Kenomagnathus: what you can do with only 2 bones

Spindler 2020
reports on a new basal pelycosaur, Kenomagnathus scottae (ROM 43608; Upper Pennsylvanian, Late Carboniferous, Garnett, KS, USA; Figs 1-3) known from a single lacrimal and maxilla (with teeth) exposed in medial view (Fig. 1).

Figure 1. Kenomagnathus in situ from Spindler 2020. The halo of organic matter is interesting.

From the abstract:
“This is the oldest known diastema in synapsid evolution, and the first reported from a faunivorous member that lacks a precanine step, aside from Tetraceratops. This unique precanine morphology occurred independently from similar structures in Sphenacodontoidea.” 

See Spindler’s freehand drawing
of the ‘true diastema’ (Fig. 2). 

Figure 2. Kenomagnathus maxilla and lacrimal with the rest of the skull restored in lateral view. Note the deeper jugal (cyan), though not as deep as in Ophiacodon (Figs. 3, 4). For that reason the mandible of Ophiacodon was used in this restoration. Spindler’s freehand drawing indicates a deeper orbit, shallower jugal and smaller naris along with a larger mandible.

It is worth noting
that maxillary teeth shrink toward the naris in Ophiacodon (Fig. 3). A diastema may be present in Pantelosaurus (formerly Haptodus saxonicus, Fig.3). These pertinent taxa were not illustrated in Spindler 2020.

Figure 3. Pertinent synapsid skulls to scale. The origin of the Pelycosauria + Therapsida is marked by phylogenetic miniaturization, as in so many other clade origins. Note the depth of the jugal in basal taxa here.

Spindler’s freehand restoration
increased the size of the orbit and decreased the depth of the restored jugal. So this is yet another cautionary tale highlighting the danger in using freehand drawings in scientific studies.

The shallow jugal depth in the Spindler freehand restoration
is a key oversight. When repaired (Fig. 2) the semi-deep jugal of Kenomagnathus transitionally links deeper jugal Ophiacodon (Fig. 3) to shallower jugal Pantelosaurus and Haptodus (Fig. 3) at the base of Pelycosauria + Therapsida in the large reptile tree (LRT, 1642+ taxa). While running the risk of ‘Pulling a Larry Martin’, there are so few traits to consider here (Fig. 1) and none contradict the present hypothesis of interrelationships. All that puts Kenomagnathus in the lineage of synapsids leading to therapsids, mammals, primates and humans.

---

References
Spindler F 2020. A faunivorous early sphenacodontian synapsid with a diastema. Palaeontologia Electronic 23(1):a01. doi: https://doi.org/10.26879/1023
https://palaeo-electronica.org/content/2020/2905-early-sphenacodontian-diastema

Helodus: a skull without sutures

Decades prior to PAUP and MacClade,
Professor Moy-Thomas 1936 reasoned that Helodus simplex (Fig. 1; Agassiz 1838; Early Carboniferous, 300mya, 30cm long) ) was close to the ancestry of the clade Holocephalii (ratfish, chimaeras and kin; Fig. 2), which we looked at yesterday. In complete accord, the large reptile tree (LRT, 1641 taxa; subset Fig. 3) fully supports that nesting using modern software: PAUP and MacClade. So… belated well done, Professor Moy-Thomas!

Figure 1. Helodus skull drawings from Moy-Thomas 1936 show no skull sutures. Colors are applied with blending edges to show were bones are based on tetrapod homologs. Gill bars are missing from these diagrams, so were added in light blue here.

In Helodus
the skull bones are all fused together, so suture estimates are provided here (Fig. 1) based on phyogenetic bracketing. Note the tiny premaxillary teeth and complex maxillary teeth. Tabulars appear to be absent. Note the coosified cervicals / anterior dorsals extending to the notochord and first dorsal spine.

Moy-Thomas 1936
considered the anatomy of Helodus in detail. From the abstract:

1. “The skull is found to be holostylic, and to have many characters in common with the skull of the Holocephali, but in some respects is less specialized.
2. The pectoral fins, with their long metapterygium, small propterygium, and fused anterior radials, resemble very closely those of the Holocephali.
3. The pelvic and unpaired fins, and general body shape are found to resemble those of the Holocephali.
4. It is concluded that the Cochliodonts are almost certainly closely related to the ancestors of the Holocephali, and the relatively unspecialized condition of the teeth gives support to the view that the holostylic condition of the jaws is primitive for the group. It is suggested that all the Bradyodonts were holostylic, that the hyomandibular may never have been suspensory, and that they may have diverged from the true Selachii before the hyomandibular played a part in the jaw suspension.”

FIgure 2. Ratfish (chimaera) and Heterodontus to scale.

 

Taking one phylogenetic step further back from Helodus,
yesterday we looked at Heterodontus (Fig. 2), the Chondrichthyes taxon phylogenetically ancestral to Helodus and the Holocephalii.

Figure 3. Subset of the LRT focusing on basal chordates, vertebrates and bony fish not related to tetrapods. Chondrichthyes are in pink.

 

References
Agassiz L 1838. Recherches Sur Les Poissons Fossiles. Tome III (livr. 11). Imprimérie de Petitpierre, Neuchatel 73-140.
Moy-Thomas JA 1936. On the Structure and Affinities of the Carboniferous Cochliodont Helodus simplex. Cambridge University Press. 73(11):488–503.

wiki/Chondrichthyes
wiki/Chimaera
wiki/Cladoselache
wiki/Chondrosteus
wiki/Symmoriida
wiki/Horn_shark
wiki/Helodus not listed in English

The origin of the Holocephali (= chimaeras, ratfish)

Didier 1995 reports,
“There are two hypotheses on the origin of Holocephali (Bonaparte 1832). The first and most generally accepted scenario is that holocephalans have evolved from some lineage of bradyodont sharks. The second hypothesis suggests that holocephalans are most closely related to placoderms.”

FIgure 1. Ratfish (chimaera) and Heterodontus to scale.

By contrast
in the large reptile tree (LRT, 1635+ taxa, Fig. 3) the few tested holocephalians (Chimaera (Fig. 1) and Belantsea)) arise from the horn shark, Heterodontus, (Fig. 1). More primitive taxa include Falcatus (Fig. 2), Iniopteryx, and Akmonistion.

Figure 2. Falcatus traced with DGS methods with reconstructed freehand image applied from Lund 1985.

Didier 1995 reports,
“The adductor muscles of Heterodontus also lie anterior to the eye and superficially they resemble chimaeroid fishes in this respect. I interpret this as a convergent feature of heterodontids and chimaeroids.” That is the only mention of Heterodontus (Fig. 1) in the text. Squalus is the outgroup taxon in Didier’s figure 46 cladogram.

Figure 3. Basal vertebrates and  tetrapods.

References
Bonaparte CL 1832. Iconografia delle fauna italica per le quattro classi degli animali vertebrati. Tomo III. Pesci. Roma. [Issued in puntata (installments), without pagination; total of 556 pp., 78 pls.
Didier DA 1995. Phylogenetic Systematics of Extant Chimaeroid Fishes (Holocephali, Chimaeroidei). American Museum Novitates 3119:86pp.

Crayssac basal pterosaur tracks? …or tenrec tracks?

Earlier we looked at Mazin and Pouech 2020
who claimed they had discovered “the first non-pterodactyloid pterosaurian trackways.” At the time, only the abstract was available to discuss and criticize.

Nine years ago
Peters 2011 published anurognathid tracks, which makes them the first non-pterodactyloid pterosaurian trackways published. Notable by its exclusion, Mazin and Pouech 2020 did not cite, “A catalog of pterosaur pedes for trackmaker identification” (Peters 2011), confirming Dr. S. Christopher Bennett’s threat, “You will not be published. And if you are published, you will not be cited.”

Now that I have seen the paper and the tracks,
(Figs. 1, 2) let’s determine what sort of tetrapod made those tracks named, Rhamphichnus crayssacensis, because they don’t look like other pterosaur tracks, as workers (see below) acknowledge.

Diagnosis from Mazin and Pouech 2020:
“Quadrupedal trackway with tridactyl digitigrade manus-prints and pentadactyl plantigrade to digitigrade pes-prints. Subparallel manus digit-prints orientated anteriorly. Pentadactyl pes-prints with more or less divergent digit prints. Pedal digit V divergent and postero-laterally rejected. Manus trackway slightly to clearly wider than the pes trackway.”

Distinct from typical pterosaur manus tracks:

1. tridactyl digit prints are subparallel (rather than widely splayed)
2. digits are oriented anteriorly (rather than laterally to posteriorly)
3. digits sometimes include additional medial and lateral impressions (never seen in other pterosaur tracks)
4. no claw marks are present (that seems wrong based on Fig. 1)
5. the manus impression is just anterior to the pes impression (rather than laterally and posteriorly, as in other pterosaur tracks)

Figure 1. Images from Mazin and Pouech 2020. Some manus tracks have at least four digits.

There are many
basal and derived pterosaurs with pedal digit 2 (or 2 and 3) the longest, distinct from Triassic pterosaurs. These were all examined and rejected as potential trackmakers matching Rhamphichnus for various reasons.

I also looked at 1600+ non-pterosaur trackmakers
due to the many unexpected traits (see list above) present in the Rhamphichnus tracks.

First and foremost,
the pterosaur antebrachium (radius + ulna) could not be pronated to produce anteriorly-oriented Rhamphichnus tracks. Due to folding and flying issues, pterosaurs, like birds, do not have the ability to pronate and supinate the wing. That’s why all pterosaur manus tracks are oriented laterally with fingers at full extension, impressing into the substrate. That manus digit 3 is often rotated posteriorly is a clue to its lepidosaurian ancestry. These facts form the hypothesis of a secondarily quadrupedal configuration for some, but not all pterosaurs.

One overlooked trackmaker stood out as a good match
for Rhamphichnus: the tenrec, Tenrec (Fig. 2), a small digitigrade quadrupedal mammal currently restricted to Madagascar. The medial and lateral manual digits are shorter than 2-4, which are parallel in orientation.

Figure 2. Rhamphichnus tracks compared to a Tenrec trackmaker. The brevity of pedal digit 5 is a mismatch, but a related taxon, Leptictidium, likewise reduces pedal digit 5.

One of those Tenrec sisters,
Rhynchocyon, greatly reduces manual digits 1 and 5, but pedal digit 3 is the longest.

Another Tenrec sister,
Leptictidium (Fig. 3), has a pes with a reduced pedal digit 5, but a short digit 2, but the manus is also a good match for Rhamphichnus. So there is great variation in the pes of tenrec clade members. Still, a small tenrec-like mammal remains a more parsimonious trackmaker than any Late Jurassic pterosaur. They were able to pronate the manus!

Figure 3. Elements of Leptictidium from Storch and Lister 1985.

Due to taxon exclusion,
Mazin and Pouech 2020 did not consider alternative trackmakers for the pterosaur-beach Rhamphichnus tracks that don’t match other pterosaur tracks or extremities. Now we’re stuck with an inappropriate name for these Late Jurassic tenrec tracks.

A late Jurassic tenrec?
The large reptile tree (LRT, 1637 taxa) supports the probability that a sister to Tenrec was present in in the Late Jurassic based on the coeval presence of derived members of Glires (Multiturberculata). Placental mammal fossils remain extremely rare in the Mesozoic, but these impressions add to their chronology.

It is worth repeating, due to the subject matter,
the Crayssac pterosaur beach still includes the pes of the JME-SOS 4009 specimen attributed to Rhamphorhynchus, as mentioned earlier. Here it is again (Fig. 4).

Figure 4. Crayssac track different from all others. Inset: Pes of Rhamphorhynchus muensteri JME-SOS 4009, no. 62 in the Wellnhofer catalog

twitter.com/Mark Witton mistakenly reports:
“Turns out we’ve been over-thinking it (pedal digit 5): it just lays flat on the ground during walking, like a regular toe.”

“For one, the walking fingers face forward, not sideways, as in pterodactyloids. This seems weird, but it turns out that non-ptero wing fingers fold roughly perpendicular to the walking digits.”

These basic bungles by a PhD pterosaur worker
demonstrate the dominance of myth-making among purported experts due to accepting published results like a journalist, without testing them, like a scientist. Dr. Witton’s 2013 pterosaur book is full of similar mistakes reviewed here in a seven-part series.

---

References
Mazin J-M and Pouech J 2020.The first non-pterodactyloid pterosaurian trackways and the terrestrial ability of non-pterodactyloid pterosaurs. Geobios 16 January 2020. PDF
Peters, D. 2011. A Catalog of Pterosaur Pedes for Trackmaker Identification
Ichnos 18(2):114-141. http://dx.doi.org/10.1080/10420940.2011.573605

https://pterosaurheresies.wordpress.com/2020/01/18/first-non-pterodactyloid-pterosaurian-trackways-ever-described-no/

 

Phylogeny of Desmostylia: Matsui and Tsuihiji 2019

Matsui and Tsuihiji 2019 bring us their views
on the phylogeny of taxa within their Desmostylia, an order of large aquatic (Pacific rim) mammals, best known from the early Oligocene (31mya) to the late Miocene (7.25mya). Traditionally there are relatively few taxa in the clade Desmostylia (Neoparadoxia (Fig. 2), Paleoparadoxia, Behemotops (Fig. 2), Desmostylus (Fig. 3) and a few others not as well represented in the fossil record.

Unfortunately the authors’ views
are too restricted with too few taxa under consideration. Desmostylians are not the extinct taxa Matsui and Tsuihiji think they are when more taxa are included (Fig. 1).

From their abstract:
“Background.
Desmostylia is a clade of extinct aquatic mammals with no living members.”

False. Although this is traditional thinking in the large reptile tree (LRT, 1638+ taxa; subset Fig. 1) living desmostylians include mysticete (baleen) whales. Their ancestors include oreodonts, mesonychids, hippos, cambaytheres and anthracobunids.

Figure 1. The Merycoidodon cladogram includes hippos, whales and a number of extinct taxa. Traditional desmostylians are in medium blue here.

 

The abstract continues:
“Today, this clade is considered belonging to either Afrotheria or Perissodactyla.”

Their figure 1 shows desmosytlians arising from either: (a) Equus, the horse, in the Perissodactyla hypothesis; (b) Elephas, the elephant, in the Afrotheria hypothesis, or (c) Procavia, the hyrax, in their Paenungulatomorpha hypothesis. In other words, the authors have no idea. Genomic studies deliver false positives like Afrotheria. and you can’t used genomic studies of deep time fossils. In the LRT, which uses traits, desmostylians arise from mesonychids, hippos and anthracobunids as we learned earlier here.

“In the currently-accepted taxonomic scheme, Desmostylia includes two families, 10
to 12 genera, and 1314 species. There have been relatively few phylogenetic analyses
published on desmostylian interrelationship compared to other vertebrate taxa, and
two main, alternative phylogenetic hypotheses have been proposed in previous studies.
One major problem with those previous studies is that the numbers of characters and
OTUs were small.”

So Matsui and Tsuihiji, are studying the details without understanding the big picture… which could affect the details. Better to have a firm foundation built, then afterwards add whatever decorations to your structure.

Figure 2. Rorqual evolution from desmostylians, Neoparadoxia, the RBCM specimen of Behemotops, Miocaperea, Eschrichtius and Cetotherium, not to scale.

The abstract continues:
“Methods.
In this study, we analyzed the phylogenetic interrelationship of Desmostylia
based on a new data matrix that includes larger numbers of characters and taxa than in
any previous studies. The new data matrix was compiled mainly based on data matrices
of previous studies and included three outgroups and 13 desmostylian ingroup taxa.
Analyses were carried out using five kinds of parsimonious methods.”

Their three outgroups were: [1] Anthracobune, [2] Pezosiren + Moeritherium and [3] all three. Again, they are playing ‘pin the tail on the donkey‘ when the actual outgroups are available online (Fig. 1). No blind guesswork is necessary.

Figure 3. Taxa in the lineage of right whales include Desmostylus, Caperea and Eubalaena. The tiny bit of jugal posterior to the orbit (in cyan) is found in all baleen whales tested so far. The frontals over the eyes are just roofing the eyeballs in Desmostylus, much wider in Caperea and much, much longer in Eubalaena.

The abstract continues:
“Results.
Strict consensus trees of the most parsimonious topologies obtained in all
analyses supported the monophyly of Desmostylidae and paraphyly of traditional
Paleoparadoxiidae. Based on these results, we propose phylogenetic definitions of the
clades Desmostylidae and Paleoparadoxiidae based on common ancestry.”

That’s fine, but incomplete. This is another classic case of knowing so much about a few taxa, while knowing nothing about their ancestors, sisters and descendants, all based on taxon exclusion. That’s why the LRT is here. The LRT is the first instrument to report that the clade Cetacea was diphyletic. Members of the Odontoceti (toothed whales) are not related to members of the Mysticeti (baleen whales), contra traditional studies that exclude pertinent taxa.

Some workers insist
that I add characters to the LRT, but as you can see, adding more traits to the LRT would not reveal the ancestors, sisters or descendants of the Desmostylia. Only more taxa solve this problem. More taxa create greater resolution and nest all enigmas. More characters do not and cannot do that, so enough with that lame request.

If you want to add more or different traits
to your more focused studies, by all means, do so! But keep that taxon list complete. Let the LRT be your guide.

---

References
Matsui and Tsuihiji 2019. The phylogeny of desmostylians revisited: proposal of new clades based on robust phylogenetic hypotheses. PeerJ:e7430 http://doi.org/10.7717/peerj.7430

Looking for the sternal complex in a tiny pterosaur

All pterosaurs have a sternal complex
(sternum + interclavicle + wrap-around clavicles), even the flightless ones. This tiny specimen (Fig. 1) probably had a sternal complex, but where is it? As everyone knows, it should be between the elbows, but it’s not there.

Figure 1. Tiny pterosaur mistakenly named Pterodactylus? pulchellus. I cannot find the sternal complex here. It should be between the elbows. That tiny red triangle under the mid-humerus is the ventral coracoid.

Pterodactylus? pulchellus BM NHM 42735 is the same size as the closely related Gmu-10157 specimen, but has a longer rostrum. The BM NHM specimen is one node closer to the common ancestor of cycnorhamphids + ornithocheirids in the large pterosaur tree (LPT, 242 taxa). The sternal complex appears to be missing or displaced in this otherwise undisturbed tiny specimen. Soft tissue confirms the narrow chord wing membrane and dual uropatagia. Pedal digit 5 remained long.

Figure 2. The GMU 10157 specimen and the P? pulchellus BM NHM 42735 specimens to scale and full size.

These tiny adults,
(Fig. 2) derived from slightly larger scaphognathids (Fig. 3) are transitional taxa undergoing phylogenetic miniaturization at the genesis of Cycnorhamphidae + Ornithocheiridae. They have not been given novel generic names by established workers because the traditionalists among them consider these to be babies/juveniles of larger, undiscovered taxa. Thus they have remained relatively ignored, despite their pristine preservation and sometimes gravid condition.

Figure 3. Click to enlarge. Taxa in the lineage of Cycnorhamphidae + Ornithocheiridae in the LPT.

The key to finding the missing sternal complex
on this relatively undisturbed specimen is to look to the only area of the skeleton that is slightly disturbed (Fig. 4). The gastralia basket is expanded beyond its natural contours in the BM NHM specimen and that’s where I find (thanks to DGS) a displaced sternal complex, separated from the coracoids and jammed back into the stomach, surrounded by gastralia, almost hidden from view.

Figure 4. Here the sternal complex of the BM NHM 42735 specimen is colored indigo.

Not sure how that happened during taphonomy,
but there you go: mystery solved!

Figure 5. Shenzhoupterus skull in situ with sternum in blue.

Earlier a sternal complex was found beneath the skull
of Shenzhoupterus (Figs. 5, 6) using the same techniques, contra traditional reconstructions (Lü, Unwin, Xu and Zhang 2008; see skull diagram insert matching no other pterosaur skull morphology in Fig. 6). Despite its derived state, the newly reconstructed Shenzhoupterus skull (Fig. 6 standing skeleton) bears all the hallmarks of sister taxa.

Figure 6. Shenzhoupterus reconstructed alongside original interpretation of skull.

While we’re on this subject,
Shenzhoupterus does not nest with azhdarchoids, as originally hypothesized, but with tiny Nemicolopterus, between dsungaripterids and tapejarids in the LPT—and neither of these clades are related to azhdarchids in the LPT, contra traditional thinking that excludes tiny taxa and large swathes of congeneric taxa.

---

References
both of the tiny taxa listed above await description and publication other than in:
Wellnhofer P 1970. Die Pterodactyloidea (Pterosauria) der Oberjura-Plattenkalke Süddeutschlands. Abhandlungen der Bayerischen Akademie der Wissenschaften, N.F., Munich 141: 1-133.

Shenzhoupterus was first described in:
Lü J, Unwin DM, Xu L and Zhang X 2008. A new azhdarchoid pterosaur from the Lower Cretaceous of China and its implications for pterosaur phylogeny and evolution. Naturwissenschaften 95 (9): online (preprint). doi:10.1007/s00114-008-0397-5. PMID 18509616.

The spiny dogfish (Squalus) enters the LRT

Slow, low and slinky,
the spiny dogfish shark (Squalus acanthias, Linneaus 1758; up to 1m in length; Figs. 1, 2) enters the large reptile tree (LRT, 1634+ taxa) basal to the angel shark (Squatina, Fig. 4) and the eagle ray (Aetobatus, Fig. 5) on one branch, and basal to the classic sharks, the mako (Isurus) and hammerhead (Sphyraena) on the other branch. On the ancestral side, Squalus is derived from two other small slinky sharks each with a terminal mouth, Cladoselache (Fig. 3) and Chlamydoselachus.

Figure 1. The spiny dogfish, Squalus acanthi as, in vivo.

Spiny dogfish are bottom-dwellers.
A tiny spine precedes each dorsal fin. They exude a mild venom, a precursor trait to descendant sting rays. The rostrum is elongate, making the jaw underslung. The lateral gill slits are set low, anterior to the pectoral fins. The anal fin is absent. Note the straightening of the heterocercal tail.

Figure 2. Skull of Squalus acanthi as with DGS colors added according to tetrapod skulls.

Squalus retains a gill bar
lateral to the jaws (red). The orbit is enormous and so is the naris, so this hunter relies on sight and smell.

Figure 3. Classic reconstruction of Cladoselache, a shark-like taxon basal to higher sharks and rays. Note the presence of dorsal fin spines and an absent anal fin.

Figure 4. Squatina in vivo, lateral view. The large pectoral and pelvic fins give Squatina a broad, ray-like appearance in dorsal view. Note the lack of an anal fin. Distinct from Squalus, Squatina has a terminal mouth and retracted nares.

Figure 5. The spotted eagle ray, Aetobatus in vivo. Venom spines appear here, as well.

PS: Quick Reminder
The LRT nests guitarfish, like Rhinobatos, and skates with Isurus, not with this clade. And nests Manta with Rhincodon, the whale shark, not with this clade, despite the massive convergence.

---

References
Linnaeus C 1758. Systema naturæ per regna tria naturæ, secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis. Tomus I. Editio decima, reformata.

wiki/Squalus