Liaoningosaurus: perhaps not an ankylosaur

This one goes back several years… with several updates!
Xu, Wang and You 2001 described what they thought was a juvenile ankylosaur, Liaoningosaurus paradoxes (Early Cretaceous, Yixian Formation) featuring “a large bony plate (somewhat shell-like) shielding the abdomen.” They tested it against only 13 other taxa and nested it outside the nodosaurs and outside the ankylosaurs…with no taxa between it and Stegosaurus.

Figure 1. Several specimens of Liaoningosaurus crushed flat plus a lateral view based on original holotype tracings.

Figure 1. Several specimens of Liaoningosaurus crushed flat plus a lateral view based on original holotype tracings.Note the lizard-like sprawling limbs in situ, a product of taphonomic crushing. Like all dinos, this one also had vertical limbs. Only a few small osteoderms are identified.

Xu et al. report:
“Diagnosis. An ankylosaurian that differs in having: shell-like ventral armour, trapezoidal sternum with slender and distally pointed posterolateral process and short medial articular margin, and pes more than twice as long as manus.”

Perhaps Xu et al. focused on ankylosaurs and nodosaurs
because all the specimens of Liaoningosaurus I have seen in publications or online (Fig. 1) have been crushed flat, with ribs spread out like ankylosaur ribs. Moreover, the pelvis was very wide, with limbs beneath the ilia, like those in ankylosaurs.

A closer look
(Fig. 1) reveals the ribs would have enclosed a deeper chest, not a wider one, though not as relatively deep as in Stegosaurus. Other more primitive stegosaurs likewise had a shorter, rounder torso cross-section.

the limbs are preserved sprawling, like those of the horned lizard, Phrynosoma. No dinosaur had sprawling limbs, so it’s okay to bring in both the limbs and the ribs (Fig. 1).

basal stegosaurs also have a very broad pelvis with limbs rotating beneath the ilium. Considering how closely ankylosaurs and stegosaurs match each other in so many traits, it is a tribute to the LRT that it recovers them in separate clades, separated by bipedal agile taxa like Lesothosaurus and Heterodontosaurus. It is unlikely that ankylosaurs ever reared up on their hind limbs, but stegosaurs appear to be able to do this.

Osteoderms are rare in Liaoningosaurus,
which is odd for an armored ankylosaur.

Ankylosaur teeth and stegosaur teeth greatly resemble one another and also resemble Liaoningosaurus teeth (Fig. 1), despite the great difference in size.

There are five digits on the manus
in Liaoningosaurus (Fig. 1) and metacarpal #5 is as long as #4. Unfortunately, ankylosaurs and kin in the LRT lack a preserved manus, but a look through the Princeton Field Guide to Dinosaurs (Paul 2010) finds no similar ankylosaur manus.

Arbour et al. 2014 report, “Examination of the holotype of Liaoningosaurus paradoxus, IVPP V12566, indicates that the ventral “plastron” is better interpreted as epidermal scales, because the broken edges do not reveal any bony histology.” Readers will note that the odd ventral plate (closeup in Fig. 2) does not appear in other Liaoningosaurus specimens (Fig. 1), but they are exposed dorsally.

Figure 2. Liaoningosaurus ventral patch. Note the scales.

Figure 2. Liaoningosaurus ventral patch. Note the scales.

The large reptile tree (LRT, 1005 taxa) includes several more ornithischian taxa, though fewer taxa with armor. In the LRT Liaoningosaurus nests between Scutellosaurus and Stegosaurus, several nodes away from the other armored ornithischians, Minmi and Scelidosaurus. No skull traits were tested in Liaoningosaurus due to the low resolution of the available images.

The armored ornithischians (stegosaurs and ankyosaurs) are so similar
to one another they are traditionally nested in one clade: Thyreophora. By contrast, the LRT separates ankylosaurs from stegosaurs. Here the few hind limb traits that separate Liaoningosaurus from Scelidosaurus and/or Minmi and ally it with Scutellosaurus and/or Stegosaurus include the following:

  1. Tibia/femur ratio not less than 1:1
  2. Fibula not appressed to tibia
  3. Fibula diameter not > half tibia diameter
  4. Metatarsus not compact
  5. Metatarsal 1 < half metatarsal 3
  6. Metatarsal 1 not > half metatarsal 4
  7. Metatarsals 2 and 3 align beyond p1.1
  8. Pedal 4 length <  metatarsal 4

Perhaps better imagery
of the skull and other parts will add to or modify this list and nesting.

The addition of a basal ankylosaur
with these traits would nudge Liaoningosaurus toward the ankylosaurs. In the LRT ankylosaurs were derived from large, armored, lumbering Scelidosaurus. By contrast, the stegosaurs were derived from small, agile Lesothosaurus and Scutellosaurus. So finding a small armored dinosaur with the above list of traits, even if it is a juvenile, should suggest taking a close look at its stegosaur affinities, despite the initial appearance of a wide round horned-lizard-like torso.

Ji et al. 2016 found fish within the torso (but not restricted to the gut) of a Liaoningosaurus suggesting a fish diet, rather than an herbivorous one.

Xu et al. 2001 reported, “Liaoningosaurus has an unusual combination of characters and it might (for example) represent a third ankylosaur lineage.” Perhaps one closer to stegosaurs. Xu et al. 2001 also report, “all manual and pedal unguals claw-shaped.” At present the manual unguals do not appear to be claw-shaped, with the the exception of #3, as in stegosaurs… AND as in ankylosaurs.

Xu X, Wang XL and You HL 2001. A juvenile ankylosaur from China. Naturwissenschaften 88:297. doi:10.1007/s001140100233
Ji Q, Wu X, Cheng Y, Ten F, Wang X and Ji Y 2016. Fish-hunting ankylosaurs (Dinosauria, Ornithischia) from the Cretaceous of China. Journal of Geology, 40(2).
Thompson RS, Parish JC, Maidment SCR and Barrett PM 2011. Phylogeny of the ankylosaurian dinosaurs (Ornithischia: Thyreophora). Journal of Systematic Palaeontology. 10 (2): 301–312. doi:10.1080/14772019.2011.569091
Arbour VM, Burns ME, Bell PR and Currie PJ 2014. Epidermal and dermal integumentary structures of ankylosaurian dinosaurs. Journal of Morphology, 275(1): 39-50. doi:10.1002/jmor.20194


Are the armored ornithischian dinosaurs diphyletic?

The Thyreophora are the armored ornithischians, a putative clade that includes Scutellosaurus (Fig. 1), Scelidosaurus (Fig. 2), ankylosaurs, nodosaurs and stegosaurs. This clade makes sense and is universally accepted because all the members are armored with ossified scutes and many members swung their clubbed or spiked tails to defend themselves. The Eurypoda is a derived clade within the Thyreophora that includes the ankylosaurs and stegosaurs. It does not include basal taxa like Scutellosaurus and Scelidosaurus.

According to Wikipedia
“Thyreophora was first named by Nopcsa in 1915. Thyreophora was defined as a clade by Paul Sereno in 1998, as “all genasaurs more closely related to Ankylosaurus than to Triceratops“. Thyreophoroidea was first named by Nopcsa in 1928 and defined by Sereno in 1986, as “ScelidosaurusAnkylosaurus, their most recent common ancestor and all of its descendants”. Eurypoda was first named by Sereno in 1986 and defined by him in 1998, as “Stegosaurus, Ankylosaurus, their most recent common ancestor and all of their descendants”.

Figure 1. Scutellosaurus is a small armored ornithischian transitional between Lesothosaurus and Stegosaurus.

Figure 1. Scutellosaurus is a small armored ornithischian transitional between Lesothosaurus and Stegosaurus. This is a new reconstruction for the skull, which is largely unknown. This one is not as tall, in accord with the cervical vertebrae. This long whip-like tail may have been the precursor to a shorter spiked tail in the related Stegosaurus.

At odds with this
the large reptile tree splits two putative thyreophorans, Scutellosaurus and Scelidosaurus (Fig. 3). Readers have understandably wondered about and questioned this heretical split.

the origin of the armored Ornithischia could use a few more fossils. Traditional paleontologists don’t yet recognize basal ornithischians like Chilesaurus and Daemonosaurus, which are relatively large taxa. Jeholosaurus and Haya are small and late survivors of a basal radiation. Scelidosaurus and ankylosaurs are closer to this base. More derived taxa include some smaller forms like Lesothosaurus and Scutellosaurus separated from Scleidosaurus by several taxonomic nodes that include the basalmost heterodontosaurs, ceratopsians and camptosaurs/iguanodonts/hadrosaurs.

Traditional paleontologists
like Holz 2011, nest Eocursor and Fabrosaurus (Lesothosaurus) at the base of the Ornithischia. By contrast, Butler et al. 2011 nested Pisanosaurus and Heterodocntosauridae at the base of the Ornithischia. Not sure what the outgroup is to either of these studies as neither employed Daemonosaurus or Chilesaurus. More here at Wikipedia.

Figure 2. Scleidosaurus is an armored basal ornithischian nesting between Pisanosaurus and Emausaurus + ankylosaurs.

Figure 2. Scleidosaurus is an armored basal ornithischian nesting between Pisanosaurus and Emausaurus + ankylosaurs.

the wide-bodied, short-skulled ankylosaur, Minmi, to the large reptile tree nests it with the large bulky armored Emausaurus and Scelidosaurus.

tall-bodied, long-skulled Stegosaurus nests it with the much smaller, short-legged, armored Scutellosaurus. So the armored ornithischian split remains here (Fig. 3). The Thyreophora appears to be diphyletic based on the few basal taxa included here.

Figure 3. Phytodinosauria. Purple nodes are traditional thyreophorans.

Figure 3. Phytodinosauria. Purple nodes are traditional thyreophorans. Moving the two taxa back together increases the MPT score by at least 9 steps, a relatively low number which is a tribute to their many convergent traits. Deletion of Scutellosaurus does not change the tree topology.

There are problems
with scoring some members of the armored ornithischians. For instance, Scutellosaurus is known from disarticulated and broken bones and only a few skull bones. Perhaps some day more complete skulls will be found. On the other hand, ankylosaurs fuse skull bones and cover them with additional ossifications of armor. So sutures are difficult to determine and the temporal skull fenestra disappear.

Scelidosaurus has tall neural spines, but ankylosaurs do not. Scutellosaurus does not have tall neural spines, but Stegosaurus does. So there is a curious mix and match of individual traits.

Stegosaurs are tall and narrow in the torso, a trait further emphasized with those tall plates. Anklyosaurs are the opposite, with an extremely wide and low torso and generally low plates, but Crichtonsaurus (Fig. 4) has tall plates. It has not been added to the large reptile tree yet.

Deletion of Scutellosaurus does not change the nesting of Stegosaurus or the rest of the tree topology.

Figure 2. Chrictonsaurus has long legs, tall plates and long legs like a Stegosaurus, but a skull and other armor like an ankylosaur.

Figure 4. Crichtonsaurus has long legs, tall plates and long legs like a Stegosaurus, but a skull and other armor like an ankylosaur. It has not been added to the large reptile tree yet.

The addition of new basal taxa
might bring the armored ornithischians back together again, but at present they are separated on the larger reptile tree. Additional material for Scutellosaurus would be much appreciated and interesting to compare to the best guess in figure 1.

Butler RJ, Jin L-Y, Jun C, Godefroit P 2011. The postcranial osteology and phylogenetic position of the small ornithischian dinosaur Changchunsaurus parvus from the Quantou Formation (Cretaceous: Aptian–Cenomanian) of Jilin Province, north-eastern China. Palaeontology 54 (3): 667–683.
Holtz TR Jr 2012. Dinosaurs: The Most Complete, Up-to-Date Encyclopedia for Dinosaur Lovers of All Ages, Winter 2011