Any major gaps left in the vertebrate family tree?

Not in the LRT.
While new vertebrate taxa are being published every week, categorically none of these are completely new and unheard of. New taxa are all falling into or between established clades.

There are no large gaps or weird enigmas
in the vertebrate fossil record, according to the the large reptile (LRT, 1663+ taxa). We know where turtles, catfish, snakes, whales and pterosaurs came from. Sure, I’d like to find someone report a short-fingered bat in the Cretaceous (Fig. 1), but that won’t come as a surprise when it happens. We already have the bookend taxa for that discovery.

Figure 1. Subset of the LRT focusing on the clade of colugos, pangolins and bats.

Now only microevolution separates one taxon from another
and one clade from another. Every taxon in the LRT has sisters and ancestors back to Cambrian chordates. All sisters are more or less visually similar to one another.

Now all we have to do
is to continue slipping new taxa between established taxon pairs already in the LRT.

The only issue that remains is one that may always remain…
We don’t have, nor will we ever have, transitional fossils from the genesis of every transition (Fig. 1). Most of these are lost to time, or were never fossilized. What we do have are later-to-extant representatives of these transitions. And that’s okay.

That’s where systematics and taxonomy stands in Spring 2020.
The hard work is done. Fossil and extant taxa nest together. Taxon exclusion has been minimized in the LRT due to its wide gamut.

Unfortunately, paleo moves at a snail’s pace,
so there are still workers who cling to invalid hypotheses like pterosaur are archosaurs, caseids are pelycosaurs, or reptiles began with Hylonomus. Vancleavea is still on several archosauriform taxon lists. Multituberculates are still considered egg-laying mammals.

All of this nonsense
can stop now. Just add taxa. The LRT provides a suggestion list.

Run your own tests
to validate the LRT or invalidate it. Don’t trust it. Test it. Just make sure your observations are insightful and true, your reconstructions (show your work!) minimize freehand influences, and your taxon list is wide enough to include all possible candidates. Then share it with us when you have something to present.

A good scientist
attempts to falsify his own and other conclusions. To that end, scoring changes and reevaluations have been a part of the LRT since its inception nearly a decade ago.

Thank you
for your readership, your suggestions and your criticisms.

Straight out of Star Wars: Satyrichthys, the armored sea robin

One of the strangest fish in the sea
is the armored sea robin, Satyrichthys (Fig. 1). Based on phylogenetic bracketing, that’s the palatine + lacrimal + jugal + postorbital creating a face mask of bony armor. Ancestral taxa, like the sea robin, Prionotus (Fig. 4) and the thread fin, Polydactylus (Fig. 3), have progressively smaller circumorbital bones.

Figure 1. Satyrichthys skull with DGS applied and overall diagrams. That preoperculum spike is shared with Prioonotus (Fig. 5).

The Satyrichthys skull
(Fig. 1) looks strikingly like a Star Wars Legion T-47 Airspeeder (Fig. 2) IMHO. In paleontology, we call this an unrelated ‘convergence.’

Figure 2. Star Wars air speeder model.

Satyrichthys rieffeli (originally Peristethus rieffeli Kaup 1859, 1873; Kawai T 2013; Fig. 1) is the extant armored sea robin with massively developed external palatine + lacrimal + jugal + postorbital extending far anterior to the small, weak, mouth.

Phylogenetically that face-mask started off innocently enough
as a circumoribital ring (palatine + lacrimal + jugal + postorbital) with a slight bump to the front on Polydactylus (Fig. 3). That ring evolved to cover more and more of the face (e.g. Prionotus, Fig. 4) until it became an all enclosing mask, as in Satyrichthys (Fig. 1).

Figure 3. Primitive Polydactylus skull has only a small, fused circumorbital ring. The former lacrimal is here the palatine given comparisons to outgroup Seriola zonata (Fig. 5). 

Polydactylus oxtonemus (Girard 1858; up to 23cm, some species up to 2m; Fig. 3) is the extant Atlantic threadfin. A perciforme (perch family), here Polydactylus (above) is also a shallows-dwelling relative of the sea robin Prionotus (below). Note the five thread-like rays/feelers anterior to the pectoral fin arising from the coracoid. These drag along the sea floor sensing prey. A second dorsal fin and forked tail distinguish this taxon from its sisters.

Figure 4. The sea-robin, Prionotus, has a more extensive circumorbital ring/face-mask. The palatine here may instead by relegated to just the ventral rim of the naris if the lacrimal extends anteriorly, as in Satyrichthys.

Prionotus evolans (Linneaus 1766; 40cm; Fig. 4 is the extant striped sea robin, a scorpionfish that uses a set of finger-like flexible spines (homologous with the thread fin fins (Fig. 2) of its large pectoral fin to walk on the seafloor. With a long straight snout, it looks more like it’s barracuda-like relatives, but descends from a last common ancestor, the banded rudderfish, Seriola zonata (Fig. 5). We looked at missing cheekbones in the banded rudder fish earlier here.

Figure 5. Gregory 1933 did not illustrate a jugal and lacrimal for Seriola zonata, but the cladogram indicates they should be there. We find them in the photo. Note the anteriorly projecting palatine retained in derived taxa like Satyrichthys.

The phylogenetic connection of rudder fish, threadfins and sea robins
appears to be a novel hypothesis of interrelationships. If you know of a prior citation, please let me know so I can promote it.

---

References
Girard CF 1858. Notes upon various new genera and new species of fishes, in the museum of the Smithsonian Institution, and collected in connection with the United States and Mexican boundary survey: Major William Emory, Commissioner. Proceedings of the Academy of Natural Sciences of Philadelphia. 10: 167-171.
Kaup JJ 1859. Description of a new species of fish, Peristethus rieffeli. Proceedings of the Zoological Society of London, 1859, 103–107, 8 pl.
Kaup JJ 1873. Über die Familie Triglidae nebst einigen Worten über die Classification. Archiv für Naturgeschichte, 39, 71–93.
Kawai T 2013. Revision of the peristediid genus Satyrichthys (Actinopterygii: Teleostei) with the description of a new species, S. milleri sp. nov. Zootaxa, 3635 (4): 419–438.
Linnaeus C von 1758. Systema naturæ per regna tria naturæ, secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis. Tomus I. Editio decima, reformata.
Linneaus C von 1766. Systema naturæ per regna tria naturæ, secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis. Tomus I. Editio duodecima, reformata. pp. 1–532. Holmiæ. (Salvius)
Sulak KJ 1975. The systematics and biology of Bathypterois (Pisces, Chlorophthalmidae) with a revised classification of benthic mystophiform fishes. University of Miami Press, 398 pp. also:  Galathea Report. 1977; 14:49pp.

wiki/Polydactylus
wiki/Flying_gurnard
wiki/Triglidae-SeaRobin

How primitive are megapodes?

Earlier the large reptile tree (LRT, 1663+ taxa) nested megapodes (like Megapodius) at a more primitive node than any other living bird, except the kiwi (Apteryx) and ratites, like (like Struthio). You might remember, a toothed bird clade restricted to the Early and Late Cretaceous was derived from toothless Crypturus (Fig. 1) in the LRT.

Figure 1. Megapodius is the extant bird nesting at the base of all neognathae (all living birds except ratites). And it looks like a basal bird, not too this… not too that.

With that in mind
and hoping to understand the reemergence of previously lost teeth in Early Cretaceous birds, I checked out Clark 1960, who reported on megapode embryology.

To set the stage, Clark wrote,
“Young birds are exceedingly precocious, being able to fly on the day of hatching and feeding actively only a few days after hatching.” He then referenced Portmann (1938, 1951, 1955) who listed several reptile-like characters of megapodes:

1. no egg tooth (megapodes hatch by kicking their way out of the shell. The ‘egg tooth’ of chickens temporarily appears on the top of the beak, not the rim);
2. lack of down feathers in embryos or nestlings;
3. lack of parental care;
4. primitive method of incubation (by solar heat, fermentation, vulcanism);
5. long incubation period (8 weeks for Leipoa);
6. large number of eggs laid;
7. slow growth to adult size (especially for Alectura);
8. primitive structure of the brain;
9. eggs usually not turned and yet hatch relatively successfully;

Clark added to Portmann’s #9
a general lack of movement of the embryo until just before hatching. This may be related to the use of fermentation as a heat source for incubation. Clark notes, “the presence of aerobic bacteria should presumably greatly deplete the available oxygen supply.” Moving embryos might have suffocated for lack of oxygen. Clark also noted: relatively large yolks, as in reptiles.

I never found a tooth thread
connecting Late Jurassic teeth in stem birds to the reemergence of teeth in Early Cretaceous crown birds (Fig. 2) following Apteryx, ratites and megapodes. Even so, every other trait indicated a transition. The above authors further support the extreme primitive nature of megapodes. Ratites no longer bury their eggs. Kiwis dig burrows.

Figure 1. Click to enlarge. Toothed birds of the Cretaceous to scale.

In the post-cladistic era
Dekker and Brom 1992 wrote, “Among megapodes, four different incubation-strategies may be distinguished:

1. mound-building,
2. burrow-nesting between decaying roots of trees,
3. burrow-nesting at volcanically heated soils, and
4. burrow-nesting at sun-exposed beaches.”

Dekker and Brom employed a cladogram
originally published by Cracraft and Mindell (1989), which mistakenly nested megapodes with galliforms (chickens and kin) due to taxon exclusion. Dekker and Brom wrote, “We conclude that similarities shared with reptiles and kiwis are due to convergence.” That traditional nesting is not confirmed by the LRT due to taxon exclusion. Burying and burrowing are primitive, but give no clue as to how Early Cretaceous birds redeveloped small teeth at first, large teeth later. Neither does megapode embryology. Perhaps that’s why this novel hypothesis of interrelationships has never appeared elsewhere. 

---

References
Clark GA Jr. 1960. Notes on the embryology and evolution of the megapodes (Aves: Galliformes). Postilla 45:1–7.
Cracraft J and Mindell DP 1989. The early history of modern birds: a comparison of molecular and morphological evidence.— In: B. Fernholm, K. Bremer & H . Jörnvall, eds. The Hierarchy of Life: Molecules and Morphology in Phylogenetic Analysis: 389-403. Amsterdam, New York, Oxford.
Dekker RWRJ and Brom TG 1992. Megapode phylogeny and the interpretation of the incubation strategies. xxx 19–31.  Zoologische Verhandelingen  278(2): 19–31.
Portmann A 1938. Beitrage zur Kenntnis der postembryonalen Entwick- lung der Vogel. Rev. Suisse Zool., 45: 273-348.
Portmann A 1951. Ontogenesetypus und Cerebralisation in der Evolution der Vogel und Sauger. Rev. Suisse Zool., 58: 427-434.
Portmann A 1955. Die postembryonale Entwicklung der Vogel als Evolu- tionsproblem. Acta XI Congr. Int. Orn., 1954. Pp. 138-151.

Gogosardina: the genesis of the squamosal

Choo, Long and Trinajstic 2009 brought us
a small, Late Devonian actinopterygian, Gogosardina coatesi (Figs. 1, 2; holotype WAM 07.12.2) known from four crushed and incomplete specimens. One contains conodont elements lodged among the branchial arches.

Figure 1. Gogosardina from Choo, Long and Trinajstic 2009 shown at full size if shown on a typical 72dpi computer monitor. Gray areas indicate missing bones on skull.

Here
(Fig. 2) a few skull bones are relabeled according to their tetrapod homologs, as in all taxa entered into the large reptile tree (LRT, 1663+ taxa). The skull is nearly identical to coeval and similarly-size Mimipiscis with slightly rotated premaxilla, a straighter anterior maxilla, a higher naris and only a partial ‘razor back’ ridge anterior to the dorsal fin. The skull is proportionally larger as well. Both have a large pineal opening between the frontals (yes, the frontals), distinct from almost all fish. The excurrent naris is confluent with the orbit. This entire clade lacks postparietals.

Figure 2. Gogosardina soul from Choo, Long and Trinajstic 2009. New labels in red. The intertemporal anchors the large hyomandibular in all fish.

Choo, Long and Trinajstic considered Gogosardina to be
a stem actinopterygian. No cladogram of relationships was published then. Wikipedia lists Gogosardina among the Palaeonisciformes (Hay 1902). In the LRT Gogosardina nests between Cheirolepis and Mimipiscis, all basal to the extant anchovy, Engraulis., which is not traditionally considered to be a paleonisciform.

Figure 3. Pteronisculus shows how the jugal splits to form a jugal and squamosal, a bone that will ultimately take over for the preopercular in this clade.

Clade member, Pteronisculus
(Fig. 3) splits the jugal into four parts. The posterior two become the single squamosal in Strunius (Fig. 4), Onychodus (Fig. 5) and all later lobefins and ultimately tetrapods.

Figure 4. Strunius shows the next step in the enlargement of the squmosal and the two bones making up the preopercular.

In Onychodus
(Fig. 5) the squamosal is beginning to take over the preopercular (= postsquamosal). Thereafter the postsquamosal is a vestige until it disappears in most tetrapods, only to reappear in a few basal tetrapods undergoing reversals.

Figure 5. Onychodus continues the enlargement of the squamosal and the reduction of the preopercular (post squamosal) in our tetrapod lineage.

Wikipedia reports,
“The Palaeonisciformes (Hay 1902) are an extinct order of early ray-finned fishes (Actinopterygii) which began in the Late Silurian and ended in the Late Cretaceous. It is not a natural group, but is instead a paraphyletic assemblage of the early members of several ray-finned fish lineages.”

Figure x. Updated subset of the LRT, focusing on basal vertebrates = fish.

With regard to anchovies, Wikipedia reports, 
“Clupeiformes (Goodrich 1909) is the order of ray-finned fish that includes the herring family, Clupeidae, and the anchovy family, Engraulidae.” 

The LRT nests few traditional clupeiformes,
but the wolf herring, Chirocentrus, nests elsewhere (at a more basal node, along with toothy Trachinocephalus) apart from the anchovy, Engraulis. So this seems to be a paraphyletic clade based on these two disparate taxa.

Putting related taxa in phylogenetic order
helps us visualize the less dramatic processes of evolution that no one ever talks about, like the origin of the squamosal. which ultimately creates the dentary-squamosal jaw joint in mammals and humans.

---

References
Choo B, Long JA and Trinajstic K 2009. A new genus and species of basal actinopterygian fish from the Upper Devonian Gogo formation of Western Australia. Acta Zoologica (Stockholm) 90 (Supp 1):194–210.

wiki/Gogosardina (not online yet)
wiki/Palaeonisciformes
wiki/Clupeiformes

Can volant fossil vertebrates inspire mechanical design?

Martin-Silverstone, Habib and Hone 2020 review volant fossil taxa,
in their hope to “synthesise key elements to provide an overview of those cases where fossil flyers might provide new insights for applied sciences.”

Caveat
readers should note, these authors have been responsible for some of the current pterosaur myth-making (e.g. pterosaur quadrupedal catapult launch) in the academic literature. (To see the entire list, enter the keywords “Silverstone”, “Habib” or “Hone” in the white box above).

Even so, let’s start with a fresh slate
and see what they have to say.

The authors report,
“Soaring is a form of passive flight (though as with gliding, is often a behaviour of powered fliers) which involves using external sources of lift.”

“Change ‘often’ to ‘always’. Soaring only comes to those who have excelled at powered flight earlier. Soaring is the next step for the highest, longest-range flyers.

“Unique fossil-only bauplans have also been described, such as the nonavialan dinosaurs Yi qi and Ambopteryx.” 

Not unique. Misinterpreted, as detailed here. Both Yi qi and Ambopteryx are derived from specific Late Jurassic Solnhofen birds (specimens traditionally assigned to Archaeopteryx), in the large reptile tree (LRT, 1663+ taxa) which makes them avialan dinosaurs. The proper phylogenetic context must be the foundation.

On the same subject, later in their text, “the recently discovered Yi and Ambopteryx show a melange of features – notably an enlarged wrist bone supporting an apparently small membranous wing, but also a flight surface composed of feathers.” This is a myth. That ‘wrist bone’ is either a radius or an ulna, depending on which wing is under consideration as corrected and detailed here.

Figure 1. Above: freehand image from Martin-Silverstone 2020 of Quetzalcoatlus northropi wing (based on the humerus shape). Pink arrows call out errors. Below: Traced image of Q. sp. wing after firsthand examination in the Wann Langston lab where the fossils were kept years ago.

Credit where due:
In the authors’ illustration of the pterosaur wing (Fig. 1), they correctly located the pteroid on the radiale, but incorrectly placed the medial carpal there, too. Free fingers 1–3 are too large and appear to be on top of one another, with their palmar surfaces facing anteriorly, following Bennett (2008, Fig. 2). In reality the palmar surfaces faced ventrally in flight with only metatarsal 3 attached to the wing finger, as in all other tetrapods (Fig. 2). That makes the free fingers point laterally while quadurpedal, as ichnites show. The wing membrane illustration (above) mistakenly extends to the hind limbs. This is the myth of the bat-wing pterosaur promoted by several Bristol professors.

Ironically,
the authors chose a flightless pterosaur, Quetzalcoatlus, to model their volant wing.

Figure 2. Pterosaur hands, dorsal view, the two opposing hypotheses.

Continuing onward to the bottom of their Figure 1
(Fig. 3 below), the authors mislabeled the left and right wings in this dorsal view (with scapulae indicating the dorsal side) of the BSP 1937 I 18 specimen of Pterodactylus.

Figure 2. This is Figure 1B of Martin-Silverstone et al. 2020 where they mislabel the left and right wings of BSP 1937 I 18. The authors labeled this specimen “Aerodactylus”, but it nests in the midst of several Pterodactylus specimen.  Colors added to show the extent of the wing membrane. See figure 4 for an animation of a similar fossil. I did not color the uropatagia behind each knee. You can see those plainly here.

The lower arrow pointing to the ‘membrane’
(‘m‘ in Fig. 2) just barely points to the trailing edge of the membrane, just missing the space behind the elbow, where, as Peters (2002) showed (and see Fig. 4) the wing membrane stretched only between the elbow and wing tip, contra Martin-Silverstone, et al. (Fig. 1). The upper arrow points to the biceps (light red), not the propatagium membrane (yellow).

Figure 4. This is the Vienna specimen of Pterodactylus, which preserves soft tissue membranes as in Fig. 3.

The authors labeled the BSP 1937 I 18 pterosaur, ‘Aerodactylus‘.
According to Wikipedia, “Aerodactylus is a dubious pterosaur genus containing a single species, Aerodactylus scolopaciceps, previously regarded as a species of Pterodactylus.”

In the large pterosaur tree (LPT) the BSP 1937 I 18 specimen nests between several other Pterodactylus specimens.

The authors report, 
“It is therefore the evolution of more extreme vane asymmetry, rather than slight asymmetry, that was critical to avian flight.”

According to the LRT, it is the elongation of locked down corticoids (and the clavicle in bats because they lack a coracoid) marks the genesis of flapping, which is more critical to avian flight.

The authors report, 
“The largest pterosaurs reached in excess of 10 m in wingspan, 250 kg in weight, and had skulls perhaps 3 m long, vastly exceeding any other known flying animal in size and weight.”

Actually the largest pterosaurs, like the largest birds, were flightless, as shown earlier here.

With regard to pterosaur wing membranes, the authors report, 
“All fossils that have relevant portions preserved and undistorted show the membrane attaching to the lower leg or ankle.” 

Actually, none of them do, including their Figure 1 (Fig. 2 above). The authors referenced Elgin, Hone and Frey 2011, another botched paper discussed earlier here. You might remember, the authors employed a fictional “shrinkage” to explain away all the fossils that did no fit their preconception, but all matched the observations in Peters 2002.

The authors report,
“Mechanical considerations indicate that pterosaur wings must have had a concave posterior margin to avoid aeroelastic instability.”

Why guess, hope and assume when you can observe? The aktinofibrils are there to avoid aeroelastic instability.

The authors report, 
“Proper tensioning of membrane wings in pterosaurs would have been impossible with a convex posterior margin, because of the single-spar construction.”

Tension between the elbow and wing tip (Peters 2002) is supported by fossil evidence (Figs. 2, 3).

The authors report,
“It has been suggested that the largest pterosaurs were secondarily flightless, but more recent work suggests that the maximum launch-capable body mass for pterosaurs may have been high, owing to the high maximum lift coefficient of their wings and their potential for quadrupedal launch.”

This is Habib’s claim based on imagined and falsified ‘evidence’ argued here. Habib’s hypothesis was based on an imagined elastic catapult potential in the wing knuckle pressed against the ground, but pterosaurs never do this according to track evidence. Click here to see the doctored evidence presented by Habib 2008.

The authors also cite the PhD thesis of C. Palmer, University of Bristol 2016. One of his first assignments as a PhD (October 2016 ) must have been to place an seeking a student, “to investigate the effectiveness of the quadrupedal launch [of pterosaurs] and by comparing it with the bipedal launch of birds, test if it was one of the factors that enabled pterosaurs to become much larger than any bird, extant or extinct.” You can read more about that advert here.

Wait a minute… since that quad launch hypothesis was a subject in Palmer’s PhD dissertation (according to the Martin-Silverstone, et al. citation, why was he advertising for someone else with less experience to take on this task? Let’s remember, students and PhD candidates have the least experience in the field. Most of the myth-making in pterosaurs comes out of universities in Southern England, evidently where students have to produce what their professors demand, or fail.

Please note: In the advert the Bristol bunch were not testing the hypothetical quad launch of pteros against the hypothetical bipedal launch of pteros. For them, quad launch was/is ‘a given’ that must be proved, despite the danger to the pterosaur, the criticism from colleagues and the lack of evidence.

At this point,
I’m only halfway through the paper. The rest we’ll save for later, if necessary. For now, some concluding remarks.

The authors stated their goal in lofty language, 
“A robust understanding of the origin of flight and the evolution of morphologies related to flight performance provides critical context for the constraints and optimisation of biological traits that can inspire mechanical design.”

The problem is, the authors have collected and presented invalid data. They have avoided putting the origin of bats, birds and pterosaurs into their proper phylogenetic context by showing the origin of flapping. How can the authors hope to emulate a pterosaur mechanically if they are freehand designing their own fictional pterosaur (Fig.1) and not looking carefully at specimens under their nose (Fig.2)? A scientist should always be trying to falsify a claim. I don’t see that here. By ignoring the literature (and the evidence) that falsifies a claim, these three authors are not acting like scientists.

With regard to mechanical pterosaurs,
the Stanford pterosaur project did not fair as well as simpler ornithopter designs.

The famous MacCready mechanical flying pterosaur
(Figs. 5, 6), was ostensibly modeled on the smaller Quetzalcoatlus specimen (Figs. 5, 6), but MacCready extended the wingspan to make his model fly. For a discussion on mechanical pterosaurs, it’s a little strange that the keyword, “MacCready” yields no results in their PDF.

Figure 5. The Macready flying model compared to Q. sp. Perhaps it has always been overlooked that the neck proportions were changed, even though the body included the weight of the motor, batteries, radio and controls. The wingspan is longer on the flying model than on the real genus.

We looked at arguments against
the hypothesis of giant volant pterosaurs here. The first thing that pterosaurs do when they give up flying is to shorten the distal wing phalanges, a fact overlooked by Martin-Silverstone, Habib and Hone. The keyword, “vestigial” does not appear in their PDF. The keyword, “distal” appears, but not in regards to pterosaur wing phalanges.

Figure 6. Paul MacCready’s flying pterosaur model had longer wings than Q. sp., with its vestigial distal wing phalanges. Here the model and its inspiration are shown to the same length.

Once again, and true to Professor Bennett’s curse,
“You will not be published, and if you are published, you will not be cited,” my published papers on the origin of pterosaurs from fenestrasaurs (Peters 2000), the origin and shape of pterosaur wings (Peters 2002), and the origin and orientation of the pteroid (Peters 2009) were not cited by these authors. As good scientists they should have cited these papers, discussed the presented data, constructed arguments, and most importantly, attempted to falsify their own hypotheses with faithful and precise observations unsullied by invented excuses (‘shrinkage’). Only when they get their pterosaurs right will they have a good basis for discussing mechanical equivalents. And please cite the work of inventor Paul MacCready.

PS
Citation #76 in Martin-Siverstone, et al. (Zakaria et al. 20160 discusses several mechanical aspects of pterosaurs. They copied the bad pterosaur bauplan from Elgin, Hone and Frey 2011 (Fig. 7) then provided an optimized wing plan with a narrower chord from their studies (Fig. 8) that more closely matched the actual wing shape of pterosaurs in Peters (2002).

Figure 7. Above problems with the Elgin, Hone and Frey (2011) pterosaur wing model with corrections proposed by Peters (2002).

All I can say is,
it’s a topsy-turvy world out there where bad data rules the day.

Figure 8. When Zanzaria et al. 2016 used math to model the optimum pterosaur wing, they found a narrow chord (red), as in figure 7, worked better than the ‘actual shape’ actually invented by Elgin, Hone and Frey 2011 wing (black).

Added a few days later:
From the Scientific American article that promoted four-fingered tenrec tracks as Crayssac pterosaur tracks: “Elizabeth Martin-Silverstone, a pterosaur expert at the University of Bristol in England, who did not take part in the work, says the fossil is the ‘final nail in the coffin of the idea that basal pterosaurs were awkward and clumsily walking around—and definitely of the idea that early pterosaurs might have been bipedal.” Not only did they walk on all fours, “but they moved around quickly and with style,’ she adds.” Martin-Silverstone is not using critical thinking. Four fingers and anteriorly-oriented manus tracks invalidated these as possible pterosaur tracks. Many pterosaurs and their fenestrasaur tritosaur lepidosaur ancestors were bipeds. See keyword “Rotodactylus” in the white box above.

References
Bennett SC 2008. Morphological evolution of the forelimb of pterosaurs: myology and function. Pp. 127–141 in E Buffetaut and DWE Hone eds., Flugsaurier: pterosaur papers in honour of Peter Wellnhofer. Zitteliana, B28.
Elgin RA, Hone DWE and Frey E 2011. The extent of the pterosaur flight membrane. Acta Palaeontologica Polonica 56 (1), 2011: 99-111. doi: 10.4202/app.2009.0145
Habib M 2008. Comparative evidence for quadrupedal launch in pterosaurs. Pp. 161-168 in Buffetaut E, and DWE Hone, eds. Wellnhofer Pterosaur Meeting: Zitteliana B28
Mazin J-M, Billon-Bruyat J-P and Padian K 2009. First record of a pterosaur landing trackway. Proceedings of the Royal Society B doi: 10.1098/rspb.2009.1161 online paper
Martin-Silverstone E, Habib MB and Hone DWE 2020. Volant fossil vertebrates: Potential for bio(-)inspired flight technology. Trends in Ecology & Evolution (advance online publication) doi: https://doi.org/10.1016/j.tree.2020.03.005
https://www.sciencedirect.com/science/article/abs/pii/S016953472030080X
Palmer C 2011. Flight in slowmotion: aerodynamics of the pterosaur wing. Proc. R. Soc. Lond. B Biol. Sci. 278, 1881–1885.
Peters D 2002. A New Model for the Evolution of the Pterosaur Wing – with a twist. – Historical Biology 15: 277–301.
Peters D 2009. A reinterpretation of pteroid articulation in pterosaurs. Journal of Vertebrate Paleontology 29:1327-1330.
Prondvai E and Hone DWE 2009. New models for the wing extension in pterosaurs. Historical Biology DOI: 10.1080/08912960902859334
Sharov AG 1971. New flying reptiles fro the Mesozoic of Kazakhstan and Kirghizia. Trudy of the Paleontological Institute, Akademia Nauk, USSR, Moscow, 130: 104–113 [in Russian].
Unwin DM and Bakhurina NN 1994. Sordes pilosus and the nature of the pterosaur flight apparatus. Nature 371: 62-64.
Zakaria MY. et al. 2016. Design optimization of flapping ornithopters: the pterosaur replica in forward flight. J. Aircraft 53: 48–59
Zittel KA 1882. Über Flugsaurier aus dem lithographischen Schiefer Bayerns. Palaeontographica 29: 7-80.

http://reptileevolution.com/pterosaur-wings.htm
http://reptileevolution.com/pterosaur-wings2.htm

In Memorium: paleontologist Robert L. Carroll

Figure 1. Robert L. Carroll in his younger days.

Robert L. ‘Bob’ Carroll (1938-2020):
a warm-hearted, kind, and knowledgeable professor, always eager to answer a question.

Earlier, we looked at the impact of his major work from 1988, the textbook ‘Vertebrate Paleontology.’ That ‘must-have’ volume was a prime resource for many students and professors for decades. Some considered it ‘The Bible’ of our profession.

We all enter science to make a contribution. Carroll made his in small and large ways, not only by describing and illustrating many of his own discoveries, but by working with others to bring them all together between book covers in the pre-cladistic era. His work will remain on our library shelves. ReptileEvolution.com was built on that foundation and stands on the shoulders of this giant.

---

References
Use key word “Carroll” to see the index of all the taxa RL Carroll helped describe and covered in this blogpost.

A few days later this link goes into detail on RL Carroll’s career.

Headline: “Vertebrate palaeontologist who recognized and described the oldest known ancestor of all reptiles birds and mammals; the origins of terrestrial vertebrates, the origin of various amphibians such as frogs and salamanders.” 

Subhead: “Any high-school kid can go out and make fossil discoveries.”

Caveat: Some of those hypotheses have been superseded by more recent discoveries (e.g. “Hylonomus lyelli, shown here, is the oldest known reptile (315 million years)”… “Another paleontological mystery: where did turtles come from? Nobody knows.”)

Early Triassic Elessaurus: another overlooked Cosesaurus sister!

Cosesaurs (basal fenestrasaurs)
(Figs. 1–3) are popping up everywhere lately!

Figure 1. MIddle Triassic Cosesaurus. Double its size and it would be close to Early Triassic Elessaurus. See figure 6.

Earlier we looked at a tiny Early Cretaceous cosesaur skull in amber
originally mistaken for a bird/dinosaur: Oculudentavis.

Figure 2. Elessaurus hind limb elements gathered together to scale. Some original scale bars were off by 2x.

Today De-Oliveira et al. 2020 bring us
a new Early Triassic cosesaur, Elessaurus gondwanoccidens (UFSM 11471, Figs. 2, 3) known from a hind limb, pelvis, partial sacrum and proximal caudal vertebrae. Cosesaurus was a derived tanystropheid, close to Langobardisaurus, but was not included in the De-Oliveira taxon list.

Figure 3. Elessarus pes compared to Cosesaurus to scale and x2. Note differences between original tracing and DGS tracing. Digit 5 was not lost. It is tucked beneath the metatarsals and was not scored in the LRT.

In the De-Oliveira et al. published cladogram
(Fig. 4) Elessaurus nests basal to the Tanystropheidae (= Macrocnemus, Amotosaurus, Tanystropheus, Tanytrachelos and Langobardisaurus) a clade they derive from Protorosaurus and Trilophosaurus. in one cladogram (Fig. 4), but not in the other (Fig. 5). The authors reported, “In addition, the new specimen presents some features only found in more specialized representatives within Tanystropheidae, such as the presence of a well-developed calcaneal tuber with a rough lateral margin.” 

Figure 4. Published cladogram by De-Oliveira et al. 2020. Note different nesting than their SuppData cladogram in figure 5.

By contrast, in the De-Olveira et al. SuppData cladogram
Elessaurus nests without resolution within the Rhynchosauria (Fig. 5). Distinct from the first analysis (Fig. 4), this (Fig. 5) included the drepanosaur ancestor, Jesairosaurus, and the derived macrocnemid, Dinocephalosaurus. The authors reported, “In this second analysis, Elessaurus adopts different positions among the MPTs, it is recovered, e.g. within Archosauriformes, as a sister-taxa of Allokotosauria+Archosauriformes and an early rhynchosaur.” This is a red flag indicating major problems in scoring and taxon exclusion.

Figure 5. Cladogram from DeOliveira et al. 2020 with colors added to show distribution and mixing of Lepidosauromorpha and Archosauromorpha clades in the LRT. Many of these purported sister taxa do not look alike! Strangely, here Elessaurus nests with rhynchosaurs, not tanystropheids (Fig. 2). The taxon in pink had to be looked up and revised here.

By contrast, 
in the large reptile tree (LRT, 1661+ taxa), Elessaurus nests as a derived tanystropheid, alongside Cosesaurus (Fig. 1). a smaller Middle Triassic taxon omitted from the original study. The authors mistakenly considered tanystropheids to be archosauromorphs, again due to taxon exclusion. Despite the many traits that converged with archosauromorph protorosaurs, tanystropheids are tritosaur lepidosaurs, derived from Huehuecuetzpalli (Fig. 7) and Tijubina. In the LRT, adding taxa does not create chaos. New taxa neatly take their place within the current tree topology. By omitting Cosesaurus, the authors omitted the most similar taxon to Elessaurus and all the added information included herein.

The cladogram by De-Oliveira et al. shuffles lepidosauromorphs
with archosauromorphs without an understanding of their Viséan split. As a result, several purported ‘sisters’ in figure 5 do not look alike, but apparently nested together by default. Based on these ‘odd bedfellows’ I suspect the authors borrowed another worker’s cladogram without checking scores or examining results.

From the abstract:
“The origin and early radiation of Tanystropheidae, however, remains elusive.”

This is false. We know the origin of Tanystropheidae back to Cambrian chordates. Taxon exclusion by the authors prevents them from recovering both distant and proximal sister taxa.

“Here, a new Early Triassic archosauromorph is described and phylogenetically recovered as the sister-taxon of Tanystropheidae.”

By contrast, in the LRT Elessaurus is a derived fenestrasaur close to Cosesaurus, a taxon excluded by the authors. We know cosesaur tracks (= Rotodactylus, Fig. 6) go back to the Early Triassic and some were much larger than Cosesaurus.

Figure 6. Click to enlarge. Scaling a quadrupedal Cosesaurus to the larger Rotodactylus tracks from Haubold 1983. Quadrant represents center of balance in the closeup foot. Graphic representation of a butt joint is nearby.

More from the abstract:
“The new specimen, considered a new genus and species, comprises a complete posterior limb articulated with pelvic elements. It was recovered from the Sanga do Cabral Formation (Sanga do Cabral Supersequence, Lower Triassic of the Parana Basin, Southern Brazil), which has already yielded a typical Early Triassic vertebrate assemblage of temnospondyls, procolophonoids, and scarce archosauromorph remains. This new taxon provides insights on the early diversification of tanystropheids and represents further evidence for a premature wide geographical distribution of this clade. The morphology of the new specimen is consistent with a terrestrial lifestyle, suggesting that this condition was plesiomorphic for Tanystropheidae.”

Likewise, Cosesaurus and related fenestrasaurs in the LRT are terrestrial taxa, distinct from other tanystropheids, all arising from tritosaur lepidosaurs like Tijubina and Huehuecuetzpalli.

Figure 7. The father of all pterosaurs and tanystropheids, Huehuecuetzpalli, a late survivor in the Early Cretaceous from a Late Permian radiation.

Larger quadrupedal cosesaurs, 
like Elessaurus, had two sacrals (Fig. 1). Smaller bipedal cosesaurs (Fig. 1) had four. Both had anterior processes on the ilium, not longer than the acetabulum width, distinct from non-fenestrasaur tanystropheids.

PS
Figure 2 in De-Oliveira has a scale bar problem in their figure 2 (explained here in Fig. 2).

---

References
De-Oliveira TM, Pinheiro FL, Da-Rosa AAS, Dias-Da-Silva S and Kerber L 2020.
A new archosauromorph from South America provides insights on the early diversification of tanystropheids. PLoS ONE 15(4): e0230890

Dibothrosuchus: a new ancestor for Gracilisuchus and Scleromochlus

According to Wikipediia,
“Dibothrosuchus is a genus of sphenosuchian, a type of basal crocodylomorph, the clade that comprises the crocodilians and their closest kin. It is known from several partial skeletons and skulls. These fossils were found in Lower Jurassic rocks of Yunnan, China.  Dibothrosuchus was a small terrestrial crocodylomorph.”

Here
in the updated crocodyomorph portion of the large reptile tree (LRT, 1658+ taxa; Fig. 6) Dibothrosuchus (Figs. 1, 2) nests among the most basal bipedal crocodylomorphs (phylogenetically far from Sphenosuchus (Fig. 7).

Figure 1. Dibothrosuchus skull fossil with colors added. Note the differences in this skull and the illustrated one in figure 2.

More specifically
an earlier sister to Late Jurassic Dibothrosuchus arose from a sister to Middle Jurrassic Junggarsuchus and Late Triassic Pseudhesperosuchus, (Fig. 3) and gave rise to smaller Middle Triassic Gracilisuchus (Fig. 4, 5), and Late Triassic Scleromochlus, Saltopus and Lagosuchus.

Figure 2. Images from Wu et al. 1993, colors and hind limbs added. Traditionally the postorbital is considered fused to the postfrontal. Compare to figure 1.

Dibothrosuchus elaphros (Early Jurassic, Simmons 1965; Wu and Chatterjee 1993) is known from an incomplete skeleton, lacking hind limbs and distal tail. In the LRT Dibothrosuchus nests at the base of bipeds AND was derived from bipeds, so phylogenetic bracketing indicates Dibothrosuchus was a biped, too.

Figure 3. A new reconstruction of the basal bipedal croc, Pseudhesperosuchus based on fossil tracings. Some original drawings pepper this image. Note the interclavicle, missing in dinosaurs and the very small ilium, only wide enough for two sacrals. The posterior dorsals are deeper than the anterior ones.

As in descendant taxa,
the rostrum of Dibothrosuchus was perforated between the premaxilla and maxilla (Figs. 1, 2). The lateral temporal fenestra was elaborated with a quadrate that had three dorsal heads for a strong articulation with the skull roof. The cervicals and their ribs were quite robust. As in few other tetrapods, the cervicals and their ribs were deeper than the skull. The proximal carpals were longer than the metacarpals.

Figure 4. Present reconstruction of Gracilisuchus with skull based on Romer 1971. See figure 4 for an updated on that skull. Inset at upper right shows the broken femur (blue on the digram) and likely proximal carpals (green on the diagram).

Gracilisuchus stipanicicorum (Romer 1972; Butler et al. 2014; Ladinian, Middle Triassic, ~230 mya, 30 cm long; holotype PULSR8) is a basal crocodilomorph. It was derived from a sister to Dibothrosuchus and preceded both Saltopus and Scleromochlus.

Figure 5. Gracilisuchus skull updated with new colors. Skull image from Butler et al. 2014. Note the tall fenestra separating the premaxilla from the maxilla, as in Dibothrosuchus (Fig. 2).

Gracilisuchus was originally considered
an ornithosuchid by Romer (1972). Others thought it nested between Parasuchus and Stagonolepis (Benton and Clark 1988), as the sister to Postosuchus (Juul 1994) or Postosuchus and Erpetosuchus (Benton and Walker 2002). Butler et al. (2014) nested Turfanosuchus, Gracilisuchus and Yonghesuchus together in a clade. Yonghesuchus is close (Fig. 5), but other omitted taxa are closer to the other two. Turfanosuchus is also close, but nests at the base of the Poposauria in the LRT (Fig. 6).

Figure 6. Subset of the LRT focusing on the Crocodylomorpha, dorsal scutes, elongate proximal carpals, bipedality and clades.

The Gracilisuchus hind limb paper by Leuona and Desojo (2011)
is about PVL 4597 (Fig. 7), a different genus, the last common ancestor of all archosaurs (crocs + dinos).

Figure 7. Taxa from the croc subset of the LRT to scale. Click to enlarge.

Lagosuchus talampayensis 
(Romer 1971) is a smaller specimen found on the same slab as the Gracilisuchus holotype. In the LRT Lagosuchus nests with Saltopus, both derived from a sister to Scleromochlus, which was derived from Gracilisuchus (Fig. 7)… so all are members of the same tiny bipedal clade within Crocodylomorpha, all derived from a Middle to Early Triassic sister to Late Jurassic Dibothrosuchus.

This is an update
from blogpost #100 in October 2011 on basal bipedal crocs. Current blogpost # is something over 3000.

---

References
Benton MJ and Clark JM 1988. Archosaur phylogeny and the relationships of the Crocodilia in MJ Benton (ed.), The Phylogeny and Classification of the Tetrapods 1: 295-338. Oxford, The Systematics Association.
Butler RJ, Sullivan C, Ezcurra MD, Liu J, Lecuona A and Sookias RB 2014. New clade of enigmatic early archosaurs yields insights into early pseudosuchian phylogeny and
the biogeography of the archosaur radiation. BMC Evolutionary Biology 14:1-16.
Juul L 1994. The phylogeny of basal archosaurs. Palaeontographica africana 1994: 1-38.
Lecuona A and Desojo, JB 2011. Hind limb osteology of Gracilisuchus stipanicicorum(Archosauria: Pseudosuchia). Earth and Environmental Science Transactions of the Royal Society of Edinburgh 102 (2): 105–128.
Lecuona A, Desojo JB and Pol D 2017. New information on the postcranial skeleton of Gracilisuchus stipanicicorum (Archosauria: Suchia) and reappraisal of its phylogenetic position. Zoological Journal of the Linnean Society XX:1–40.
Parrish JM 1993. Phylogeny of the Crocodylotarsi, with reference to archosaurian and crurotarsan monophyly. Journal of Vertebrate Paleontology 13(3):287-308.
Romer AS 1971. The Chañares(Argentina) Triassic reptile fauna. Two new bu incompletely known long-limbed pseudosuchians. Breviora 378:1–10.
Romer AS 1972. The Chañares (Argentina) Triassic reptile fauna. An early ornithosuchid pseudosuchian, Gracilisuchus stipanicicorum, gen. et sp. nov. Breviora 389:1-24.
Simmons DJ 1965. The non-therapsid reptiles of the Lufeng Basin, Yunnan, China. Fieldiana Geology. 15: 1–93.
Wu X-C and Chatterjee S 1993. Dibothrosuchus elaphros, a crocodylomorph form the Lower Jurassic of China and the phylogeny of the Sphenosuchia. Journal of Vertebrate Paleontology 13:58-89.

wiki/Gracilisuchus
wiki/Dibothrosuchus

Former Gracilisuchus specimens: now closer to Trialestes

Over the last several weeks
the large reptile tree (LRT, 1660+ taxa, subset Fig. 1) was updated once again with a focus on the Crocodylomorpha. Two congeneric taxa known from a few scraps were eliminated. More insightful identification of skull bones (Figs. 1, 5) settled old issues. Over the next several posts some of the newly recovered hypothetical interrelationships will be presented for review.

We’ll start here
with a new nesting in the LRT (subset Fig. 1) for the small specimens (MCZ4116 and MCZ4118, Fig. 2) formerly assigned to Gracilisuchus (Figs. 4, 5). Now they nest either as hatchling Trialestes (Fig. 3), or, just as likely, as phylogenetically miniaturized Middle Triassic predecessors to the much larger and highly derived Late Triassic basal crocodylomorph, Trialestes. In either case, now Trialestes and its tiny doppelgänger nest together in the LRT, closer to each other than either is to any other taxon, despite a magnitude or two difference in size (Fig. 3). Gracilisuchus nests several nodes away in the next clade (Fig. 1).

Figure 1. Subset of the LRT focusing on the Crocodylomorpha, dorsal scutes, elongate proximal carpals, bipedality and clades. Images changes every 5 seconds.

Hatchling? Trialestes? (MCZ 4116, MCZ 4118, originally Gracilisuchus, Brinkman 1981; Middle Triassic; Fig. 2). These two specimens have a taller, narrow skull than Gracilisuchus (Figs. 4, 5) and a long list of other distinct traits and proportions that nest them with the very much larger Trialestes (Fig. 3) in the LRT (Fig. 1).

Figure 2. The former Gracilisuchus specimens MCZ4116 and MCZ4118 (Middle Triassic) with colors added.

Trialestes romeri (Bonaparte 1982) = Triassolestes (Reig, 1963/Tillyard 1918) Carnian, Late Triassic ~235 mya) is known from scattered parts here reconstructed and restored (Fig. 3). Clark, Sues and Berman (2000) redescribed the known parts and admitted the possibility that this taxon combined dinosaurian and crocodylomorph characters.

Figure 3. Trialestes (Late Triassic)  reconstructed. At upper left is MCZ4116 to scale.

Quadrupedal Trialestes
is indeed different than most basal bipedal crocodylomorphs (see Pseudhesperosuchus), but it has elongate proximal carpals (Fig. 3) and a long list of other croc clade traits. The elongate ilium is typical of bipedal taxa indicating a bipedal ancestry. Additional sacrals that would have filled out the sacral set between the ilia (Fig. 3) are not known, but likely were present.

Figure 4. Present reconstruction of Gracilisuchus with skull based on Romer 1971. See figure 4 for an updated on that skull.

In Trialestes
the vertebral centra had excavated lateral surfaces, for bird-like air sacs. The radius was longer than the humerus, a character otherwise known only in dinosaurs. The long radiale was slightly shorter than the ulnare. The fingers were tiny, another indicator of a bipedal ancestry. The pelvis was semi-perforated with a well-developed supraacetabular crest, as in basal dinosaurs. The femoral head was inturned, indicating an erect posture. The ankle joint had a crocodile normal configuration and a functionally pentadactyl pes.

Figure 5. Gracilisuchus (Middle Triassic) skull updated with new colors. Compare to figure 2.

The MCZ 4116 and MCZ 4118 specimens 
are coeval with Gracilisuchus in the Middle Triassic and similar in size, but share more traits in the LRT with highly derived Late Triassic Trialestes. As we’ve seen before, new morphologies often express their genesis in phylogenetically miniaturized taxa. That may be the case with the MCZ specimens, appearing millions of years before the much larger Trialestes. More discoveries, like an adult Trialestes in the Middle Triassic, will someday settle this ontogenetic and phylogenetic issue. This blogpost is where this issue starts. If this is not a novel hypothesis of interrelationships, let me know so I can promote the older citation.

Updates have been a continuing feature
of the LRT since its origin nine years ago, along with the steady addition of taxa to the present total of 1658 taxa, plus several hundred taxa in the pterosaur and therapsid cladograms. Correcting mistakes is standard practice in every science and every correction is another rewarding moment of discovery. Holding on to outdated and invalid hypotheses has been an acknowledged problem in paleontology.

---

References
Bonaparte JF 1982. Classification of the Thecodontia. Geobios Mem. Spec. 6, 99-112
Brinkman D 1981. The origin of the crocodiloid tarsi and the interrelationships of thecodontian archosaurs. Breviora 464: 1–23.
Clark JM, Sues H-D and Berman DS 2000. A new specimen of Hesperosuchus agilis from the Upper Triassic of New Mexico and the interrelationships of basal crocodylomorph archosaurs. Journal of Vertebrate Paleontology 20(4):683-704.
deFranca MAG, Bittencourt JdS and Langer MC 2013. Reavaliação taxonomica de Barberenasuchus brasiliensis (Archosauriformes), Ladiniado do Rio Grande do Sul (Zona-Assembleia de Dinodontosaurus). Palaenotogia em Destaque Edição Especial Octubro 2013: 230.
Irmis RB, Nesbitt SJ and Sues H-D 2013. Early Crocodylomorpha. Pp. 275–302 in Nesbitt, Desojo and Irmis (eds). Anatomy, phylogeny and palaeobiology of early archosaurs and their kin. The Geological Society of London. doi:10.1144/SP379.24.
Kischlat EE 2000. Tecodôncios: a aurora dos arcossáurios no Triássico. Pp. 273–316 in Holz and De Ros (eds.). Paleontologia do Rio Grande do Sul. Porto Alegre: CIGO/UFRGS.
Lecuona A, Ezcurra MD and Irmis RB 2016. Revision of the early crocodylomorph Trialestes romeri (Archosauria, Suchia) from the lower Upper Triassic Ischigualasto Formation of Argentina: one of the oldest-known crocodylomorphs. Papers in Palaeontology (advance online publication). DOI: 10.1002/spp2.1056
Reig, OA 1963. La presencia de dinosaurios saurisquios en los “Estratos de Ischigualasto” (Mesotriásico Superior) de las provincias de San Juan y La Rioja (República Argentina). Ameghiniana 3: 3-20.
Riff D et al. 2012. Crocodilomorfos: a maior diversidade de répteis fósseis do Brasil. TERRÆ 9: 12-40, 2012.
Zanno LE, Drymala S, Nesbitt SJ and Schneider VP 2015. Early Crocodylomorph increases top tier predator diversity during rise of dinosaurs. Scientific Reports 5:9276 DOI: 10.1038/srep09276.

wiki/Trialestes

Genesis of air breathing in basal tetrapods

The genesis of limbs and toes 
from lobes and fins gets most of the publicity in transitional fish-tetrapods.

Today we look at the less popular transition
from water breathing with gills to air breathing with a nose and lungs.

Like most fish,
Onychodus (Fig. 1) drew in oxygenated water by opening its mouth. At this moment, the gill covers are closed to prevent backdraft. Closing the mouth and raising the basihyal (medial bone between the mandibles) until it presses against the solid palate reduces the mouth volume, pushing that mouthful of  water posteriorly past the gills where oxygen and carbon dioxide are transferred. At this time the gill covers are open to permit that water to exit, completing the cycle. The dual nares have nothing to do with respiration at this point, only olfaction, with water passively entering the anterior opening and passively exiting the posterior opening (Fig. 1). The air bladder arising from the gut tube anterior to the stomach is not involved in respiration at this stage.

Among lobefin fish,
coelacanths, like Latimeria, have this primitive system.

Figure 1. Onychodus is typical of most fish having dual external nares strictly for olfactory sensing. Gill covers are part of the respiratory apparatus.

Among lobefin lungfish (Late Silurian to present),
like Kenichthys (Fig. 2), Youngolepis, Polypterus (the extant bichir) and Howidipterus, oxygen-poor water, supplemented by gulps of dry air, once again enters the mouth and is passed back over the gills and out the gill covers. Both the incurrent and excurrent nares migrate ventrally. (Not sure why.) Worthy of a Nature article, the excurrent opening is parked on the jaw margin between the premaxilla and maxilla in Kenichthys, so half the excurrent exited outside the mouth, while the other half exited inside the mouth (see ventral view in Fig. 2), all passively. (Not sure why this migration took place either, except that with the lips sealed inhalation and exhalation can still take place… slowly… in and out of both openings, perhaps to retain mouth moisture during aestivation (hibernation in dry mud.) Note the pinprick size of each opening.

Figure 2. Kenichthys Images from Zhu and Ahlberg 2004, colors added. The authors made a convincing argument that Kenichthys represented a transitional taxon between Youngolepis and Eusthenopteron. Note the lack of vomer fangs and a distinctly different set of skull sutures in Kenichthys, which does not nest with Eusthenopteron in the LRT.

Among basal lobefin crossopterygians (Early to Late Devonian),
like Gogonasus, Eusthenopteron, and elongate, flattened Cabonnichthys, Elpistostege, Tiktaalik and Panderichthys the tiny excurrent nasal opening just barely enters the rim of the mouth cavity and is thereafter considered a choana. The tiny external incurrent opening is thereafter considered a naris. Based on their tiny sizes, both remain useless for respiration. Large gill covers and a solid palate are retained for traditional water respiration supplemented by dry air gulping as needed.

Figure 3. Panderichthys palates. Note the lateral line below the naris is not continuous, contra Lombard and Bolt.

When the gill covers disappear in fossil taxa
that signals the genesis of air-breathing from mouth to paired air bladders (now called ‘lungs’) rather than past the disappearing gills. According to the LRT, this occurred twice (if we don’t count the ontogenetic transformation of juvenile tadpoles (with gills) to adult frogs (with lungs) and other similar basal tetrapods).

In clade one: primitive Koilops retained and operculum (gill cover). Derived, but lobe-finned Tiktaalik and Spathicephalus did not have an operculum.

In clade two: weak limbed, four-fingered Trypanognathus (Fig. 4), Deltaherpeton, Collosteus, Pholidogaster, Greererpeton and Ossinodus, all lacked an operculum.

Figure 4. Animation of air-breathing (tidal ventilation) in basal tetrapods with weak lungs inflated by contraction and expansion of the throat sac, rather than gill irrigation powered by the reduced ceratobranchials, still present at right. Air-tight nose flaps had to be present in order for this system to work. 

Clade two exceptions: Robust-limbed, eight-fingered Acanthostega (Fig. 5) and Ichthyostega retained tiny gill covers (operculum) as adults. And they had primitive tiny nares and choana, still not suitable for air-breathing. These convergent exceptions are here considered reversals due to a suite of derived traits nesting these two famous taxa apart from more primitive tetrapods and apart from each other in the LRT.

Figure 5. the MGUH VP 8160 specimen attributed to Acanthostega. Note the many similarities to Ymeria. Note the spiracle openings surrounded by the supratemporals. This provides an accessory respiration opening, convergent with bottom-dwelling skates and rays from the shark clade.

The signal that air-breathing respiration through the nostrils had begun
(Fig. 4) is when the nares and choana of fossil taxa enlarge to handle the larger volume of tidal ventilation coming through them. The nares also migrate higher on the skull so that they are at least partly visible in dorsal view. The internal nares are fully inside the mouth, which must be able to seal shut to divert air through the nares, rather than leaking past the lips. Gill covers are absent. Air-tight nose flaps had to be present in order for this system to work. The pterygoids reduce and retreat posteriorly (Fig. 4), creating large, pliable openings in the formerly solid palate (Fig. 3), expanding the potential volume of the mouth.

According to the LRT,
(subset Fig. 5) the enlargement and migration of the nares and choana occurred several times because several clades of derived basal tetrapods retained tiny lateral nares and choana despite having fully developed limbs.

Figure 5. Subset of the LRT focusing on basal tetrapods and their narial openings.

Dorsal ribs
Basal tetrapods depend on an expanding and contracting the gular sac for tidal ventilation of the lungs, mimicking their lobe-finned ancestors. These same basal tetrapods (Fig. 6) were all low and wide with relatively straight, laterally-oriented ribs incapable of expanding and contracting the torso and lungs. Not until dorsal ribs elongated and started curling around the inside of an increasingly round (in cross-section) torso where they able to expand and contract the volume of the torso and the lungs inside. In that way mobile ribs gradually replaced a mobile throat sac for tidal ventilation.

Figure 6. Dorsal and ventral views of Panderichthys and several basal tetrapods demonstrating the low, flat skulls and bodies with small limbs and relatively straight ribs, all to scale. Note the brevity of the tail in thee taxa.

The irony is
we know of Ichthyostega-grade tetrapods walking on land in the Middle Devonian. By that I mean, we know of tetrapods with relatively large limbs and supernumerary digits capable of elevating the belly off the substrate. Phylogenetic analysis indicates the trackmaker was a mouth-breather with tiny lateral nares. This was a short-lived experiment (as far as we know at present) leaving only Late Devonian descendants, like Icthyostega, that disappeared by the Early Carboniferous.

The longer lasting clade,
the one that produced all the other tetrapods including reptilomorphs, living amphibians and microsaurs, all had a long, low, flat body and skull with smaller 4-fingered limbs not capable of elevating the belly off the substrate, like Greererpeton and Trimerorhachis (Fig. 6). Only later, and by convergence did descendants rise off their belly with stronger limbs, mimicking those pioneer Middle Devonian tetrapod trackmakers.

---

References
Schoch RR and Voigt S 2019. A dvinosaurian temnospondyl from the Carboniferous-Permian boundary of Germany sheds light on dvinosaurian phylogeny and distribution. Journal of Vertebrate Paleontology. DOI: 10.1080/02724634.2019.1577874.xxx

This blogpost comes not in response to a new academic paper, but to revisiting some of the taxa in the the large reptile tree (LRT, Figs. 5, 6) at this transition. Thanks to reader Dave M for the impulse to reexamine these taxa.