Changzhousaurus, new pennaraptoran from Early Cretaceous of China (free pdf)

[Ben Creisler]

Ben Creisler

bcre...@gmail.com

A new paper:

Free pdf:

Changzhousaurus sinensis gen. et sp. nov.

XU Xing (2026)
A new feathered dinosaur from Early Cretaceous of northern China highlighting the complexity of early pennaraptoran evolution and comments on several relevant conceptual and methodological issues.
Vertebrata Palasiatica (advance online publication)
DOI: 10.19615/j.cnki.2096-9899.260616.
https://www.vertpala.ac.cn/EN/10.19615/j.cnki.2096-9899.260616

Recent discoveries of early-diverging pennaraptoran fossils have shed light on the origin of birds and, in particular, the evolution of defining avian features such as pennaceous feathers and flight capability. Here I report a new pennaraptoran dinosaur based on a fossil recovered from the Lower Cretaceous Jiufotang Formation of western Liaoning, China. Despite exhibiting a combination of derived features observed across distinct pennaraptoran lineages, this new taxon is likely an early-diverging deinonychosaurian. Most notably, it possesses exceptional plumage characteristics: it represents the first known early-diverging pennaraptoran to bear both large pedal feathers and highly elongated rectrices; these elongated rectrices are substantially more abundant than those of other early-diverging pennaraptorans, bearing a superficial resemblance to the tail plumes of peacocks; and its wing feathers form the proportionally largest feathered wings among non-avialan pennaraptorans—even with relatively short bony forelimbs—indicating a decoupling of forelimb skeletal length and feathered wing surface area. This discovery underscores the complexity of early pennaraptoran evolution and raises several conceptual and methodological issues in pennaraptoran research. These issues include how to recover a robust pennaraptoran phylogeny, how to infer the aerial behavior and habitat ecology of early-diverging pennaraptorans, and how to define feathers and birds. I briefly address these issues in this paper.

[Mickey Mortimer]

Unfortunately, Figures 3 and 13 are missing, Figures 9 and 10 are the same, and all figures have terrible pixelated resolution.

Mickey Mortimer

[Tim Williams]

Mickey Mortimer <therizino...@gmail.com> wrote:

> Unfortunately, Figures 3 and 13 are missing, Figures 9 and 10 are the same, and all figures have terrible pixelated resolution.

Hopefully these will be rectified in the final version.  I'm especially intrigued by the morphologies of _Changzhousaurus_' feathers .

I'm concerned with the substance of the text.  In particular, the author's skepticism of what he calls the "uniformitarian" approach to ecomorphology.  In his words: "uniformitarian approach, which means that ancient life follow the same rules of physics, chemistry and biology derived from living organisms.  More specifically, these studies applied two widely used methods (i.e., ecomorphological inference and biomechanical analysis) developed under the uniformitarian framework, and both methods are heavily dependent on “rules” derived from living animals."

Ummm… yes.  Because those same rules invoke physics, chemistry and biology that apply to all life on Earth.  What follows in the manuscript seems to be special pleading for small, early-diverging pennaraptorans - that they are somehow exceptional to the usual 'rules'.  The manuscript seems to be saying that comparing their feather morphologies and pedal proportions to modern birds is misleading, because modern birds are too specialized for flight and arboreality.  

But - when it comes to the ecologies and behaviors of tetrapods, the same physical rules apply.  This is the case for powered flight using feathered wings (and membranous wings, in the case of scansoriopterygids); or to gripping a tree branch with the hands or feet.  That's why quantitative and qualitative comparisons using living analogs are so important.  And crown birds are the closest surviving relatives of _Archaeopteryx_, _Anchiornis_, _Microraptor_,  _Changzhousaurus_, etc.

"Characters including a reversed hallux, elongated penultimate pedal phalanges and curved pedal claws are widely regarded as indicators of arboreality in early-diverging pennaraptorans including Archaeopteryx (Feduccia, 1999)."

You're on thin ice indeed when you cite Feduccia's book "The origin and evolution of birds" to support your case.  Feduccia's earlier (1993) study using claw curvature to prove arboreality has been refuted so many times it's just not funny any more (e.g., Glen and Bennett, 2007; Dececchi and Larsson, 2011; Fowler et al., 2011).

The author later argues that _Anchiornis_ was capable of flapping flight based on its "overall phenotypic suite".  Same for the "bat-winged" scansoriopterygids.  I think the argument here is that the overall appearance of a fossil animal can be used to override detailed biomechanical or aerodynamic studies.  Again, this strikes me as special pleading.

"Specifically, _Changzhousaurus sinensis_ is derived from arboreal pennaraptoran ancestors characterised by proportionally elongated penultimate phalanges and larger, more distally positioned halluces, and it secondarily modified its pedal apparatus for a more terrestrial environment."

This could have been written by GSP.  Personally, I'm open to the hypothesis that many early-diverging pennaraptorans were secondarily flightless.  But proposing that they were also secondarily terrestrial (= descended from arboreal ancestors) is an entirely different matter.  

"_Changzhousaurus sinensis_ represents the first documented early-diverging pennaraptoran to concurrently possess both highly elongated tail feathers and large pedal feathers. Although this study does not aim to systematically investigate the functional significance of these plumage features, a purely communicative function for both pedal and tail feathers appears biologically implausible"

I don't see why a display function for these feathers is implausible at all.  Especially because earlier in the manuscript the tail feathers of _Changzhousaurus_ are compared directly to the tail fan of peacocks.

"Bony forelimb length has long been widely adopted as a proxy for wing size and aerial locomotor capacity in fossil taxa. Despite its proportionally short bony forelimbs, _Changzhousaurus sinensis_ bears the largest proportional feathered wings among non-avialan pennaraptorans, demonstrating significant morphological decoupling between skeletal forelimb dimensions and soft-tissue wing size. This finding indicates that a proportionally short bony forelimb cannot be reliably interpreted as conclusive evidence for small wing size or absent flight capability in early-diverging pennaraptorans."

On the contrary, I'd say it can.  If biomechanical or aerodynamic studies indicate that the forelimbs were too weak or too short for powered flight, then the most parsimonious explanation is that the animal didn't fly.  

[Mickey Mortimer]

For anyone curious, when scored in the Lori matrix (with the figures being poor resolution and the cranial figure missing), Changzhousaurus emerges sister to Xiaotingia then Mei within Sinovenatorinae in Troodontidae, with troodontids and archaeopterygids (including anchiornithines) as deinonychosaurs. 

This makes sense, as Xu says "a broad groove appears to be present on the posterior half of the dentary, a feature known in many troodontids and the anchiornithine Xiaotingia", "The forelimb proportion is similar to that of some early-diverging paravians such as in Xiaotingia: the radius and manus are about 95% and 115% of the humeral length, respectively, compared to 90% and 115% in Xiaotingia", "The radius is bowed anteriorly, and with the posteriorly bowed ulna, form a relatively broad forearm (Figs. 9, 10), a derived feature only known in Xiaotingia (STM 27-2)", "Metacarpals are similar in general morphology to those of Xiaotingia: Metacarpal II [note Xu uses the II-III-IV manual digit hypothesis for maniraptorans] is about 30% of metacarpal III length, metacarpal IV is laterally bowed, about the same in robustness as metacarpal III, and slightly longer than the latter", "The ilium is shallow in lateral view (Fig. 7), with the acetabulum-level height to the anteroposterior length ratio of 0.18, compared to 0.17 in Xiaotingia", "The pubic peduncle is large, with the anteroposterior length about 25% of the iliac length, compared to ... about 30% in Xiaotingia", "The pubic peduncle is proportionally wide in lateral view, about 2.4 times as wide anteroposteriorly as high dorsoventrally measuring at the mid-length of the peduncle, compared to about 2.2 in Xiaotingia", "The cuppedicus fossa has a large lateral exposure below the preacetabular process as in unenlagiines, early-diverging dromaeosaurids, some early-diverging avialans such as Sapeornis, and the anchiornithine Xiaotingia", "The ischium is a slender and strap-like bone (Fig. 7) as in some microraptorines and Xiaotingia." Indeed, this seems to be Xu's conclusion, where he says 'Particularly noteworthy is that Changzhousaurus sinensis shares with the anchiornithine Xiaotingia a few derived features. For example, both taxa have an anteriorly bowed radius (also present in Buitreraptor), a large and wide iliac pubic peduncle (also in Rahonavis), a slender, straight, strap-like ischial shaft with thickened anterior margin (also in some microraptorines), and ischial pubic peduncle long and distally tapered and ischial iliac peduncle short and robust and aligned with the ischial shaft." 

... And "The antorbital fossa is poorly defined anteriorly and ventrally, and the possible interfenestra bar is flushed with the external surface of the maxilla as in Byronosaurus (Makovicky et al., 2003) and Mei long holotype", "As preserved, the forearm forms an angle of about 75 degrees with the humerus, and of about 90 degrees with the manus, the latter often seen in early-diverging paravians such as Archaeopteryx, Microraptor, and Mei (Makovicky et al., 2026)", "The small distal carpal is identified as the distal carpal 4, and it attaches to the laterodistal corner of the ‘semilunate’ carpal as in some early-diverging deinonychosaurian specimens such as Mei long holotype (Xu et al., 2014b)", "Proportionally long and robust metacarpal IV is also seen in scansoriopterygids, halszkaraptorines, Mei, and many enantiornithines", "Elongate manual digit IV is also known in scansoriopterygids, halszkaraptorines, and Mei", "small rounded maxillary fenestra (probably in Mei long holotype)", "maxillary tooth row extending posteriorly about preorbital bar level (at least in Mei long holotype)."

... AND "In summary, Changzhousaurus sinensis is most likely an early-diverging deinonychosaurian, and its discovery provides additional evidence for the monophyly of Deinonychosauria."

Xu didn't attempt a phylogenetic analysis, but barring changes from the cranial figure Changzhousaurus seems to simply be a sinovenatorine instead of a weird pennaraptoran mix.

Mickey Mortimer