Toxodon and Pyrotherium are wombats

Earlier the LRT nested two taxa, Vintana and Zalambdalestes, with the wombat, Vombatus. I did not know then that this would start a trend among enigmatic and unusual taxa.

Today
the enigmatic South American ‘ungulates’ Toxodon (Fig. 1) and Pyrotherium (Fig. 2) move over from the decimated Notoungulata and Ungulata to nest with marsupial wombats. They lack epipubes and have only three molars (per side x4), but other related marsupials also share those traits. It’s no longer a rule that marsupials have to have four molars.

Figure 1. Toxodon was a notoungulate placental. Now it's a wombat marsupial.

Figure 1. Toxodon was a notoungulate placental. Now it’s a wombat marsupial with only three toes on each foot. Note the septomaxilla and posterior placement of the jaw glenoid.

Toxodon platensis
(Owen 1837; Pliocene-Pleistocene 2.6-.016 mya; 2.7 m in length) lost the medial and lateral toes on all four hoofed extremities. That makes it look like a perissodactyl ungulate. The high neural spines and low position of the skull remind one of North American bison. Prehistoric humans hunted them, based on the arrow tips found with skeletons.

Figure 2. Pyrotherium was another notoungulate. Now it is another wombat. From what I've seen, I think the dorsal view is fairly damaged. Note the septomaxilla (orange) and posterior placement of the jaw joint. I show the fossil and drawings because not all the parts match up.

Figure 2. Pyrotherium was another notoungulate. Now it is another wombat. From what I’ve seen, I think the dorsal view is fairly damaged. Note the septomaxilla (orange) and posterior placement of the jaw joint. I show the fossil and drawings because not all the parts match up.

Pyrotherium sorondoi
(Ameghino 1889, 1894, 1895; Early Oligocene, 29-21 mya; 3m long); data here comes only from its skull, but post-cranial pieces are known indicating a graviportal stance. Wikipedia called Pyrotherium an ungulate. Other workers (i.e. Shockley et al. 2004) had trouble nesting it, too. It converges with Arsinoitherium and elephants. Provided with a wide gamut of mammals to nest with, Pyrotherium nests with Toxodon close to Vombatus.

Figure 3. Two large wombats, Vombatus and Phascolonus for comparison.

Figure 3. Two large wombats, Vombatus and Phascolonus for comparison. These wombats have epipubes.

The wombats
are becoming much more diverse than earlier imagined. Credit ‘taxon inclusion’ for this insight and others.

References
Ameghino F 1889Contribución al conocimiento de los mamíferos fósiles de la República Argentina, obra escrita bajo los auspicios de la Academia Nacional de Ciencias de la República Argentina para presentarla a la Exposición Universal de Paris de 1889. Actas Academia de Ciencias. de Córdoba 6:11027.
Ameghino F 1894. Sur les oiseaux fossiles de Patagonie; et la faune mammalogique des couches à Pyrotherium. Boletin del Instituto Geographico Argentino 15:501-660.
Ameghino F 1895. Premiére contribution à connaissance de la fauna mammalogique de couches à Pyrotherium. Boletín Instituto Geográfico Argentino 15:603660.
Owen R 1837. Description of the cranium of the Toxodon platensis. Proceedings of the Geological Society of London 2:541-542.
Shockey BJ & Anaya F 2004. Pyrotherium macfaddeni, sp. nov. (late Oligocene, Bolivia) and the pedal morphology of pyrotheres. Journal of Vertebrate Paleontology. 24 (2): 481–488. doi:10.1671/2521.

wiki/Pyrotherium
wiki/Toxodon

New paper on stem archosauromorpha: Foth et al. 2016

When Foth et al. 2016 report,
“Here, we analyse the cranial disparity of late Permian to Early Jurassic archosauromorphs and make comparisons between non-archosaurian archosauromorphs and archosaurs (including Pseudosuchia and Ornithodira) on the basis of two-dimensional geometric morphometrics.” we are immediately ready for a bogus report based on the antiquated inclusion of the clades listed above.

Foth et al. 2016 set up their study
based on traditional phylogenies, not the large reptile tree [my comments follow]:

  1. “Living birds and crocodylians, as well as their extinct relatives including pterosaurs and non-avian dinosaurs, comprise the extraordinarily diverse and successful crown clade Archosauria.” [pterosaurs are lepidosaurs]
  2. “non-archosaurian archosauromorphs (i.e. taxa on the stem lineage leading towards archosaurs) formed a species rich component of Triassic ecosystems (>90 valid species) and achieved high morphological diversity, including highly specialized herbivores (Azendohsaurus, rhynchosaurs), large apex predators (erythrosuchids), marine predators with extremely elongated necks (tanystropheids), armoured crocodile-like forms (dosewellids, proterochampsids ), and possibly even turtles).” [Azendosaurus, rhynchosaurs, tanystropheids and turtles are all lepidosauromorphs].

The Foth et al. cladogram includes the following taxa
that have nesting problems:

  1. Tanystropheidae [should be in Tritosauria, Lepidosauria]
  2. Allokotosauria (a new paraphyletic ‘clade’ by Nesbitt et al. 2015 nesting between Protorosaurus and Prolacerta) – Pamelaria [Protorosauria], Azendohsaurus, Trilophosaurus [Rhynchocephalia]
  3. Rhynchosauria [should be in Rhynchocephalia, Lepidosauria]
  4. Pterosauria [should be in Tritosauria, Lepidosauria]
  5. and the archosauriforms could use a lot of work! It’s all mixed up in there.

The rest of the paper
discusses the large amount of  cranial disparity in this clade. No wonder there is so much cranial disparity, they have thrown in so many unrelated taxa!!! As a referee I would have sent this manuscript back to the authors. The sister taxa do not demonstrate a gradual accumulation of character traits. They really need to expand their taxon list. They are missing SO many transitional taxa.

By contrast
there is not so much cranial disparity in the archosauriform subset of the LRT because they are more closely related to each other. In fact, the differences between sisters have been minimalized by taxon inclusion, creating the microevolution between taxa that even Creationists support.

References
Foth C, Ezcurra MD, Sookias RB, Brusatte SL and Butler RJ 2016. Unappreciated diversification of stem archosaurs during the Middle Triassic predated the dominance of dinosaurs. BMC Evolutionary Biology, 2016, Volume 16, Number 1, Page 1 online here.

Nesbitt SJ, Flynn JJ, Pritchard AC, Parrish MJ, Ranivoharimanana L and Wyss AR 2015. Postcranial osteology of Azendohsaurus madagaskarensis (?Middle to Upper Triassic, Isalo Group, Madagascar) and its systematic position among stem archosaur reptiles. Bulletin of the American Museum of Natural History. 398: 1–126.

Say ‘No’ to Notoungulata. Turns out, it’s not a clade.

This is where testing takes us out on yet another limb
because the Notoungulata has been a clade for over 100 years (Roth 1903). If you are unfamiliar with this clade, as I was… you can get a basic education on the Notoungulata here,

According to
Darin Croft, PhD, Paleomammalogist, the Notoungulata is a clade of diverse South American Tertiary mammals all united, “by characters of their ear region and their teeth, including the presence of a loph on their upper molars known as the ‘crochet‘.” 

The clade Notoungulata produced taxa convergent with:

  1. buffalo/rhinos: Toxodon
  2. elephants/hippos: Pyrotherium. Astrapotherium
  3. chalicotheres: Homalodotherium
  4. peccarys: Thomashuxleya
  5. rabbits:  Protypotherium
  6. and others

Unifortunately
as I keep adding notoungulates to the large reptile tree (Fig. 1; 808 taxa at present), they keep nesting not with each other, but with a diverse selection of placental and marsupial mammals. That should not be happening, unless a false paradigm is present.

Figure 1. A selection of purported notoungulates (in amber) were added to the LRT and they did not nest together. That means they're not a clade.

Figure 1. A selection of purported notoungulates (in amber) were added to the LRT and they did not nest together. That means they’re not a clade.

Are dental characters
over-emphasized in traditional studies? including the Notoungulata? Do the cusps and valleys of molars trump the rest of the mammal’s morphology? Is it more reasonable to posit that teeth might be converging in these disparate taxa? For instance, Homalodotherium is an excellent sister to Chalicotherium, but I did not test the tooth cusps. And Toxodon is an excellent wombat. We’ll take closer looks at those taxa in future blogs.

Questioning basic assumptions is okay
because this is Science and everything is up for discussion. If a test does not deliver promised results, it’s okay to wonder why.

If we were to take the Notoungulata at face value
we would have to accept the wide range of morphologies within this one clade, as in the clade that includes tenrecs and whales. Unfortunately, testing shows that the range of body types in the Notoungulata is more readily matched by other clades, including the wombats within the marsupials and the chalicotheres within the placentals.

The loss of the clade ‘Notoungulata’
follows a list of other clades that have been lost based on the results of the LRT.

  1. Amniota is now a junior synonym for Reptilia
  2. Ornithodira is now a junior synonym for Reptilia
  3. Parareptilia is now a junior synonym for Reptilia
  4. Pterodactyloidea is paraphyletic
  5. Allotheria:  no mammals form a clade between Metatheria and Eutheria
  6. And others…

In counterpoint,
several new clades have been erected, resurrected or revised here:

  1. Archosauromorpha
  2. Lepidosauromorpha
  3. Enaliosauria = plesiosaurs + ichthyosaurs and their kin
  4. Tritosauria – a previously unrecognized squamate clade
  5. Prosquamata – another previously unrecognized squamate clade
  6. Fenestrasauria (goes back 16 years to Peters 2000, but still not used in academic publication
  7. Tenreccetacea = tenrecs + whales
  8. And several others…

Share your thoughts on this matter,
if you wish…

References
Billet G 2011. Phylogeny of the Notoungulata (Mammalia) based on cranial and dental characters. Journal of Systematic Palaeontology 9:481-497.
Cifelli RL 1993. The phylogeny of the native South American ungulates; pp. 195-216 in F. S. Szalay M J Novacek, and MC McKenna (eds.), Mammal Phylogeny: Placentals. Springer-Verlag, New York.
Roth S 1903. Los Ungulados Sudamericanos. Anales del Museo de La Plata (Sección Paleontológica). 5: 1–36. OCLC 14012855.
Scott WB 1932. Mammalia of the Santa Cruz Beds. Volume VII, Paleontology. Part III. Nature and origin of the Santa Cruz Fauna with additional notes on the Entelonychia and Astrapotheria. ; pp. 157-192 in W. B. Scott (ed.), Reports of the Princeton University Expeditions to Patagonia, 1896-1899. Princeton University, E. Schweizerbart’sche Verlagshandlung (E. Nägele), Stuttgart.

wiki/Notoungulata

Macrauchenia: a South American perissodactyl

Figure 1. Macrauchenia museum mount.

Figure 1. Macrauchenia museum mount.

Figure x. Macrauchenia cladogtam. Tapir and Chalicotherium are perissodactyls.

Figure x. Macrauchenia cladogtam. Tapir and Chalicotherium are perissodactyls.

Famous for its assumed
elephant-like proboscis, arising from a dorsal narial opening (Fig 2), Macrauchenia was a long-legged grazing ungulate with three toes on each manus and pes. In the large reptile tree it nests with Chalicotherium, which, in turn, nests with Tapirus, (the tapir) an extant perissodactyl with a short flexible trunk. A recent analysis of collagen sequences (Welker et al. 2015) found the same relationship. Not sure why this needed resolution… tapirs also have a trunk, dorsal narial opening, three hooves per foot AND some still live in South America. I guess that used to be considered ‘convergence.’ Here the LRT calls it ‘homology.’

Discovered by Charles Darwin in 1834,
and published by Richard Owen in 1836, Macrauchenia patachonica (Pliocene 7mya to Pleistocene .02mya; 3m length) was otherwise similar to a camel in proportions with a horse-like skull. Macrauchenia was an herbivore with a full arcade of short teeth in its jaws (Fig. 2). The last premolar looks like a molar, but, like other premolars, it is slightly larger than the other teeth and all sister taxa have 3 molars per side.

Figure 2. Macrauchenia skull in several view (from Owen 1836?) with bones colorized here. Note the dorsal extension of the premaxilla.

Figure 2. Macrauchenia skull in several view (from Owen 1836?) with bones colorized here. Note the dorsal extension of the premaxilla. The fossa posterior to the naris could anchor large proboscis muscles. 6 premolars and 3 molars appear to have been present. Not sure about the palatine here.

Wikpedia reports, “Macrauchenia was a long-necked and long-limbed, three-toed South American ungulate mammal, typifying the order Litopterna.  Early forms are near the condylarths, to such an extent that the litopterns might be considered merely as surviving and diversely specialized condylarths.” The LRT did not nest Macrauchenia with the basal Condylartha, but that is still a monophyletic clade that now includes all hoofed and edentate mammals — along with all the original basal condylarths.

Thanks to reader SBJ
for suggesting a number of South American mammals to add to the LRT. This is number one of several to come.

References
Owen R 1838. Description of Parts of the Skeleton of Macrauchenia patachonica. In Darwin, C. R. Fossil Mammalia Part 1 No. 1. The zoology of the voyage of H.M.S. Beagle. London: Smith Elder and Co.
Welker F et al. 2015. Ancient proteins resolve the evolutionary history of Darwin’s South American ungulates. Nature. doi:10.1038/nature14249. ISSN 0028-0836.
wiki/Macrauchenia

Is this the footprint of Arizonasaurus?

Figure 1. Synaptichnium MNA V3425. Arrow points to direction of movement and aligns with sagittal plane. PILs and pads added.

Figure 1. Synaptichnium MNA V3425. Arrow points to direction of movement and aligns with sagittal plane. PILs and pads added. The pink manus track is another specimen.

The middle Triassic Moenkopi formation
in Arizona has provided a rich trove of fossils. An excellent footprint (MNA V3425, Fig. 1) was recently published online here and attributed to Arizonasaurus, a likely bipedal carnivorous archosauriform (Fig. 2). Arizonasaurus was derived from basal Rauisuchia, like Vjushkovia, and is most closely related to Yarasuchus and Qianosuchus according to the large reptile tree.

Figure 2. Arizonasaurus. Not sure which of the two mandibles is correct here, so both are presented. Note, neither manus nor pes is preserved in the specimen.

Figure 2. Arizonasaurus. Not sure which of the two mandibles is correct here, so both are presented. Note, neither manus nor pes is preserved in the specimen.

According to the online article,
“Paleontologist Christa Sadler has written a book, “Dawn of the Dinosaurs,” about the archosaurs of the Middle and Late Triassic in the region. Unusually detailed footprints of the large reptile, or something like it, are preserved in a slab of Moenkopi sandstone in the collections repository at the Museum of Northern Arizona, where Sadler has studied. MNA  [Museum of Northern Arizona] Colbert Collections Curator of Vertebrate Paleontology David Gillette, Ph.D., says the footprints were discovered in Wupatki National Monument in 1973.”

Figure 3. Manus impression of man v3245. Note the heavy scales here.

Figure 3. Manus impression of man v3245. Note the heavy scales here.

The LRT currently doesn’t include ichnites (footprints)
but let’s see what happens this time, since the track is so precisely imprinted. Unfortunately, Arizonasaurus does not preserve the manus or the pes (Fig. 1). Nevertheless, out of 801 candidate taxa, MNA 3425 nests with a sister to Arizonasaurus, Decuriasuchus, and is similar to the pes of other Arizonasaurus sisters, Qianosuchus and Nandasuchus, all Middle Triassic taxa. So, phylogenetic bracketing works, at least to this extent. And it just shows you don’t need a long list of character traits to successfully nest some taxa.

Figure 3. Scaly palms of two crocodilians. Digit 1 is on the left in both specimens.

Figure 4. Scaly palms of two crocodilians. Digit 1 is on the left in both specimens.

Notes on the scaly palm of MNA V3425
Dinosaur footprints do not have large scale impressions. By contrast, croc hands and feet do have large scales (Fig 3). The sisters to Arizonasaurus, Qianosuchus and Yarasuchus, both have short limbs, a long rostrum and a general crocodile-like build. Likewise Decuriasuchus was long-bodied, quadrupedal with a large foot and a presumably small hand (not preserved). In similar fashion, Arizonasaurus likely also had a large foot and small hand based on its pectoral and pelvic girdles and femur (Fig. 2), but was a likely biped.

Figure 5. Decuriasuchus does not preserve the manus, but it was probably small based on the forelimb.

Figure 5. Decuriasuchus does not preserve the manus, but it was probably small based on the forelimb.

Belated apologies
to those who tried [or continue to try] to access www.reptileevolution.com yesterday and today. Eviidently the server is down, wherever it is. I can’t access it either to make updates and repairs. Hopefully the RepEvo website will be restored soon. :  )

 

Peltephilus, the horned armadillo, enters the LRT

Updated June 18, 2021
with the realization that the premaxilla had a ventral / palatal exposure as in Holmesina, a taxon added later.

Peltephilus ferox is close to the base of the armadillos and a sister to Holmesina.

Figure 1. Peltephilus skull, manus and pes. There are 3 premolars and 3 molars present, the standard pattern for sister taxa. Extant armadillos have long, narrow rostra with an elongate premaxilla and 7 cone-shaped identical teeth none of which extend below the orbit.  Two species / specimens are shown here. Line art is from Vizcaino and Farina 1997,

Figure 1. Peltephilus skull, manus and pes. There are 3 premolars and 3 molars present, the standard pattern for sister taxa. Extant armadillos have long, narrow rostra with an elongate premaxilla and 7 cone-shaped identical teeth none of which extend below the orbit.  Two species / specimens are shown here. Line art is from Vizcaino and Farina 1997,

According to
Wkiipedia, “Peltephilus ferox (Ameghino 1887‭; 1.5 m long) the horned armadillo, is an extinct species of dog-sized, armadillo xenarthran mammal which first inhabited Argentina during the Oligocene epoch, and became extinct in the Miocene epoch. Notably, the scutes on its head were so developed that they formed horns protecting its eyes. Aside from the horned gophers of North America, it is the only known fossorial horned mammal. Although it had traditionally been perceived as a carnivore because of its large, triangular-shaped teeth, Vizcaino and Farina argued in 1997 that Peltephilus was a herbivore.”

Postcranially
Peltephilus had transverse bands of ossified armor along the back. It had short legs and large claws, ideal for digging or ripping open ant colonies.

References
Ameghino F 1894. Enumeration Synoptique des especes de mammiferes fossiles des formations Eocenes de Patagonie. Boletin de la Academia Nacional de Ciencias en Cordoba (Republica Argentina) 13:259-452
Ameghino F 1897. Mamiferos Cretaceos de la Argentina. Segunda contribucion al conocimiento de la fauna mastologica de las capas con restos de Pyrotherium. – Boletin Instituto Geografico Argentino 18:406-521.
Vizcaino SF and  Farina RA 1997. Diet and locomotion of the armadillo Peltephilus: a new view. Lethaia, 30, 79-86.

wiki/Peltephilus

Look no further for a Xenarthran ancestor: It’s been Barylambda all along.

…and for aardvarks,
like Orycteropus (Fig.1), the relationship to Barylambda (Fig. 2) is even closer (Fig. 3).

Figure 3. Orycterpus, the extant aardvark, is a living sister to Barylambda from the Paleocene.

Figure 1. Orycterpus, the extant aardvark, is a living sister to Barylambda from the Paleocene.

A suitable ancestor
for the odd mammal clade Xenarthra has been long sought. As it turns out, that ancestor (Fig. 2) has been hiding in plain sight for several decades.

Figure 1. Barylambda looks like a large ground sloth for good reason. It is a sister to the direct ancestor and nests at the base of the Xenarthra along with Orycteropus, the aardvark.

Figure 2. Barylambda looks like a large ground sloth for good reason. It is a sister to the direct ancestor and nests at the base of the Xenarthra along with Orycteropus, the aardvark.

Figure 3. Subset of the large reptile tree showing the nesting of Barylambda with Orycteropus and Xenarthra.

Figure 3. Subset of the large reptile tree showing the nesting of Barylambda with Orycteropus and Xenarthra.

When Barylambda was added to the large reptile tree (LRT; Fig. 3) it nested as a sister to Orycteropus and basal to the Xenarthra (sloths, anteaters and armadillos). That should not come as any big surprise, considering how workers have portrayed and described Barylambda as capable of assuming a tripodal posture, like a ground sloth, with its robust tail.

Often compared to ground sloths, Barylambda (Patterson 1933; originally Titanoides, renamed by Patterson 1937CNHM P-14637, a partial skeleton; late Paleocene, 55 mya) now nests with aardvarks, sloths, anteaters and armadillos. Not sure why this relationship has been overlooked or missed for over 75 years. Certainly many have seen the similarity in overall build. Wikipedia describes current uncertain relationships between Xenarthra and other mammal clades. In any case, it’s good to find yet another ancestor for a former enigma clade using a verifiable analysis.

Note the robust build
of Barylambda, the deep tail chevrons, the tiny premaxillary teeth. the out-turned ilia, the low jugals, the lack of large canine-like canine teeth (in females), all traits that point to aardvark and edentate morphologies. No other tested taxa nest closer. Here Barylambda nests next to, but not with, the basal condylarths in the LRT, Onychodectes. Ectoconus and Pantolambda.

Wikipedia describes Barylambda as a genus of herbivorous pantodont mammal the size of a pony, but with five-toed plantigrade hands and feet. The unguals (claws) were not sharp, but small and rounded. Three species are known. We should look for a smaller sister to Barylambda with a longer rostrum in the Paleocene for more direct ancestors to anteaters and armadillos.

The LRT (now 801 taxa) does not recover the same
cladogram topology as O’Leary et al. 2013, which reported, “Our tree suggests that Placentalia first split into Xenarthra and Epitheria.” I don’t find Barylambda in their taxon list. More criticism of that recent paper on mammal relationships can be found here.

A final thought
Barylambda is a big taxon. Evolutionary novelty most often occurs among small taxa. We might expect that Barylambda and Orycterpus AND the members of the Xenarthra are all descendant from a smaller ancestor in the early Paleocene. Or… perhaps more likely, the novelty we see in Orycteropus AND members of the Xenarthra are based on several smaller descendants of Barylambda, some of them armored, starting in the mid Paleocene when Barylambda was probably common and widespread, rather than rare at its genesis.

References
O’Leary, MA et al. 2013. The placental mammal ancestor and the post-K-Pg radiation of  placentals. Science 339:662-667. abstract
Patterson B 1933. A new species of the amblypod Titanoides from western Colorado. American Journal of Science, 25:4 15-425.
Patterson B 1937. A new genus, Barylambda, for Titanoides faberi, Paleocene amblypod. Geological Series, Field Museum of Natural History, 6:229-231. online
Simons EL 1960. The Paleocene Pantodonta. Transactions of the American Philosophical Society, New Series 50(6):1-81

wiki/Barylambda

 

 

 

The number of molars in marsupials and placentals

Here
the large reptile tree (subset Fig. 1) divides mammals into pre-therians (monotremes and kin), and therians (marsupials + placentals). Note: there are no allotherians here. They all nest elsewhere.

Typical molars
are multi-rooted teeth that erupt only once in mammals as they approach adulthood. Some molars are not multi-rooted, but ever-growing and lack enamel (xenarthrans). Most molars have several cusps, but a few (i.e. Stylinodon, Vintana, Glyptotherium) do not.

Traditional paleontology
holds that marsupials have four molars while placentals have three (see below). I tested that tradition in the LRT and found that you can’t always count on this rule. Turns out there is a mix of molar numbers in marsupials and placentals (Fig. 1).

Figure 1. Molar numbers in mammals. Four molars is the basal number. A few taxa add molars. Several lose one molar for a total of three. A few have fewer than three molars. Among xenartharans the number of molars is difficult to ascertain due to the transformation of all the teeth into similar often non-molar shapes.

Figure 1. Molar numbers in mammals. Four molars is the basal number. A few taxa add molars. Several lose one molar for a total of three. A few have fewer than three molars. Among xenartharans the number of molars is difficult to ascertain due to the transformation of all the teeth into similar often non-molar shapes.

According to Wikipedia: “The early marsupials have… three premolars and four molars. In other groups [derived marsupials] the number of teeth is reduced. Marsupials in many cases have 40 to 50 teeth, significantly more than placental mammals … and they have more molars than premolars.”

In the LRT
the situation is a little different. Four molars is the basal number for all mammals. You’ll find four molars in Sinoconodon. By contrast, you’ll find that Kuehneotherium, Amphitherium and Akidolestes increase that number to six while another sister, highly derived Ornithorynchus sheds all molars as an adult. Juramaia had three molars. In tiny, but adult, Hadrocodium only two molars were present.

Among marsupials, the creodonts beginning with Hyaenodon had only three molars. The odd wombats Vintana and Zalambdalestes also had three molars. In Vintana a tooth anterior to the molar row is vestigial and non-working.

In placentals, four molars remain the basal number. The fourth molar is lost by convergence in several clades (Fig. 1), but retained in Asioryctes, flying lemurs, Henkelotherium + Nambaroo, tenrecs + whales and Onychodectes at the base of the Condylarthra. Glyptotherium appears to have had six molars. Bradypus appears to have had four. Only the anterior one of five teeth in Bradypus is distinct from the others and they are all ventral to the jugal, hence the molar designation. By contrast, in the related  Peltephilus, the molars are vestiges and three premolars appear to be present.

Among pretherians
With maturation, the anterior premolars of Morganucodon are shed and not replaced, and a diastema forms behind the upper canine that is elongated over time as premolars are lost.

All this is very interesting
and points to the importance of establishing a cladogram of relationships before establishing phylogenetic ‘rules.’ for various traits.

 

 

A new nesting for two ‘odd bedfellow’ mammals

An ‘odd bedfellow’
is a taxon that doesn’t appear to fit where it currently nests. This has been called a “by default nesting” here because the more attractive true sisters are not present in the cladogram. That problem is called “taxon exclusion’ and often accompanies smaller more focused studies that do not glean their taxon lists from prior wide gamut, overarching studies, like this one. Some of those smaller studies include taxa that should not be included. Others exclude pertinent taxa.

For instance
pterosaurs don’t look like any archosaur. And fenestrasaurs (pterosaur sisters) are almost never included in pterosaur studies that usually favor archosaurs. Turtles don’t look like any archosauriform. Mesosaurs don’t look like any so-called anapsid(basal reptile). Caseasaurs don’t look like pelycosaurs. Vancleavea does not look like any archosaur. And the list goes on. These are traditional problems that continue to plague tetrapod paleontology.

This time in the large reptile tree
Vincelestes (Fig. 1) did not look enough like any basal placental and/or Carnivora. Ernanodon (Fig. 3) did not look enough like any pangolin or pangolin ancestor. The problem was, without more attractive taxa in the cladogram they nested by default in the large reptile tree where they did as ‘odd bedfellows.’ We’ve seen this before in a test when we deleted all current pterosaur and turtle sisters and they both nested with sauropterygians. The fact that there was massive convergence between certain marsupials and placentals just made the problem worse.

Figure 1. Vincelestes skull in two views. Now it nests with the creodont marsupial, Oxyaena (figure 2).

Figure 1. Vincelestes skull in two views. Now it nests with the creodont marsupial, Oxyaena (figure 2).

So
when creodonts were recently added to the large reptile tree, Hyaenodon and Oxyaena nested not with placentals, where traditional paleontologists thought should nest, but with marsupials, despite lacking four molars and epipubes. Their basal sister, Thylacinus, has very poorly developed epipubes and is losing molar 4. So that made the marsupial/placental convergence problem that much more difficult to solve.

Figure 2. Oxyaena, a credont/marsupial nests with Vincelestes.

Figure 2. Oxyaena, a credont/marsupial nests with Vincelestes. This marsupial lacks epipubes and has only three molars.

So now, with more taxa in the Marsupialia
Vincelestes now nests with the creodont/marsupial Oxyaena. Ernanodon now nests between the creodont/marsupial Hyaenodon and Oxyaena. Both look much more like their creodont sisters now. I think the ‘odd bedfellow’ problem has been solved.

Figure 3. Ernanodon used to nest poorly with pangolin ancestors, but not nests between the creodonts Hyaenodon (figure 4) and Oxyaena (figure 2).

Figure 3. Ernanodon used to nest poorly with pangolin ancestors, but not nests between the creodonts Hyaenodon (figure 4) and Oxyaena (figure 2).

The convergence problem in marsupials
has just gotten worse as four former placentals now nest with marsupials. We no longer can just count molars and look for epipubies. But we can still look for the septomaxilla, which were overlooked in these taxa. And the jugal extends back to the jaw joint, as in other marsupials (Oxyaena may be an exception, or the posterior jugal may be taphonomically missing).

Figure 1. Hyaenodon horrid us was the size of a large dog. This carnivorous marsupial was formerly considered a creodont.

Figure 4. Hyaenodon horrid us was the size of a large dog. This carnivorous marsupial was formerly considered a creodont. Now it nests with Ernanodon.

I have often encouraged detractors
to find poorly nested taxa. Well… here again, they had their chance and missed it. It’s good to clean up such problems. This is the process we call Science. And once again this demonstrates the importance of taxon inclusion in cladistic analysis. Every single taxon added to the large reptile tree (now at 798) reduces the chances that key sisters are missing when attempting to correctly nest new additions. It just keeps getting better and more complete. And problem solving, making little discoveries, like these, are wonderfully rewarding.

PS
I actually added a character to the LRT that scores the number of molars. We’ll look at that distribution in a future blogpost.

References
Bonaparte JF 1986. Sobre Mesungulatum houusayi y nuevos mamíferos Cretácicos de Patagonia, Argentina [On Mesungulatum houssayi and new Cretaceous mammals from Patagonia, Argentina]. Actas del IV Congreso Argentino de Paleontología y Biostratigrafía 2:48-61
Cope ED 1874. Report upon vertebrate fossils discovered in New Mexico, with descriptions of new species. Chief of Engineers Annual Report, Appendix, S. 589-606, U.S. Government Printing Office, Washington.
Ding SY 1979. A new edentate from the Paleocene of Guangdong. Vertebrata PalAsiatica 17:57–64. [Chinese 57–61; English 62–64].
Gaudrey A 1878. Les enchaînements du monde animal dans les temps géologiques mammifères tertiaires. F. Savy. (ed.) Paris 28pp. online here.
Horovitz I 2003. The type skeleton of Ernanodon antelios is not a single specimen. Journal of Vertebrate Paleontology. 23 (3): 706–8.
Kondrashov P and Agadjanian AK 2012. A nearly complete skeleton of Ernanodon (Mammalia, Palaeanodonta) from Mongolia: morphofunctional analysis. Journal of Vertebrate Paleontology. 32 (5): 983–1001. doi:10.1080/02724634.2012.694319.
Laizier L and de Parieu J 1838. Description et determination d’une machoire fossile appartenant a un mammifere jusqu’a pressent inconnu, Hyaenodon leptorhynchus. Comptes-rendus hebdomadaires des séances de l’Académie des Sciences, Paris 7:442.
Scott WM 1895. The osteology of Hyaenodon. Academy Natural Sciences Philadelphia Journal 9:499-536. online here.
Ting S, Wang B and Tong Y-S 2005. The type specimen of Ernanodon antelios. Journal of Vertebrate Paleontology 25(3):729-731.

wiki/Ernanodon
wiki/Vincelestes
wiki/Oxyaena
wiki/Hyaenodon

Immaturity in the Latest Devonian: Acanthostega

This blogpost is somewhat outside of the reptile arena,
but marginally pertinent, as you’ll see, nevertheless.

A recent paper by Sanchez et al. 2016
reports that none of the known Acanthostega specimens (Fig. 1) are fully mature, even at six years old. As we learned earlier, these Latest Devonian lobe fins and basal tetrapods were not involved in the transition to land, which occurred in the early Middle Devonian, as documented by footprints. But, as late survivors of that transitional phase, they do give us a good view as to what happened tens of millions of years earlier.

Figure 1. Size comparisons of lobe-find fish, their eggs and embryos. Latimeria at top, Ichthyostega in the middle, Acanthostega at bottom with hypothetical adult with reduced tail fin, egg and hatchling also shown. Figure 1. Size comparisons of lobe-find fish, their eggs and embryos. Latimeria at top, Ichthyostega in the middle, Acanthostega at bottom with hypothetical adult with reduced tail fin, egg and hatchling also shown.

Figure 1. Size comparisons of lobe-find fish, their eggs and embryos. Latimeria at top, Ichthyostega in the middle, Acanthostega at bottom with hypothetical adult with reduced tail fin, egg and hatchling also shown.

The Sanchez abstract reports
“A long early juvenile stage with unossified limb bones, during which [Acanthostega] individuals grew to almost final size, was followed by a slow-growing late juvenile stage with ossified limbs that lasted for at least six years in some individuals. The late onset of limb ossification suggests that the juveniles were exclusively aquatic,”

Hypothetical hatchlings
based on hip dimensions and analogy with Latimeria (Fig. 1) suggest adults were 5-6x longer than juveniles. In Latimeria the ratio is 9x. Coelocanths can live for 48 years. Gestation can be 12-13 months. No information exists for the age of sexual maturity, but it could be as late as 20 years in Latimeria.

So the Sanchez et al. discoveries
seem to make sense in terms of phylogenetic bracketing. Lobefins take a long time to mature and they live a long time thereafter.

References
Sanchez S, Tafforeau  P, Clack JA & Ahlberg PE 2016. Life history of the stem tetrapod Acanthostega revealed by synchrotron microtomography. Nature (advance online publication) doi:10.1038/nature19354