Regrettably, the authors ignore 99% of the current research on the evolution of birds and the origin of avian flight. Instead, they focus on 'outlier' studies by a very small number of authors (such as Czerkas, Feduccia, Ruben) that are not representative of the field. The 'phylogenetic' work of James and Pourtless (2009) has been thoroughly discredited. Further, claims that certain well-supported hypotheses are controversial or problematic (such as endothermy in non-avian dinosaurs, or the dinosaurian origin of birds) are misleading; the studies cited in support of such claims are poorly founded and could even be deemed vexatious.
The authors clearly need to do more reading. The abstract opens with: "The oldest feathers known to date have been found in archosaurs capable to fly." This is blatantly incorrect, and is the first of many erroneous statement made throughout this manuscript. Contrary to what the authors assert, filamentous integumentary structures are found in non-avian theropods such as compsognathids, basal tyrannosaurs, ornithomimosaurs, and maniraptorans.
"The body of theropods was covered mostly by typical reptilian scales" is also untrue considering Sciurumimus, compsognathids, ornithomimids, Dilong, Yutyrannus, etc., even granted the authors are following Martin in considering maniraptorans to be non-theropodan. Note that the filamentous structures in the above taxa are the same type as those the authors consider to be feathers in scansoriopterygids (minus the retrices of Epidexipteryx), so it would be special pleading to consider that family to have a feathered covering but theropods to not. Indeed, later in the paper they consider Tianyulong's structures to be integumentary even though they've been studied far less than e.g. Sinosauropteryx's. Also, there's no evidence the structures in Juravenator or tyrannosaurids are reptilian scales instead of feather derivatives like on birds' feet.
Tim's comment accurately describes the entire introduction.
It's nowhere mentioned that Scansoriopteryx and Epidexipteryx may not preserve obvious remiges because the specimens are juveniles. Zhongornis actually preserves remiges and retrices (Gao et al., 2008:780), contra the authors.
The presence of a propatagium in pennaraptorans has no obvious correlation to a plagiopatagium ever being present. The authors certainly mention no anatomical or developmental codependence. Similarly, the presence of a slight patagium on the thigh of young doves as described by Beebe (1915; which the authors cite incorrectly as using "evolution" in its title instead of "ancestry", a mistake probably caused by copying Czerkas and Feduccia's 2014 bibliography) is assumed to be a rudiment without reason, as since Beebe's time zoological science has realized ontogeny does not recapitulate phylogeny. In the same vein, there's no reason to assume Yi's membrane wing was reflective of scansoriopterygids as a whole since the other genera don't have 'styliform elements' to support a patagium. Let alone reflective of the basal bird condition. Nor is there a reason to connect any femoral wing as described by Beebe to the elongate tibiotarsal and metatarsal feathers of Anchiornis and Microraptor. Indeed, Beebe himself (pg. 45) said there was "no connection" between his columbid femoral wings and feather-footed pigeon breeds. Finally, there was no described methodology in placing the centers of mass or lift on the reconstruction of Yi, which is itself largely guesswork as the angle of the styliform processes and posterior body morphology are mostly unknown (e.g. Scansoriopteryx has an Archaeopteryx-like tail that could provide significant lift if it had comparable retrices, while Epidexipteryx has a tiny tail with streamer-like feathers). Compare these assertions to the quantitative aerodynamic work done on Yi being presented at SVP this year (which we can't discuss due to the embargo).
The two Praeornis structures have not been shown to be feathers or archosaurian, so have little to bear on the topic. Epidexipteryx's (at least basally unbarbed) retrices are said to suggest that structure was primitive to feathers, but it and other scansoriopterygids have filaments too. There's no reason to consider scansoriopterygids to have a more ancestral condition than oviraptorosaurs or paravians, as all analyses place them in Pennaraptora surrounded by taxa with barbed remiges and retrices. That the authors have to cite papers from 1922 and 1955 show that they ignore (as Tim said) "99% of the current research" that use molecular data to inform the current consensus that feathers and (non-bird) scales are only homologous at a basic level and one did not evolve from the other. "The reason of the replacement of the membrane by feathers requires more study but one possibility is the role of sexual selection" is an empty statement, which echoes my point above there's no reason to think Yi's patagium had anything to do with the evolution of modern bird flight.
The authors show an unfamiliarity with current phylogenetic thought, as contra them, Cau (2018: fig. 3) did recover scansoriopterygids nested within taxa with feathers, at least as close to modern birds as anchiornithids. Even his Bayesian analysis (fig. 6) recovered them as oviraptorosaurs, and thus still birds under the authors' definition. As stated above, all recent analyses find scansoriopterygids somewhere in Pennaraptora. And since some (e.g. Brusatte et al., 2014) recover them as basal oviraptorosaurs, there's no temporal conflict inherent in quality cladistic results. The authors undercut their own points by emphasizing multiple times they "do not consider scansoriopterygids as direct ancestors of the main lineage of birds." Then there's the unique argument the ceratosaur Limusaurus is a bird contrary to every study and numerous characters throughout its anatomy, cited to the main author's previous paper here. Also the strange assertion (shared with Martin) that taxa like ornithomimosaurs could be dinosaurs or birds, even though the authors think these are as distantly related as Deinosuchus and Hesperornis.
The discussion of endothermy is highly simplistic ("the typical metabolic rates in theropods would have been a little lower than those of extant birds and mammals"), Bajdek's ideas regarding digestive relevance have not been published and were thoroughly rejected in reviews on this site, and Ruben's respiratory turbinate idea also never took hold (ironically the most recent data suggests Tyrannosaurus did have respiratory turbinates- Witmer and Ridgely, 2010).
James and Pourtless (2009) actually found birds to most parsimoniously be theropod dinosaurs, but obscured this in their figures. Longisquama emerged as a pennaraptoran and Effigia as an ornithomimosaur (they never figure or mention that latter result...), which along with many problems shows it's just a bad analysis. And it still finds theropodan birds, even when it scores every dinosaur (even Herrerasaurus) as unknown for every manual character. As for the partially closed acetabula of some maniraptorans, ankylosaurids fully closed their acetabulum but evolved from other dinosaurs with fully open acetabula, so Anchiornis and Scansoriopteryx could do it too.
In conclusion, the paper suffers from several major flaws. As Tim says, it treats the vast majority of work that informs the consensus as comparable to fringe papers written by the same few holdouts. It shows an unfamiliarity with the literature, whether it be Zhongornis' remiges, Cau's placement of scansoriopterygids among what the authors call birds, or getting the title of Beebe's paper right. There's a lazy vagueness that backs their claims, from why Yi's centers of balance and gravity are placed where they are, what "flight" is defined as, what "a little lower" metabolic rate is and how that relates to endothermy and growth rate, etc.. A great example is "in this paper we have chosen to use a broader term “archosaur” to call the arboreal ancestor of birds." Is that Archosauria the crown group as most recently used, the classic Archosauria that includes proterosuchids and is equivalent to the current Archosauriformes, something else...? And based on what? Any decent modern paper would say "birds are as close to crusotarsans as erythrosuchids based on characters xyz, but outside Loricata based on lacking characters abc", but the authors (and all Birds Are Not Dinosaurs authors...) just say they're derived from some Triassic, arboreal archosaur-y thing despite having a huge range of well described complete specimens from all over Archosauromorpha.
But most of all, the fatal flaw in this paper is that it's only a bunch of unevidenced assertions to reach its conclusion. IF a propatagium implies a plagiopatagium and/or the femoral skin of baby doves is a remnant of the ancestral bird condition, and IF scansoriopterygids are outside Pennaraptora unlike every analysis, and IF they truly lack remiges as adults, and IF Yi's patagia are symplesiomorphic for birds, and IF our untested and unquantified model of Yi's physics is accurate, THEN birds primitively may have had uropatagia. And uropatagia relate to the tibiotarsal and metatarsal wings of anchiornithines and microraptorians... because they're on the legs? Honestly, Yi having metatarsal wings like those groups would be more phylogenetically plausible and possibly solve the authors' asserted aerodynamic problem. Oh, but the authors "do not consider scansoriopterygids as direct ancestors of the main lineage of birds", which leaves us with IF a propatagium implies a plagiopatagium and/or the femoral skin of baby doves is a remnant of the ancestral bird condition, THEN birds primitively may have had uropatagia. A lot of work to do there. Everything involving birds being dinosaurs or not is even more tenuous, but that's not the main thrust of the paper, so I'll leave it at that.
New References- Brusatte, Lloyd, Wang and Norell, 2014. Gradual assembly of avian body plan culminated in rapid rates of evolution across the dinosaur-bird transition. Current Biology. 24(20), 2386-2392.
Gao, Chiappe, Meng, O'Conner, Wang, Cheng and Liu, 2008. A new basal lineage of Early Cretaceous birds from China and its implications on the evolution of the avian tail. Palaeontology. 51(4), 775-791.
Witmer and Ridgely, 2010. The Cleveland tyrannosaur skull (Nanotyrannus or Tyrannosaurus): New findings based on CT scanning, with special reference to the braincase. Kirtlandia. 57, 61-81.
The biggest issue here is the idea that flight evolving from a distinct wing structure into another wing structure, that also predates that "first" structure, is more likely than flight evolving twice in closely related animals. We already know flight evolved twice in ornithodirans, I see no reason to rule out the idea that among maniraptorans and early pterosaurs it may have appeared numerous times.
Let alone to go to such lengths just to deny that.
The authors also promote a number of misconceptions (feathers evolving from traditional reptilian scales is an obvious one), and refer to concrete relations in paleontology (such as birds being dinosaurs) as "controversial".
There is one point where I must agree with the implications of the authors statements however, and that is the assertion that wings did not evolve for gliding. Feathers may predate powered flight, but the idea of feathered wings used for gliding is a dated one. Of all the gliding animals known not one evolved into a flying lineage, and only a handful of gliding animals actually glide using their limbs (and they always use pre-existing structures to do so).
Microraptor is not a valid example here, as the nature of its locomotion is debated. Some evidence points to powered flight (it gains more lift at higher speed, something not true of known gliding animals). This is also supported by similarly controversial evidence for flight muscle attachments in juvenile Deinonychus.
"Bird-like" wings are incredibly versatile for locomotion, they aid an animal in everything from climbing to balance. The idea that they evolved for anything but locomotion seems ludicrous, but that locomotion was almost certainly not flight, and definitely not gliding.
You can also choose to receive updates via daily or weekly email digests. If you are following multiple preprints then we will send you no more than one email per day or week based on your preferences.
Note: You are now also subscribed to the subject areas of this preprint and will receive updates in the daily or weekly email digests if turned on. You can add specific subject areas through your profile settings.
Usage since published - updated daily