(C) PLOS One
This story was originally published by PLOS One and is unaltered.
 . . . . . . . . . .



Dipteran flight diversity is shaped by aerodynamic constraints, scaling, and evolutionary trade-offs [1]

['Camille Le Roy', 'Experimental Zoology Group', 'Wageningen University', 'Wageningen', 'The Netherlands', 'Ilam Bharathi', 'Thomas Engels', 'Ism', 'Cnrs', 'Aix-Marseille Université']

Date: 2026-07 

Flight has been a key innovation in insect evolution, yet the selective and mechanistic pressures shaping their flight motor systems remain poorly understood. Here, we present a comprehensive comparative analysis of flight in Diptera (true flies), integrating morphology, wingbeat kinematics, and aerodynamics within a phylogenetic framework. We quantified morphology in 133 species spanning the Dipteran phylogenetic and size range, and for a subset of 46 species, we combined high-speed stereoscopic videography with computational fluid dynamics (CFD) to characterize wingbeat kinematics and aerodynamic performance, respectively. Our results reveal that morphology is strongly structured by phylogeny, whereas wingbeat kinematics are broadly conserved across Diptera, reflecting dominant aerodynamic constraints. Two early-diverged lineages, Culicomorpha (mosquitoes and midges) and Tipulomorpha (crane flies), exhibit strikingly divergent kinematics and aerodynamics, suggesting lineage-specific selective pressures. Combining these data with scaling analyses shows that maintaining in-flight weight support across the dipteran size range requires systematic allometric adjustments in wing morphology, wingbeat kinematics, and flight musculature. Smaller dipterans achieve weight support through relatively larger wings and higher wingbeat frequencies, whereas larger dipterans achieve the same aerodynamic requirement through increased investment in flight musculature to sustain the necessary mechanical power output. These size-dependent trait combinations highlight how different morphological and kinematic adaptations evolved in response to the shared physical requirements of hovering flight across Diptera. Mosquitoes and midges represent an extreme case, exhibiting a pronounced aerodynamic–acoustic trade-off with disproportionately high wingbeat frequencies, large flight musculature and increased aerodynamic and acoustic power, consistent with selection favoring acoustic signaling during in-swarm mating. By integrating comparative morphology, kinematics, and aerodynamics across a major insect radiation, our study uncovers the interplay between physical scaling laws, aerodynamic constraints, and ecological pressures in shaping the evolution of animal flight. These findings provide a mechanistic framework for understanding how complex locomotor systems diversify under multiple selection pressures.

Funding: This work was supported by an NWO Vidi research grant to F.T.M. (I/VI.Vidi.193.054; www.nwo.nl ). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data Availability: All data underlying the findings of this study are fully available without restriction. The complete datasets used in the analyses are deposited in the Dryad Digital Repository under https://doi.org/10.5061/dryad.gxd25480s . The repository includes wing morphology, wingbeat kinematics, and CFD-derived aerodynamic forces and power for all studied species. It also contains all custom code used to generate the study figures from the deposited data, as well as CFD parameter files, microscope images, and representative high-speed videos for our studied species. In addition, the repository provides an interactive tool that allows users to filter, visualize, and export the processed data or selected subsets. This tool is also accessible via an online interface at https://insectflight.eu . DNA barcode data for all analyzed specimens are publicly available through the BOLD System under project DIPTR via https://portal.boldsystems.org/result?query=DIPTR . The version of the WABBIT code used for the numerical simulations is archived on Zenodo and available via https://doi.org/10.5281/zenodo.20234346 . No ethical or legal restrictions apply to the sharing of these data.

Copyright: © 2026 Le Roy et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Introduction

Insects are the most species-rich class of animals, comprising an estimated 5 million species—about 80% of all known animal species [1]. Their remarkable radiation may be partly attributed to their flight ability [2], which has led to a striking diversity of aerial lifestyles [3]. Despite the pivotal role that flight has played in the evolution and ecological importance of insects, our understanding of the evolutionary processes that generate the diversity of their flight kinematics, morphology, and associated aerodynamics remains limited.

This is in part because insect flight is remarkably complex and challenging to quantify [4], limiting the feasibility of large comparative evolutionary flight studies. While the flight of large vertebrate fliers can be addressed using classical aeronautical principles and often can be inferred based on morphology [5–7], that of insects heavily relies on complex high-frequency flapping wing motions, resulting in unsteady aerodynamic mechanisms that fall outside the realm of conventional aerodynamic theory [4,8,9]. Precisely quantifying the diversity of wing morphology and wingbeat kinematics, as well as the underlying unsteady aerodynamics, is therefore crucial to understand the evolution of flight in insects.

Fundamentally, flying insects generate aerodynamic forces through the reciprocal motion of the wings swept at a high angle-of-attack, but the overall wingbeat kinematics may greatly vary among taxa. The extent to which the evolution of wingbeat kinematics is channeled by aerodynamic constraints, and how these constraints mediate flight divergence between species, is poorly understood. Although bounded by physical laws, variations in wingbeat kinematics among flying insects can be influenced by distinct evolutionary pressures. On one hand, neutral divergence along phylogenetic tree branches may lead to different flight kinematics between species. On the other hand, selective pressures acting on flight can strongly constrain the evolution of wingbeat kinematics, overcoming the pattern expected under neutral evolution.

Because flight is among the most energetically costly forms of locomotion [10], minimizing its energetic demands may represent one of the strongest selective pressures shaping the insect flight motor system. Cost-reducing selective pressures may be especially strong in highly aerial species, such as long-distance migrants [11]. Conversely, adaptation to some ecological niche may promote the evolution of specialized flight abilities, leading to wingbeat kinematics that diverge from those predicted by energetic minimization alone. For example, the evolution of aerial predatory habits in some groups (e.g., dragonflies or robber flies) may promote adapted flight kinematics and morphology enabling extremely agile flight [12,13]. In contrast for prey species, natural selection may favor traits that enhance escape success, such as the rapid escape maneuverability of flies and mosquitoes [14,15], or unpredictable flight behavior in mosquitoes and Lepidoptera [16,17]. Sexual selection may also exert a powerful influence on the evolution of flight. In hoverflies, for example, maintaining a highly stable stationary flight is crucial for guarding a mating territory [18,19], which may drive the evolution of specialized hovering flight kinematics through male-male competition for females [19,20]. Moreover, many insect species mate in swarms, and in some of the swarming species, males locate females by detecting the auditory flight tones produced by their beating wings [18,21,22]. It has been suggested that the high-frequency wingbeat produced by swarming mosquitoes and midges (Culicomorpha) results from sexual selection, as it maximizes audibility during mating interactions [21,23,24].

Flying insects span an impressive size range, from the tiniest 10 μg fairy-fly to the largest 100 g Goliath beetle, covering seven orders of magnitude in body mass. Physical scaling laws strongly shape how the flight motor system functions across this range. One important effect is that, under geometric similarity, aerodynamic force production does not scale in proportion to body weight across sizes. Maintaining weight support at small body sizes, therefore, requires allometric adjustments in wing morphology and/or wingbeat kinematics [20]. As a consequence, the flight style that minimizes aerodynamic power differs across the size spectrum.

Aerodynamic scaling across sizes is expressed by the Reynolds number (Re), which quantifies the relative importance of inertial and viscous forces [25]. Flying insect wings operate at Re from tens to several thousands (Re ) [26]. This range places most species in the intermediate Re-regime, between a high-Re inertial regime where inertial forces dominate, and a low-Re laminar-viscous regime where viscous forces dominate [27]. These flow regimes strongly affect the aerodynamic drag coefficient ( ) of a flapping insect wing. In the high-Reynolds-number inertial regime, the drag coefficient is relatively low and approaches Reynolds-number independence ( ). In contrast, tiny insects operating near the laminar–viscous regime experience elevated drag coefficients because viscous forces dominate and the laminar boundary layer thickens. Under classical Blasius laminar boundary-layer theory, drag coefficients scale here inversely with Reynolds number as [27]. In the intermediate regime, drag coefficients therefore follow a scaling that lies between these two asymptotic limits. As a consequence, drag coefficients increase progressively as insect size decreases [27,28].

This Re-dependent shift alters how drag, lift, and aerodynamic power scale with body size, and thus constrains the biomechanical strategies available to sustain powered flight. For example, tiny ptiliids beetles have evolved very lightweight bristled wings that still function effectively as solid surfaces at the low Reynolds numbers in which they operate [29]. In contrast, large flying insects such as Morpho butterflies experience Reynolds numbers high enough to permit efficient gliding, and have evolved wing shapes adapted for this flight mode [30].

The diversity of insect flight kinematics likely results from the interplay between shared ancestry, physical scaling laws, aerodynamic constraints, and ecological pressures. Disentangling the relative role of these factors on the evolution of such a complex trait is a major challenge necessitating a coupled comparative evolutionary, biomechanical, and aerodynamic perspective. The integration of these approaches has nevertheless rarely been attempted.

Contrasting with the huge number of flying insect species, flight has been comprehensively studied in only a handful of them, mostly with the aim to understand the aerodynamic mechanisms involved [14,31–36]. Although providing valuable insights into flight mechanics [9], such studies generally consider single (model) species, or at most a few species in comparison, and are thus of limited evolutionary relevance. In contrast, more extensive comparative evolutionary studies on insect flight mostly focused on morphology only, or simple flight kinematics parameters such as wingbeat frequency, and simplified aerodynamic modeling [3,37–39]. This prevented us from fully appreciating the rich diversity in morphology, wingbeat kinematics, and aerodynamics across taxa. The combination of detailed biomechanics studies and comparative phylogenetics is thus needed to understand the mechanistic and selective forces shaping the diversity of insect flight.

To address this knowledge gap, we performed a broad-scale comparative study of detailed morphology, wingbeat kinematics, and aerodynamics, focusing on the order Diptera (true flies). Flies represent one of the largest radiation of flying animals [40]. With ~160,000 named species spread across ca. 180 families [41], and an estimated >1 million undescribed species, they represent ~15% of all animal species [1,42]. Dipteran insects are found nearly everywhere on earth, inhabiting ecosystems ranging from the arctic tundras to arid deserts [43,44]. They have been described as the ecologically most diverse animal order [45], as they comprise a wide array of trophic specializations (including but not limited to saprophages, predators, blood-feeders, herbivores, or pollinivores), as well as highly diverse sexual behaviors among species (including swarming, mate guarding, or male-male territorial contest) [46]. Moreover, the phylogenetic relationships are relatively well-resolved for most Dipteran taxa [41,42,47], enabling to investigate the macro-evolutionary processes shaping trait diversity across species (e.g., [48–50]). Finally, flight is central to many of dipterans’ essential behaviors for survival and reproduction, making this group an ideal model for addressing the evolutionary and biomechanical drivers of flight diversity.

Diptera are regarded as highly specialized fliers, with a remarkably sophisticated sensory-motor system. Their flight is powered by asynchronous indirect flight muscles that drive rapid wingbeats, while tiny steering muscles at the wing base provide precise flight control and steering [51,52]. Another key trait in Diptera is the presence of halteres, which are specialized sensory and flight control organs derived from the hindwings [48,53]. These club-shaped structures confer precise control over body rotations during flight. The combination of specialized sensory and motor systems makes Dipteran insects among the most agile flying animals [52,54].

Despite the high flight specialization of Diptera, marked differences in wingbeat kinematics exist among taxa. Most notably, the flight of mosquitoes (Culicidae) is characterized by low wingbeat amplitudes and high wingbeat frequencies [35,55]. In contrast, other similarly-sized dipterans such as fruit flies (Drosophila) exhibit much larger wingbeat amplitude but lower frequencies [32]. Moreover, mosquitoes and midges exhibit slender body forms and high-aspect-ratio wings, while fruit flies possess stouter bodies and low-aspect-ratio wings. In addition, the relative mass of flight musculature varies significantly between these taxa, from ~30% of total body mass in Drosophila to ∼55% in similarly-sized Chironomous species [56,57], suggesting substantial variation in relative flight power availability. Mosquitoes and fruit flies, therefore, represent two extremes in Dipteran flight motor systems, and also correspond to relatively early and recent diverged lineages, respectively. Beyond these highly divergent taxa, flight kinematics across most other dipteran groups remain poorly characterized. In particular, the coupling between aerodynamic constraints, energetic efficiency, and additional selective pressures has not been systematically resolved.

In this study, we aim to uncover the evolutionary and biomechanical drivers of flight diversity in Diptera. For this, we quantified morphology, flight kinematics, and aerodynamic properties across a broad phylogenetic and size range of Diptera. Hereby, we use hovering flight as a standardized benchmark because it provides a consistent aerodynamic reference state across species. We then applied physical scaling laws within a phylogenetically informed comparative framework to evaluate how morphology, kinematics, and aerodynamics vary with size and lineage.

To provide baselines against which we assess allometric deviations and lineage-specific adaptations, we distinguish between geometric, kinematic, and dynamic similarity. Here, geometric similarity refers to proportional scaling of morphology, kinematic similarity to similarity of wingbeat angles and motion patterns without time rescaling, and dynamic similarity to fluid dynamic similarity whereby aerodynamic force coefficients are invariant with size.

Our investigation is structured around the following three objectives:

(i) We ask whether the evolutionary diversification of Diptera flight was driven by large plasticity in wingbeat kinematics, or whether aerodynamic constraints channeled the evolution of flight kinematics. Such channeling would promote greater morphological than kinematic diversity across the order. (ii) We test how the physical requirement for powered flight is maintained across the vast size range of Diptera, evaluating whether this is achieved through allometric scaling of morphology, kinematics, or a combination of both. More specifically, we examine three key requirements for powered flight: generating sufficient aerodynamic force for weight support, minimizing aerodynamic power consumption, and ensuring that flight muscles can deliver the mechanical power needed for sustained flapping. (iii) We investigate whether lineage-specific selective pressures, such as selection for acoustic signaling in mosquitoes, promote the evolution of traits that depart from energy cost minimization. Such pressures may generate a trade-off between minimizing aerodynamic power and meeting the demands imposed by other sexual or ecological selection.

By contrasting theoretical expectations from physiological and aerodynamic models and the observed flight diversity, our work highlights how phylogeny, sexual selection, and ecological pressures influenced the evolution of the Dipteran flight motor system. This comparative, phylogenetically informed, and size-aware approach enables us to evaluate lineage-specific traits against baselines defined by similarly sized dipterans. In doing so, it separates traits that are extreme in absolute terms from those that are unusually extreme for their size, and allows trait associations to be evaluated as broad comparative patterns instead of species-specific observations.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3003473

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/