Wortman 1902 was first: Three origins for placental mammals hypothesis

Like amniotes, placentals are defined by their mode of reproduction,
not their bones, skin or metabolism.Therefore fossils mammals present a phylogenetic problem. Some workers resolve that issue with genes. Others with teeth. The presence of novel epipubic bones is restricted to marsupials, so that helps. Unfortunately, not all marsupials have epipubic bones.

The last common ancestor
of all placentals in the large reptile tree (LRT, 2324 taxa) is tiny extinct Late Cretaceous Asiatherium.

That’s a phylogenetic problem because Asiatherium is an ancestor of many extant marsupials (= with pouches) that, in turn, are ancestors of placentals.

The LRT recently recovered three –
now reduced to two – mammal clades that reproduce with a placenta: Placentalia 1 and Placentalia 2. In Placentalia 2 extant tree shrews, here represented by extant Ptilocercus and Tupaia, are dual last common ancestors.

Wortman 1902 was the first to split the Placentalia into three clades
Matthew 1906 wrote, “In Wortman’s view the Carnivora. Creodonta, Insectivora, etc., arise each as a separate branch from the Cretaceous marsupials. This view is only held conjecturally in the case of other groups, but is quite speciticallv stated in regard to Creodonts and true Carnivores.”

Wortman had that hypothesis first. That was long before software-assisted phylogenetic analysis, Internet data gathering and a century of additional fossil recovery.

Matthew 1906 argued against splitting the Placentalia
“I do not think, however, that the evidence, even as stated by Wortman, supports this view, and quite naturally he is inclined to lay emphasis upon the marsupialoid features of the creodont skull. On the contrary, I think it is safe to say that if we set aside superficial and adaptive characters, and rest principally upon deep-seated resemblances such as are found in the characters of the base of the skull, the dental and dorsolumbar formula, etc., we find every known creodont very much nearer to the modern Carnivora than to the modern marsupials. On the other hand, the little that is known of Cretaceous marsupials bears distinctly the marsupial stamp in every detail and does not show any essential approach to the early-placentals.”

As you can see, Matthew’s argument depends on listing traits, which can and do converge. Phylogenetic analysis employs the last common ancestor method.

Genes cannot help in deep time studies.

Getting back to the LRT…
In Placentalia 1 the situation is less clear than in Placentalia 2. The extinct genus Vulpavaus, is the last common ancestor of Placentalia 1 in the LRT. It’s a phylogenetic bracketing assumption that Vulpavus reproduced with a placenta because extant descendant taxa from three clades, Eupleres, Talpa and Lemur, all reproduce with a placenta.

That leaves several fossil taxa in a gray zone
between known marsupials, like Chironectes and Thylacinus and basal placentals like Eupleres, Talpa and Lemur,

Sinopa grangeri
(Fig 1) is one of those gray zone Proplacentalia 1 taxa. Others include Chriacus + Kopidodon, four members of the Palaeosinopa clade and two more members of the  Sinopa clade.

These could all be marsupials or all placentals. Some could be placentals. Some could be transitional in some way we don’t know yet.

Matthew 1906 discussed carnivorous marsupials
in his paper on the osteolgy of Sinopa (Fig 1). Matthew listed several traditional, typical, standard ‘marsupial’ traits “showing that they are far removed from Sinopa or from any of the Creodonta. In Sinopa in Oxyaena and Hyaenodon and probably in Patriofelis, the only other creodonts in which the dorsolumbar formula is known, it is twenty, as in all Carnivora, and these genera have all evidently descended from primitive carnivore ancestors, whose principal distinctions are given above.”

After testing in the LRT these three taxa are either marsupials or members of the gray zone Proplacentalia 1.

There are several clades of carnivorous marsupials.
A clade with only two tested members, Thylacoleo and Wakaleo, is derived from omnivorous to herbivorous sugar gliders in the LRT. Most other carnivorous marsupials are associated with members of the Creodonta.

Wikipedia reported that creodonts
are extinct carnivorous placentals that lived from the Paleocene to Miocene. “Originally thought to be a single group of animals ancestral to the modern Carnivora, this order is now usually considered a polyphyletic assemblage of two different groups, the oxyaenids and the hyaenodontids, not a natural group. Most modern paleontologists agree both “creodont” families are related to Carnivora, but are not their direct ancestors. It is still unclear how closely the two families are related to each other.”

Here in the LRT, the Tasmanian wolf, Thylacinus, a recently extinct marsupial, is a member of the Creodonta. No members give rise to placental descendants.

In the LRT creodonts go back
to Early Cretaceous Vincelestes.

The Proplacentalia 1 clade
would appear to document a rapid radiation of placentals following the asteroid impact 66 mya – except that Eupleres and Fossa were isolated on Madagascar prior to the Late Jurassic splitting of that island from several continents and a subcontinent.

Meanwhile, the Proplacentalia 2 clade
has multituberculate members going back to the Middle Jurassic, all arboreal.

That indicates a near simultaneous dual origin for placental reproduction in mammals.

References
Halliday TJD, Upchurch P and Goswami A 2015. Resolving the relationships of Paleocene placental mammals. Biological Reviews. 92 (1): 521–550.
Matthew WD 1906. The Osteology of Sinopa, a Creodont Mammal of the Middle Eocene.
Proc. U. S. Nat. Mus., Vol. XXX, pp. 203-233, pl. XVI.
Wortman JL 1901-1903. Studies of Eocene Mammalia in the Marsh Collection, Peabody Museum. Am. Jour. Sci. 11:333–348.

wiki/Dasyuromorphia

Source: https://pterosaurheresies.wordpress.com/2024/08/22/wortman-1902-was-first-three-origins-for-placental-mammals-hypothesis/