Antarctanax: a late-surviving basal synapsid, not a dino ancestor

Please see the notes in the following comments section. Most importantly after publication the authors report an errant scale bar, nearly doubling the apparent size of one of the pedes. 

Peecook, Smith and Sidor 2019
bring us news of a Early Triassic amniote from the Transantarctic Mountains, Antarctanax shackletoni (Figs. 1, 2), “known from a partial postcranial skeleton including cervical and dorsal vertebrae, a humerus, and both pedes.” 

Figure 1. Antarctanax manus and pes in situ with original tracing and color added here.

Figure 1. Antarctanax manus and pes in situ with original tracing and color added here.

if the scale bars are correct, and they seem to be, the smaller ‘pes’, the one surrounded by cervicals, is really a manus (Figs. 1, 2). Furthermore, the small manus matches the small humerus and radius. Added later: The scale were not correct, as noted at top.

Figure 2. Antarctanax manus and pes compared to those of Cabarzia and Aerosaurus, two basal synapsids.

Figure 2. Antarctanax manus and pes compared to those of Cabarzia and Aerosaurus, two basal synapsids. As you can see, basal synapsids rather quickly evolved similarly sized hands and feet.

The authors mislabeled
the robust, displaced metatarsal 5 as metatarsal 1, which lies beneath it (colored orange, Figs. 1, 2). Perhaps a reconstruction would have helped expose this error before submission.

The authors report,
“Our inclusion of A. shackletoni in phylogenetic analyses of early amniotes finds it as an archosauriform archosauromorph.” Their cladogram based on Ezcurra et al. 2014 nested Antarctanax in an unresolved polytomy with the basal archosauriforms, Proterosuchus, Erythrosuchus and Euparkeria. Their cladogram based on Ezcurra 2016 nested Antarctanax in an unresolved polytomy with other basal archosauriforms, FugusuchusSarmatosuchus. I am not aware of a manus or pes preserved for these two taxa. Of the above listed taxa, Proterosuchus (Fig. 3) comes closest, but has a hooked metatarsal 5 and metacarpal 3 is the longest, distinct from Antarctanax.


Figure 3. Synaptichnium compared to a slightly altered pes of Proterosuchus. Note a reduction of one phalanx in pedal digit 4 to match one less pad in the ichnite. The last two (or three phalanges) of pedal 4 are unknown in Proterosuchus.

This time it is not taxon exclusion, but bad timing.
When the manus and pes of Antarctanax are added to the large reptile tree (LRT, 1395 taxa), Antarctanax nests with basalmost synapsids, like Cabarzia (Figs. 2, 4) and Aerosaurus (Fig. 2). Aerosaurus was included in Ezcurra et al. 2014 and tested by Peecook, Smith and Sidor 2019. You’ll have to ask the authors why Antarctanax did not nest closer to Aerosaurus. Cabarzia trostheidei (Spindler, Werneberg and Schneider 2019, Fig. 3) could have influenced their thinking and scoring, but it was published only a few weeks ago, too late to include in their submission.

Figure 1. Cabarzia in situ and tracing distorted to fit the photo from Spindler, et al. 2019. Inserts show manus and pes with DGS colors and reconstructions. Scale bar = 5 cm.

Figure 4. Cabarzia in situ and tracing distorted to fit the photo from Spindler, et al. 2019. Inserts show manus and pes with DGS colors and reconstructions. Scale bar = 5 cm.

Peecock, Smith and Sidor did not provide a reconstruction
of Antarctanax, but online Discover magazine provided an in vivo painting and crowned it, “Dinosaur Relative Antarctanax.” According to the LRT, Antarctanax was a late-surviving (Early Triassic) basal member of our own lineage, the Synapsida, with a late Carboniferous genesis.

Therapsid synapsids were plentiful in Antarctica in the Early Triassic.
The headline should have focused on the unexpected presence of this sprawling, pre-pelycosaur, basal synapsid in the Mesozoic, surviving the Permian extinction event in this Antarctic refuge, alongside a closer relative of mammals, Thrinaxodon.

Ezcurra MD, Scheyer TM and Butler RJ 2014. The origin and early evolution of Sauria: reassessing the Permian saurian fossil record and the timing of the crocodile-lizard divergence. PLoS ONE 9:e89165.
Ezcurra MD 2016. The phylogenetic relationships of basal archosauromorphs, with an emphasis on the systematics of proterosuchian archosauriforms. PeerJ 4:e1778.
Peecook BR, Smith RMH and Sidor C 2019. A novel archosauromorph from Antarctica and an updated review of a high-latitude vertebrate assemblage in the wake of the end-Permian mass extinction. Journal of Vertebrate Paleontology e1536664 (16 pages) DOI: 10.1080/02724634.2018.1536664
Spindler F, Werneberg R and Schneider JW 2019. A new mesenosaurine from the lower Permian of Germany and the postcrania of Mesenosaurus: implications for early amniote comparative osteology. PalZ Paläontologische Gesellschaf

2 thoughts on “Antarctanax: a late-surviving basal synapsid, not a dino ancestor

  1. Hi Dave,
    I think you should know that I’ve been recently corresponding with Brandon Peecook (not Peecock) about some of the oddities you have brought up. Apparently the scale bar was not properly scaled in figure 4 of the paper (the one showing the right pes), causing it to appear significantly larger than the left (which you identify as a manus).

    Fortunately, table 3 of the paper shows discrete measurements of every identifiable tarsal and phalange, making it clear that the differences in size boil down to a few millimeters at most. When a size difference does exist, it’s usually a case of differences in preservation (a lot of the phalanges seem to have unfinished or crushed bone), or visibility due to the overlapping bones. This means it’s unlikely that the left pes is a manus, due to its similar proportions compared to the right.

    On a related note, this makes it possible to directly compare the metatarsal V in the left pes to the robust element which you identify as mt V (and they identify as mt I) in the right pes. The left mt V is clearly of an unusual shape, with a pronounced medial process and no clear flexor tubercules (as described in the text). The robust element in the right pes, on the other hand, is much more typical in its hourglass-like shape and easily visible flexor tubercules. It’s quite unlikely that it’s the metatarsal V, instead it’s likely the metatarsal I as the authors have argued. I cannot identify a bone under it (which you think is the “true” metatarsal I).

    In fact, a brighter picture ( shows that the supposed bone has a rough texture similar to that of the surrounding sediment, as well as apparently scoring due to mechanical preparation. I assume that the sediment was left there as a base for the metatarsal above it, so that it’s not left suspended in midair. Plus, I doubt that none of the three different people working on the specimen would have overlooked a bone in that space. There’s only a single layer of overlap here, it’s not like this is a Green River fish or a Jehol fossil. The bit you portray as poking out distally looks like a broken first phalanx to me, and apparently that’s the interpretation of the describers as well, since they list two phalanges on the first digit of both feet in table 3.

    Admittedly the metatarsal I in the picture of the left pes looks much thinner than the robust metatarsal in the right one, but this is because the right mt I was preserved in a different orientation, with the right mt I visible in ventral view and the left one visible laterally, due to a partial collapse in the block (you can clearly see a discreet change in height based on the line of shadows here:×1024.jpg).The right mt I also might look wider than it did in life due to compression of the block. None of this is last paragraph is conjecture, these are the reasons given to me by Peecook during my conversation with him.

Leave a Reply

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

You are commenting using your 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.