Multituberculate origins: Two views

A recent paper by Csiki-Sava et al. 2018
described a new multituberculate, Litovoi. The authors also produced a cladogram of multituberculates (Fig. 1).

Long have I wondered
which taxa were considered outgroups for the multituberculates in modern paleo-thinking. Thanks to Csiki-Sava et al. now we know they nested Haramiyavia as the outgroup (Figs. 1, 2).

Or is that solution possible
only due to taxon exclusion?

Figure 1. Cladogram of multituberculate origins and interrelations by Csiki et al. 2018.

Figure 1. Cladogram of multituberculate origins and interrelations by Csiki-Sava et al. 2018.

By contrast
the large reptile tree (LRT, 1201 taxa) nested multis with rodents and plesiadapids (Fig. 2). Haramiyavia nested far distant, as a pre-mammal, not far from Pachygenelus. While the .nex files include all the details, the illustration of skulls (Fiji. 3) compares the two hypotheses of relationships.

Figure 2. Cladogram of multituberculate origins according to the LRT.

Figure 2. Cladogram of multituberculate origins according to the LRT

Taking the skulls from Figure 2
(Fig. 3) one can compare the traditional hypothesis of multituberculate origins with that recovered by the LRT, offering a sort of short hand of all the data scores. One should appear to demonstrate a gradual accumulation of traits. The other should appear to not do so well. Which outgroup lineage appears to have more multituberculate traits in your judgement?

Figure 3. Comparing multituberculate origins: Cziki-Sava et al. vs. LRT.

Figure 3. Comparing multituberculate origins: Cziki-Sava et al. vs. LRT. Gray background taxa are multituberculates. The morphological gap between Haramiyavia and multis is great, much greater than the gap between Paramys and multis.

The closest living relative of long extinct multituberculates,
according to the LRT is Daubentonia, the aye-aye (Figs. 4, 5), once considered a lemur-like primate, but here nesting with extinct Carpolestes and the multis. No other primate, living or extinct (Plesiadapis is also a rodent relative in the LRT), has such a suite of bony traits, including those very large, rodent-like (due to homology) incisors.

Figure 1. Daubentonia was considered a primate for over 150 years. Here it nests with Plesiadapis, rodents and rabbits.

Figure 4. Daubentonia was considered a primate for over 150 years. Here it nests with Plesiadapis, rodents and rabbits.

According to Wikipedia
“The multituberculates existed for about 166 or 183 million years, and are often considered the most successful, diversified, and long-lasting mammals in natural history. They first appeared in the Jurassic, or perhaps even the Triassic, survived the mass extinction in the Cretaceous, and became extinct in the early Oligocene epoch, some 35 million years ago. The oldest known species in the group is Indobaatar zofiae from the Jurassic of India, some 183 million years ago, and the youngest are two species, Ectypodus lovei and an unnamed possible neoplagiaulacid, from the late Eocene/Oligocene Medicine Pole Hills deposits of North Dakota. If gondwanatheres are multituberculates (all tested taxa are not in the LRT), then the clade might have survived even longer into the Colhuehuapian Miocene in South America, in the form of Patagonia peregrine.”

Figure 2. Skeleton of Daubentonia (aye-aye). Like other plesiadapids, it convergences with the lemuroid primates.

Figure 5. Skeleton of Daubentonia (aye-aye). Like other plesiadapids, it convergences with the lemuroid primates.

Employing taxon inclusion,
the LRT presents a heretical and more parsimonious hypothesis of multituberculate origins (Figs 2, 3) that tests Haramiyavia and over 1000 other possible candidates.

To test this hypothesis,
simply add the above suggested relevant taxa to your favorite wide gamut phylogenetic analysis and run. Let me know if your analysis then confirms the LRT—or do you find yet another origin/set of outgroups for the multituberculates? Haramiyavia has very few multi traits, far fewer than rodents and Daubentonia.

References
Csiki-Sava ZVremir MMeng JBrusatte SL and Norell MA 2018. Dome-headed, small-brained island mammal from the Late Cretaceous of Romania. 

https://en.wikipedia.org/wiki/Haramiyavia
https://en.wikipedia.org/wiki/Multituberculata

9 thoughts on “Multituberculate origins: Two views

  1. “one can compare the traditional hypothesis of multituberculate origins with that recovered by the LRT, offering a sort of short hand of all the data scores. One should appear to demonstrate a gradual accumulation of traits. The other should appear to not do so well. Which outgroup lineage appears to have more multituberculate traits in your judgement?”

    The subjective appearance of this is something nobody in phylogenetics considers important, since the algorithm in PAUP/TNT will guarantee it is true for the examined characters. How many papers do you see arguing their topology is correct because of the gradual accumulation of traits? The same amount arguing for their correctness based on the low number of MPTs. Cite a phylogeneticist who argues these are important for topological accuracy.

  2. Maximum parsimony = minimum change = gradual accumulation of traits. So, I cite all phylogeneticsts who use PAUP and kin. Subjective appearance = shorthand for examining hundreds of characters one at a time. It looks like a duck… quacks like a duck… etc. etc.

    And you know that.

    • Maximum parsimony has nothing to do with gradual vs. rapid change. You’re just falsely equating concepts like your previous claim that the LRT “resolved” controversies. Ditto subjective appearence, which is the broad similarity in form, missing the minutiae of braincase foramina, dental cusp morphology, molecular details, etc.. You just write those off as unimportant, but you have no valid reason for doing so other than they disagree with the LRT.

  3. Okay. Here goes: Mickey said, “Maximum parsimony has nothing to do with gradual vs. rapid change.” Indeed, it does. By definition. That’s why I look like my parents. Not someone else’s parents.

    • Forgive me, but I really don’t see how the speed of the change has anything very much to do with the degree of resemblance. Although a very rapid divergence might require some justification, e.g. heavy selection pressure in favour of novel adaptations.

      • Speed is not mentioned here. Only taxon exclusion. When I talk about gradual accumulation of traits, that may still take tens of millions of years. To my knowledge (willing to be wrong here) prior workers have mentioned that multituberculates were rodent-like, but no one has tested them generically with rodents, carpolestids, plesiadapids and other relatives in the clade Glires.

        On the other hand, if it had been established by others that multis and rodents were close to one another, and I had instead proposed that Haramiyavia was a closer relative, despite the dissimilarities, what would your arguments be in support of that relationship?

        Make sure we talk about the news (the taxa) here, and not judge the bearer of the news.

  4. Mickey, You may be experiencing yet another bout of cognitive dissonance. Imagine that the illustration was produced by Dr. John Wible… easier to imagine the veracity of the claims then, isn’t it? Hard to imagine that so many workers for so long got it wrong (turtle, whale, reptile, pterosaur origins, archosaur membership, Chilesaurus, Lagerpeton, etc. etc.) almost always due to taxon exclusion. A good way to test the veracity of any claim is to repeat the experiment and tell us what resulted.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.