
==== Front
Proc Natl Acad Sci U S A
Proc Natl Acad Sci U S A
PNAS
Proceedings of the National Academy of Sciences of the United States of America
0027-8424
1091-6490
National Academy of Sciences

38466860
202401482
10.1073/pnas.2401482121
commCommentaryevolutionEvolution418
437
Commentary
Biological Sciences
Evolution
Inferring aerial behavior in Mesozoic dinosaurs: Implications and uncertainties
Xu Xing xing.xu@ivpp.ac.cn
a b c 1
aCentre for Vertebrate Evolutionary Biology, School of Life Sciences, Yunnan University, Chenggong, Kunming 650504, China
bSouthwest United Graduate School, Kunming 650092, China
cKey Laboratory of Vertebrate Evolution and Human Origins, Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of Sciences, Beijing 100044, China
1Email: xing.xu@ivpp.ac.cn.
11 3 2024
19 3 2024
11 9 2024
121 12 e2401482121Copyright © 2024 the Author(s). Published by PNAS.
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ This article is distributed under Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND).

The National Natural Science Foundation of China 42288201 Xing Xu Yunnan Revitalization Talent Support Program 202305AB350006 Xing Xu
==== Body
pmcThe last few decades have witnessed tremendous efforts to infer whether particular Mesozoic dinosaurs could fly and how the seemingly volant ones moved through the air (1, 2). Both questions are directly relevant to one of the most astonishing events in vertebrate evolution, namely the transition from land to air along the line to modern birds (3–5). In PNAS, Kiat and O’Connor provide the results of their quantitative analyses bearing on whether dinosaur flight had a single origin or multiple origins, and where the origin(s) of flight occurred in phylogenetic terms, based on a novel approach to extracting functional signals from the flight feathers of both extinct and living dinosaurs (6).

The authors’ analyses suggest that theropod flight had a single origin at the base of the Pennaraptora (Fig. 1A), a group containing not only all birds but also several other dinosaur clades including the oviraptorosaurs, dromaeosaurs, and troodontids. Such results are not surprising, given that a suite of other important avian characteristics also originated at the pennaraptoran node (3). However, other studies (7) have supported a multiple-origins scenario for flight in pennaraptorans (Fig. 1B). Several factors have contributed to making the question of single flight origin versus multiple origins controversial, including variation across flight performance studies in both methods and results, and uncertainties in ancestral state reconstruction derived from limited taxon sampling, phylogenetic uncertainties, and differing model choices.

Fig. 1. Two scenarios for theropod flight origins. (A) Theropod flight originated at the base of the Pennaraptora, suggesting numerous secondary flight losses among early pennaraptorans. (B) Pennaraptorans evolved aerial capabilities multiple times early in their evolutionary history.

Ecomorphological inference, based on the form-function relationship established by qualitative and quantitative analyses of key morphological features and aerial behaviors of living birds, is one approach that has been widely used in efforts to understand flight in extinct dinosaurs. Many morphological features, such as flight feather shape and limb proportions, tend to not only differ between volant and flightless birds but also vary among volant birds with contrasting flight capabilities and styles, making them potentially useful in assessing volancy and flight mode in extinct dinosaurs (1). However, the same body part, or even the same morphological condition, sometimes leads different investigators to conflicting assessments of aerial capability. For example, flight feather vane asymmetry has been suggested to indicate volancy in Archaeopteryx (8), but the flight feathers of this iconic taxon might have slender rachises, potentially a sign that Archaeopteryx flew poorly or even not at all (9). Furthermore, even the reliability of asymmetrical vanes as an indicator of aerial locomotor ability has been called into question (10).

The authors’ analyses suggest that theropod flight had a single origin at the base of the Pennaraptora, a group containing not only all birds but also several other dinosaur clades including the oviraptorosaurs, dromaeosaurs, and troodontids.

Resolving the debate will require researchers to come up with better data and methods, and this is exactly what Kiat and O’Connor do in their new study (6). The authors present a more comprehensive dataset than those available in previous studies, incorporating flight feather shape data from a larger number of species and innovatively using data on numbers of rectrices and remiges to draw inferences about flight capability. More importantly, they estimate rates of evolutionary loss for these flight-related features and use these results to explain why some secondarily flightless birds still sport such hallmarks of volancy as a wing equipped with asymmetrical flight feathers. These results are then used to infer whether some early-diverging pennaraptoran dinosaurs were fliers and when those inferred to be secondarily flightless might have lost their flight capabilities, a question previously left untouched.

However, this ecomorphological perspective has some limitations. Stem birds and their close dinosaur relatives are hugely different in the anatomy of the flight apparatus compared to extant flying dinosaurs. For example, many stem birds lack a prominent sternal keel, a structure that plays a key role in flight in modern birds. They also differ significantly from extant flying birds in shoulder girdle morphology, and in the configuration of the wing in general and the flight feathers in particular. Even more surprisingly, some early-diverging pennaraptorans, namely scansoriopterygid theropods, possess no large flight feathers but instead have membranous wings (11). Although these differences would suggest a lack of flight capability in these extinct species under a strict ecomorphological approach, many studies have suggested that these differences may instead merely indicate that early-diverging pennaraptorans and extant flying birds used significantly different modes of aerial locomotion (11, 12). The established flight–morphology relationship based on data from extant birds strictly applies only to powered flight, and the absence of the anatomical correlates of volancy seen in modern birds does not rule out the possibility that some early-diverging pennaraptorans might have been passive flyers or even active flyers that flew in a different manner from extant taxa.

Biomechanical analysis represents another major approach to studying the potential aerial behaviors of extinct dinosaurs. All volant vertebrates follow the same aerodynamic laws: They must be able to generate lift and thrust to counteract gravity and air resistance, respectively, and must persistently overcome these countervailing forces in order to qualify as powered fliers. Volant vertebrates are further constrained by the rules of biology, which limit, for example, the amount of force their muscles can produce per unit of physiological cross-sectional area. In practice, biomechanical analysis can take many different forms. For example, one could estimate such parameters as wing loading (i.e., body mass divided by wing area) and specific lift (i.e., the amount of lift generated per unit of body mass) for extinct species, and compare the results with those obtained from extant volant animals whose aerial behaviors are well known. One could build a digital or a physical model of a taxon of interest and determine such aerodynamic parameters as thrust, lift, and airspeed for a given gliding posture or pattern of flight kinematics, in software or in a wind tunnel. One could even build a robotic animal and fly it in realistic environments.

The biomechanical approach, although not implemented by Kiat and O’Connor (6), has yielded important results that have improved our understanding of the evolution of aerial behavior in extinct dinosaurs (1, 2). Nevertheless, it also has several limitations. First, aerodynamic analyses normally rely on a model or a set of parameters that represents an organism’s whole body, but such a representation may be difficult to achieve given the rarity of relatively complete fossils, particularly with respect to soft tissue preservation. Second, even where it is possible to estimate such parameters as airspeed, body mass, and wing shape for a particular species, the estimates may be less than fully accurate or reliable for a variety of reasons. Finally, assumptions about the range of gliding postures and/or patterns of flapping kinematics that a volant taxon could have utilized are also subject to uncertainty.

The ecomorphological and biomechanical approaches ultimately represent applications of the principle that the present is the key to the past when attempting to understand the behavior of extinct species. Both approaches, in current research practice, are therefore heavily dependent on data from living birds (13). The huge differences between stem and crown birds render use of modern birds as a living analog in functional analyses commensurately challenging and sometimes even misleading. Furthermore, environmental factors such as air density could have had a significant effect on the aerial behavior of various taxa. Indeed, there is evidence suggesting that the density of the air changed during Earth’s history and that aerial behaviors were particularly widespread in periods when air density was relatively high (14).

The best way to solve the living analog problem is to adopt a holistic approach using multiple lines of evidence, and multiple methods, to infer the flight capabilities and modes of extinct dinosaurs. For example, functional data from living birds are misleading in the case of scansoriopterygids, making it desirable to consider volant animals possessing membranous wings when attempting to reconstruct the flight behavior of these taxa. Another way to address this modern analog problem is to infer behaviors of some extinct species based on established knowledge of others. For example, the possibility of arboreality in early-diverging pennaraptorans, including the earliest-diverging birds such as Archaeopteryx, has been rejected by many studies based on data on the pedal morphology of extant flying birds (15). However, modern arboreal bird taxa are highly specialized animals, a consideration that calls into question their relevance as living analogs for early-diverging forms. One possible solution would be to instead use pedal morphology data from the scansoriopterygids, which are highly likely to have been arboreal, given that early-diverging pennaraptorans are more similar in general morphology to each other than to extant birds.

Despite ongoing challenges, some highly credible results have been obtained concerning the flight capabilities and styles of early birds and their close dinosaur relatives, particularly in relation to trends in flight evolution. For example, a trend toward improved flight capability along the line to crown birds is widely accepted, based on various analyses (1, 4). Also, there is general agreement that some form of aerial behavior appeared before the origin of birds (3, 5, 16), perhaps even arising independently in multiple lineages (7).

A single-origin scenario for dinosaur flight would significantly affect our conventional understanding of birds and other pennaraptoran dinosaurs. For example, many familiar dinosaurs, such as Velociraptor, Troodon, and Oviraptor, would have to be regarded as secondarily flightless (i.e., descended from volant animals). This would not be particularly surprising, however, for several reasons. First, flightlessness has evolved many times among different crown bird lineages; second, it would have been relatively easy for early birds to revert to the flightless condition, given that they were less specialized for volancy than crown birds; and finally, the Mesozoic aerial space was highly competitive because of the presence of pterosaurs alongside volant pennaraptoran dinosaurs. Also, acceptance of a single-origin scenario would highlight an interesting question: How to define birds and distinguish them from non-birds. Technical terms for grouping animals are relatively easily created and disseminated among the academic community, but only terms that belong to the vernacular can leave a deep mark in popular culture. To the general public, birds are feathered animals whose bodies are highly specialized for flight. If flight, pennaceous feathers, and some other salient avian features indeed originated at the base of the Pennaraptora, wouldn’t it be better for us to call all pennaraptoran dinosaurs birds, just like we called Archaeopteryx a bird when this iconic species was first discovered?

The author thanks Dr. Corwin Sullivan for commenting on and editing the manuscript, and the National Natural Science Foundation of China (Grant No. 42288201) and Yunnan Revitalization Talent Support Program (202305AB350006) for supporting his research.

Author contributions

X.X. designed research; performed research; analyzed data; and wrote the paper.

Competing interests

The author declares no competing interest.

See companion article, “Functional constraints on the number and shape of flight feathers,” 10.1073/pnas.2306639121.
==== Refs
1 M. Pittman , “Methods of studying early theropod flight” in Pennaraptoran Theropod Dinosaurs: Past Progress and New Frontiers, M. Pittman, X. XU, Eds. (American Museum of Natural History, New York, 2020), pp. 277–294.
2 C. Sullivan, X. Xu, J. O’Connor, Complexities and novelties in the early evolution of avian flight, as seen in the Mesozoic Yanliao and Jehol Biotas of Northeast China. Palaeoworld 26 , 212–229 (2016).
3 X. Xu , An integrative approach to understanding bird origins. Science 346 , 1253293 (2014).25504729
4 Z. H. Zhou, The origin and early evolution of birds: Discoveries, disputes, and perspectives from fossil evidence. Naturwissenschaften 91 , 455–471 (2004).15365634
5 S. L. Brusatte, J. K. O’Connor, E. D. Jarvis, The origin and diversification of birds. Curr. Biol. 25 , R888–R898 (2015).26439352
6 Y. Kiat, J. K. O’Connor, Functional constraints on the number and shape of flight feathers. Proc. Natl. Acad. Sci. U.S.A. 121 , e2306639121 (2024).38346196
7 R. Pei , Potential for powered flight neared by most close avialan relatives, but few crossed its thresholds. Curr. Biol. 30 , 4033–4046 (2020).32763170
8 A. Feduccia, H. B. Tordoff, Feathers of Archaeopteryx: Asymmetric vanes indicate aerodynamic function. Science 203 , 1021–1022 (1979).17811125
9 R. L. Nudds, G. J. Dyke, Narrow primary feather rachises in Confuciusornis and Archaeopteryx suggest poor flight ability. Science 328 , 887–889 (2010).20466930
10 P. Shipman, Take Wing-Archaeopteryx and Evolution of Bird Flight (The Guernsey Press Co. Ltd., London, 1998), p. 398.
11 X. Xu , A bizarre Jurassic maniraptoran theropod with preserved evidence of membranous wings. Nature 521 , 70–73 (2015).25924069
12 D. F. Voeten , Wing bone geometry reveals active flight in Archaeopteryx. Nat. Commun. 9 , 923 (2018).29535376
13 F. J. Serrano, L. M. Chiappe, Independent origins of powered flight in paravian dinosaurs? Curr. Biol. 31 , R359–R373 (2021).
14 R. Dudley, Atmospheric oxygen, giant Paleozoic insects and the evolution of aerial locomotor performance. J. Exp. Biol. 201 , 1043–1050 (1998).9510518
15 C. L. Glen, M. B. Bennett, Foraging modes of Mesozoic birds and non-avian theropods. Curr. Biol. 17 , R911–R912 (2007).17983564
16 V. Allen, K. T. Bates, Z. H. Li, J. R. Hutchinson, Linking the evolution of body shape and locomotor biomechanics in bird-line archosaurs. Nature 497 , 104–107 (2013).23615616
