Asiatherium enters the LRT: mammal nomenclature issues follow

Everyone agrees
that Asiatherium (Figs, 1,2) nests close to Monodelphis, Caluromys and placentals. Trofimov and Szalay 1994 agreed. So did Denyer, Regnault and Hutchinson 2020. So did the large reptile tree (LRT, 1729+ taxa, subset Fig. 3).

Figure 1. Asiatherium in situ from Szalay and Trofimov 1996.

Asiatherium reshetovi (Trofimov and Szalay 1994, Szalay and Trofimov 1996; PIN 3907; Late Cretaceous; 80mya; Figs. 1, 2) is a key Mongolian metathere ancestral to monodelphids and Caluromys, which is ancestral to placentals. It is derived from Triassic sisters to extant late survivors, Didelphis, Gilronia and Marmosops.

Figure 2. Asiatherium skull slightly modified (longer lateral view premaxilla to match dorsal and ventral views) from Szalay and Trofimov 1996. Colors added here.

The problem is,
according to results recovered by the LRT, mammal clade nomenclature needs to go back to basics. Several modern mammalian clade names are found to be junior synonyms of traditional clades in the LRT.

Prototheria (Gill 1872) is a junior synonym
for Monotremata (Bonaparte 1837) in the LRT.

According to Wikipedia, “Prototheria is a paraphyletic subclass to which the orders Monotremata, Morganucodonta, Docodonta, Triconodonta and Multituberculata have been assigned, although the validity of the subclass has been questioned.”

In the LRT Morganucodon is a a marsupial (see below). Docodon is a taxon within Monotremata. Triconodon is a taxon within Monotremata. Multituberculata is a clade within the placental clade Glires (Fig. 4). So, the clade Monotremata is monophyletic and has precedence.

Theria (Parker and Haswell 1897) is a junior synonym
of Marsupialia (Illiger 1811). Metatatheria (Thomas Henry Huxley 1880) is also a junior synonym of Marsupialia.

The late-surviving basalmost marsupial in the LRT (Fig. 4), Ukhaatherium (Fig. 3), has epipubic (marsupial) bones. That long rostrum indicates this taxon is close to monotremes.

Figure 3. Ukhaatherium in situ.

Unlike the monophyletic clade Monotremata,
a series of nested marsupial clades are present. The last of these gives rise to Placentalia, only one of several that lose the pouch (Fig. 4). New names are proposed here where appropriate:

1. Marsupialia = Ukhaatherium and kin + all descendants (including placentals)
2. Paleometatheria = Morganucodon and kin + all descendants.
3. Didelphimetatheria = Eomaia and kin + all descendants
4. Phytometatheria = Marmosops and kin + all descendants
5. Carnimetatheria = Asiatherium and kin + all descendants
6. Transmetatheria = Caluromys and kin + all descendants
7. Placentalia = Vulpavus and kin + all descendants

Figure 4. Subset of the LRT cladogram of basal Mammalia. Note the new names proposed here.

Basal marsupial taxa are omnivores. 
Derived phytometatheres are herbivores. Derived carnimetatheres are carnivores to hyper-carnivores. Transmetatheres (Carluromys) and basal Placentalia remain omnivores.

In the LRT Eutheria (Gill 1872) is a junior synonym
of Placentalia (Owen 1837). Omnivorous civets like Nandinia are basal placentals. Carnivora is a basal placental clade following basal placental civets.

Competing cladograms
Denyer, Regnault and Hutchinson 2020 recently looked at the marsupial patella, or more specifically the widespread absence or reduction of the kneecap. The authors concluded, “metatherians independently ossified their patellae at least three times in their evolution.”

Unfortunately, Denyer et al. tested Caenolestes, the ‘shrew opossum’. Not surprisingly it nested close to placentals in their cladogram. Caenolestes was earlier nested in the LRT within the placental clade, Glires, closer to shrews than to opossums. It has no pouch, but converges with marsupials in several aspects. Inappropriate taxon inclusion, like Caenolestes, occurs due to taxon exclusion. Excluded taxa would have attracted and removed the inappropriate taxon. Taxon exclusion plagues Denyer et al.

Historically, you may remember,
Bi et al. 2018, while presenting Early Cretaceous Ambolestes, suffered from massive taxon exclusion and traditional bias in attempting to produce a cladogram of mammals. Bi et al. recovered Sinodelphys (Early Cretaceous) and Juramaia (Late Jurassic) as ‘eutherians’. In the LRT both are monotremes.

Other basal mammal cladograms
depend too much on tooth traits. Convergence in tooth traits creates problems, as documented earlier. We’ll look at this problem in more detail soon.

The above subset of the LRT appears to be a novel hypothesis
of interrelationships. If not, please provide a citation so I can promote it.

---

References
Bi S, Zheng X, Wang X, Cignetti NE, Yang S, Wible JR. 2018. An Early Cretaceous eutherian and the placental marsupial dichotomy. Nature 558(7710):390395 DOI 10.1038/s41586-018-0210-3.
Denyer AL, Regnault S and Hutchinson JR 2020. Evolution of the patella and patelloid in marsupial mammals. PeerJ 8:e9760 http://doi.org/10.7717/peerj.9760
Szalay FS and Trofimov BA 1996. The Mongolian Late Cretaceous Asiatherium, and the early phylogeny and paleogeography of Metatheria. Journal of Vertebrate Paleontology 16(3):474–509.
Trofimov BA and Szalay FS 1994. New Cretaceous marsupial from Mongolia and the early radiation of Metatheria. Proceedings of the National Academy of Sciences 91:12569-12573

New study on Thylacosmilus atrox: “not a marsupial saber-tooth predator”??

Janis et al. 2020
bring us some heretical views regarding Thylacosmilus, the famous saber-toothed marsupial (Fig. 1).

Figure 1. Thylacosmilus skull. Note the deep maxillae in dorsal contact containing giant canine roots. These are not present in Patagosmilus.

For those in a hurry:
The Janis et al. study includes a phylogenetic analysis of placental sabertooth cats that nests the saber-toothed marsupial, Thylacosmilus (Fig. 1), at the base of the clade (Fig. 2). In a way, that is true, but this is missing so many transitional taxa that we’re left with apples and oranges. So, that’s not going to work because Janis et al. are testing analogy and convergence, rather than homology. It’s better to test apples and apples, even when dealing with stress tests, etc.

Figure 2. Cladogram from Janis et al. 2020. Note the lack of marsupial taxa, other than Thylacosmilus at the base. Be wary whenever the taxon under review nests at the base of the cladogram.

Lacking here
is a phylogenetic analysis that includes the closest marsupial relatives of Thylacosmilus: 1. Schowalteria (Fig. 3), 2. Vincelestes + Conorytes, 3. Huerfanodon 4. Monodelphis + Chironectes. That’s how they line up in the large reptile tree (LRT, 1698+ taxa). You need related taxa to decipher the phylogenetic bracketing of Thylacosmilus based on homology, not analogy. The last two taxa are extant. One is an omnivore, the other an aquatic carnivore. Among the extinct taxa, Schowalteria, Vincelestes, Conoryctes and Huefanodon all appear to have been marsupial saber-toothed predators, contra the Janis et al. headline. Vincelestes goes back to the Early Cretaceous.

Figure 3. Showalteria. Not much there, but enough to nest it with Thylacosmilus. Is this a predator? According to Wikipeia,.. no.

Janis et all. 2020 wrote:
“While the superficial appearance of Thylacosmilus atrox resembles that of placental saber- tooths, its detailed anatomy makes this animal an ecomorphological puzzle, and the analyses performed here show it to be unlike other carnivores, saber-toothed or otherwise”

That’s because they omitted related taxa…where are the comparable marsupials?

“While we can demonstrate that T. atrox could not have been a predator in the mode proposed for the saber-toothed feliform carnivorans, it is challenging to propose an alternative mode of life.”

That’s because they omitted related taxa…where are the comparable marsupials?

“We note that, while there is often the temptation to shoehorn an extinct animal into the ecomorphological role of an extant one (see Figueirido, Martín-Serra & Janis, 2016)—or even, as in this case, the proposed ecomorphological role another extinct animal—T. atrox may well have had no analogs in the extant or extinct fauna.

That’s because they omitted related taxa…where are the comparable marsupials?

“We extend this discussion of extinct animals without living analogs in the conclusions. Here we present some ecomorphological hypotheses for T. atrox that align with the peculiarities of its anatomy.”

1. We note that the unusual subtriangular shape of these canines makes them appear more like a claw than a blade; and, like a claw, they appear well-adapted for pulling back.
2. Our biomechanical study shows that both the skull and the canines of T. atrox are better in resisting pull back stresses than those of S. fatalis.
3. The small infraorbital foramen of T. atrox supports the hypothesis that its canines were not used for killing prey, as it would not require such careful and precise positioning of the canines.
4. The postcanine teeth of T. atrox exhibit blunted tip wear, unlike the shearing wear on the teeth of carnivores that specialize on flesh
5. T. atrox was clearly not a bone-crusher: this type of diet is contraindicated by the DMTA analysis and the lack of cranial specializations (including evidence for powerful jaw adductors) seen in extant bone-crushers
6. T. atrox had less powerful jaw adductor and head depressor muscles than placental saber-tooths. The cervical and caudal cranial anatomy are not indicative of the ability for extreme head elevation and forceful head depression, as observed in the anatomy of placental saber-tooths, implicated in those carnivores for a predatory head strike.
7. The virtual absence of incisors (certainly the absence of a stout incisor battery) in T. atrox is challenging for the hypothesis of a cat-like mode of feeding, as it would have been unable to strip flesh from a carcass or transport its prey.

Janis et al. wonder:
“Could [soft internal organs] have been the preferred diet of the marsupial saber-tooth?”

Janis et al. propose:
“Incisor loss or reduction in mammals is correlated with the use of a protrusible tongue in feeding, as seen in myrmecophageous mammals.”

Janis et al. propose:
“T. atrox has been ‘‘shoehorned’’ (see Figueirido, Martín-Serra & Janis, 2016) into the saber-tooth ecological role: much less attention has been paid to the way in which this animal differs in its morphology from placental saber-toothed predators, making a similar type of predatory behavior unlikely.

“We advance the suggestion that it was not an active predator, but rather relied on the use of existing carcasses, deploying its large canines for carcass opening rather than for killing, a hypothesis supported by our biomechanical analyses that show superior performance in a ‘‘pull-back’’ scenario.”

“Thylcosmilus atrox was a very different type of carnivorous mammal than the placental saber-tooths: the oft-cited convergence with placentals such as Smilodon fatalis deserves a rethink, and the ‘‘marsupial saber-tooth’’ may have had an ecology unlike any other known carnivorous mammal.”

Figure 4. Maximum gape of Thylacosmilus. At upper left is the placental, Smilodon, for comparison.

Geology = Environment
Thylacosmilus was found in a vast Miocene tidal flat environment with a wide variety of terrestrial and shalllow aquatic taxa. So that doesn’t help much.

Saving the best for last: getting back to Late Cretaceaous Schowalteria
Wikipeda report, “It is notable for its large size, being among the largest of Mesozoic mammals,^[ as well as its specialization towards herbivory, which in some respects exceeds that of its later relatives.” This is based on tooth wear, with nearly all crowns gone (Fig. 3) in the only specimen known. Traditionally Schowalteria has been allied with styinodonts (Carnivora close to seals, but not to sea lions) and to taeinodonts (which the LRT found to be poylphyletic). Certainly, what little is known of Schowalteria is similar to these taxa. Others have omitted these LRT sisters in their analyses. Janis et al. omit the word ‘sister’ from their text.

Figure 5. Patagosmilus to scale alongside Hadrocodium. These sabetooth taxa are not directly related to Thylacosmilus in the LRT.

Patagosmilus has been called a sister to Thylacosmilus,
but the LRT nested Patagosmilus (Fig. 5) with the tiny basal therian, Hadrocodium (Fig. 5) as we learned earlier here. Taxon inclusion produces surprises like this.

At times like this it’s good to test.
Deletion of Thylacosmilus changes nothing in the LRT. Deletion of Schowalteria changes nothing in the LRT.

In my opinion
it was great that Janis et al. 2020 tested Thylacosmilus with its placental analogs, but they should also have tested Thylacosmilus with its marsupial homologs.

---

References
Janis CM, Figueirido B, DeSantis L, Lautenschlager S. 2020. An eye for a tooth: Thylacosmilus was not a marsupial ‘‘saber-tooth predator’’. PeerJ 8:e9346 http://doi.org/10.7717/peerj.9346

https://en.wikipedia.org/wiki/Ituzaing%C3%B3_Formation

https://pterosaurheresies.wordpress.com/2018/12/20/marsupial-sabertooth-taxa-are-polyphyletic/

SVP abstracts – Ambolestes and the origin of placentals

Bi S-D et al. 2019 discuss Early Cretaceous Ambolestes
(Figs. 1, 2) and the Early Mesozoic marsupial/placental split.

Figure 1. Ambolestes tracing from Bi et al. 2018.

From the abstract:
“Extant placental and marsupial mammals are the dominant vertebrates in many ecosystems, which makes the placental-marsupial dichotomy a significant event in Earth’s history.”

The large reptile tree (LRT, 1592 taxa) splits placentals from marsupials as shown below (Figs. 3, 4). The Early Cretaceous marsupial Bishops splits from the placental outgroup taxon, the extant marsupial Caluromys (Fig. 6). More timely, derived placental multituberculates, like Megaconus (Fig. 5), have been found in Middle Jurassic strata. That means a long line of undiscovered small, arboreal, placentals extends back to the Late Triassic/Earliest Jurassic.

Figure 2. Ambolestes skull reconstructed. Jaw tips restored.

Bi et al. continue:
“Molecular estimates of the divergence of placentals and marsupials (and their broader clades Eutheria and Metatheria) fall primarily in the Jurassic.”

Since Early Jurassic Megazostrodon is the proximal outgroup for all mammals, and Early Triassic Morganucodon is a marsupial, and Middle Jurassic Megaconus the LRT supports a Late Triassic split for placentals and marsupials.

Figure 3. Select basal cynodonts and mammals set chronologically. The divergence times for placentals (Eutheria), marsupials (Metatheria) and monotremes (Mammalia) are estimated here. Note the large gaps of time in which fossils are not known.

Bi et al. continue:
“In support, the oldest purported eutherian, Juramaia, is reported to be from the early Late Jurassic (160 million-years ago) of Liaoning Province, northeastern China.”

In the LRT (subset Fig. 1) Juramaia nests as a basal prototherian, an egg laying basal mammal.

“The oldest purported metatherian, Sinodelphys, is 35 million-years younger from the
Early Cretaceous Jehol Biota also in Liaoning Province, northeastern China.”

In the LRT Sinodelphys is another monotreme.

“In 2018, we reported a new eutherian, Ambolestes zhoui, also from the Jehol Biota. The fossil, a nearly complete skeleton, preserves anatomical detail unknown from contemporaneous eutherians including the hyoid apparatus and ectotympanic. The complete hyoid is the first known for any Mesozoic mammaliaform, and the ectotympanic resembles that in some extant didelphid marsupials.”

In the LRT (Fig. 1) Ambolestes (Figs. 3, 4) is a metathere/marsupial close to the extant Virginia opossum, Didelphis.

Figure 4. Subset of the LRT focusing on the Kynodontia and Mammalia. Non-eutherian taxa in red were tested in the LRT but not included because they reduce resolution. Eutherian taxa in red include a basal pangolin and derived xenarthran, clades that extend beyond the bottom of this graphic. The pink clade proximal to mammals was considered mammalian by Lautenschlager et al. due to a convergent mammalian-type jaw joint.

Bi et al. continue:
“In our phylogenetic analysis concentrating on the eutherian-metatherian dichotomy, the closest relative of Ambolestes was Sinodelphys, and both fell within Eutheria.”

As shown above, the LRT does not confirm that hypothesis of interrelationships.

Figure 5. Subset of the LRT focusing on Glires and subclades within.

Bi et al. continue:
“With Sinodelphys as a eutherian, postcranial differences formerly thought to indicate different invasions of a scansorial niche by meta and eutherians in Jehol are only variations among the early members of the placental lineage. Additionally, the earliest known metatherians are approximately 15 million years younger than previously thought and their
fossils, isolated teeth and fragmentary jaws, are from North America. Our tree results in a 50 million-year ghost lineage for Metatheria, accepting the 160 million-years age for Juramaia. 

The LRT confirms a 210 mya origin for Metatheria, starting with Morganucodon, so no ghost is necessary.

Figure 6. Caluromys, the largest of the mouse opossums, to scale with its LRT sister, Vulpavus, a basal member of Carnivora and Placentalia.

Bi et al. continue:
“A possibility raised elsewhere is that the age of Juramaia is incorrect; rather than Late Jurassic, perhaps it is from the Early Cretaceous Jehol Biota. In our study, Juramaia is in a clade with Albian/Aptian Prokennalestes and Late Cretaceous eutherians by having a more molariform ultimate upper premolar. In contrast, Ambolestes, as in the outgroups, has a non-molariform ultimate upper premolar. Although resolution of this intriguing debate is not currently possible, our understanding of the issues has been furthered by the discovery of Ambolestes.”

As shown above, the LRT does not confirm the Bi et al. hypothesis of interrelationships.

---

References
Bi S-D et al. 2019. The Early Cretaceous eutherian Ambolestes and its implications for the Eutherian/Metatherian dichotomy. Journal of Vertebrate Paleontology abstracts.

Revisiting the predatory metathere clade, Sparassodonta

Figure 1. A traditional selection of sparassodont metatheres from Forasieppi, MacPhee and del Pino 2019

In their study of the cranium
of the South American sparassodont sabertooth, Thylacosmilus, Forasieppi, MacPhee and del Pino 2019 report, “Sparassodonta is the group that includes the common ancestor of Patene and all its descendants. Undisputed records of Sparassodonta, including ones for Patene simpsoni, begin in the Early Eocene (Itaboraian) and extend through to the Pliocene (Chapadmalalan), when the last of them disappeared.”

FIgure 2. Cladogram of the traditional Sparassodonta from Babot and Forasiepi 2016. Taxa also found in the LRT are colored. Compare to Figure 2. The Babot and Forasiepi 2016 cladogram includes tooth only and mandible only taxa.

A recent cladogram of Sparassodonta and its outgroups
(Fig. 1) was published in Babot and Forasiepi 2016 (Fig. 2). This cladogram is distinct from the large reptile tree (LRT, 1530 taxa), so no confirmation here.

FIgure 3. Subset of the LRT focusing on the clade Metatheria (Marsupialia). Taxa shared with Babot and Forasiepi 2016 are colored. Compared to Figure 1. This cladogram includes relatively well-known and complete taxa.

Patene is known
from a tiny partial maxilla and mandible. I have not added it to the LRT. Mayulestes has just been downloaded, awaiting testing.

Figure 4. Thylacosmilus compared to Vincelestes separated by tens of millions of years. The both have maxillae conjoined dorsally to house the large canines.

According to the LRT,
taxa missing from the Babot and Forasiepi tree include Vincelestes (a sister taxon to Thylacosmilus in the LRT) and a long list of other carnivorous marsupials. Hadrocodium is not included in Babot and Forasiepi. It attracts the other sabertooth, Patagosmilus (Fig. 5), as we learned earlier here.

Figure 5. Patagosmilus to scale alongside Hadrocodium. These sabetooth taxa are not directly related to Thylacosmilus in the LRT.

FIgure 6. Borhyaena skull cracked and angled to match the glenoid to the jaw joint, distinct from the original illustration (above).

Late note: added the same evening as the original post:
Mayulestes ferox (Fig. 1) was just now added to the LRT, and it nests at the base of the Masrasector + Borhyaena clade. Nothing else changed. Thylacosmilus is still not related to  the dasyurids, including the creodonts and borhyaenids.

---

References
Babot J and Forasiepi AM 2016. Mamíferos predadores nativos del Cenozoico sudamericano: evidencias filogenéticas y paleoecológicas. Contributions del MACN 6 Historia evolutiva y paleobiogeografica de los vertebros de America del Sur. Agnolin FL et al. editors.
Forasiepi AM, MacPhee RDE and Hernandez del Pino 2019. Caudal cranium of Thylacosmilus atrox (Mammalia, Metatheria, Sparassodonta, a South American predaceous sabertooth. Bulletin of the American Museum of Natural History 433:1–64.

Hegetotherium and Pachyrukhos: not ungulates and not notoungulates

Hegetotherium and Pachyrukhos
were recently redescribed by Seoane and Cerdeño 2019 who considered them to be members of the Notoungulata, an invalidated polyphyletic clade with former members now nesting in various hooved marsupial and hooved placental clades.

Traditionally notoungulates are considered placentals and ungulates.
Seoane and Cerdeño report, “Hegetotheriidae is one of the most derived clades in the Order Notoungulata, the most abundant and diverse group of South American native ungulates.” The key word in there is ‘diverse’. Notungulates are SO diverse some of them are not related to the others.

Figure 1. Image from Cassini 2013. Pink taxa are marsupials. Others are placentals.

We looked at Hegetotherium,
earlier. In the large reptile tree (LRT, 1517 taxa) Hegetotherium nests with Mesotherium and Interatherium at the base of the marsupial Toxodon clade, derived from the wombat (Vombatus) clade. All those are derived from the Paedotherium clade (e.g. Paedotherium, Phalanger, Petaurus and Thylacoleo). In Seoane and Cerdeño, Paedotherium is a taxon nesting close to Pachyrukhos, a taxon not included in the LRT.

The Seoane and Cerdeño taxon list also includes
Paedotherium, Mesotherium and Hegetotherium, but not Interatherium, Toxodon, Phalanger, Petaurus and Thylacoleo. So taxon exclusion and lacking a wide gamut viewappear to be twin problems here. Seoane and Cerdeño did not realize the taxa in their study were marsupials close to phalangers and wombats. They assumed, by tradition, they were dealing with placental ungulates, close to cows and deer.

---

References
Cassini G 2013. Skull Geometric Morphometrics and Paleoecology of Santacrucian (Late Early Miocene; Patagonia) Native Ungulates (Astrapotheria, Litopterna, and Notoungulata). Ameghiniana 50 (2):193–216. DOI: 10.5710/AMGH.7.04.2013.606
Seoane FD and Cerdeño E 2019. Systematic revision of Hegetotherium and Pachyrukhos (Hegetotheriidae, Notoungulata) and a new phylogenetic analysis of Hegetotheriidae. Journal of Systematic Palaeontology http://dx.doi.org/10.1080/14772019.2018.1545146

Ever hear of Palaeothentes?

Currently there is no Wikipedia page for this taxon.
Even so, I found it to be far more important at filling gaps and shaking up paradigms than it seemed at first. The small, dull-looking taxa tend to be like that, as readers now know.

You could find this rat-sized Miocene taxon
(Fig. 1) in Carroll’s 1988 book, Vertebrate Paleontology, now well-worn and in pieces due to constant page flipping and scanning. Today’s research has revealed several more precise and more recent resources.

Figure 1. Not a marsupial, and not a shrew opossum, Palaeothentes nests in the LRT at the base of the Apatemys + Trogosus clade next to the clade of living shrew opossums within Glires.

According to Abello and Candella 2010,
Palaeothentes minutes (Ameghino 1887) is a paucituberculatan (details below) from the Santa Cruz Formation The results indicate that Palaeothentes would have been an agile cursorial dweller, with leaping ability, similar to the extant paucituberculatan Caenolestes fuliginosus and the didelphid Metachirus nudicaudatus.”

Okay, so now we have a problem.
In the large reptile tree (LRT, 1445 taxa) Caenolestes is not a marsupial. It nests with Rhyncholestes and more distantly Apatemys and more distantly, the extant tree shrew, Tupaia and the extant shrew, Scutisorex. As noted earlier, shrew opossums are placental shrews, not marsupial opossums in the LRT.

Wikipedia reports,
“Like several other marsupials, they do not have a pouch, and it appears that females do not carry the young constantly, possibly leaving them in the burrow.”
That’s describes most rodent/rabbit/tree shrew mothers and their young.

Wikipedia reports,
“Paucituberculata is an order of South American marsupials. Although currently represented only by the eight living species of shrew opossums, this order was formerly much more diverse, with more than 60 extinct species named from the fossil record, particularly from the late Oligocene to early Miocene epochs.”

Let’s solve that problem
by adding Palaeothentes to the LRT. Doing so recovers this taxon at the base of the Apatemys + Trogosus clade, next to the clade that includes Caenolestes, within Glires, far from Marupialia.

I suspect taxon exclusion
is the cause for the present lack of confirmation for traditional consensus. Many PhDs over several decades have followed tradition in nesting and testing shrew opossums with marsupials without testing them against apatemyids apparently. That’s why the LRT is here, to test taxa that have never been tested together before.

But wait! There’s a novel twist here~~~~~~!
Carroll 1988 reports, “Caenolestids have long been recognized as being very distinct from other South American marsupials, but they share with them a highly distinctive pattern of the spermatozoa, which become paired within the epididymis. Paired sperm are not known in any placental groups or among the Australian marsupials.” 

Sorry.
Physical traits have to trump genes and sperm. It just has to be that way because the LRT includes fossil taxa, which never preserve sperm. There have to be rules that all participants abide by. Interesting that the gene for paired spermatozoa is localized to one continent, just as genes separate other placentals into afrotheres and laurasiatheres. By the way, “The data show that paired spermatozoa exhibit a significant motility advantage over single spermatozoa in a viscous medium” according to Moore and Taggart 1995, who tested Monodelphis, a South American opossum.

Finally
we have a last common ancestor for arboreal Apatemys (Eocene, North America) and terrestrial Trogosus (Eocene, North America), two former enigma taxa with little to no relationship with other better known mammal clades. All members of Glires had their genesis sometime in the Jurassic, based on the presence of highly derived multituberculates (clade: Glires) in the Jurassic.

Wikipedia considers
apatemyids and trogosinae (Tillodontia) to be members of the Cimolesta, “an extinct order of non-placental eutherian mammals.” This bungling of the mammal family tree is due to taxon exclusion and the lack of a phenomic (trait-based) wide gamut cladogram that includes all the taxa present in the LRT. Paleontology needs to toss off a wide range of useless tradition with a reptile revolution led by someone out there confirming (or refuting) the widest gamut cladogram presently available: the LRT.

---

Palaeothentes lemoinei (Ameghino 1887, MPM-PV 3566; Miocene) was considered a prehistoric shrew opossoum (clade: Paucituberculata) but here nests next to shrew opossums, at the base of the Apatemys + Trogosus clade within Glires. The skull is 2x wider than tall, the canines are still large, the last premolar is large with a flat occlusal surface and the nasals split to form a zigzag suture with the frontals.

---

References
Abello MA and Candela AM 2010. Postcranial skeleton of the Miocene Marsupial Palaeothentes(Paucituberculata, Palaeothentidae): Paleobiology and Phylogeny. Journal of Vertebrate Paleontology 30(5):1515-1527.
Ameghino F 1887. Enumeracions sistematicad e las especies de mamiferos
fosiles coleccionados por Carlos Ameghino en los terranos eocenos de la Patagonia austral y depositados en el Museo La Plata. Boletin Museo de La Plata, 1:1-26.
Carroll RL 1988. Vertebrate Paleontology and Evolution. W. H. Freeman and Co. New York.
Forasiepi AM, Sánchez-Villagra MR, Schmelzle T,  Ladevèze S and Kay RF 2014. An exceptionally well-preserved skeleton of Palaeothentes from the Early Miocene of Patagonia, Argentina: new insights into the anatomy of extinct paucituberculatan marsupials. Swiss Journal of Palaeontology, 133(1):1-21.
Moore HD and Taggart DA 1995. Sperm pairing in the opossum increases the efficiency of sperm movement in a viscous environment. Biol. Reprod. 52(4):947-53.
Osgood WH 1921. A monographic study of the American marsupial, Caenolestes. Field Museum of Natural History, Zoological series 14:1–156.

wiki/Apatemyidae
wiki/Paucituberculata
https://en.wikipedia.org/wiki/Shrew_opossum
wiki/Vertebrate_Paleontology_and_Evolution

Paucituberculata -Trouessart 1898, Ameghino 1894

---

There was some news
about Palaeothentes recently (see below). Note, the experts consulted here consider this genus a marsupial.

New Bolivian Marsupials from the Middle Miocene

Dactylopsila, the striped possum, enters the LRT

Dactylopsila trivirgata (Gray 1858) is the extant striped possum (Fig. 1), closely related to the sugar glider, Petaurus and the marsupial lion, Thylacoleo (below), according to the large reptile tree (LRT, 1412 taxa). Dactylopsila is an arboreal marsupial with a prehensile tail the size and proportions of a placental squirrel. The fourth finger is elongated and used to extract beetles and caterpillars from tree bark, analogous to the extant aye-aye, Daubentonia. Dactylopsila, also eats leaves, fruit and small vertebrates.

By convergence
Dactylopsila has similar teeth and overall proportions to the extinct arboreal placental Apatemys (Fig. 2).

Figure 1. Dactylopsila skull in 3 views, plus in vivo. Comparisons to the extinct arboreal placental Apatemys (figure 2) are intriguing, showing convergence.

For comparison, we recently looked at Apatemys
here as it relates to the extant shrew opossums Caenolestes and Rhyncholestes, now nesting as apatemyid placentals in the LRT, rather than as traditional didelphid marsupials. The convergence is powerful here. Despite the phylogenetic distance, only 12 extra steps are needed to nest caenolestids with basal didelphids.

Figure 2. Apatemys skull in situ and reconstructed shares several similar traits with the extant striped opossum, Dactylopsila, including a squirrel-like size, elongate fingers and similar teeth.

The nesting of Dactylopsila
close to Petaurus (Fig. 3) is not controversial.

Figure 3. Subset of the LRT showing the nesting of Dactylopsila, the striped opossum, with Petaurus the sugar glider and Thylacoleo, the marsupial lion.

The problem continues to be
the traditional nesting of the marsupial lion, Thylacoleo (Fig. 4), as a member of the wombats (Vombatiiformes), rather than the Phalangeriformes and Petauroidea, as recovered by the LRT (Fig. 3), which points to a bigger problem…

Nowhere in traditional taxon lists
will you find interatheres, toxodontids and creodonts. All these taxa need to be tested in traditional metathere trees because the LRT has tested them and they nest with metatheres. It’s a good time for a confirmation or a refutation. PhD students… are you looking for a good subject to write about for your dissertation?

Figure 4. Large Thylacoleo skeleton compared to small Petaurus skeleton to scale. Dactylopsila is similar in size to Petaurus.

Here, again,
is where tradition, opinion and bias have, so far, trumped testing. Taxon exclusion needs to be tested with taxon inclusion. The list of taxa needing testing is provided by the LRT.

---

References
Gray JE 1858. List of species of Mammalia sent from the Aru Islands by Mr A.R. Wallace to the British Museum. Proceedings of the Zoological Society of London. 26: 106–113.

wiki/Striped_possum – Dactylopsia trivirgata

Shrew opossums (caenolestids) are shrew-like marsupials

Updated July 18, 2022
with a shifting of two members of the Paucituberculata, shrew-like Caenolestes and Rhyncholestes, nesting with the marsupial, Acristotherium, in the LRT.

According to Wikipedia,
“The family Caenolestidae contains the seven surviving species of shrew opossum: small, shrew-like marsupials that are confined to the Andes mountains of South America.”

Figure 1. Caenolestes skull and in vivo. It sure looks more like a shrew than an opossum. Skull images from Digimorph.org and used with permission. Colors added. See figure 2 for updated image of skull.

The trouble is
tested caenolestids, Caenolestes (Fig. 1) and Rhyncholestes (Fig. 2), do not have a pouch. Even so, they nest with the marsupial, Acristotherium, in the LRT.

On the traditional side,
Dr. Darren Naish reported online for Tetrapod Zoology/Scientific American in 2015, “Incidentally, the most frequently used name for the group – shrew-opossums – might not be a particularly good one, seeing as they don’t look much like shrews, don’t live like shrews, and don’t act like shrews. And they’re not technically opossums, either, but perhaps we can let that go.”

Figure 2. Shrew-like shrew opossums, Caenolestes and Rhyncholestes, nest with the marsupial, Acristotherium, in the LRT.

Rhyncholestes raphanurus (Osgood, 1924; long-nosed shrew-opossum, Chilean shrew opossum, extant; snout-vent length 20cm).

Genetically
Wikipedia reports. “Genetic studies indicate that they are the second most basal order of marsupials, after the didelphimorphs” (Nilsson et al. 2010). That’s exactly where the LRT documents them. Even so, don’t trust genes. Test traits.

According to AnimalDiversity.org, “In general, members of family Caenolestidae can be distinguished from other marsupial groups by their unique dentition. Their lower middle incisors are large and have a forward slope; likewise, they have a reduced number of incisors. The dental formula for genus Caenolestes is: I 4/3, C 1/1, P 3/3, M 4/4, 46 teeth total. Shrew opossums have short robust limbs, each containing 5 digits; their middle 3 digits are shorter than the outside two. Their humeri are extremely heavy; in comparison, their femurs are relatively slender. Members of family Caenolestidae have unusual lip flaps, they may function as a method of preventing debris from interfering with their whiskers or they may help prevent ingestion of unwanted debris. Similar to other marsupials, Caenolestid females have 2 uteri and 2 vaginas. Members of genus Caenolestes lack a pouch but do have 4 mammae, 2 on either side of their abdomen.”

As a matter of fact, a double vagina sometimes occurs in humans.

Rhyncholestes raphanurus (Osgood, 1924; long-nosed shrew-opossum, Chilean shrew opossum, extant; snout-vent length 20cm), nests in the LRT between the shrew-mole, Uropsilus, and a large living shrew, Scutisorex, all within the placental clade, Glires. Wikipedia and other sources consider this shrew-like South American mammal a marsupial, but Wiki also notes that Rhyncholestes lacks a marsupium (pouch).

Caenolestes fuliginosus (originally Hyracodon fuliginosus Tomes 1863).

---

References
Marsh OC 1872. Preliminary description of new Tertiary mammals. Part II. American Journal of Science 4(21):202-224.
Nilsson MA, et al. (6 co-authors) 2010. Tracking Marsupial Evolution Using Archaic Genomic Retroposon Insertions”. PLoS Biology. 8 (7): e1000436. doi:10.1371/journal.pbio.1000436
Osgood WH 1924. Field Mus. Nat. Hist. Publ., Zool. Ser. 14:170.

tetrapod-zoology/you-never-hear-much-about-shrew-opossums/
wiki/Shrew_opossum = Caenolestidae
animaldiversity.org/accounts/Caenolestes_fuliginosus/
wiki/Apatemyidae
wiki/Rhyncholestes
wiki/Caenolestes
wiki/Paucituberculata
wiki/Uterus_didelphys

Numbat genesis in the Early Jurassic

Coelocanth. Tuatara. Numbat.
Name three taxa that have not changed much in hundreds of millions of years.

Figure 1. Myrmecobius, the living numbat, has remained essentially unchanged for nearly 200 million years based on the LRT. Note the loss of posterior molars and the simplification of the remaining anterior molars. Orange arrow point to palatal pits that receive the long lower canines.

Extant numbats
(genus: Myrmecobius, Fig. 1) nest in the large reptile tree (LRT, 1412 taxa; subset Fig. 2) basal to Early Cretaceous Anebodon, Middle Jurassic Docofossor and the extant marsupial mole (genus: Notoryctes). All arise from the extant Dasycercus (Fig. 3). So that provides an interesting cladogram with members that span from the Triassic to the present. That means some extant taxa had nearly identical ancestors that shared the planet with the first dinosaurs and pterosaurs.

Figure 2. Subset of the large reptile tree focusing on the basal phytometatheria, including extant numbats (Myrmecobius), basal to Middle Jurassic Docofossor.

Myrmecobius fasciatus (Waterhouse 1841) is the extant numbat. Here it nests between Dasycercus and Anebodon. Since an ancestral taxon, Docofossor, is known from the Middle Jurassic, a sister to Myrmecobius had its genesis in the Early Jurassic. The molars are narrow and simplified. This is a marsupial termite eater, convergent with placental termite- and ant-eaters. Over each orbit is the reappearance of an old bone, the postfrontal. The canine is smaller. The jugal is straighter.

Figure 3. Dasycercus, the extant mulgara, is the carnivorous phylogenetic ancestor to the clade that includes numbats, Docofossor and kin in the LRT.

What’s interesting are the molars in Myrmecobius.
Take a good look (Fig. 1). The molars are narrow and simplified because this taxon eats termites (or vice versa). A phylogenetic descendant, Docofossor (Fig. 5) was considered a docodont based on its simple tooth morphology. Another phylogenetic descendant, Anebodon, was considered a symmetrodont based on its tooth morphology.

The LRT results remind us
not to put so much emphasis on tooth morphology. The LRT makes mammal systematics so much simpler by nesting taxa according to all their tested traits, not just a few, rather plastic, dental traits.

Figure 4. Dasycercus in vivo. This is the extant mulgara, a carnivorous nocturnal basal phytomarsupial with origins in the Early Jurassic.

Dasycercus cristicauda (originally ‘Chaetocercus‘ Krefft 1867; Peters 1875; 22cm + 13 cm tail) is the extant mulgara, considered a dasyurid marsupial. Here carnivorous, nocturnal Dasycercus nests apart from Dasyurus between Anebodon and Myrmecobius at the base of the herbivorous clade of marsupials. The pouch is reduced to two lateral folds of skin.

Figure 5. Docofossor in situ with DGS tracings. This Middle Jurassic taxon nests as a derived descendant of Dasycercus and Myrmecobius in the LRT.

Docofossor brachydactylus (Luo et al. 2015; Middle Jurassic, 160 mya; BMNH 131735; 9cm in precaudal length) was originally considered a member of the Docodontidae along with Docodon and Haldanodon outside of the Mammalia. Here it nests as a Jurassic sister to Anebodon and Notoryctes. Broad, short-fingered hands, larger than the feet, along with other traits mark Docofossor as a digging animal, similar to moles like Talpa and Chrysochloris.

---

References
Bi S-D, heng X-T, Meng J, Wang X-L, Robinson N and Davis B 2016. A new symmetrodont mammal (Trechnotheria: Zhangheotheriidae) from the Early Cretaceous of China and trechnotherian character evolution. Nature Scientific Reports 6:26668 DOI: 10.1038/srep26668
Gadow H 1892. On the systematic position of Notoryctes typhlops. Proc. Zool. Soc. London 1892, 361–370.
Luo Z-X, Meng QJ, Ji Q, Liu D, Zhang Y-G, Neande AI 2015.Evolutionary development in basal mammaliaforms as revealed by a docodontan. Science. 347 (6223): 760–764.
Peters WCH 1875. Sitzungsberichte der Gesellschaft Naturforschender Freunde zu Berlin 1875: 73.
Stirling EC 1888. Transactions of the Royal Society, South Australia 1888:21
Stirling EC 1891. Transactions of the Royal Society, South Australia 1891:154
Tate GHH 1951. The banded anteater, Myrmecobius Waterhouse (Marsupialia). American Museum Novitates 1521, 8 pp.
Waterhouse GR 1836. Myrmecobius fasciatus. Proc. Zool. Soc. London 4: 69–131.
Waterhouse GR 1841. Description of a new genus of mammiferous animals from Australia, belonging probably to the order Marsupialia. Trans. Zool. Soc., London2, aricle. 11, p 149.

wiki/Dasycercus
wiki/Myrmecobius

Hey! Some of those Miocene ‘ungulates’ are marsupials!

More heresy today
courtesy of taxon inclusion.di

Cassini 2013
looked at several traditional Miocene South American ‘ungulates’ (Fig. 1) unaware that these taxa do not nest in monophyletic clade any more specific than Theria in the large reptile tree (LRT, 1401 taxa). Cassini was reporting the results of “an ecomorphological study based on geometric morphometrics of the masticatory apparatus.” So he was working from prior cladograms and focusing on the mechanics of eating.

Figure 1. Image from Cassini 2013. Pink taxa are marsupials. Others are placentals.

Earlier
here and here the traditional clade Notoungulata was splintered by the LRT into several clades, some among the marsupials, other among the placentals.

Traditional ‘Litopterna’
Pony-like (single toe), tusked Diadiaphorus (Fig. 1) nests at the base of the hyrax clade. Theosodon (Fig. 1) nests as a derived taxon. Also included, but not listed: Chalicotherium and other chalicotheres.

Considering its member taxa,
the clade Litopterna (Ameghino 1889) is a junior synonym for Chalicotheridae (Gill 1872) in the LRT.

Considering its member taxa,
the clade Astrapotheria (Lydekker 1894) is a junior synonym for Meniscotheriinae (Cope 1882) and both nest within Phenacodontidae (Cope 1881).

Interatheriidae (Ameghino 1887) traditionally includes Interatherium, Protypotherium, Miocochilius and other taxa listed here. In the LRT Interatherium nests close to the ancestry of the Toxodon clade + the kangaroo clade within Metatheria. By contrast, Protypotherium and Miocochillus nest with Homalodotherium deep within the Eutheria. Homalodotherium traditionally nests with the the metatherian Toxodon. According to the LRT, all the above taxa developed similar enough traits by convergence that all were mistakenly lumped together in the invalid placental clade ‘Notoungulata’.

This is not the first time
metatherians were split from convergent eutherians. Most creodonts are marsupial predators, phylogenetically distinct from their traditional sisters in the clade Carnivora, within the clade Eutheria (Placentalia).

New taxa added to the LRT:

1. Hegetotherium (Fig. 1) nests with Mesotherium between Interatherium and the Toxodon clade in the Metatheria. 
2. Diadiaphorus (Fig. 1), the horse-mimic, nests at the base of the hyrax – elephant clade.

References
Cassini G 2013. Skull Geometric Morphometrics and Paleoecology of Santacrucian (Late Early Miocene; Patagonia) Native Ungulates (Astrapotheria, Litopterna, and Notoungulata). Ameghiniana 50 (2):193–216. DOI: 10.5710/AMGH.7.04.2013.606