Reassessing Maiopatagium: now it’s a Jurassic porcupine!

Modified August 21, 2018 with the note that a procoracoid and coracoid were likely present in Jurassic rodents. These traits appear to be atavisms since taxa between prototheres and Jurassic rodents do not have these bones. 

Another case of taxon exclusion…remedied.
Earlier we looked at the Jurassic mammal, Maiopatagium, a putative glider, surrounded by a deep halo of long, straight hair. Meng et al. 2017 nested Maiopatagium between Sinoconodon and Haldanodon, taxa more primitive than mammals.

By contrast
the large reptile tree (LRT, 1235 taxa) nested Maiopatagium with Vilevolodon and Shenshou, two Jurassic arboreal rodents.

Now with 24 more taxa,
and several new ones from the rodent clade, the LRT nests Maiopatagium with the only tested porcupine, the small arboreal Coendou.

Figure 1. Subset of the LRT focusing on Scandentia + Glires. Yellow-green taxa are Jurassic in age.

Figure 1. Subset of the LRT focusing on Scandentia + Glires. Yellow-green taxa are Jurassic in age.

With this nesting
that halo of long straight hair on Maiopatagium
(Fig. 4) takes on a new identity as a pelage of still soft pre-quills, similar to a closely related taxon, Chinchillanesting with a former enigma taxon, Neoreomys

To no one’s surprise,
the guinea pig (genus: Cavia) nests with the pig-sized capybara (genus: Hydrochoerus). All but Maiopatagium are widely recognized members of the Hystricomorpha clade of rodents. The presence of Maiopatagium in this rodent clade supports the previously reported Jurassic radiation and dispersal of rodents (Fig. 1) currently represented by  a few specimens not widely recognized as rodents, nor tested against rodents. Porcupines and chinchillas were not in the Meng et al. taxon list.

Shifting Maiopatagium in the LRT to Sinoconodon adds 49 steps. Shifting to the more primitive Haldanodon adds 58 steps.

Distinct from all extant and extinct rodents Maiopatagium was reported to have a small coracoid and pro-coracoid, traits that disappear in therian mammals. This could be an atavism (= reversal) or it could be a misinterpretation of a crushed process of the scapula that appears in other hystricomorphs (Fig. 2). Vilevolodon has a protothere-like pro-coracoid and coracoid and it is medial to the scapula, not lateral as shown below. Interesting that similar structures appeared medial and lateral to the shoulder joint by convergence.

Figure 2. Possible source for the coracoid and procoracoid in Maiopatagium as crushed parts of the acromion process on other hystricomorphs.

Figure 2. Possible source for the coracoid and procoracoid in Maiopatagium as crushed parts of the acromion process on other hystricomorphs. At left is Hydrochoerus, the capybara. Above right is Cavia, the guinea pig. Lower right is Maiopatagium from Meng et al. 2017. Crushing would tend to break this fragile process. 

When the skull of Maiopatagium
nests with rodents we should consider the possibility that it may have included a large braincase (Fig. 3) not figured or restored by Meng et al. 2017 (Fig. 4).

Figure 3. Maiopatagium skull revised with extended, rodent-like cranium. Compared to figure 4. The anterodorsal naris is a hystricomorph trait. So is the premaxilla-frontal contact overlooked by Meng et al. 

Figure 3. Maiopatagium skull revised with extended, rodent-like cranium. Compared to figure 4. The anterodorsal naris is a hystricomorph trait. So is the premaxilla-frontal contact overlooked by Meng et al.

I was never able to see the gliding membrane
distinct from the halo of long hairs on Maiopatagium (Fig. 4) as described by Meng et al. 2017. No related taxa in the LRT are gliders.

Figure 2. Maiopatagium images from Meng et al. with the addition of a braincase restored here.

Figure 4. Maiopatagium images from Meng et al. with the addition of a braincase restored here. The pes has a new reconstruction (Fig. 5) than shown here.

The porcupine Coendou prehensilis
(Fig. 5) is the closest living relative to Maiopatagium in the LRT. Yes, the tooth shapes are distinctly different, but tooth shapes are highly variable and these taxa are separated by 160 million years. The limbs are longer in the Jurassic taxon and the hair has not yet turned into quills. The LRT does not test every trait. However, traits in the LRT nest Maiopatagium as a primitive porcupine and less likely to glide than originally figured.

Figure 4. Coendou, the extant prehensile-tailed porcupine, nests with the Jurassic Maiopatagium in the LRT. No other taxon nests closer among the 1268 tested.

Figure 5. Coendou, the extant prehensile-tailed porcupine, nests with the Jurassic Maiopatagium in the LRT. No other taxon nests closer among the 1268 tested.

The side-by-side alignment of the calcaneum and astragalus
figured by Meng et al. (Fig. 4) is yet another pre-therian trait (see Eomaia for the first shift to the therian state). Rodents don’t have this type of ankle (Fig. 5), so when you see it in the rodent clade we might count this as an atavism… possibly because Maiopatagium could have been hanging from branches or descending tree trunks head first and rotating the ankle, as squirrels do. The other possibility is a misinterpretation of the tarsals by Meng et al. An alternate reconstruction is shown here (Fig. 6).

In the porcupine pes,
please note the large flat bone arising from the medial tarsals (Fig. 5). The chinchilla does not have this disk, but Maiopatagium does (Fig. 6). It is an atavism arising from digit zero on the pes. Atavisms like this form the spur on the screamers.

Figure 6. The pes and tarsus of Maiopatagium traced and reconstructed with DGS methods compared to original art by Meng et al. 2017 (drawing).

Figure 6. The pes and tarsus of Maiopatagium traced and reconstructed with DGS methods compared to original art by Meng et al. 2017 (drawing). The porcupine, Coendou, also has a small digit 1 and a medial disk (tarsal zero) arising from the tarsus. The calcaneum appears to be crushed into several pieces, so the ‘calcar’ may be a broken artifact. No sister taxa have the Meng et al. ankle. Tarsal 5 and the lateral centrale (cuboid) are also separate.

Added almost a day later:
the pes of another specimen, BMNH1133 (from Meng et al. 2017, Fig. 7) compared to Rattus the rat. Pretty similar when reconstructed, aren’t they?

Figure 7. Another pes from Meng et al. 2017, this time reconstructed and compared to Rattus the rat. All the bones are there in just about the same shape and interrelation.

Figure 7. Another pes from Meng et al. 2017, this time reconstructed and compared to Rattus the rat. All the bones of the tarsus are there in just about the same shape and interrelation. The digits differ in proportion. Note the matching of the tibia-fibula width to a typical narrowly stacked astragalus and calcaneum.

 

References
Kermack KA, Kermack DM, Lees PM and Mills JRE 1998. New multituberculate-like teeth from the Middle Jurassic of England. Acta Palaeontologica Polonica 43(4):581-606.
Meng Q-J, Grossnickle DM, Liu D, Zhang Y-G, Neander AI, Ji Q and Luo Z-X 2017.
New gliding mammaliaforms from the Jurassic. Nature (advance online publication)
doi:10.1038/nature23476

wiki/Hydrochoerus
wiki/Maiopatagium
wiki/Coendou
raftingmonkey.com/Neoreomys
wiki/Brazilian_guinea_pig
wiki/Chinchilla

1 thought on “Reassessing Maiopatagium: now it’s a Jurassic porcupine!

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.