, as the clavicle and interclavicle ossified portions were also tiny.
Figure 2. The SMNS 51827 Dorygnathus with its long torso, really tiny sternal complex and open ventral area without gastralia. What’s going on here with regard to diet and flying abilities? On a secondary note, note the butt joint between the pubis and prepubis. This is not a rotating bone.
Let’s talk about that -really- tiny sternal complex in the SMNS 51827 specimen
All Dorygnathus have a tiny sternal complex, but the one in SMNS 51827 is the smallest (Fig. 2) I’ve seen. Generally, when one part shrinks, another expands. In this case as the chest region shrank, a more flexible belly region expanded. Why? Hard to say. But the solution is not convergent with Rhamphorhynchus, which shared long teeth, but had a large sternal complex and a short gastralia-filled stiff belly.
The manual claws of SMNS 51827 were large (Fig. 3), trenchant and preserved (Fig. 3) as if grasping an object with a circular cross section, like a tree trunk of appropriate diameter.
Figure 3. The SMNS specimen of Dorygnathus, largely complete and exposed ventrally. Note those hands, posed as if grappling a tree trunk of that diameter. Pertinent bones colorized using DGS. Surprise to one and all… DGS reveals that enough gastralia to fill the belly appear to be missing here! Just a few (in pink) show up here. Left side ribs in blue. Right side ribs in green. Sternal complex in purple and blue. See figure 2 for tracings without in situ fossil.
The loss of ventral ossification permitted the dorsal vertebrae to be more flexible. It also permitted the belly to be more expandable. Vertebrae #14 is longer than the others. It happens to be the vert that anchored the scapulae. The entire torso is relatively elongated in SMNS 51827 (Fig. 4).
Figure 4. The elongated torso, small sternal complex and long fingers distinguish this Dorygnathus specimen SMNS 51827. Were those rake-like teeth used for raking in vegetation? Good question. Maybe. Maybe not.
The last time we saw such a long torso, such a small sternal complex and such long fingers was in Longisquama (Fig. 4), the sister to the Pterosauria (they shared a recent common ancestor).
Figure 4. Longisquama on a tree trunk.
Actually the sternal complex is smaller in Dorygnathus, just large enough to anchor the coracoids. Perhaps we can imagine Dorygnathus had a similar lifestyle, leaping from tree trunk to tree trunk, weakly flapping and mostly gliding due to its smaller than typical pectoral anchors but normal-sized wings.
Figure 1. Archaeopteryx and several other basal birds. Here Archaeopteryx is a relative giant. This is an older illustration, predating all of the recent and now, not so recent, finds from China. The wings, sternum and tail are derived in the smaller birds. That’s the smallest Archaeopteryx, the Eichstätt specimen.
An analogy may also be drawn with Archaeopteryx, a primitive bird that likewise had substantial wings, stem-like coracoids and large grappling claws, but very little sternum to anchor flight muscles.
In summary and one lifestyle struck off the “probable” list
Apparently, and distinct from most other pterosaurs, Dorygnathus was a long-waisted, large-clawed, weak flyer. Perfectly adapted to grappling tree trunks, it may have barely gained altitude as it weakly flapped, gliding from tree to tree. So, a lifestyle spent dipping for fish over the open sea seems less probable.
Figure 6. Pioneer herbivores in their own clades, both with large procumbent teeth, not unlike Dorygnathus. Daemonosaurus is a dinosaur. Stenocybus is a basal therapsid leaning toward the galesaurid/dicynodont branch.
If not a fish diet, then what?
I’m still wondering what the alternative diet would be, considering those teeth, that Middle Jurassic forest and that belly. According to Padian (2008) no stomach contents are known. With those raking teeth are we looking at a high-browsing herbivore? Angustinaripterus had even larger raking teeth, but no post-crania is known. Notably some reptile clades that adopted herbivory had basal forms that developed procumbent teeth, like Daemonosaurus (among the Dinosauria) and Stenocybus (among the basal Therapsida (Fig. 6) (galeasaurids and dicynodonts).
Better Flyers Millions of Years Later
The several descendant clades of Dorygnathus all developed a larger sternal complex and larger flight muscles, as did the several descendant clades of Archaeopteryx. Even the flightless pterosaur, JM SoS 2428, had a large sternal complex!
Better Flyers Among Other Jurassic Pterosaurs
Similarly the clades that splintered off earlier (Dimorphodontia and Eudimorphdon ranzii + Campylognathoides + Rhamphorhynchus) also all developed a much larger sternal complex. Maybe the Dorygnathus clade just got off to a slow start with regard to flying, slowed down a bit during their Dorygnathus phase, then really took off thereafter.
Jones TD et al 2000. Nonavian Feathers in a Late Triassic Archosaur. Science 288 (5474): 2202–2205. doi:10.1126/science.288.5474.2202. PMID 10864867.
Martin LD 2004. A basal archosaurian origin for birds. Acta Zoologica Sinica 50(6): 978-990.