- Email: [email protected]
A century and a half of research on the evolution of insect flight
Arthropod Structure & Development xxx (2017) 1e6
Contents lists available at ScienceDirect
Arthropod Structure & Development journal homepage: www.elsevier.com/locate/asd
A century and a half of research on the evolution of insect flight David E. Alexander University of Kansas, Department of Ecology & Evolutionary Biology, 1200 Sunnyside Avenue, Rm. 2041 Lawrence, KS 66045-7534, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 2 June 2017 Received in revised form 7 November 2017 Accepted 18 November 2017 Available online xxx
The gill and paranotal lobe theories of insect wing evolution were both proposed in the 1870s. For most of the 20th century, the paranotal lobe theory was more widely accepted, probably due to the fundamentally terrestrial tracheal respiratory system; in the 1970s, some researchers advocated for an elaborated gill (“pleural appendage”) theory. Lacking transition fossils, neither theory could be definitively rejected. Winged insects are abundant in the fossil record from the mid-Carboniferous, but insect fossils are vanishingly rare earlier, and all earlier fossils are from primitively wingless insects. The enigmatic, isolated mandibles of Rhyniognatha (early Devonian) hint that pterygotes may have been present much earlier, but the question remains open. In the late 20th century, researchers used models to study the interaction of body and protowing size on solar warming and gliding abilities, and stability and glide effectiveness of many tiny adjustable winglets versus a single, large pair of immobile winglets. Living stoneflies inspired the surface-skimming theory, which provides a mechanism to bridge between aquatic gills and flapping wings. The serendipitously discovered phenomenon of directed aerial descent suggests a likely route to the early origin of insect flight. It provides a biomechanically feasible sequence from guided falls to fully-powered flight. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Directed aerial descent Flight evolution Gill theory Insect flight Paranotal lobe theory Pleural appendage theory
1. Earliest theories Well before Darwin and Wallace proposed the concept of evolution by natural selection, natural historians were attempting to explain the source or origin of insect wings. As reviewed in detail by Crampton (1916), authors throughout the early 1800s suggested that wings were modified from such structures as legs or gills. Not surprisingly, these suggestions were largely proposed on a background of Special Creation. Although at least some were based on sound anatomical work, they were not described in a way we would recognize as fitting a modern, evolutionary framework. The first scientific description of the origin of insect wings in a modern evolutionary context apparently was published just less than 150 years ago. In his 1870 animal anatomy book, Carl Gegenbaur proposed that insect wings evolved from tracheal gills similar to those present on modern-day aquatic insect larvae (Gegenbaur, 1870). Indeed, Gegenbaur devotes a full page to describing an evolutionary scenario: “The wings must be regarded as homologous with the lamellar tracheal gills … It is quite clear
E-mail address: [email protected].
that we must suppose that the wings did not arise as such, but were developed from other organs which had another function, such as tracheal gills; … Every increase of surface area increases the respiratory value of the organ, and so leads toward its future function. …” (from the English translation: Gegenbaur, 1878, p. 247). This description of an evolutionary change in function, and a possible mechanism to drive such a change, appears remarkably modern, particularly given that the first edition of Darwin's On the Origin of Species (Darwin, 1860) was less than a decade old when Gegenbaur was writing. Interestingly, Gegenbauer's hypothesis may have been at least partly inspired by suggestions from many decades earlier that insect wings shared various characteristics with tracheal gills (Oken, 1809e1811). Soon after Gegenbaur's book was published, Fritz Müller published a series of papers on termites (e.g., Müller, 1873a, b). Müller observed lateral tergal lobes on the thorax of certain termite nymphs. He concluded that these lobes were incipient wings (in the context of the recapitulation theory, widely accepted at the time) and since the lobes did not contain obvious tracheae, he rejected Gegenbaur's contention that wings arose from tracheal gills (Müller, 1875). Whereas Gegenbaur outlined a possible pathway for the evolution of wings, Müller rejected tracheal gills as a source for wings but did not immediately
https://doi.org/10.1016/j.asd.2017.11.007 1467-8039/© 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Alexander, D.E., A century and a half of research on the evolution of insect flight, Arthropod Structure & Development (2017), https://doi.org/10.1016/j.asd.2017.11.007
2
D.E. Alexander / Arthropod Structure & Development xxx (2017) 1e6
propose an alternative evolutionary pathway. (In a footnote, Müller said he planned to publish a more detailed comparison of his views versus Gegenbaur's, but apparently he never did, possibly because other authors made his case for him.) Thus, before 1880, a theory based on tracheal gills and a theory based on lateral tergal lobes were already in print, and variations on these theories would remain the two major insect-wing-origin theories into the 21st century. At about the time Gegenbaur and Müller were proposing that insect wings arose from tracheal gills or tergal lobes, researchers proposed a number of other possible origins for insect wings. Plateau (1871), for example, suggested that wings arose from hypertrophied spiracles. Jaworowski (1896) built on earlier suggestions that wings arose from legs by suggesting that wings and legs had a common origin, both being variations on dermal outgrowths that he believed were originally respiratory, and that gave rise to all arthropod appendages. None of these theories seemed to gain as much acceptance as those of Gegenbaur and Müller; by 1900, most entomologists seem to have settled on either the gill theory or the tergal lobe theory (Crampton, 1916). The original form of Gegenbaur's “gill theory” of wing evolution involved gills enlarging for improved gas exchange, then a partial transition to terrestrial life, in which gills might have aided gliding, or steering during leaps, and then with improvements in musculature and articulations, they became wings that could flap for powered flight. Woodworth, in his monograph on insect wing veins (Woodworth, 1906), devotes several pages to greatly elaborating and refining the gill theory. He points out that true tracheal gills e used primarily for gas exchange e would not make effective wing precursors, and some other intermediate stage would have been needed. He suggested that the stiffened covers that form part of (or replace) some gills in immature mayflies would have formed a better source for the evolution of wings. He also points out that the gill theory has the advantage of starting out with an appendage that already possesses a moveable articulation. Such an appendage would thus have no need to evolve an articulation and associated musculature from scratch. Early in the 20th century, Crampton (1916) reviewed the previous studies addressing insect wing origins. He seems to have coined the term “paranotal lobes” for Müller's tergal expansions, and the theory has been known as the “paranotal lobe theory” ever since. Crampton produced a detailed list of the evidence in favor of both the gill theory and the paranotal lobe theory, as well as listing many authors who had argued in favor of the former (e.g., Lubbock, 1873; Graber, 1877; Lang, 1888; Simroth, 1891; Pratt, 1897; Osborn, 1905) or the latter (e.g., Huxley, 1877; Pancritius, 1884; Korschelt and Heider, 1891; Packard, 1898; Powell, 1904). Although Crampton called the gill theory a “fascinatingly clever one,” and said “the logic of its appeal is almost irresistible,” (Crampton, 1916), he concluded that the weight of evidence e e.g., wings not being serially homologous with abdominal tracheal gills, and strong evidence for aerial respiration being primitive in insects (even aquatic ones) e was against the gill theory. He also described evidence favoring the paranotal theory e widespread occurrence of leaping ability and of paranota on the prothorax of extant insects e so he argued that the paranotal theory was a better fit to the available evidence. He also pointed out that evolving a new articulation for the wings should not be seen as a major stumbling block, given that tracheal gills would have also had to evolve a new articulation at some point (Crampton even describes the oribatid mites, often touted as examples of arthropods that have evolved a novel articulation for structures very much like paranotal lobes, which have been mentioned in connection with the paranotal lobe theory in books as recently as those by Dudley (2000) and Alexander (2015).)
2. Mid-20th century For the next five decades or more, the paranotal lobe theory seems to have been generally accepted (Snodgrass, 1931, 1935; Forbes, 1943; Wigglesworth, 1963; Flower, 1964). This acceptance may have been due as much to the general recognition that the insect respiratory system is of fundamentally terrestrial origin, as to Crampton's arguments. In one of the very few papers supporting an aquatic origin for insect flight during this period, Grant (1945) proposed a somewhat naïve model that actually has more in common with the “cursorial” theory for the evolution of flight in birds than with the Gegenbaur-Woodworth tracheal-gill theory. In the second half of the 20th century, some prominent scientists began arguing in support of a variation on the gill, or more generally, pleural appendage theory. Wigglesworth, who had earlier published a model of insect flight evolution that more or less took the paranotal lobe theory for granted (Wigglesworth, 1963), seems to have changed his mind and become a supporter of the pleural-appendage theory (Wigglesworth, 1973, 1976). He described mayfly larvae that use some of their modified gills for both covers and ventilation, or even as swimming paddles. He suggested that both abdominal gills and wings were modified coxal exites, and proposed that thoracic gill covers might have evolved into paddles. Such paddles then might have been used aerodynamically by semi-aquatic insects stranded by drying ponds and rivers, and blown into the air by winds and updrafts. In this way, those insects which managed to control their flightpath and return to water would have had an advantage, selecting for better aerial control, and eventually flapping. -Peck (1978, 1983, In an extensive series of papers, Kukalova 1985, 1987, 1997, 2008) used her interpretations of various fossils to theorize on the origin of wings. She developed a scheme to homologize all appendages of all arthropods, which included unrecognized leg segments, including two basal to the coxa that are either lost or fused with the body wall in extant insects. She described both wings and abdominal gills as exites of one of these hypothetical basal leg segments and explicitly stated that they are -Peck, 2008). She focused serially homologous structures (Kukalova much more on pattern that on process, and generally accepted elements of theories describing how thoracic gill covers could evolve -Peck, into flapping wings proposed by earlier authors (Kukalova 1978). Curiously, the arguments of both Wigglesworth and -Peck are somewhat reminiscent of the theory of Kukalova Jaworowski (1896), in that all three authors view legs, wings and gills as all being derived from a common source. -Peck, In spite of the arguments of Wigglesworth and Kukalova the paranotal lobe theory seems to have been more widely accepted throughout the second half of the 20th century. Most authors seemed to either take the paranotal lobe theory for granted (Wootton, 1976; Bitsch, 1994) or to actively argue in favor of it (Rasnitsyn, 1981; Quartau, 1986; Dudley, 2000). 3. Fossils Unfortunately, the fossil record has so far offered little help in -Peck understanding how insect flight arose. (Although Kukalova based her arguments heavily on fossils, later workers saw much less detail in the same fossils, e.g [Rasnitsyn and Novokshonov, thoux and Briggs, 2008], 1997; Deuve, 2001; Boxshall, 2004; Be and few other authors have placed such emphasis on fossil evidence as a basis for theories of wing origins.) Indeed, the lack of any transition fossils between the earliest primitively flightless hexapods and later, fully-volant fossil species has often led paleoentomologists to lament the lack of an “Archaeopteryx” for insects (e.g., Grimaldi and Engel, 2005).
Please cite this article in press as: Alexander, D.E., A century and a half of research on the evolution of insect flight, Arthropod Structure & Development (2017), https://doi.org/10.1016/j.asd.2017.11.007
D.E. Alexander / Arthropod Structure & Development xxx (2017) 1e6
Several fairly complete specimens of Collembola are known from the Rhynie chert, dated to approximately 400 mya (Whalley and Jarzembowski, 1981). Fossils of a head capsule and some other fragments of a basal species of Archeaognatha have been found in deposits from Quebec, dating to roughly 390 mya (Labandeira et al., 1988). Fragmentary remains of what appear to be sclerites from Archaeognatha or Zygentoma, found in upstate New York, have been dated to about 380 mya (Shear et al., 1984). These are obviously not winged insects, and they shed little or no light on the evolution of flight. Also found in the Rhynie chert were a pair of isolated mandibles, named Rhyniognatha hirsti but not originally described in detail. After languishing in obscurity for many decades, these mandibles were re-described by Engel and Grimaldi (2004), who interpreted them as being most similar to pterygote mandibles. If these were actually pterygote mandibles, their presence would push the origin of insect flight back at least 50 million years before the earliest unequivocal winged insects in the fossil record. Unfortunately, only the mandibles were preserved, and according to Michael Engel (personal communication), the preservation conditions of the Rhynie chertdboiling mineral springsdwould have made preservation of delicate structures like wings extremely unlikely. More recently, Haug and Haug (2017) re-examined Rhyniognatha using different methods. They concluded that their evidence more strongly supported Rhyniognatha being a centipede, but they could not rule out the possibility of it being a pterygote insect. Thus, Rhyniognatha offers a tantalizing suggestion that winged insects were present much earlier than other fossils suggest, but lacking more of the body, no definitive conclusion is possible. After a long gap, winged insects appear “fully-fledged” in the fossil record during the late Carboniferous at approximately 318 mya (Grimaldi and Engel, 2005), with fully-formed, fully-articulated wings, obviously capable of powered e flapping e flight. Indeed, in some late Carboniferous outcrops, fossils of isolated insect wings are the most numerous type of animal fossil. Some of the earliest common winged insect fossils are from the nowextinct Paleodictyoptera, with their enigmatic prothoracic winglets (Fig. 1). Arguments as to the possible function of these winglets, whether or not they possessed a movable articulation,
3
and their relevance to discussions about the evolution of wings, are ongoing. Similar arguments surround the potential mobility of the large wing pads common on immatures of Paleozoic insects: could they have been flapped? Haug et al. (2016) recently reviewed this controversy and re-examined fossils of immature insects described by -Peck and others. They used new photographic methods to Kukalova study development and potential mobility of the wing pads. They concluded that many late-instar nymphs may indeed have had moveable wing pads, although the broad-based fold-like attachment would have limited the pads to simple up-and-down movements. 4. Models and bounded ignorance Lacking transition fossils, and seeing little progress from discussions of traditional wing origin theories, in the late 20th century various researchers looked for new ways to approach the question. Some used experiments to try to tease apart the physical features and constraints that might have favored the acquisition of wings. Others looked at living insects to find behaviors that might have a bearing on the evolution of flight. Kingsolver and Koehl (1985), for example, used a series of physical models to study both thermal and aerodynamic (gliding) properties for “protowings” of various sizes, using bodies of different sizes. They discovered a thoughtprovoking overlap in performance between the sizes of protowings that worked as effective solar collectors and as effective gliding wings for models of larger size (Fig. 2). Their term “bounded ignorance,” from a later review article (Kingsolver and Koehl, 1994), neatly captures this approach: use experiments to determine what is physically possible, and to illustrate the physical constraints, without knowing what the actual structure looked like. Wootton and Ellington (1991) took a similar approach, also using physical models. Much earlier, Flower (1964) had shown that insect-sized cylinders could achieve surprisingly good glide angles (up to 45 ) if they could be stabilized at the proper falling angle. Wootton and Ellington tested model cylinders with either a series of nine pairs of small adjustable winglets down each side (representing abdominal gills), or a single pair of larger, fixed winglets near the front (representing paranotal lobes), each version with and
Fig. 1. Reconstruction of a palaeodictyopteran, based on several fossil specimens of Stenodictya. Redrawn from Kukalova (1970).
Please cite this article in press as: Alexander, D.E., A century and a half of research on the evolution of insect flight, Arthropod Structure & Development (2017), https://doi.org/10.1016/j.asd.2017.11.007
4
D.E. Alexander / Arthropod Structure & Development xxx (2017) 1e6
Fig. 2. Models of insects with winglets or protowings of various lengths, used in aerodynamic and thermal heating experiments. Redrawn from Kingsolver and Koehl (1985).
without long tail filaments (Fig. 3). Their main goal was to see if these structures could give the cylinders the necessary stability to fall at an angle that generated lift, and hence, horizontal motion. They found various combinations that produced a stable glide, including the small-winglets model when the incidence angle of the hindmost pair of winglets was carefully adjusted, and the single-pair model when provided with long tail filaments. They thus showed that both configurations could convert a fall into a steep glide, although the small-winglets model achieved a slightly shallower glide angle. Again, this gliding performance was best for larger models. Focussing on behavior rather than models, Marden and Kramer (1994) observed surface-skimming behavior in extant stoneflies (Plecoptera), where adults of some species are too weak to fly, but use flapping wings as propellers to skate over the surface of water, much like an Everglades airboat. They found that even with shortened wings, the stoneflies could still locomote over the water surface. Marden and Kramer suggested that surface skimming might be an intermediate stage in the evolution of flight. If an aquatic insect with mobile gill covers evolved an air-breathing adult stage, it might be able to skim across the water surface even with small, weak winglets. Selection for faster skimming would lead to longer wings and faster, stronger muscles. These improvements might then allow hops off the water at first, leading eventually to powered flight. Marden has since published extensively on this “surface skimming” theory of insect wing evolution
Fig. 3. Simplified models to test gliding stability and glide angle, representing the pleural appendage theory (left) and the paranotal lobe theory (right). Redrawn from Wootton and Ellington (1991).
(e.g., Marden and Kramer, 1995; Marden et al., 2000; Marden, 2003, 2013; Marden and Thomas, 2003). It is essentially a modification of the gill theorydan aquatic insect with gills or gill-covers exapted to function as propellers for skimmingdand so is subject to the counterargument that the evidence favors a terrestrial origin for insect flight. In a recent example of the bounded-ignorance approach, Yanoviak and Dudley (2017) compared the aerial behavior of a species of arboreal archeognathan with that of the semi-aquatic immatures of a species of mayfly that routinely leap from vertical surfaces as an escape behavior. These authors showed experimentally that the bristletails were capable of directed glides and aerial maneuvers, whereas the mayflies showed no directional control or orienting behaviordindeed, they tumbled end-overenddduring their leaps. Yanoviak and Dudley concluded that the immature mayflies’ gills played no aerodynamic role, and they suggested that gills were unlikely to have played a significant role in the evolution of flight. 5. Directed aerial descent In a classic case of scientific serendipity, Steven Yanoviak noticed that when some worker ants were dislodged from their perches high in the canopy of tropical trees, they seemed to be able to land back on the trunk more often than not. Robert Dudley (co-organizer of this symposium), an insect flight expert, happened to be at the same field station in Panama where Yanoviak was working. Along with ant expert Michal Kaspari, they performed experiments to quantify this behavior and discovered that some wingless ants, without any overt aerodynamic specializations, could indeed steer their falls and alight back on the tree trunk rather than falling helplessly to the forest floor (Yanoviak et al., 2005). They called this phenomenon “directed aerial descent,” and it has become an important new concept in discussions about the evolution of wings, as noted elsewhere in this symposium. Directed aerial descent (DAD) turns out to be very widespread among wingless arthropods, now having been described in spiders, bristletails, silverfish and various insect nymphs, in addition to both New World and Old World tropical arboreal ants (Hasenfuss, 2002; Yanoviak et al., 2009; Dudley and Yanoviak, 2011). A somewhat similar behavior is even used by some arboreal lizards (Oliver, 1951). The crucial insight from the DAD concept is that arboreal animals can have a significant behavioral head start on gliding before evolving any wing-like structures or other evident aerodynamic specializations. If arboreal animals were subjected to selection pressure to extend the horizontal distance they covered while falling, or perhaps to improve their maneuverability, those animals capable of performing DAD would be more likely to evolve anatomical aerodynamic modifications than animals without such a useful exaptation. In a recent review article, Dudley and Yanoviak fleshed out the connection between DAD and flight evolution with a proposed sequence of acquisition of behaviors leading to powered flight (Dudley and Yanoviak, 2011); this sequence is shown in Fig. 4. A key point is that any of these stages can be a completely viable longterm state for a given species; they are not necessarily mere brief waypoints on the way to evolving flight. Indeed, for each step in this sequencedexcept number 7, elaboration of wings and maneuversdexamples among modern animals can be found living in that stage. Although stimulated primarily by research focused on the evolution of flight in insects, the sequence proposed by Dudley and Yanoviak might be just as relevant for reconstructing the evolution of flapping flight in flying vertebrates. An arboreal origin for flight in bats and pterosaurs is generally accepted, so the relevance of this
Please cite this article in press as: Alexander, D.E., A century and a half of research on the evolution of insect flight, Arthropod Structure & Development (2017), https://doi.org/10.1016/j.asd.2017.11.007
D.E. Alexander / Arthropod Structure & Development xxx (2017) 1e6
Generalized Biomechanical Scenario for the Acquisition of Aerial Behaviors and Flight 1. Arboreality; residence on elevated substrate 2. Jumping (either volitional or via startle reflex); falling 3. Aerial righting and landing reflexes 4. Parachuting (drag-based descent) 5. Directed aerial descent (lift-based and drag-based; steep glide angles) 6. Gliding (predominantly lift-based; shallow glide angles) 7. Elaboration of wings and maneuvers 8. Flapping flight Fig. 4. Sequence of evolutionary steps leading from living in trees with no aerodynamic ability to completely powered, flapping flight. Based on data in Dudley and Yanoviak (2011).
sequence is obvious for them. Moreover, the recent discoveries of small, feathered, probably arboreal dinosaurs in the same lineage that contains birds (Godefroit et al., 2013; Xu et al., 2014) suggests that this sequence probably applies to the evolution of bird flight as well. In conclusion, I should note that the original concept for this presentation came from the insect section of my recent book on the evolution of animal flight (Alexander, 2015), although the final version of this review contains quite a bit of new material. I have avoided arguments based on evolutionary developmental biology (“evo-devo”), because of my impression that early results purporting to support one theory (Averof and Cohen, 1997) could be interpreted in other ways (Niwa et al., 2010; Engel et al., 2013) and did not throw as much light on the issue as some authors suggested. More recent evo-devo work has raised the intriguing possibility that insect wing evolution involved components of both paranotal lobes and pleural appendages (Niwa et al., 2010; Clark-Hachtel and Tomoyasu (2016); Elias-Neto and Belles, 2016); a recent study of wing pad structure from nymphal specimens of Paleodictyoptera (Prokop et al., 2017) lends some support to such a hybrid notalpleural theory of wing origins. Readers should find that the evodevo papers in this symposium provide a more detailed and timely description of the wing origin scenarios supported by evolutionary developmental evidence. Acknowledgements: I would like to thank Robert Dudley and Günther Pass for inviting me to speak in this symposium, and also to thank Robert for helpful comments on the manuscript and Günther for help with interpreting some of the articles written in German. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References Alexander, D.E., 2015. On the Wing: Insects, Pterosaurs, Birds, Bats and the Evolution of Animal Flight. Oxford University Press, New York. Averof, M., Cohen, S.M., 1997. Evolutionary origin of insect wings from ancestral gills. Nature 385, 627e630. thoux, O., Briggs, D.E.G., 2008. How Gerarus lost its head: stem-group orthoptera Be and paraneoptera revisited. Syst. Entomol. 33, 529e547. Bitsch, J., 1994. The morphological groundplan of hexapoda - critical-review of recent concepts. Ann. Soc. Entomol. Fr. 30, 103e129. Boxshall, G.A., 2004. The evolution of arthropod limbs. Biol. Rev. 79, 253e300. Clark-Hachtel, C.M., Tomoyasu, Y., 2016. Exploring the origin of insect wings from an evo-devo perspective. Curr. Opin. Insect Sci. 13, 77e85. Crampton, G.C., 1916. The phylogenetic origin and the nature of the wings of insects according to the paranotal theory. J. N. Y. Entomol. S. 24, 1e39. Darwin, C., 1860. On the Origin of Species. J. Murray, London. Deuve, T., 2001. The epipleural field in hexapods. Ann. Soc. Entomol. Fr. 37, 195e231. Dudley, R., 2000. The Biomechanics of Insect Flight: Form, Function, Evolution. Princeton University Press, Princeton, New Jersey. Dudley, R., Yanoviak, S.P., 2011. Animal aloft: the origins of aerial behavior and flight. Integr. Comp. Biol. 51, 926e936.
5
Elias-Neto, M., Belles, X., 2016. Tergal and pleural structures contribute to the formation of ectopic prothoracic wings in cockroaches. Roy. Soc. Open Sci. 3, e160347. https://doi.org/10.1098/rsos.160347. Engel, M.S., Davis, S.R., Prokop, J., 2013. Insect wings: the evolutionary developmental origins of nature's first flyers. In: Minelli, A., Boxshall, G., Fusco, G. (Eds.), Arthropod Biology and Evolution: Molecules, Development, Morphology. Springer Verlag, Berlin, pp. 269e298. Engel, M.S., Grimaldi, D.A., 2004. New light shed on the oldest insect. Nature 427, 627e630. Flower, J.W., 1964. On the origin of flight in insects. J. Insect. Physiol. 10, 81e88. Forbes, W.T.M., 1943. The origin of wings and venational types in insects. Am. Midl. Nat. 29, 381e405. Gegenbaur, C., 1870. Grundzüge der vergleichenden Anatomie. Wilhelm Engelmann, Leipzig. Gegenbaur, C., 1878. Elements of Comparative Anatomy. Macmillan and Co., London. Godefroit, P., Cau, A., Dong-Yu, H., Escuillie, F., Wenhao, W., Dyke, G., 2013. A Jurassic avialan dinosaur from China resolves the early phylogenetic history of birds. Nature 498, 359e362. Graber, V., 1877. Die Insekten. R. Oldenbourg, Munich. Grant, C., 1945. More on the origin of flight. Entomol. News 56, 243e245. Grimaldi, D.A., Engel, M.S., 2005. Evolution of the Insects. Cambridge University Press, New York. Hasenfuss, I., 2002. A possible evolutionary pathway to insect flight starting from lepismatid organization. J. Zool. Syst. Evol. Res. 40, 65e81. Haug, C., Haug, J.T., 2017. The presumed oldest flying insect: more likely a myriapod? PeerJ 5 e3402. https://doi.org/10.7717/peerj.3402. Haug, J.T., Haug, C., Garwood, R.J., 2016. Evolution of insect wings and development - new details from palaeozoic nymphs. Biol. Rev. 91, 53e69. Huxley, T.H., 1877. A manual of the anatomy of invertebrated animals. J. & A. Churchill, London. Jaworowski, A., 1896. Die Entwickelung des Spinnapparates bei Trochosa singoriensis €nge und der Flügel bei den Laxm. mit Berücksichtigung der Abdominalanha Insekten. Jenaische Z. für Naturwiss. 30, 39e74. Kingsolver, J.G., Koehl, M.A.R., 1985. Aerodynamics, thermoregulation, and the evolution of insect wings: differential scaling and evolutionary change. Evolution 39, 488e504. Kingsolver, J.G., Koehl, M.A.R., 1994. Selective factors in the evolution of insect wings. Annu. Rev. Entomol. 39, 425e451. Korschelt, E., Heider, K., 1891. Lehrbuch der vergleichenden Entwicklungsgeschichte der wirbellosen Thiere. Gustav Fischer, Jena. -Peck, J., 1978. Origin and evolution of insect wings and their relation to Kukalova metamorphosis, as documented by fossil record. J. Morphol. 156, 53e125. -Peck, J., 1983. Origin of the insect wing and wing articulation from the Kukalova arthropodan leg. Can. J. Zool. 61, 1618e1669. -Peck, J., 1985. Ephemeroid wing venation based upon new gigantic Kukalova Carboniferous mayflies and basic morphology phylogeny and metamorphosis of pterygote insects (Insecta, Ephemerida). Can. J. Zool. 63, 933e955. -Peck, J., 1987. New Carboniferous Diplura, Monura, and Thysanura, the Kukalova hexapod ground plan, and the role of thoracic lobes in the origin of wings (Insecta). Can. J. Zool. 65, 2327e2345. -Peck, J., 1997. Arthropod phylogeny and “basal” morphological structures. Kukalova In: Fortey, R.A., Thomas, R.H. (Eds.), Arthropod Relationships. Chapman and Hall, London, pp. 269e279. -Peck, J., 2008. Phylogeny of higher taxa in Insecta: finding synapomorphies Kukalova in the extant fauna and separating them from homoplasies. Evol. Biol. 35, 4e51. , J., 1970. Revisional study of the order palaeodictyoptera in the upper Kukalova carboniferous shales of commentry, France. Part III. Psyche 77, 1e44. Labandeira, C.C., Beall, B.S., Hueber, F.M., 1988. Early insect diversification: evidence bec. Science 242, 913e916. from a lower devonian bristletail from que Lang, A., 1888. Lehrbuch der vergleichenden Anatomie. Gustav Fischer, Jena. Lubbock, J., 1873. On the origin of insects. Journal of the linnean society of london. Zoology 11, 422e425. Marden, J.H., 2003. The surface-skimming hypothesis for the evolution of insect flight. Acta Zool. Cracov 46, 73e84. Marden, J.H., 2013. Reanalysis and experimental evidence indicate that the earliest trace fossil of a winged insect was a surface-skimming neopteran. Evolution 67, 274e280. Marden, J.H., Kramer, M.G., 1994. Surface-skimming stoneflies: a possible intermediate stage in insect flight evolution. Science 266, 427e430. Marden, J.H., Kramer, M.G., 1995. Locomotor performance of insects with rudimentary wings. Nature 377, 332e334. Marden, J.H., O'Donnell, B.C., Thomas, M.A., Bye, J.Y., 2000. Surface-skimming stoneflies and mayflies: the taxonomic and mechanical diversity of twodimensional aerodynamic locomotion. Physiol. Biochem. Zool. 73, 751e764. Marden, J.H., Thomas, M.A., 2003. Rowing locomotion by a stonefly that possesses the ancestral pterygote condition of co-occurring wings and abdominal gills. Biol. J. Linn. Soc. 79, 341e349. Müller, F., 1873a. Beitr€ age zur Kenntniss der Termiten. I. Die Geschlechtstheile der Soldaten von Calotermes. Jenaische Z. für Med. Naturwissenschaft VII 333e358. Müller, F., 1873b. Beitr€ age zur Kenntniss der Termiten. III. “Die Nymphen mit kurzen me form” (Lespe s). Jenaische Z. Flügelscheiden” (Hagen) “nymphes de la deuxie für Med. Naturwissenschaft VII 451e463. Müller, F., 1875. Beitr€ age zur Kenntniss der Termiten. IV. Die Larven von Calotermes rugosus Hag. Jenaische Z. für Naturwiss. IX 241e264.
Please cite this article in press as: Alexander, D.E., A century and a half of research on the evolution of insect flight, Arthropod Structure & Development (2017), https://doi.org/10.1016/j.asd.2017.11.007
6
D.E. Alexander / Arthropod Structure & Development xxx (2017) 1e6
Niwa, N., Akimoto-Kato, A., Niimi, T., Tojo, K., Machida, R., Hayashi, S., 2010. Evolutionary origin of the insect wing via integration of two developmental modules. Evol. Dev. 12, 168e176. Oken, L., 1809 e 1811. Lehrbuch der Naturphilosophie. Friedrich Frommann, Jena. Oliver, J.A., 1951. “Gliding” in amphibians and reptiles, with a remark on an arboreal adaptation in the lizard. Anolis carolinensis Voigt. Am. Nat. 85, 171e176. Osborn, H., 1905. The origin of the wings of insects. Proc. Ohio State Acad. Sci. 4, 333e339. Packard, A.S., 1898. A Text-book of Entomology. Macmillan, New York. Pancritius, P., 1884. Notiz über Flügelentwicklung bei den Insecten. Zool. Anz 7, 370e373. Plateau, F., 1871. Qu'est-ce que l'aile d'un insecte. Stettin Entomol. Ztg. 32, 33e42. Powell, P.B., 1904. Development of the wings of certain beetles and some studies of the origin of the wings. J. N. Y. Entomol. S. 12, 237e243. Pratt, H.S., 1897. Imaginal discs in insects. Psyche 8, 15e20. , M., Nel, A., Ho € rnschemeyer, T., Krzemin ska, E., Krzemin ski, W., Prokop, J., Pecharova Engel, M.S., 2017. Paleozoic nymphal wing pads support dual model of insect wing origins. Curr. Biol. 27, 263e269. Quartau, J.A., 1986. An overview of the paranotal theory on the origin of the insect wings. Publicaçoees do Inst. Zool. “Dr. Augusto Nobre” 194, 1e42. Rasnitsyn, A.P., 1981. A modified paranotal theory of insect wing origin. J. Morphol. 168, 331e338. Rasnitsyn, A.P., Novokshonov, V.G., 1997. On the morphology of Uralia maculata (Insecta: Diaphanopterida) from the early permian (kungurian) of ural (Russia). Entomol. Scand. 28, 27e38. Shear, W.A., Bonamo, P.M., Grierson, J.D., Rolfe, W.D.I., Smith, E.L., Norton, R.A., 1984. Early land animals in North America: evidence from Devonian age arthropods from Gilboa, New York. Science 224, 492e494.
Simroth, H., 1891. Die Entstehung der Landtiere, ein biologischer Versuch. Wilhelm Engelmann, Leipzig. Snodgrass, R.E., 1931. Morphology of the insect abdomen: Part I: general structure of the abdomen and its appendages. Smithson. Misc. Collect 85, 1e128. Snodgrass, R.E., 1935. Principles of Insect Morphology. McGraw-Hill Book Co., New York. Whalley, P., Jarzembowski, E.A., 1981. A new assessment of Rhyniella, the earliest known insect, from the Devonian of Rhynie, Scotland. Nature 291, 317. Wigglesworth, V.B., 1963. Origin of wings in insects. Nature 197, 97e98. Wigglesworth, V.B., 1973. Evolution of insect wings and flight. Nature 246, 127e129. Wigglesworth, V.B., 1976. The evolution of insect flight. In: Rainey, R.C. (Ed.), Insect Flight. Blackwell Scientific Publications, Oxford, U.K, pp. 255e269. Woodworth, C.W., 1906. The wing veins of insects. Univ. Calif. Publ. Entomol. 1, 1e152. Wootton, R.J., 1976. The fossil record and insect flight. In: Rainey, R.C. (Ed.), Insect Flight. Blackwell Scientific Publications, Oxford, U.K, pp. 235e254. Wootton, R.J., Ellington, C.P., 1991. Biomechanics and the origin of insect flight. In: Rayner, J.M.V., Wootton, R.J. (Eds.), Biomechanics in Evolution. Cambridge University Press, Cambridge, pp. 99e112. Xu, X., Zhou, Z., Dudley, R., Mackem, S., Chuong, C.-M., Erickson, G.M., Varricchio, D.J., 2014. An integrative approach to understanding bird origins. Science 346, 1341. Yanoviak, S.P., Dudley, R., 2017. Jumping and the aerial behavior of aquatic mayfly larvae (Myobaetis ellenae, Baetidae). Arthropod Struct. Dev. https://doi.org/ 10.1016/j.asd.2017.06.005. Yanoviak, S.P., Dudley, R., Kaspari, M., 2005. Directed aerial descent in canopy ants. Nature 433, 624e626. Yanoviak, S.P., Kaspari, M., Dudley, R., 2009. Gliding hexapods and the origins of insect aerial behaviour. Biol. Lett. 5, 510e512.
Please cite this article in press as: Alexander, D.E., A century and a half of research on the evolution of insect flight, Arthropod Structure & Development (2017), https://doi.org/10.1016/j.asd.2017.11.007
Contents lists available at ScienceDirect
Arthropod Structure & Development journal homepage: www.elsevier.com/locate/asd
A century and a half of research on the evolution of insect flight David E. Alexander University of Kansas, Department of Ecology & Evolutionary Biology, 1200 Sunnyside Avenue, Rm. 2041 Lawrence, KS 66045-7534, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 2 June 2017 Received in revised form 7 November 2017 Accepted 18 November 2017 Available online xxx
The gill and paranotal lobe theories of insect wing evolution were both proposed in the 1870s. For most of the 20th century, the paranotal lobe theory was more widely accepted, probably due to the fundamentally terrestrial tracheal respiratory system; in the 1970s, some researchers advocated for an elaborated gill (“pleural appendage”) theory. Lacking transition fossils, neither theory could be definitively rejected. Winged insects are abundant in the fossil record from the mid-Carboniferous, but insect fossils are vanishingly rare earlier, and all earlier fossils are from primitively wingless insects. The enigmatic, isolated mandibles of Rhyniognatha (early Devonian) hint that pterygotes may have been present much earlier, but the question remains open. In the late 20th century, researchers used models to study the interaction of body and protowing size on solar warming and gliding abilities, and stability and glide effectiveness of many tiny adjustable winglets versus a single, large pair of immobile winglets. Living stoneflies inspired the surface-skimming theory, which provides a mechanism to bridge between aquatic gills and flapping wings. The serendipitously discovered phenomenon of directed aerial descent suggests a likely route to the early origin of insect flight. It provides a biomechanically feasible sequence from guided falls to fully-powered flight. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Directed aerial descent Flight evolution Gill theory Insect flight Paranotal lobe theory Pleural appendage theory
1. Earliest theories Well before Darwin and Wallace proposed the concept of evolution by natural selection, natural historians were attempting to explain the source or origin of insect wings. As reviewed in detail by Crampton (1916), authors throughout the early 1800s suggested that wings were modified from such structures as legs or gills. Not surprisingly, these suggestions were largely proposed on a background of Special Creation. Although at least some were based on sound anatomical work, they were not described in a way we would recognize as fitting a modern, evolutionary framework. The first scientific description of the origin of insect wings in a modern evolutionary context apparently was published just less than 150 years ago. In his 1870 animal anatomy book, Carl Gegenbaur proposed that insect wings evolved from tracheal gills similar to those present on modern-day aquatic insect larvae (Gegenbaur, 1870). Indeed, Gegenbaur devotes a full page to describing an evolutionary scenario: “The wings must be regarded as homologous with the lamellar tracheal gills … It is quite clear
E-mail address: [email protected].
that we must suppose that the wings did not arise as such, but were developed from other organs which had another function, such as tracheal gills; … Every increase of surface area increases the respiratory value of the organ, and so leads toward its future function. …” (from the English translation: Gegenbaur, 1878, p. 247). This description of an evolutionary change in function, and a possible mechanism to drive such a change, appears remarkably modern, particularly given that the first edition of Darwin's On the Origin of Species (Darwin, 1860) was less than a decade old when Gegenbaur was writing. Interestingly, Gegenbauer's hypothesis may have been at least partly inspired by suggestions from many decades earlier that insect wings shared various characteristics with tracheal gills (Oken, 1809e1811). Soon after Gegenbaur's book was published, Fritz Müller published a series of papers on termites (e.g., Müller, 1873a, b). Müller observed lateral tergal lobes on the thorax of certain termite nymphs. He concluded that these lobes were incipient wings (in the context of the recapitulation theory, widely accepted at the time) and since the lobes did not contain obvious tracheae, he rejected Gegenbaur's contention that wings arose from tracheal gills (Müller, 1875). Whereas Gegenbaur outlined a possible pathway for the evolution of wings, Müller rejected tracheal gills as a source for wings but did not immediately
https://doi.org/10.1016/j.asd.2017.11.007 1467-8039/© 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Alexander, D.E., A century and a half of research on the evolution of insect flight, Arthropod Structure & Development (2017), https://doi.org/10.1016/j.asd.2017.11.007
2
D.E. Alexander / Arthropod Structure & Development xxx (2017) 1e6
propose an alternative evolutionary pathway. (In a footnote, Müller said he planned to publish a more detailed comparison of his views versus Gegenbaur's, but apparently he never did, possibly because other authors made his case for him.) Thus, before 1880, a theory based on tracheal gills and a theory based on lateral tergal lobes were already in print, and variations on these theories would remain the two major insect-wing-origin theories into the 21st century. At about the time Gegenbaur and Müller were proposing that insect wings arose from tracheal gills or tergal lobes, researchers proposed a number of other possible origins for insect wings. Plateau (1871), for example, suggested that wings arose from hypertrophied spiracles. Jaworowski (1896) built on earlier suggestions that wings arose from legs by suggesting that wings and legs had a common origin, both being variations on dermal outgrowths that he believed were originally respiratory, and that gave rise to all arthropod appendages. None of these theories seemed to gain as much acceptance as those of Gegenbaur and Müller; by 1900, most entomologists seem to have settled on either the gill theory or the tergal lobe theory (Crampton, 1916). The original form of Gegenbaur's “gill theory” of wing evolution involved gills enlarging for improved gas exchange, then a partial transition to terrestrial life, in which gills might have aided gliding, or steering during leaps, and then with improvements in musculature and articulations, they became wings that could flap for powered flight. Woodworth, in his monograph on insect wing veins (Woodworth, 1906), devotes several pages to greatly elaborating and refining the gill theory. He points out that true tracheal gills e used primarily for gas exchange e would not make effective wing precursors, and some other intermediate stage would have been needed. He suggested that the stiffened covers that form part of (or replace) some gills in immature mayflies would have formed a better source for the evolution of wings. He also points out that the gill theory has the advantage of starting out with an appendage that already possesses a moveable articulation. Such an appendage would thus have no need to evolve an articulation and associated musculature from scratch. Early in the 20th century, Crampton (1916) reviewed the previous studies addressing insect wing origins. He seems to have coined the term “paranotal lobes” for Müller's tergal expansions, and the theory has been known as the “paranotal lobe theory” ever since. Crampton produced a detailed list of the evidence in favor of both the gill theory and the paranotal lobe theory, as well as listing many authors who had argued in favor of the former (e.g., Lubbock, 1873; Graber, 1877; Lang, 1888; Simroth, 1891; Pratt, 1897; Osborn, 1905) or the latter (e.g., Huxley, 1877; Pancritius, 1884; Korschelt and Heider, 1891; Packard, 1898; Powell, 1904). Although Crampton called the gill theory a “fascinatingly clever one,” and said “the logic of its appeal is almost irresistible,” (Crampton, 1916), he concluded that the weight of evidence e e.g., wings not being serially homologous with abdominal tracheal gills, and strong evidence for aerial respiration being primitive in insects (even aquatic ones) e was against the gill theory. He also described evidence favoring the paranotal theory e widespread occurrence of leaping ability and of paranota on the prothorax of extant insects e so he argued that the paranotal theory was a better fit to the available evidence. He also pointed out that evolving a new articulation for the wings should not be seen as a major stumbling block, given that tracheal gills would have also had to evolve a new articulation at some point (Crampton even describes the oribatid mites, often touted as examples of arthropods that have evolved a novel articulation for structures very much like paranotal lobes, which have been mentioned in connection with the paranotal lobe theory in books as recently as those by Dudley (2000) and Alexander (2015).)
2. Mid-20th century For the next five decades or more, the paranotal lobe theory seems to have been generally accepted (Snodgrass, 1931, 1935; Forbes, 1943; Wigglesworth, 1963; Flower, 1964). This acceptance may have been due as much to the general recognition that the insect respiratory system is of fundamentally terrestrial origin, as to Crampton's arguments. In one of the very few papers supporting an aquatic origin for insect flight during this period, Grant (1945) proposed a somewhat naïve model that actually has more in common with the “cursorial” theory for the evolution of flight in birds than with the Gegenbaur-Woodworth tracheal-gill theory. In the second half of the 20th century, some prominent scientists began arguing in support of a variation on the gill, or more generally, pleural appendage theory. Wigglesworth, who had earlier published a model of insect flight evolution that more or less took the paranotal lobe theory for granted (Wigglesworth, 1963), seems to have changed his mind and become a supporter of the pleural-appendage theory (Wigglesworth, 1973, 1976). He described mayfly larvae that use some of their modified gills for both covers and ventilation, or even as swimming paddles. He suggested that both abdominal gills and wings were modified coxal exites, and proposed that thoracic gill covers might have evolved into paddles. Such paddles then might have been used aerodynamically by semi-aquatic insects stranded by drying ponds and rivers, and blown into the air by winds and updrafts. In this way, those insects which managed to control their flightpath and return to water would have had an advantage, selecting for better aerial control, and eventually flapping. -Peck (1978, 1983, In an extensive series of papers, Kukalova 1985, 1987, 1997, 2008) used her interpretations of various fossils to theorize on the origin of wings. She developed a scheme to homologize all appendages of all arthropods, which included unrecognized leg segments, including two basal to the coxa that are either lost or fused with the body wall in extant insects. She described both wings and abdominal gills as exites of one of these hypothetical basal leg segments and explicitly stated that they are -Peck, 2008). She focused serially homologous structures (Kukalova much more on pattern that on process, and generally accepted elements of theories describing how thoracic gill covers could evolve -Peck, into flapping wings proposed by earlier authors (Kukalova 1978). Curiously, the arguments of both Wigglesworth and -Peck are somewhat reminiscent of the theory of Kukalova Jaworowski (1896), in that all three authors view legs, wings and gills as all being derived from a common source. -Peck, In spite of the arguments of Wigglesworth and Kukalova the paranotal lobe theory seems to have been more widely accepted throughout the second half of the 20th century. Most authors seemed to either take the paranotal lobe theory for granted (Wootton, 1976; Bitsch, 1994) or to actively argue in favor of it (Rasnitsyn, 1981; Quartau, 1986; Dudley, 2000). 3. Fossils Unfortunately, the fossil record has so far offered little help in -Peck understanding how insect flight arose. (Although Kukalova based her arguments heavily on fossils, later workers saw much less detail in the same fossils, e.g [Rasnitsyn and Novokshonov, thoux and Briggs, 2008], 1997; Deuve, 2001; Boxshall, 2004; Be and few other authors have placed such emphasis on fossil evidence as a basis for theories of wing origins.) Indeed, the lack of any transition fossils between the earliest primitively flightless hexapods and later, fully-volant fossil species has often led paleoentomologists to lament the lack of an “Archaeopteryx” for insects (e.g., Grimaldi and Engel, 2005).
Please cite this article in press as: Alexander, D.E., A century and a half of research on the evolution of insect flight, Arthropod Structure & Development (2017), https://doi.org/10.1016/j.asd.2017.11.007
D.E. Alexander / Arthropod Structure & Development xxx (2017) 1e6
Several fairly complete specimens of Collembola are known from the Rhynie chert, dated to approximately 400 mya (Whalley and Jarzembowski, 1981). Fossils of a head capsule and some other fragments of a basal species of Archeaognatha have been found in deposits from Quebec, dating to roughly 390 mya (Labandeira et al., 1988). Fragmentary remains of what appear to be sclerites from Archaeognatha or Zygentoma, found in upstate New York, have been dated to about 380 mya (Shear et al., 1984). These are obviously not winged insects, and they shed little or no light on the evolution of flight. Also found in the Rhynie chert were a pair of isolated mandibles, named Rhyniognatha hirsti but not originally described in detail. After languishing in obscurity for many decades, these mandibles were re-described by Engel and Grimaldi (2004), who interpreted them as being most similar to pterygote mandibles. If these were actually pterygote mandibles, their presence would push the origin of insect flight back at least 50 million years before the earliest unequivocal winged insects in the fossil record. Unfortunately, only the mandibles were preserved, and according to Michael Engel (personal communication), the preservation conditions of the Rhynie chertdboiling mineral springsdwould have made preservation of delicate structures like wings extremely unlikely. More recently, Haug and Haug (2017) re-examined Rhyniognatha using different methods. They concluded that their evidence more strongly supported Rhyniognatha being a centipede, but they could not rule out the possibility of it being a pterygote insect. Thus, Rhyniognatha offers a tantalizing suggestion that winged insects were present much earlier than other fossils suggest, but lacking more of the body, no definitive conclusion is possible. After a long gap, winged insects appear “fully-fledged” in the fossil record during the late Carboniferous at approximately 318 mya (Grimaldi and Engel, 2005), with fully-formed, fully-articulated wings, obviously capable of powered e flapping e flight. Indeed, in some late Carboniferous outcrops, fossils of isolated insect wings are the most numerous type of animal fossil. Some of the earliest common winged insect fossils are from the nowextinct Paleodictyoptera, with their enigmatic prothoracic winglets (Fig. 1). Arguments as to the possible function of these winglets, whether or not they possessed a movable articulation,
3
and their relevance to discussions about the evolution of wings, are ongoing. Similar arguments surround the potential mobility of the large wing pads common on immatures of Paleozoic insects: could they have been flapped? Haug et al. (2016) recently reviewed this controversy and re-examined fossils of immature insects described by -Peck and others. They used new photographic methods to Kukalova study development and potential mobility of the wing pads. They concluded that many late-instar nymphs may indeed have had moveable wing pads, although the broad-based fold-like attachment would have limited the pads to simple up-and-down movements. 4. Models and bounded ignorance Lacking transition fossils, and seeing little progress from discussions of traditional wing origin theories, in the late 20th century various researchers looked for new ways to approach the question. Some used experiments to try to tease apart the physical features and constraints that might have favored the acquisition of wings. Others looked at living insects to find behaviors that might have a bearing on the evolution of flight. Kingsolver and Koehl (1985), for example, used a series of physical models to study both thermal and aerodynamic (gliding) properties for “protowings” of various sizes, using bodies of different sizes. They discovered a thoughtprovoking overlap in performance between the sizes of protowings that worked as effective solar collectors and as effective gliding wings for models of larger size (Fig. 2). Their term “bounded ignorance,” from a later review article (Kingsolver and Koehl, 1994), neatly captures this approach: use experiments to determine what is physically possible, and to illustrate the physical constraints, without knowing what the actual structure looked like. Wootton and Ellington (1991) took a similar approach, also using physical models. Much earlier, Flower (1964) had shown that insect-sized cylinders could achieve surprisingly good glide angles (up to 45 ) if they could be stabilized at the proper falling angle. Wootton and Ellington tested model cylinders with either a series of nine pairs of small adjustable winglets down each side (representing abdominal gills), or a single pair of larger, fixed winglets near the front (representing paranotal lobes), each version with and
Fig. 1. Reconstruction of a palaeodictyopteran, based on several fossil specimens of Stenodictya. Redrawn from Kukalova (1970).
Please cite this article in press as: Alexander, D.E., A century and a half of research on the evolution of insect flight, Arthropod Structure & Development (2017), https://doi.org/10.1016/j.asd.2017.11.007
4
D.E. Alexander / Arthropod Structure & Development xxx (2017) 1e6
Fig. 2. Models of insects with winglets or protowings of various lengths, used in aerodynamic and thermal heating experiments. Redrawn from Kingsolver and Koehl (1985).
without long tail filaments (Fig. 3). Their main goal was to see if these structures could give the cylinders the necessary stability to fall at an angle that generated lift, and hence, horizontal motion. They found various combinations that produced a stable glide, including the small-winglets model when the incidence angle of the hindmost pair of winglets was carefully adjusted, and the single-pair model when provided with long tail filaments. They thus showed that both configurations could convert a fall into a steep glide, although the small-winglets model achieved a slightly shallower glide angle. Again, this gliding performance was best for larger models. Focussing on behavior rather than models, Marden and Kramer (1994) observed surface-skimming behavior in extant stoneflies (Plecoptera), where adults of some species are too weak to fly, but use flapping wings as propellers to skate over the surface of water, much like an Everglades airboat. They found that even with shortened wings, the stoneflies could still locomote over the water surface. Marden and Kramer suggested that surface skimming might be an intermediate stage in the evolution of flight. If an aquatic insect with mobile gill covers evolved an air-breathing adult stage, it might be able to skim across the water surface even with small, weak winglets. Selection for faster skimming would lead to longer wings and faster, stronger muscles. These improvements might then allow hops off the water at first, leading eventually to powered flight. Marden has since published extensively on this “surface skimming” theory of insect wing evolution
Fig. 3. Simplified models to test gliding stability and glide angle, representing the pleural appendage theory (left) and the paranotal lobe theory (right). Redrawn from Wootton and Ellington (1991).
(e.g., Marden and Kramer, 1995; Marden et al., 2000; Marden, 2003, 2013; Marden and Thomas, 2003). It is essentially a modification of the gill theorydan aquatic insect with gills or gill-covers exapted to function as propellers for skimmingdand so is subject to the counterargument that the evidence favors a terrestrial origin for insect flight. In a recent example of the bounded-ignorance approach, Yanoviak and Dudley (2017) compared the aerial behavior of a species of arboreal archeognathan with that of the semi-aquatic immatures of a species of mayfly that routinely leap from vertical surfaces as an escape behavior. These authors showed experimentally that the bristletails were capable of directed glides and aerial maneuvers, whereas the mayflies showed no directional control or orienting behaviordindeed, they tumbled end-overenddduring their leaps. Yanoviak and Dudley concluded that the immature mayflies’ gills played no aerodynamic role, and they suggested that gills were unlikely to have played a significant role in the evolution of flight. 5. Directed aerial descent In a classic case of scientific serendipity, Steven Yanoviak noticed that when some worker ants were dislodged from their perches high in the canopy of tropical trees, they seemed to be able to land back on the trunk more often than not. Robert Dudley (co-organizer of this symposium), an insect flight expert, happened to be at the same field station in Panama where Yanoviak was working. Along with ant expert Michal Kaspari, they performed experiments to quantify this behavior and discovered that some wingless ants, without any overt aerodynamic specializations, could indeed steer their falls and alight back on the tree trunk rather than falling helplessly to the forest floor (Yanoviak et al., 2005). They called this phenomenon “directed aerial descent,” and it has become an important new concept in discussions about the evolution of wings, as noted elsewhere in this symposium. Directed aerial descent (DAD) turns out to be very widespread among wingless arthropods, now having been described in spiders, bristletails, silverfish and various insect nymphs, in addition to both New World and Old World tropical arboreal ants (Hasenfuss, 2002; Yanoviak et al., 2009; Dudley and Yanoviak, 2011). A somewhat similar behavior is even used by some arboreal lizards (Oliver, 1951). The crucial insight from the DAD concept is that arboreal animals can have a significant behavioral head start on gliding before evolving any wing-like structures or other evident aerodynamic specializations. If arboreal animals were subjected to selection pressure to extend the horizontal distance they covered while falling, or perhaps to improve their maneuverability, those animals capable of performing DAD would be more likely to evolve anatomical aerodynamic modifications than animals without such a useful exaptation. In a recent review article, Dudley and Yanoviak fleshed out the connection between DAD and flight evolution with a proposed sequence of acquisition of behaviors leading to powered flight (Dudley and Yanoviak, 2011); this sequence is shown in Fig. 4. A key point is that any of these stages can be a completely viable longterm state for a given species; they are not necessarily mere brief waypoints on the way to evolving flight. Indeed, for each step in this sequencedexcept number 7, elaboration of wings and maneuversdexamples among modern animals can be found living in that stage. Although stimulated primarily by research focused on the evolution of flight in insects, the sequence proposed by Dudley and Yanoviak might be just as relevant for reconstructing the evolution of flapping flight in flying vertebrates. An arboreal origin for flight in bats and pterosaurs is generally accepted, so the relevance of this
Please cite this article in press as: Alexander, D.E., A century and a half of research on the evolution of insect flight, Arthropod Structure & Development (2017), https://doi.org/10.1016/j.asd.2017.11.007
D.E. Alexander / Arthropod Structure & Development xxx (2017) 1e6
Generalized Biomechanical Scenario for the Acquisition of Aerial Behaviors and Flight 1. Arboreality; residence on elevated substrate 2. Jumping (either volitional or via startle reflex); falling 3. Aerial righting and landing reflexes 4. Parachuting (drag-based descent) 5. Directed aerial descent (lift-based and drag-based; steep glide angles) 6. Gliding (predominantly lift-based; shallow glide angles) 7. Elaboration of wings and maneuvers 8. Flapping flight Fig. 4. Sequence of evolutionary steps leading from living in trees with no aerodynamic ability to completely powered, flapping flight. Based on data in Dudley and Yanoviak (2011).
sequence is obvious for them. Moreover, the recent discoveries of small, feathered, probably arboreal dinosaurs in the same lineage that contains birds (Godefroit et al., 2013; Xu et al., 2014) suggests that this sequence probably applies to the evolution of bird flight as well. In conclusion, I should note that the original concept for this presentation came from the insect section of my recent book on the evolution of animal flight (Alexander, 2015), although the final version of this review contains quite a bit of new material. I have avoided arguments based on evolutionary developmental biology (“evo-devo”), because of my impression that early results purporting to support one theory (Averof and Cohen, 1997) could be interpreted in other ways (Niwa et al., 2010; Engel et al., 2013) and did not throw as much light on the issue as some authors suggested. More recent evo-devo work has raised the intriguing possibility that insect wing evolution involved components of both paranotal lobes and pleural appendages (Niwa et al., 2010; Clark-Hachtel and Tomoyasu (2016); Elias-Neto and Belles, 2016); a recent study of wing pad structure from nymphal specimens of Paleodictyoptera (Prokop et al., 2017) lends some support to such a hybrid notalpleural theory of wing origins. Readers should find that the evodevo papers in this symposium provide a more detailed and timely description of the wing origin scenarios supported by evolutionary developmental evidence. Acknowledgements: I would like to thank Robert Dudley and Günther Pass for inviting me to speak in this symposium, and also to thank Robert for helpful comments on the manuscript and Günther for help with interpreting some of the articles written in German. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References Alexander, D.E., 2015. On the Wing: Insects, Pterosaurs, Birds, Bats and the Evolution of Animal Flight. Oxford University Press, New York. Averof, M., Cohen, S.M., 1997. Evolutionary origin of insect wings from ancestral gills. Nature 385, 627e630. thoux, O., Briggs, D.E.G., 2008. How Gerarus lost its head: stem-group orthoptera Be and paraneoptera revisited. Syst. Entomol. 33, 529e547. Bitsch, J., 1994. The morphological groundplan of hexapoda - critical-review of recent concepts. Ann. Soc. Entomol. Fr. 30, 103e129. Boxshall, G.A., 2004. The evolution of arthropod limbs. Biol. Rev. 79, 253e300. Clark-Hachtel, C.M., Tomoyasu, Y., 2016. Exploring the origin of insect wings from an evo-devo perspective. Curr. Opin. Insect Sci. 13, 77e85. Crampton, G.C., 1916. The phylogenetic origin and the nature of the wings of insects according to the paranotal theory. J. N. Y. Entomol. S. 24, 1e39. Darwin, C., 1860. On the Origin of Species. J. Murray, London. Deuve, T., 2001. The epipleural field in hexapods. Ann. Soc. Entomol. Fr. 37, 195e231. Dudley, R., 2000. The Biomechanics of Insect Flight: Form, Function, Evolution. Princeton University Press, Princeton, New Jersey. Dudley, R., Yanoviak, S.P., 2011. Animal aloft: the origins of aerial behavior and flight. Integr. Comp. Biol. 51, 926e936.
5
Elias-Neto, M., Belles, X., 2016. Tergal and pleural structures contribute to the formation of ectopic prothoracic wings in cockroaches. Roy. Soc. Open Sci. 3, e160347. https://doi.org/10.1098/rsos.160347. Engel, M.S., Davis, S.R., Prokop, J., 2013. Insect wings: the evolutionary developmental origins of nature's first flyers. In: Minelli, A., Boxshall, G., Fusco, G. (Eds.), Arthropod Biology and Evolution: Molecules, Development, Morphology. Springer Verlag, Berlin, pp. 269e298. Engel, M.S., Grimaldi, D.A., 2004. New light shed on the oldest insect. Nature 427, 627e630. Flower, J.W., 1964. On the origin of flight in insects. J. Insect. Physiol. 10, 81e88. Forbes, W.T.M., 1943. The origin of wings and venational types in insects. Am. Midl. Nat. 29, 381e405. Gegenbaur, C., 1870. Grundzüge der vergleichenden Anatomie. Wilhelm Engelmann, Leipzig. Gegenbaur, C., 1878. Elements of Comparative Anatomy. Macmillan and Co., London. Godefroit, P., Cau, A., Dong-Yu, H., Escuillie, F., Wenhao, W., Dyke, G., 2013. A Jurassic avialan dinosaur from China resolves the early phylogenetic history of birds. Nature 498, 359e362. Graber, V., 1877. Die Insekten. R. Oldenbourg, Munich. Grant, C., 1945. More on the origin of flight. Entomol. News 56, 243e245. Grimaldi, D.A., Engel, M.S., 2005. Evolution of the Insects. Cambridge University Press, New York. Hasenfuss, I., 2002. A possible evolutionary pathway to insect flight starting from lepismatid organization. J. Zool. Syst. Evol. Res. 40, 65e81. Haug, C., Haug, J.T., 2017. The presumed oldest flying insect: more likely a myriapod? PeerJ 5 e3402. https://doi.org/10.7717/peerj.3402. Haug, J.T., Haug, C., Garwood, R.J., 2016. Evolution of insect wings and development - new details from palaeozoic nymphs. Biol. Rev. 91, 53e69. Huxley, T.H., 1877. A manual of the anatomy of invertebrated animals. J. & A. Churchill, London. Jaworowski, A., 1896. Die Entwickelung des Spinnapparates bei Trochosa singoriensis €nge und der Flügel bei den Laxm. mit Berücksichtigung der Abdominalanha Insekten. Jenaische Z. für Naturwiss. 30, 39e74. Kingsolver, J.G., Koehl, M.A.R., 1985. Aerodynamics, thermoregulation, and the evolution of insect wings: differential scaling and evolutionary change. Evolution 39, 488e504. Kingsolver, J.G., Koehl, M.A.R., 1994. Selective factors in the evolution of insect wings. Annu. Rev. Entomol. 39, 425e451. Korschelt, E., Heider, K., 1891. Lehrbuch der vergleichenden Entwicklungsgeschichte der wirbellosen Thiere. Gustav Fischer, Jena. -Peck, J., 1978. Origin and evolution of insect wings and their relation to Kukalova metamorphosis, as documented by fossil record. J. Morphol. 156, 53e125. -Peck, J., 1983. Origin of the insect wing and wing articulation from the Kukalova arthropodan leg. Can. J. Zool. 61, 1618e1669. -Peck, J., 1985. Ephemeroid wing venation based upon new gigantic Kukalova Carboniferous mayflies and basic morphology phylogeny and metamorphosis of pterygote insects (Insecta, Ephemerida). Can. J. Zool. 63, 933e955. -Peck, J., 1987. New Carboniferous Diplura, Monura, and Thysanura, the Kukalova hexapod ground plan, and the role of thoracic lobes in the origin of wings (Insecta). Can. J. Zool. 65, 2327e2345. -Peck, J., 1997. Arthropod phylogeny and “basal” morphological structures. Kukalova In: Fortey, R.A., Thomas, R.H. (Eds.), Arthropod Relationships. Chapman and Hall, London, pp. 269e279. -Peck, J., 2008. Phylogeny of higher taxa in Insecta: finding synapomorphies Kukalova in the extant fauna and separating them from homoplasies. Evol. Biol. 35, 4e51. , J., 1970. Revisional study of the order palaeodictyoptera in the upper Kukalova carboniferous shales of commentry, France. Part III. Psyche 77, 1e44. Labandeira, C.C., Beall, B.S., Hueber, F.M., 1988. Early insect diversification: evidence bec. Science 242, 913e916. from a lower devonian bristletail from que Lang, A., 1888. Lehrbuch der vergleichenden Anatomie. Gustav Fischer, Jena. Lubbock, J., 1873. On the origin of insects. Journal of the linnean society of london. Zoology 11, 422e425. Marden, J.H., 2003. The surface-skimming hypothesis for the evolution of insect flight. Acta Zool. Cracov 46, 73e84. Marden, J.H., 2013. Reanalysis and experimental evidence indicate that the earliest trace fossil of a winged insect was a surface-skimming neopteran. Evolution 67, 274e280. Marden, J.H., Kramer, M.G., 1994. Surface-skimming stoneflies: a possible intermediate stage in insect flight evolution. Science 266, 427e430. Marden, J.H., Kramer, M.G., 1995. Locomotor performance of insects with rudimentary wings. Nature 377, 332e334. Marden, J.H., O'Donnell, B.C., Thomas, M.A., Bye, J.Y., 2000. Surface-skimming stoneflies and mayflies: the taxonomic and mechanical diversity of twodimensional aerodynamic locomotion. Physiol. Biochem. Zool. 73, 751e764. Marden, J.H., Thomas, M.A., 2003. Rowing locomotion by a stonefly that possesses the ancestral pterygote condition of co-occurring wings and abdominal gills. Biol. J. Linn. Soc. 79, 341e349. Müller, F., 1873a. Beitr€ age zur Kenntniss der Termiten. I. Die Geschlechtstheile der Soldaten von Calotermes. Jenaische Z. für Med. Naturwissenschaft VII 333e358. Müller, F., 1873b. Beitr€ age zur Kenntniss der Termiten. III. “Die Nymphen mit kurzen me form” (Lespe s). Jenaische Z. Flügelscheiden” (Hagen) “nymphes de la deuxie für Med. Naturwissenschaft VII 451e463. Müller, F., 1875. Beitr€ age zur Kenntniss der Termiten. IV. Die Larven von Calotermes rugosus Hag. Jenaische Z. für Naturwiss. IX 241e264.
Please cite this article in press as: Alexander, D.E., A century and a half of research on the evolution of insect flight, Arthropod Structure & Development (2017), https://doi.org/10.1016/j.asd.2017.11.007
6
D.E. Alexander / Arthropod Structure & Development xxx (2017) 1e6
Niwa, N., Akimoto-Kato, A., Niimi, T., Tojo, K., Machida, R., Hayashi, S., 2010. Evolutionary origin of the insect wing via integration of two developmental modules. Evol. Dev. 12, 168e176. Oken, L., 1809 e 1811. Lehrbuch der Naturphilosophie. Friedrich Frommann, Jena. Oliver, J.A., 1951. “Gliding” in amphibians and reptiles, with a remark on an arboreal adaptation in the lizard. Anolis carolinensis Voigt. Am. Nat. 85, 171e176. Osborn, H., 1905. The origin of the wings of insects. Proc. Ohio State Acad. Sci. 4, 333e339. Packard, A.S., 1898. A Text-book of Entomology. Macmillan, New York. Pancritius, P., 1884. Notiz über Flügelentwicklung bei den Insecten. Zool. Anz 7, 370e373. Plateau, F., 1871. Qu'est-ce que l'aile d'un insecte. Stettin Entomol. Ztg. 32, 33e42. Powell, P.B., 1904. Development of the wings of certain beetles and some studies of the origin of the wings. J. N. Y. Entomol. S. 12, 237e243. Pratt, H.S., 1897. Imaginal discs in insects. Psyche 8, 15e20. , M., Nel, A., Ho € rnschemeyer, T., Krzemin ska, E., Krzemin ski, W., Prokop, J., Pecharova Engel, M.S., 2017. Paleozoic nymphal wing pads support dual model of insect wing origins. Curr. Biol. 27, 263e269. Quartau, J.A., 1986. An overview of the paranotal theory on the origin of the insect wings. Publicaçoees do Inst. Zool. “Dr. Augusto Nobre” 194, 1e42. Rasnitsyn, A.P., 1981. A modified paranotal theory of insect wing origin. J. Morphol. 168, 331e338. Rasnitsyn, A.P., Novokshonov, V.G., 1997. On the morphology of Uralia maculata (Insecta: Diaphanopterida) from the early permian (kungurian) of ural (Russia). Entomol. Scand. 28, 27e38. Shear, W.A., Bonamo, P.M., Grierson, J.D., Rolfe, W.D.I., Smith, E.L., Norton, R.A., 1984. Early land animals in North America: evidence from Devonian age arthropods from Gilboa, New York. Science 224, 492e494.
Simroth, H., 1891. Die Entstehung der Landtiere, ein biologischer Versuch. Wilhelm Engelmann, Leipzig. Snodgrass, R.E., 1931. Morphology of the insect abdomen: Part I: general structure of the abdomen and its appendages. Smithson. Misc. Collect 85, 1e128. Snodgrass, R.E., 1935. Principles of Insect Morphology. McGraw-Hill Book Co., New York. Whalley, P., Jarzembowski, E.A., 1981. A new assessment of Rhyniella, the earliest known insect, from the Devonian of Rhynie, Scotland. Nature 291, 317. Wigglesworth, V.B., 1963. Origin of wings in insects. Nature 197, 97e98. Wigglesworth, V.B., 1973. Evolution of insect wings and flight. Nature 246, 127e129. Wigglesworth, V.B., 1976. The evolution of insect flight. In: Rainey, R.C. (Ed.), Insect Flight. Blackwell Scientific Publications, Oxford, U.K, pp. 255e269. Woodworth, C.W., 1906. The wing veins of insects. Univ. Calif. Publ. Entomol. 1, 1e152. Wootton, R.J., 1976. The fossil record and insect flight. In: Rainey, R.C. (Ed.), Insect Flight. Blackwell Scientific Publications, Oxford, U.K, pp. 235e254. Wootton, R.J., Ellington, C.P., 1991. Biomechanics and the origin of insect flight. In: Rayner, J.M.V., Wootton, R.J. (Eds.), Biomechanics in Evolution. Cambridge University Press, Cambridge, pp. 99e112. Xu, X., Zhou, Z., Dudley, R., Mackem, S., Chuong, C.-M., Erickson, G.M., Varricchio, D.J., 2014. An integrative approach to understanding bird origins. Science 346, 1341. Yanoviak, S.P., Dudley, R., 2017. Jumping and the aerial behavior of aquatic mayfly larvae (Myobaetis ellenae, Baetidae). Arthropod Struct. Dev. https://doi.org/ 10.1016/j.asd.2017.06.005. Yanoviak, S.P., Dudley, R., Kaspari, M., 2005. Directed aerial descent in canopy ants. Nature 433, 624e626. Yanoviak, S.P., Kaspari, M., Dudley, R., 2009. Gliding hexapods and the origins of insect aerial behaviour. Biol. Lett. 5, 510e512.
Please cite this article in press as: Alexander, D.E., A century and a half of research on the evolution of insect flight, Arthropod Structure & Development (2017), https://doi.org/10.1016/j.asd.2017.11.007