Lophiaspis: an odd oreodont with tiny upper teeth and a bullet-shaped skull

Vautrin et al. 2020 published on rare Lophiaspis
(Fig 1) a less than popular and poorly known taxon.

Wikipedia reports only this:
“Lophiaspis is an extinct genus of early perissodaclotyl endemic to southern Europe during the Early to Middle Eocene, living from 55.8 to 37.2 Ma. Remains have been found from France, Spain, and Portugal.”

Figure 1. Lophiaspis maurettei from Vautrin et al. 2020. Colors and reconstruction added here. The presenced of only tiny anterior upper teeth and the flattness of the upper molars is real, not an artifact of preservation, evolving from the tiny canine plesiomorphic condition in Agriochoerus antiquus.

By contrast
in the large reptile tree (LRT, 2047 taxa) Lophiaspis maurettei (Depéret 1910; Vautrin et al. 2014, Early Eocene 43mya) nests with oreodonts. It was similar to Agriochoerus antiquus (Figs 2, 3), but with more of its anterior upper teeth reduced to vestiges. The naris opened ventrally. That’s a rare morphology for placentals, convergent with humans. The upper molars were flattened with little to no lateral exposure. In lateral view the skull was streamlined like a bullet.

Figure 2. The skull of Agriochoerus antiquus has a vestigial canine and larger incisors, the opposite of most other oreodonts. See the closeup in figure 3. Note the extension of the nasals that will evolve to produce a ventral naris in Lophiaspis.

Originally and recently
(e.g. Vautrin et al. 2020) Lophiaspis was considered a primitive perissodactyl and a relative of the basal tapir or a basal chalicothere (Bai, Wang and Meng 2018). In both studies Lophiaspis nested with what they thought was a basal tapir, Lophiodon (Fig 4, actually a marsupial), based largely on convergent tooth traits and taxon exclusion. Vautrin et al. 2014 omitted Agriochoerus and oreodonts from their study that focused only on basal perissodactyls to the exclusion of other placental mammal taxa.

Figure 3. A close up of the anterior dentition in the oreodont, Agriochoerus antiquus. The canines are orange. The upper incisors are gold. Note the upper canine is a vestige here, much smaller than the lower canine and smaller than the anterior premolar (cyan).

Bai, Wang and Meng 2018 likewise nested Lophiaspis
with perissodactyls by cherry picking taxa and omitting oredonts from their taxon list. These authors nested Lophiaspis with chalicotheres (aka Ancylopoda). These authors also nested the basal pigs, Lambdotherium and Danjiangia with brontotheres while omitting pig ancestors like Cainotherium. These authors also nested the desmostylian ancestor and derived hippo, Cambaytherium with the basal elephant, Radinskya, by omitting hippos, desmostylians and elephants. The LRT minimizes taxon exclusion by including so many more taxa that prior, more focused studies.

Vautrin et al. 2014 reported,
“The Lophiodontidae are endemic perissodactyls from Europe that flourished during the Eocene.

When more taxa are added Lophiaspis (Fig 1) is not related to Lophiodon (Fig 4) or perissodactyls. So the rest of this abstract describes an irrelevant clade.

“Despite their preponderance in the European fossil record, their exact origin and relationships within the perissodactyls remain unknown due to the rare and fragmentary material in the early Ypresian, the time of their earliest radiation. Lophiaspis maurettei is the oldest and earliest diverging lophiodontid known to date but is unfortunately poorly known.

The oreodont clade is more primitive than the clade Perissodactyla in the LRT.

“We describe here the results of new excavations of the type locality of Palette. Important new material including complete skulls, mandibles, post-cranial elements and juvenile specimens lead us to revise Lophiaspis maurettei from Palette and other localities and to describe novel morphology for this species. According to an original phylogenetic analysis, based on a revised matrix of dental, cranio-mandibular and postcranial characters, Ls. maurettei is an early diverging lophiodontid morphologically close to Protomoropus and Paleomoropus, two basal chalicotheres, known from Asia and North America, respectively.

Since the authors omitted pertinent taxa and included irrelevant taxa, their analysis fails from the start.

“Our resulting topology does not support the previously proposed inclusion of the lophiodontids within the Ceratomorpha and supports a position within the suborder Ancylopoda, close to some Eomoropidae representatives. These results imply that Ls. maurettei was restricted to Southern Europe during the early Eocene, which would be compatible with an Asian origin for lophiodontids in accordance with the evolutionary history of other perissodactyls and placental mammals.”

The lesson here is the usual lesson: Don’t cherry pick your taxon list. Let a large gamut analysis tell you which taxa to test in your more focused studies.

Figure 2. Subset of the LRT focusing on artiodactyls. Lophiodon has been added basal to liptoterns and chalicotheres.

References
Bai B, Wang Y-Q and Meng J 2018. The divergence and dispersal of early perissodactyls as evidenced by early Eocene equids from Asia. Nature Communications Biology 1:115 | DOI: 10.1038/s42003-018-0116-5 | http://www.nature.com/commsbio
Depéret C 1910. Etudes sur la famille des Lophiodontidés. Bull. Soc. géol. Fr., (4)10: 558-577.
Vautrin Q et al. (11 co-authors) 2020. New remains of Lophiaspis maurettei (Mammalia, Perissodactyla) from the Early Eocene of France and the implications for the origin of the Lophiodontidae. Journal of Vertebrate Paleontology. 40(6): e1878200.

wiki/Agriochoerus
wiki/Lophiaspis

How hyrax ancestors evolved incisor tusks

According to Wikipedia
“The tusks of hyraxes develop from the incisor teeth as do the tusks of elephants.”

Genomic and phenomic workers agree
the rabbit-like hyrax (Procavia, Fig. 1) is related to the much larger elephants and manatees. The large reptile tree (LRT, 2046 taxa, subset Fig 5) concurs.

The question asked today:
which taxa were ancestral to hyraxes? And how did those tusks first appear?

Figure 1. Skull of Procavia from Digimorph.org and used with permission. Colors added here.

According to results recovered
in the LRT, subset Fig 5), extant hyrax (genus: Procavia, Fig 1) relatives include pony-like Diadiaphorus (Fig 2, 3) from the Miocene. Note the single digit on each limb contacting the substrate.

Figure 2. Diadiaphorus skull in 3 views. Colors added here.

Typically the canines are the longest, sharpest teeth in mammals.
That is not the case with elephants and their proximal ancestors. In the elephant clade lateral incisors are the longest, sharpest teeth. That homology also appears in hyraxes and Diadiaphorus (Figs 2, 3). We need to look to older taxa to find the evolutionary change from long canines to long incisors.

Figure 3. Skeleton of Diadiaphorus. Note the Mesohippus like feet as this taxon walked on only digit 3.

Hyrax and Diadiaphorus ancestors include
Paleocene Ectoconus cedrus (Fig 4 left) and Late Paleocene Ectoconus ralstonensis (Fig 4 right).

That longest tooth change occurs within the genus Ectocion.
Ectocion cedrus (Fig 4 left) has typical small incisors and long canines. In Ectocion ralstonensis (Fig 4 right) the canines are smaller and the lateral incisors are larger, becoming robust tusks. The LRT indicates these two Ectocion species should not be considered congeneric since they nest close to, but apart from one another in the LRT (Fig 5).

Figure 4. Ectocion cedrus (left) nests basal to the extant rock hyrax, Procavia and Miocene Diadiaphorus (Fig. 3), ultimately giving rise to elephants through Late Paleocene Ectocion ralstonensis (right, aka E osbornianus). Note the small lateral incisors vs canine on E cedrus. Note the larger incisiors and smaller canines on E ralstonensis.

The LRT heretically nests elephants within the clade Perissodactyla,
if one includes chalicotheres within Perissodactyla, which is traditional. In the LRT chalicotheres are more primitive than hyraxes and elephants. This clade is closer to elepahants + hyraxes than to the clade of tapirs, rhinos and horses. In the LRT these two now separated perissdactyl clades are listed here as Perissodactyla 1 and Perissodactyla 2 awaiting unique taxonomic names that have monophyletic memberships.

Figure 5. Subset of the LRT focusing on pantodont placentals, including Harpagolestes and Ocepeia.

According to Wikipedia
“Hyraxes share several unusual characteristics with mammalian orders Proboscidea (elephants and their extinct relatives) and Sirenia (manatees and dugongs), which have resulted in their all being placed in the taxon Paenungulata. Male hyraxes lack a scrotum and their testicles remain tucked up in their abdominal cavity next to the kidneys, as do those of elephants, manatees, and dugongs Female hyraxes have a pair of teats near their armpits (axilla), as well as four teats in their groin (inguinal area); elephants have a pair of teats near their axillae, and dugongs and manatees have a pair of teats, one located close to each of the front flippers.The tusks of hyraxes develop from the incisor teeth as do the tusks of elephants; most mammalian tusks develop from the canines. Hyraxes, like elephants, have flattened nails on the tips of their digits, rather than the curved, elongated claws usually seen on mammals.”

“Through the middle to late Eocene, many different species existed,^[23] the largest of them weighing the same as a small horse and the smallest the size of a mouse.”

There is no mention of Ectocion
in the Wikipedia description of the ancestry of hyraxes, this due to taxon exclusion. Unfortunately the author(s) of the Wikipedia online article on the hyrax support the invalid genomic clade, ‘Afrotheria‘.

Ectocion cedrus
(Cope 1882a, e, Thewissen 1990; UM 86155; Paleocene, 50mya, Fig 4) This genus is known from a fragmentary skull and several partial skeletons (that I have not seen published yet). The dental morphology is lophodont with a narrow rostrum. Contra prior authors, Ectocion is not closely related to Phenacodus (Fig 5).

Ectocion ralstonensis
(Cope 1882 a, e; Granger 1915; Thewissen 1990; Late Paleocene; 11 cm skull length) has longer incisor tusks, smaller canines.

This appears to be a novel hypothesis of interrelationships.
If not please provide a citation so I can promote it here.

References
Cope ED 1882a. Contributions to the history of the Vertebrata of the lower Eocene of Wyoming and New Mexico, made during 1881. Proceedings of the American Philosophical Society: 139-197.
Cope ED 1882e. Note on Eocene Mammalia. American Naturalist 16:522.
Granger W 1915. A revision of the lower Wasatch and Wind River faunas, Part III: Order Condylarthra, families Phenacodontidae and Meniscotheriidae. Bulletin of the American Museum of Natural History 34:329-361.
Thewissen JGM 1990. Evolution of Paleocene and Eocene phenacodontidae (Mammalia, Condylarthra). University of Michigan Papers on Paleontology 29:1-107.

wiki/Hyrax
wiki/Ectocion
wiki/Radinskya

Publicity, unfortunately no mention of the taxa listed above here.

https://www.nytimes.com/2018/09/11/science/tusks-teeth-elephants-genes.html

Goodbye, Epoicotheriidae: In the LRT Alocodontulum, Xenocranium and Epoicotherium are not related to one another

Updated July 15, 2022
with a new nesting for Alocodontulum and Metacheiromys with Vulpavus ovatus.

Every hear of the clade Epoicotheriidae?
I never heard of this clade until a few days ago. Then again, I could be the last to know…

According to Wikipedia,
“Epoicotheriidae (“strange beasts”) is an extinct family of pangolin-like insectivorous mammals which were endemic to North America from the early Eocene to the early Oligocene 55.8—30.9 Ma existing for approximately 24.9 million years.^[2] Epoicotheriids were highly specialized animals that were convergent with the golden moles of Africa in the structure of their skulls and forelimbs, and would have had a similar lifestyle as subterranean burrowers.”

Presently
only three traditional epoicotheres have been added to the large reptile tree (LRT, 2046 taxa, subset Fig 6). Most are known from far less complete remains.

Rose and Emry 1983 described all three
essentially as one type of burrower when they reported, “These epoicotheres, the latest surviving palaeanodonts, have numerous fossorial adaptations and must have been predominantly subterranean. Their skeletal specializations are similar to, and equal or surpass in degree of development, those of most living fossorial mammals.”

The authors followed with a long paragraph of shared traits,
then concluded, “These traits indicate that Xenocranium and Epoicotherium were among the most specialized “rapid-scratch” diggers ever to evolve. Like many extant subterranean mammals, Xenocranium and Epoicotherium were essentially sightless, but they were specialized for low frequency sound reception.”

Only one traditional epoicothere,
Alocodontullum (Fig 1), nests in the LRT alongside Metacheiromys with Vullpavus ovatus.

Figure 1. Metacheiromys and Alocontullum to the same scale about 1/3 actual size. Both nest close to living pangolins. Lower skeleton assembled from data in Rose, Emery and Gingerich 1992.

Alocodontullum atopum
Rose, Bown and Simons 1978; Rose, Emery and Gingerich 1992; early Eocene, 45 cm in length; UM 93740) is a close relative of Metacheiromys. It differs in relative metatarsal and toe lengths and having flat-cusped teeth posterior to long dowel-like (not sharp or curved) canines. Large claws and forelimbs provided with a large olecranon process (= elbow) are traits often found in diggers. The hands and feet are relatively larger. The mandible, and presumably the missing skull, is relatively smaller.

Figure 2. Subset of the LRT focusing on pangolins and their relatives, including Alocodontullum (Fig 1), whose manus and pes is shown at right.

A second traditional epoicothere,
the eponymous Epoicotherium (Fig 3), was first described by Simpson 1927 as close to Dasypus the nine-banded armadillo. After testing in the LRT (Fig 6) Epoicotherium does nest in the Xenarthra, but basal to the armored fairy armadillo, Chlamyphorus. That nesting is one step from the unarmored aardvark Orycteropus and Dasypus, which both have a longer rostrum.

Figure 3. Epcoitherium from Simpson 1927. Colors added here.

Epoicotherium unicum
Simpson 1927; Rose and Emry 1983; Eocene and Oligocene, 40mya) this taxon was considered a scansorial insectivore, a palaeonodon and a fossorial pangolin relative close to the edentate, Xenocranium. The LRT nests it firmly at the base of all armadillos.

A third traditional epoicothere,
tiny Xenocranium (Fig 4), also nests in the edentates, but with the much larger tree sloth, Bradypus.

Figure 4. Tiny Xenocranium. These two skulls are shown life size on a 72dpi monitor. A jugal is preserved in specimen no. 58 (at right).

Xenocranium pileorivale
(Colbert 1942; Emry and Rose 1983; Late Oligocene) was originally considered a burrowing edentate, then a pangolin. Here it tests as a tiny relative of an extant tree sloth. It is the smallest known sloth. Post-crania includes relatively large claws, perhaps for burrowing. The laterally expanded squamosals (magenta) are not similar to the posteriorly expanded occiput (= cranium) seen in Epoicotherium.

Figure 5. Bradupus (tree sloth skull) from Digimorph.org and used with permission. Colors added here. Compare to figure 4. This skull is taller than wide and much larger. Xenocranium was much smaller.

Summary:
Adding taxa (Fig. 6) does not support the monophyly of the the traditional clade Epoicotheriidae. Putative burrowing traits shared by these three taxa are convergent, and perhaps not all for burrowing. Sloths now have a wider gamut of sizes and niches.

References
Emry RJ and Rose KD 1983. Extraordinary Fossorial Adaptations in the Oligocene Palaeonodonts Epoicotherium and Xenocranium (Mammalia). Journal of Morphology 175(1):33–56.
Rose KD, Bown TM and Simons EL 1978. Alocodontulum, a new name for Alocodon Rose, Bown and Simons 1977, non Thulborn, 1973. Journal of Paleontology 52(5):1162.
Rose KD, Emery RJ and Gingerich PD 1992. Skeleton of Alocodontulum atopum, an early Eocene Epoicotheriid (Mammalia, Palaenodonta) from the Bighorn Basin, Wyoming. Contributions from the Museum of Paleontology The University of Michigan 28(10):221–245. Simpson GG 1927. A North American Oligocene edentate. Ann. Carnegie Mus. 17:283-98.

Alocodontulum – Fossilworks
wiki/Epoicotheriidae Simpson 1927 (placental)55.8–30.9 Ma
wiki/Epoicotherium (Simpson, 1927)
wiki/Xenocranium (Colbert, 1942)

Scottish Dearc is another Dorygnathus, and it’s not the largest Jurassic pterosaur

Dearc sgiathanach
(Jagielska et al. 2022, NMS G.2021.6 Middle Jurassic, Figs 1, 2) was originally considered the largest Jurassic rhamphorhynchine pterosaur. It is not. Late Jurassic Sericipterus (Fig 1) was larger.

Figure 1. Dearc shown to scale with the larger Jurassic pterosaur, Sericipterus.

With taxon exclusion
and their borrowed low-resolution, 30-pterosaur phylogenetic analysis, the eleven authors estimated the upper range of wingspan of gracile-winged Dearc (Fig 1) from unrelated long-winged Rhamphorhynchus.

Jagielska et al wrote,
“Its most remarkable attribute is its size: its wingspan was ca. 1.9–3.8 m, roughly the size of the largest flying birds today.”

1.9mx 2 = 3.8m. That is a rough estimate with a 100% range. Based on published scale bars and comparisons to related Dorygnathus taxa (Fig 2), Dearc had an est 1.5m wingspan.

Jagielska et al also wrote,
“Dearc is the first Jurassic pterosaur whose wingspan can confidently be estimated at ca. 2.5 m or greater, based on a well-preserved, articulated skeleton.5 Its closest relatives, Angustinaripterus and Sericipterus, are also sizeable for Jurassic species, with wingspans previously estimated at 1.61–1.738 m extrapolated from patchy fossils.”

The authors, referees and editors should have caught this internal discrepancy, especially so since wingspan is the core of their headline grab. Angustinaripterus is known from a skull only. Sericipterus is larger in every regard.

Jagielska et al also wrote,
“To estimate wingspan, we compiled measurements of complete wingspans of two non-monofenestratans represented by large sample sizes—Rhamphorhynchus and Dorygnathus—and regressed these against the lengths of individual bones to create predictor formulas. These results demonstrate that Dearc is the largest Jurassic pterosaur yet known, consistent with the fact that its humerus and skull are the longest of any Jurassic specimens.”

Perhaps documenting this claim with to-scale graphics (Fig. 1) would have helped. The first wing phalanx of Rhamphorhynchus extends back to the elbow when folded. By contrast, in Dorygnathus the first wing phalanx typically does not extend back to the half point of the ulna.

Figure 2. Click to enlarge. Dearc (bottom row, second from left) nests in the middle of several Dorygnathus specimens in the LPT. The variety in morphology in this clade is typical for pterosaur clades.

Dearc was also considered a new genus.
Here in the large pterosaur tree (LPT, 262 taxa) Dearc (Fig 3) nests in the middle of several Dorygnathus specimens. So Dearc is not a new genus, at least not until someone volunteers to split up Dorygnathus. Older workers tend to lump. Younger workers tend to split.

Taxon exclusion
remains the number one problem in paleontology. We can’t test just one Dorygnathus. No two are alike (Fig 2).

We also can’t borrow long wings from unrelated taxa
in order to produce a superlative-laden headline.

We also need to stop borrowing cladograms.
Jagielska et al. reported they borrowed data from eleven other cladograms, not all of them original studies. Apparently none of the borrowed cladograms included a wide gamut of Dorygnathus specimens.

Jagielska et al. reported,
“The Middle Jurassic age of Dearc adds to increasing evidence that this interval—once a frustrating gap in the pterosaur record—was in fact a dynamic time of diversification, in which a variety of basal taxa and early monofenestratan lineages.”

The Middle Jurassic has never been a frustrating gap for pterosaurs.
A wide variety of Dorygnathus (Fig. 2) taxa filled the Middle Jurassic of Europe. Omitting them gave Jagielska et al. perimission to borrow cladograms and wingspans.

Postscript
Dearc is the same pterosaur discussed a few days ago when the authors announced their paper had been accepted for publication and it was supposed to be embargoed until January 2023, according to information available at the Cell Press website. I guess the editors changed their mind or there was an editorial mistake.

Figure 3. ‘Dimorphodon’ weintraubi was another large pterosaur from the Middle Jurassic.

PPS
‘Dimorphodon’ weintraubi (Fig 3) was another large Middle Jurassic pterosaur.

Figure 4. Here the CZ specimen of Campylognathoides is shown to scale with Dearc.

PPPS
The Early Jurassic CZ specimen of Campylognathoides (Fig 4) was almost as tall as Dearc with a wider wingspan and more robust wing fingers.

Figure 5. Dearc compared to the two largest specimens of Rhamphorhynchus, which have longer and more robust wing phalanges and a more robust torso with larger sternal complex.

PPPPS
The authors compared Dearc (est 1.5m wingspan) to the largest Rhamphorhynchus (est 1.7m wingspan and slightly taller and more robust, Fig 5). The authors came up with some estimates for wingspan (see above), but published no side-by-side comparative graphics (Figs 1–5), which is traditional fare in superlative stories like this one.

We have to raise our standards
Referee David Hone wrote, “It’s clearly a non-monofenestratan pterosaur and actually one that is very close to Rhamphorhynchus, enough in fact to be found to be a member of the Rhamphorhynchinae in the phylogenetic analysis that they did.”

Dearc is not very close to Rhamphorhynchus (Fig 5).

Was Hone chosen to be a referee for his expertise in pterosaurs?
Few other workers have been responsible for so much misidentification and bungling of pterosaur anatomy and phylogeny. You’ll note that Hone did not have his own wide gamut phylogenetic analysis comparable to the LPT, but gladly accepted the Jagielska et al. 2022 analysis at face value. He also accepted all the bungles listed above. This is the state of pterosaur paleontology today. That’s why we get papers riddled with omissions.

References
Jagielska N et al. (ten co-authors) 2022. A skeleton from the Middle Jurassic of Scotland illuminates an earlier origin of lage pterosaurs. Current Biology 32:1–8.

wiki/Dorygnathus
wiki/Dearc

Publicity
BBC YouTube video:

Lovely fossil suffering only from hyperbole.

Early Paleocene Alcidedorbignya: how sharp claws evolved to flat hooves

Recently discovered and described
Alcidedorbignya (Muizon and Marshall 1992 Muizon et al 2015; MHNC 8372, Figs 1, 2) as “one of the oldest and most primitive of the pantodonts” (= edentate ancestors) and “the only pantodont genus known from South America” (excluding all edentate taxa [Fig 5] known from and still living in South America).

Alcidedorbignya is also much smaller
than ancestral and descendant taxa (Fig 1). That makes it a phylogenetically miniaturized taxon. As we’ve seen many times before, such taxa often develop novel structures and niches retained by their larger descendants.

Figure 1. Onychodectes, Alcidedorbignya and Pantolambda are Paleocene terrestrial placentals having recently dropped down from their safe havens in the trees following the K-Pg extinction event.

According to Muizon et al. 2015,
“In the discussion on possible scansorial habits of Alcidedorbignya, comparison with the arboreal Dendrohyrax [a tree hyrax] may appear surprising given the fact that the postcranial anatomy of this genus shows many traits suggesting a long cursorial ancestry (Fischer 1986). Nevertheless, we regard this comparison as highly relevant since tree hyraxes climb very well without claws (also absent in Alcidedorbignya) and without the great capacity of prehension of arboreal primates (also absent in Alcidedorbignya). Their ability to climb is related to the association of digital friction pads associated to unusual carpal mobility that allows wrist (instead of elbow) supination, two features that will be discussed below. Therefore, we regard Dendrohyrax as a good potential functional analogue of Alcidedorbignya on some aspects.”

Figure 2. From Muizon et al. 2015, Alcidedorbignya posed as an arboreal mammal. Their initial bias prevented them from seeing the headline topic here: This is the first terrestrial placental with anything like hooves.

Analogs can be good. Homologs are always better.
Several direct descendants of Early Paleocene Alcidedorbignya (e.g. Pantolambda, Barylambda, glyptodonts) were large, terrestrial herbivores with broad unguals on plantigrade feet.

Closely related basal phenacodonts,
like Phenacodus (Fig. 4), had digitigrade hands and feet with broad unguals starting evolve into hooves. Among those phenacodont descendants were hyraxes, like Procavia and Dendrohyrax. The cladograms presented by Muizon et al. 2015 suffer from taxon exclusion, so that team was unable to completely resolve phenacodonts from pantodonts and unable to resolve placentals from marsupials.

Let’s not forget
many edentates had long-clawed feet and hands, distinct from the broad, dull unguals of basal edentates. This return to an arboreal niche (e.g. in sloths) or to a digging niche (e.g. armadillos and anteaters) aided by trenchant claws is a reversal. That happens when the genes that initiate ungual tip spreading do not turn on. Taking advantage of this reversal, the pantodonts Ectoconus and Titanoides did not retain or develop wide, hoof-like additions to their relatively sharp unguals. When the hoof-gene was not well established in basal taxa, it was more easily lost.

Figure 3. Manus of Alcidedorbignya from Muizon et al. 2015. PILs added. Note the sharp claws evolving ‘snow shoe’ like hooves suggesting (not proviing) a more terrestrial existence. Direct descendants among edentates are all plantigrade, many with broad round unguals. Closely related phenacodonts (Fig. 4) are digitigrade and ultimately evolve hooves.

What PILs reveal.
Parallel interphalangeal lines (PILs) are applied here (Fig. 3) to the manus of Alcidedorbignya. Typically, on joints able to flex and extend in concert the PILs are continuous. Here, exceptionally, on digits 1 and 5 the PILs are not continuous, so flexion and extension are both restricted and prevented at those joints. Medial and lateral PIL sets appear to be essentially useless for extension and flexion revealing that the ability to grasp a branch inside a circle defined by the flexion of digits 1 and 5 coming into contact (as in primates including humans) is not present here.

Figure 4. Phenacodus in a digitigrade configuratiion. Digits 1 and 5 do not support weight. This is another larger close relative of Alcidedorbignya. Note the wide, round unguals.

This is how a digitigrade manus with hooves on digits 2–4 begins.
The need for digits 1 and 5 is eliminated except for using them to help stiffen the proximal phalanges, aiding the elevation of all five proximal phalanges above the substrate during quadrupedal locomotion. Elevating short digits 1 and 5 above the substrate often eliminates any need for them, so in descendant ungulates (= phenacodonts) there are only one, two, three or four working digits plus vestiges. Sometimes digits 1 or 5 become useful again (e.g. Elephas), representing yet another reversal.

Figure 5. Subset of the LRT focusing on terrestrial and herbivorous placentals (= pantodonts + phenacodonts). Alcidedorbignya is nearly a basalmost taxon here.

Alcidedorbignya inopinata
(Muizon and Marshall 1992 Muizon et al 2015; MHNC 8372, Early Paleocene, 64 mya; scale bar is 2cm) was originally and is here considered a basal pantodont close to Pantolambda. The rostrum is convex with a smaller naris. Claw tips develop wide, snowshoe-like ‘hooves’ here.

Ectoconus ditrigonus – E majusculus
(Cope 1884 Early Paleocene, 66-63mya, sheep-sized) is a basal pantodont known from a rare complete skeleton. Pantolambda is a sister. Onychodectes is basal. Though canines were present, Ectoconus was an herbivore with five digits on all four limbs. Often shown as a digitigrade, this taxon was likely plantigrade based on phylogenetic bracketing.

Pantolambda bathmodon
(Cope 1882, Middle Paleocene, 66-63 mya, sheep-sized; AMNH 3956) is another basal condylarth, and a coeval sister to Ectoconus with a shorter skull, shorter lumbar region and shorter limbs. Pantolambda is a pantodont.

References
Cope ED 1882. On the condylarthra. Academy of Natural Sciences of Philadelphia Proceedings 1882:95-97.
Cope ED 1882. Synopsis of the Vertebrata of the Puerco Eocene epoch. Proceedings of the American Philosophical Society 20
Cope ED 1884. The Amblypoda. The American Naturalist 18 (112):6=461-471.
Simons EL 1960. The Paleocene Pantodonta. Transactions of the American Philosophical Society, New Series 50(6):1-8.
de Muizon C de and Marshall LG 1992. Alcidedorbignya inopinata (Mammalia: Pantodonta) from the early Paleocene of Bolivia: phylogenetic and paleobiogeographic implications. Journal of Paleontology 66 (3): 499-520.
deMuizon C, Billet G, Argot C, Ladeveze S and Goussard F 2015. Alcidedorbignya inopinata, a basal pantodont (Placentalia, Mammalia) from the early Palaeocene
of Bolivia: anatomy, phylogeny and palaeobiology. Geodiversitas 87(4):397-634.

wiki/Alcidedorbignya
wiki/Ectoconus
wiki/Pantolambda

What is Dearc sgiathanach? How do you pronounce Dearc sgiathanach? And have you ever seen so much publicity over an ‘accepted paper’ under embargo?

Surprisingly
a day or two later this ‘accepted paper’ was published. Details here.

Spoiler alert: Dearc is not a new genus (it nests within Dorygnathosaurus) and it is not the largest pterosaur from the Jurassic.

Quoting various publicity sources:
One: “Dearc is the biggest pterosaur we know from the Jurassic period, and that tells us that pterosaurs got larger much earlier than we thought, long before the Cretaceous period when they were competing with birds, and that’s hugely significant. With a wingspan of more than 2.5 meters (8.2 feet), it’s the biggest pterosaur ever discovered from the Jurassic period and last flapped its wings 170 million years ago. Its sharp teeth, which would have snapped up fish, still retain their shiny enamel.”

Two: “Dearc sgiathanach (pronounced jark ski-an-ach), which translates to ‘winged reptile.'”

Three: Quoting Cell Press — “Jagielska N et al. (ten co-authors) 2022. A skeleton from the Middle Jurassic of Scotland illuminates an earlier origin of large pterosaurs. Current Biology online Accepted author manuscript, 01/26/22 Embargo ends: 01/26/23.“

Unfortunately, there is little else that can be said (respecting the embargo)
of Dearc (and two other coeval Scottish pterosaurs) that hasn’t already been previewed as an SVP abstract in 2020, another SVP abstract in 2019, and a raft of superlative-laden online publicity breaking around the world today (examples below) that “leaves scientists gobsmacked.”

If you have never been gobsmacked, by all means, give it a try!

Figure 1. A different Skye pterosaur from traced from in situ specimens found online back in 2019. Has this specimen been given a name yet? Or a publication date? If so, please advise.

The little else that can be said
is the several (three maybe four at last count) pterosaur specimens coming out of Skye, Scotland all appear to be basal wukongopterids (Fig. 1), a clade often claimed to be ancestral to pterodactyloid-grade pterosaurs, but only when the actual phylogenetically miniaturized ancestors of pterodactyloid-grade pterosaurs are omitted from a wider gamut analysis (Fig. 2). With present knowledge the wukongopterid (= monofenestratan) clade produced no Cretaceous descendants.

Figure 2. Subset of the LPT showing the nesting of the first Skye pterosaur from available data in 2019, still not published officially.

Oddly, this CNN publicity
throws more superlatives at Quetzalcoatlus than at Dearc:
cnn.com/2022/02/22/europe/flying-reptile-pterosaur-fossil-skye-scotland

The Daily Mail, more superlatives
https://www.dailymail.co.uk/sciencetech/article-10537923/Worlds-largest-Jurassic-pterosaur-unearthed-Isle-Skye.html

The National, a Scottish online news source… ‘superlative’ in the headline
thenational.scot/news/19942299.superlative-pterosaur-fossil-found-isle-skye-named-dearc-sgiathanach/

WHTC.com: gobsmacked
whtc.com/2022/02/22/scottish-fossil-of-flying-reptile-leaves-scientists-gobsmacked/

Skye (Scotland) STV News… ‘world’s largest’
Jurassic pterodactyl fossil unearthed: video

Other publicity out today
google.com/search?q=Dearc+sgiathanach

Congratulations
to the authors for the publicity that surrounds the acceptance of your paper! … and as an aside, what does ’embargo’ mean in Scotland? Here it usually means, ‘keep it under wraps’, or ‘an official prohibition on any activity.’

References
Jagielska N et al. (ten co-authors) 2022. A skeleton from the Middle Jurassic of Scotland illuminates an earlier origin of large pterosaurs. Current Biology online Accepted author manuscript, 01/26/22 Embargo ends: 01/26/23.
Jagielska N and Brusatte SL 2021. Primer Pterosaurs. Cell Press Current Biology Magazine. R984 Current Biology 31, R973–R992, August 23, 2021 link here.
Jagielska N et al. (9 co-authors) 2020. An exceptionally well preserved pterosaur from the Middle Jurassic of Scotland. SVP abstracts 2020.
Martin-Silverstone E, Unwin DM and Barrett PM  2019. A new, three-dimensionally preserved monofenestratan pterosaur form the Middle Jurassic of Scotland and the complex evolutionary history of the scapulo-vertebrael articulation. Journal of Vertebrate Paleontology abstracts.
Martin-Silverstone et al. (five co-authors) 2022. A new pterosaur from Skye, Scotland and the early diversification of flying reptiles. BioRxiv preprint online here.

pterosaurheresies.wordpress.com/2021/08/24/the-usual-pterosaur-myths-repeated-by-jagielska-and-brusatte-2021/

pterosaurheresies.wordpress.com/2020/10/24/svp-abstracts-10-scottish-middle-jurassic-pterosaur-back-again-this-year/

Tiny Ocepeia: now a just weaned baby mesonychid close to Harpagolestes in the LRT

Gheerbrant et al. 2014 presented a small Paleocene mammal skull
(Figs 1, 2) they described as “the oldest Afrotherian. Within Afrotheria Ocepeia is reconstructed as more closely related to insectivore-like afroinsectiphilians (i.e., aardvarks, sengis, tenrecs, and golden moles) than to paenungulates. The remarkable character mosaic of Ocepeia makes it the first known ‘‘transitional fossil’’ between insectivore-like and ungulate-like afrotherians.”

Here
in the large reptile tree (LRT, 2046 taxa, subset Fig 3) little Ocepeia nests with the much larger, hippo-like mesonychid, Harpagolestes (Fig 1), not near the transition between insectivores and ungulates (Fig 3). There is no mention of ‘Harpagolestes‘ in the Gheerbrant et al 2014 text. And there is no such clade as ‘Afrotheria‘ which is based on untrustworthy genomic, rather than trustworthy phenomic data. Here, once again, taxon exclusion seems to be the problem. Readers will note the skull of Ocepeia does not resemble those of aardvarks, sengis, etc., but is a good, if immature, match for large mesonychids.

Gheerbrant et al. reported,
“The preserved teeth (right P3–4, M1–2) are unworn and it is possible that M3 was still in crypt or just erupting in this specimen, indicating a young individual. However, the sutures are strongly fused and in many cases hardly distinct; together with the low sagittal crest, it might suggest a young adult female.”

“The tegmen tympani is remarkably inflated and pneumatized; it forms a large and robust barrel-like bony structure. The hyperinflated tegmen tympani is known in Meniscotherium, Mesonyx, anthracotheriids, hippopotamids and cetaceans.”

That matches LRT results. This sentence is the only mention of mesonychids and hippos.

The tegmen tympani is a thin plate of the petrous part of the temporal bone that separates the intracranial compartment and middle ear.

Figure 1. Two Harpagolestes skulls to scale with the skull of Ocepeia (see figure 2 for details).

Compared to the adult skull of Harpagolestes,
the Ocepeia skull (Figs. 1, 2) lacks a large sagittal crest, a large set of procumbent canines, and worn molars. Otherwise, the LRT nests the two together and apart from 2044 competing canditate tax.

Figure 2. Reconstructed skull of Ocepeia based on µCT scans presented in Gheerbrant et al. 2014. Colors added here. Compare to Harpagolestes in figure 1.

Based on the size difference and phylogenetic nesting
(Fig 1) Ocepeia was a baby of a much larger mesonychid close to Harpagolestes. That’s the novel hypothesis proposed here. Let us know if someone else proposed this since 2014. This hypothesis, like all hypotheses, is subject to change as new data arrives.

Gheerbrant et al. reported,
“Autapomorphies include striking anthropoid-like characters of the rostrum and dentition. Besides having a basically eutherian-like skull construction, Ocepeia daouiensis is characterized by ungulate-like, and especially paenungulate-like characters of skull and dentition (e.g., selenodonty).”

Anthropoids are apes. Harpagolestes and Ocepeia are a few nodes apart from the base of the Artiodactyla, whch includes ungulates and paenungulates.

Figure 3. Subset of the LRT focusing on pantodont placentals, including Harpagolestes and Ocepeia.

Harpagolestes uintensis
(Type H. macrocephalus Wortman 1901; Scott 1888, Szalay and Gould 1966; AMNH 1892, 1878; Eocene, 51 mya; 50 cm skull length) is widely considered to be a large, hyaena-like mesonychid, like Mesonyx, but with a deeper and wider skull. Here Harpagolestes is derived from a sister to Mesonyx and basal to Hippopotamus, all plant eaters. The teeth are often heavily to extremely worn, as is typical of herbivores. A diastema just develops around the anterior premolars. The jugal has the genesis of a postorbital bar to meet that of the frontal, which is complete in Hippopotamus. Stout limb bones and heavily worn teethindicate Harpagolestes was not able to pursure prey.

Ocepeia daouiensis
(Gheerbrant et al 2001, 2014; Paleocene, 60 mya; 9 cm skull length) is a small Hippopotamus mimic from Africa. It is so small it may be a baby Harpagolestes. The pneumatized skull contains many air spaces. Slightly larger skulls have larger canines and so are considered male, but may just be older. This specimen has only a tiny canine. The molar teeth are very large and the jugal deepens below the orbit, hiding the posterior molars in lateral view. The premaxilla is transverse. The upper canine rubs against the lower large incisior creating a facet, as in hippos and mesonychids. Note the dorsal eyes and deep jugals. Ocepeia was found with aquatic taxa and was probably amphibious. The name Ocepeia derives from the initials of Office Chérifien des Phosphates (O.C.P.), the national Moroccan phosphate mining company.

References
Gheerbrant E, Amaghzaz M, Bouya B and Goussard F and Letenneur C 2014. Ocepeia (Middle Paleocene of Morocco): The Oldest Skull of an Afrotherian Mammal. PLoS ONE. 9 (2): e89739. doi:10.1371/journal.pone.0089739.
Szalay FS and Gould SJ 1966. Asiatic mesonychidae (Mammalia, Condylarthra). Bulletin of the American Museum of Natural History 132(2):127–174.

wiki/Mesonyx
wiki/Harpagolestes
wiki/Ocepeia

Pantodonts and phenacodonts are extant: The second half of placental evolution

Yesterday
we looked at the first half of placental evolution: tree shrews and their highly evolved descendants. Today no more insectivore tree-huggers, way more hooves, tusks and horns.

Wikipedia reports,
“Pantodonta is an extinct suborder (or, according to some, an order) of eutherian mammals.”

Not extinct.
When you add taxa, as in the large reptile tree (LRT, 2044 taxa, subset Fig. 1) most traditional members of the Pantodonta (= Barylambdidae, Titanoideidae and Alcidedorbignya) are basal to Xenarthra (= sloths, aardvarks, anteaters, etc.). Since edentates are not extinct, neither are the monophyletic pantodonts.

Another traditional pantodont clade,
Coryphodontoidea (= Coryphodon and kin), nest apart from the other pantodonts in the LRT. Coryphodon descendants have hooves, tusks and horns… and are likewise not extinct. Monophyletic clades never lose their original identity (e.g. birds are dinosaurs, humans are mammals, etc).

Figure 1. The second half of placental evolution, the terrestrial herbivores and their hooved, tusked and horned descendants. Plus baleen whales, plus manatees, the two unrelated marine clades that lose their hind limbs and grow flukes. A few Easter eggs are in here. We’ll visit them soon.

But wait, there’s more.
Wikipedia reports, “Xenarthrans were previously classified alongside the pangolins and aardvarks in the order Edentata (meaning toothless, because the members do not have incisors and lack, or have poorly developed, molars). Subsequently, Edentata was found to be a polyphyletic grouping whose New World and Old World taxa are unrelated, and it was split up to reflect their true phylogeny.”

That ‘true’ phylogeny was determined by gene studies.
Unfortunately we’ve seen time and again phylogenetic problems that arise from deep time gene studies. These appear to have been affected by endemic viruses restricted by geography (e.g. Afrotheria, Laurasiatheria, etc.)

By contrast,
in the LRT aardvarks, like Orycteropus, nest with traditional edentates. Pangolins nest between colugos and bats in the LRT, as shown yesterday.

Figure 2. A lineage of phenacodonts from Phenacodus to the gray whale, Eschrichtius. So phenacodonts are not extinct. Neither are mesonychids. Neither are desmostylians. And hippos are not artiodactyls.

Phenacodonts are also not extinct
Wikipedia reports, “Phenacodontidae is an extinct family of large herbivorous mammals traditionally placed in the “wastebasket taxon” Condylarthra, which may instead represent early-stage perissodactyls.”

By contrast,
in the LRT (Fig. 2) Phenacodus is basal to a smaller clade that includes extinct Astrapotherium. Phenacodus is also basal to a larger clade that includes extant placental herbivores that give birth to prococial young. Ectocion is a traditional phenacodont, but in the LRT two species nest apart from Phenacodus, closer to to perissodactyls.

Digitigrady
At this point one subtle difference between pantodonts and phenacodonts is digitigrady in the latter (Fig. 1) with evolution to walking only on the unguals in ungulates and transforming the manus into flippers in baleen whales and manatees. [This was added February 21, 2022]

Monophyletic artiodactyls now include perissodactyls
and paenungulates (= elephants, sirenians and hyracoids). Homalodotherium is the last common ancestor.

Eleven years ago
Asher and Helgen 2010 worked on nomenclature within placental mammal phylogeny. Unfortunately, Asher and Helgen didn’t create their own cladogram, nor did they borrow others. Instead they accepted genomic results and applied the principles of priority.

Asher and Helgen 2010 wrote,
“Although not all clades are fully resolved, and our understanding of many extinct radiations remains poor, several previously intractable issues surrounding the living radiations have now been settled.”

Eleven years later
the LRT completely resolved the placentals and new light has been shed on our understanding of many extinct radiations (Fig. 1).

References
Asher RJ and Helgen KM 2010. “Nomenclature and placental mammal phylogeny”. BMC Evolutionary Biology10(1):102.

Tree shrews take off: The first half of placental evolution

Distinct from current genomic cladograms
of the clade Placentalia (Fig. 3), the large reptile tree (LRT, 2044 taxa, subset Fig. 1) test traits, in a wide gamut of genus-based taxa (including fossils). The LRT documents a gradual accumulation of derived traits in full resolution leaving no large morphological gaps.

Here basal placentals are similar to
and include the lemur-sized and shaped coatimundi, Nasua, Fig. 2). Here basal placentals are also similar to marsupial outgroups, one of which, Monodelphis, also lacks a pouch, as in placentals. This hypothesis of interrelationships appears reasonable and keeps basal Mesozoic placentals hiding, feeding and mating up in the trees, away from terrestrial predatory dinosaurs.

Figure 1. The first half of placental evolution, a subset of the LRT. If you’re curious about any of these taxa, you can find their data at ReptileEvolution.com. Part 2 appears tomorrow. There are a few Easter eggs in here that we’ll look at in the near future.

By contrast,
genomic analyses (Fig. 3) nest toothless armadillos and aardvarks at the base of the Placentalia, neither of which resemble outgroup arboreal marsupials like Monodelphis. This appears unreasonable, yet is widely accepted and often published in paleontology.

Figure 2. The coatimundi (Nasua) compared to the ring-tailed lemur (Lemur). Both are basal placentals. Both are bauplans ffom which derived placentals eventually evolved, both larger and smaller forms.

Mess 2014 reflects current thinking
in placental evolution using genomic analysis (Fig. 3). They reported, “These are the novel clade Afrotheria (elephants, manatees, hyraxes, aardvark, elephant shrews, tenrecs) that originated on the African continent, Xenarthra (sloths, anteaters, armadillos) that had their main distribution area in South America, Laurasiatheria (carnivores, Cetartiodactyla, horses, pangolins, bats, Eulipotyphla) that have mainly evolved on the Northern continents, and finally Euarchontoglires (rodents, lagomorphs, primates, tree shrews, flying lemurs) that evolved more independently from the continent’s history.”

Endemic (= local, continental) viruses affect genes.
That’s why deep time genomic results don’t match deep time phenomic results (Fig. 1) often enough to be trusted or replicated using traits. As a bonus: when you use traits, you can use fossils omitted from genomic taxon lists.

Figure 3. Placental genomic cladogram borrowed by Mess 2014 dividing placentals into geographic zones. This cladogram nests bats (Chiroptera) with rhinos (Perissodactyla), aardvarks (Tubilidentata) with hyraxes (Hyracoidea) and armadillos (Xenarthra) at the origin of Placentalia. These do not match the LRT. The clade Euarchontoglires matches the LRT. So does the Hyracoidea + Proboscidea + Sirenia. So does the Perissodactyla + Artiodactyla.

Unlike all other phylogenetic analyses
of any sort, the LRT employs a wide gamut of taxa back to Cambrian worms and basal fish. The LRT is fully resolved and documents a gradual accumulation of derived traits at every node. In that way the LRT provides a hypothesis of interrelationships that model evolutionary events in all chordates, including placentals. There are no large morphological gaps in the LRT and no enigma taxa. That’s what we’re looking for. The LRT separates seals from sea lions, odontocetes from mysticetes and nests multituberculates and aye-ayes with rodents. Other studies more or less fail due to the number one problem in paleontology: taxon exclusion.

This appears to be a novel hypothesis of interrelationships.
If not, please send a citation so I can promote it here.

Tomorrow:
part 2, the seoond half of the placentals. They also start in the tree, but drop to the ground immediately following the K-Pg extinction event.

References
Mess A 2014. Placental Evolution within the Supraordinal Clades of Eutheria with the Perspective of Alternative Animal Models for Human Placentation. Advabces in Biology 2014: Article ID 639274 https://doi.org/10.1155/2014/639274

New data for the old skull of Mesonyx

…and, not surprisingly, the teeth are worn flat,
like those of an old hippo (Fig 1). Ever since Mesonyx entered the large reptile tree (LRT, 2044 taxa) it nested close to the genus, Hippopotamus, an extant herbivore, in a large clade of derived large placental herbivores that give birth to precocial young — rather than the more primitive clade of smaller placental omnivores, insectivores and carnivores with altricial (= helpless, underdeveloped) young to which we belong.

Figure 1. Engravings of Mesonyx. Colors and animation added here. Note the worn teeth, typical of an older (more mature or close to death) herbivorous mammal.

Wikipedia reports,
“They were probably active hunters.”

That’s a traditional myth,
according to phylogenetic bracketing and those worn down molars in figure 1.

Five years ago
the good folks at Paleofail.tumblr.com were a little upset that the LRT nested Mesonyx with Hippopotamus, suggesting (via phylogenetic bracketing) that Mesonyx, “was a plant eater.” If you go to that PaleoFails site you’ll see the data used for Mesonyx in 2016: a skull with unworn multicusp molars (or see Fig. 2 below). Unfortunately that was the only data that had crossed my path until now. Finding more detailed data sporting worn teeth (Fig. 1) further cements the hypothesis of mesonychid herbivory.

PalaeoFail.tumbler.com presents
“The best in failing miserably at palaeontology.”

In 2016 PaleoFails reported mesonychid herbivory
was a miserable fail at paleontology.

Unfortunately, that ‘miserable fail’
was widely ‘liked’ by dozens of pseudonymous readers at Paleofails. And that’s okay. They were holding tight to their traditions in the face of a novel heresy. As Yale professor John Ostrom lamented, it takes time and rafts of evidence (Fig. 1) to change minds in paleontology.

Some quotes from Paleofail readers follow:
synapsidgirl reblogged this from jurassicsunsets and added:
“Somebody. Please. Stop. Him. I mean, seriously, how can anyone look at that dentition and go “That there is a herbivore…”

As mentioned above, phylogenetic bracketing with herbivores infers an herbivorous diet in fossil taxa. It’s a standard method in science. Synapsidgirl joins PhD Darren Naish, amateur paleontologist Mickey Mortimer and videographer, Dino Diego, in their urge to cancel novel ideas coming out of here.

jurassicsunsets reblogged this from palaeofail and added:
“Mesonyx was a mammal, not a reptile. Its taxonomic position is a bit unclear, but it probably wasn’t sister to hippos,…“

Thanks for clearing that up, Jurassicsunsets.

justgilbertianthings reblogged this from palaeofail and added:
“I have no words for this“

Someone, please hurry! Offer this pseudonym a hankie, a lawn chair and a lemonade.
: – )

For those of you who seriously want to become a paleontologist,
achieving infamy and a dose of vilification from pseudonyms and PhDs is (regrettably) going to part of the process when you discover something that upsets the established sense of order. Just be ready for it.

If you have a beef with someone’s hypothesis,
don’t come to battle with your emotions, a smidgen of outrage and an urge to cancel. Grow up. Wipe your tears. Be professional. Come to battle with your own cladogram, fresh data and an urge to help your colleague get back on the right track. Phylogenetic analysis works. So does phylogenetic bracketing, both for professional and all other scientists.

Figure 2. The graphic and caption that earned a place of dishonor at PaleoFails: “Mesonyx, the first known mesonychid was a sister to Hippopotamus in the large reptile tree. So maybe it was a plant eater.”

Mesonyx obtusidens
(Cope 1872, Eocene, 51 mya, 1.5m long, YPM VP 013066) Widely considered a carnivore, Mesonyx had the same large, dull, canine fangs also found in Hippopotamus, an extant herbivore. Mesonychids have been linked to whales, but toothed whales like Maiacetus are desendants of tenrecs. Baleen whales are descendants of oreodonts, Mesonyx. hippos, anthracobunids and desmostylians in that order.

References
Cope ED 1872. Descriptions of some new Vertebrata from the Bridger Group of the Eocene. Proceedings of the American Philosophical Society 12:460-465.

wiki/Mesonyx