Dongoconodon: mammal? or pre-mammal?

Mammals are described by females having breasts – because infants have no teeth. Young juveniles develop milk teeth at weaning. Older larger juveniles replace those with their final set of adult teeth, including a few more erupting for the first time in larger jaws.

Having two sets of teeth in one lifetime is called diphyodonty. Mammals are known from the Late Triassic onwards.

So why in Hell would Nature publish a headline by Bi et al 2026 describing their new genus Dongoconodon (IMMNH-PV01700, Fig 1) as ‘a polyphodont mammal‘?

Mammals are definedby the last common ancestor method. So are all clades. Remember that as you read this. Everyone involved here appears to have forgotten this key to taxonomy.

Screenshot ” data-large-file=”https://pterosaurheresies.wordpress.com/wp-content/uploads/2026/10/dongoconodon588.jpg?w=584″ src=”https://pterosaurheresies.wordpress.com/wp-content/uploads/2026/10/dongoconodon588.jpg?w=588″ alt=”" class=”wp-image-98142″ srcset=”https://pterosaurheresies.wordpress.com/wp-content/uploads/2026/10/dongoconodon588.jpg 588w, https://pterosaurheresies.wordpress.com/wp-content/uploads/2026/10/dongoconodon588.jpg?w=150 150w, https://pterosaurheresies.wordpress.com/wp-content/uploads/2026/10/dongoconodon588.jpg?w=300 300w” sizes=”(max-width: 588px) 100vw, 588px” />

Here’s what the authors wrote in their abstract: “Diphyodonty (the presence of two generations of teeth) characterizes extant mammals and has long been regarded as pivotal to the evolution of other mammalian traits, including precise occlusion and determinate growth.

True.

“Nevertheless, our understanding of the pattern of dental replacement in Mesozoic mammals is incomplete.

False, but go on… I’m listening…

“Here, we report a nearly complete skeleton of an Early Cretaceous eutriconodontan that is the only mammal to our knowledge that appears to have multiple generations (polyphyodonty) at some loci in the four tooth types.

Not permitted. This is headline grabbing under the false pretense of novelty.

In the auditory apparatus, its ossified Meckel’s cartilage has slight contact with the mandible, supporting the interpretation that Meckel’s separated from the mandible before it separated from the middle ear ossicles.

Although the separation of auditory elements from the posterior jaw is part of the transition to the mammal condition it is not the describing character. Placental multituberculates show this process can reverse itself, something Bi et al were not aware of due to taxon exclusion (once again!).

Taxon exclusion remains the number one problem in paleontology.

“Reappearance of polyphyodonty and molariform replacement in the new fossil and related gobiconodontids is probably associated with their ecomorphological niche and carnivorous diet.”

This is speculative = imaginary and not supported by description, definition or cladistics.

According to the LRT this was no reappearance. Instead it was the last appearance of polyphyodonty prior to the origin of diphyodont mammals as defined by the last common ancestor method – the only method paleontologists should use.

After analysis Dongoconodon was recovered as the sister taxon to a clade including Spinolestes, Gobiconodon, and Repenomamus, which together comprise the family Gobiconodontidae. Those are all pre-mammals (= pre-last common ancestor at the Monotreme/Marsupial split) in the large reptile tree (LRT, 2340 taxa).

So Dongoconodon cannot be a mammal by description or definition.

Here’s where the problem started. Bi et al ‘pulled a Larry Martin’. See if you can spot it in their introduction: “…diphyodonty and a fully detached middle ear are widely regarded as two key diagnostic features of extant mammals and represent fundamental transitions in mammalian evolution.”

In the LRT the only diagnostic feature is diphyodonty, but that’s beyond the point.

The point is only the last common ancestor method can be used for determining clade membership. That’s the definition of a clade: The last common ancestor and all descendants. The description = diagnosis of a clade is something else.

Academics teach this in their lecture halls and textbooks. So what happened??

In the LRT Dongoconodon does not nest within the last common ancestor of the clade Mammalia = all Monotremes + all Marsupials, their last common ancestor and its descendants.

Never use one to a dozen traitsfor determining clade membership. That’s called ‘Pulling a Larry Martin‘ after a KU professor who used to tease and teach students by asking them about traits, then revealing a convergent trait somewhere else on the Tree of Life.

And Nature’s editors fell for it. So did the six authors. So did the referees.

Consider this a cautionary tale. Take a lesson from their collective mistake.

Academics: you should know better. Remember what you learned in school.

Readers: sorry you had to hear this from an unpaid unschooled amateur.

Final note: Dongocondoon (Fig 1) is a great specimen: fully articulated, sub-adult and well preserved!

References

Bi S et al (5 co-authors) 2026. A polyphyodont, semiaquatic eutriconodontan mammal from the Early Cretaceous of China. NatureCommunications 17, 10032 (2026). https://doi.org/10.1038/s41467-026-77339-w

wiki/Dongoconodon

Source: https://pterosaurheresies.wordpress.com/2026/10/05/dongoconodon-mammal-or-pre-mammal/