Powered flight has evolved only four times in the history of the Earth, once among insects and three times among vertebrates. Insects were the first active fliers on the planet, and powered flight must be considered one of the major innovations in insect evolution. The first winged insects occur in the fossil record in the Early Carboniferous 350 million years ago. By this time, land plant evolution had been under way for more than 50 million years, and the first real forests had developed. One might speculate that the first forests provided an extensive three dimensional environment that promoted the evolution of wings in order to enable the insects to move more freely in this new, complex habitat. The first fauna of winged insects we observe 350 million years ago is quite diverse, indicating that their evolution had been going on for some time. Intriguingly, the fossil Rhyniognatha from the late Devonian I presented in the previous video displays some features in its mandibular structure, for example, the triangular shape that resembles those seen in winged insects. Furthermore, molecular phylogenetic studies that have been used to try to date events in insect evolution indicate that winged insects arose much earlier than indicated by the fossil record, suggesting that Rhyniognatha of 400 million years age might fall inside the time range of winged insects. However, you will recall that Rhyniognatha is a very incomplete fossil known mostly from its mouthparts, and no other actual fossil wings occur until 350 million years ago, so the evidence for an earlier occurrence of winged insects is at best indirect. Many hypotheses have been proposed to explain the origins of wings in insects. Here I will only mention two. One is the paranotal lobe hypothesis. Paranotal lobes are short outgrowths from the sides of the dorsal plate, the notum, in the thoracic segments. They can be observed in the living silverfish which are otherwise wing- and flightless. The paranotal lobes might allow a controlled descend during falling. It can be easily imagined that further development of the lobes might have enabled insects to glide. Eventually, articulations at the base of the paranotal lobes in the mid- and hind thorax might have evolved, and some of the muscles in the thorax becoming involved and enlarged to make possible real powered flight. The other hypothesis for wing evolution is the leg exite or gill hypothesis. The aquatic nymphs of mayflies living today have a series of gills along the sides of the abdomen, one pair of gills per segment. The gills are movable and have tracheal branches internally. The pattern of tracheae superficially resembles wing venation. No corresponding outgrowths are observed in any of the thoracic segments in mayflies or any other living insects. The gill hypothesis suggests that both the abdominal gills and the wings on the thoracic segments are derived from leg exites, that is, outgrowths from the basal parts of the legs in the thorax and abdomen, the abdomen having otherwise lost other traces of legs. In other words, thoracic wings and abdominal gills are serially homologous structures that have been adapted for different purposes; this is a recurrent theme in arthropod evolution in general. These and other hypotheses about wing evolution in insects have been discussed for decades. Recently, gene expression studies have been included as evidence for one hypothesis or another. Still, we have no final answer to exactly how insect wings evolved. Regardless of what formed the precursors of the wings were derived from, it is evident that even a small "proto-wing" would have given an aerodynamic advantage. Initially, it would probably have been used for gliding, but natural selection could have gradually expanded it and improved on the flight mechanism, eventually making powered flight possible. One of the first examples of fully winged insects from the Carboniferous are the Paleodictyoptera or "super-suckers", a name that is much easier to remember. They have a pair of wings on the mid- and hind thorax as is common for most winged insects, but in addition some Paleodictyoptera have short outgrowths resembling paranotal lobes from the dorsal part of the anterior thorax segment. These outgrowths have been interpreted as homologues of the wings on the posterior thorax segments. These outgrowths are not known from any living winged insects. The name "super-suckers" comes from the comparatively large size of these insects and what appears to be a well-developed proboscis on the head. This is the first instance of the formation of a proboscis known from the fossil insect record. It is thought that the paleodictyopterans fed by inserting their proboscis in plants, sucking liquid from them. Possible feeding scars from the proboscises have been identified in fossils of contemporary plants. The super-suckers were a prominent element in the insect fauna in the last part of the Paleozoic until they went extinct at the end of the Permian. Fossil evidence from the Carboniferous and onwards suggests that not only Paleodictyoptera were feeding on plant parts. Apart from external feeding and piercing-and-sucking, fossil plant material also displays signs of having been bored by insects or other arthropods; also gall formation is in evidence from around 300 million years ago. These various forms of feeding damage can be observed on leaves, stems, seeds, trunks and roots. It seems that before the end of the Paleozoic, the majority of the insect-plant interactions we know today have been established, and terrestrial ecosystems every bit as complex as the ones we see today are in place. Of course, the composition of these ecosystems with regard to exactly which taxa filled out the different ecological roles was very different from those in the present, both with regard to plants and animals. Subsequent development in the terrestrial ecosystems consisted primarily of a succession of taxa in established ecological roles as some groups went extinct while others rose to dominance. Insect groups that are still present today were also represented among the earliest winged insects. These include the Odonata, or dragonflies and damselflies. An early lineage of these, the Protodonata or griffin flies, includes the largest insects known. Like their living relatives, they seem to have been aerial predators, having the legs directed forward to be able to catch other insects in flight. Some of the griffin flies reached a wing span of 70 cm. Interestingly, the time of occurrence of these monster insects in the late Carboniferous or early Permian coincided with record high oxygen concentration during the history of insect evolution, 35% of atmospheric content compared to about 20% today. To understand how these two events might be correlated, we need to consider how insects handle gas exchange. Most insects breathe through tracheae. These are essentially air-filled tubes that extend from the surface of the insect throughout the body to the tissues, for example muscles that they need to supply with oxygen. This is very different from the situation in land vertebrates, where oxygen is transferred from the lungs to the circulatory system and conveyed in the blood to the tissues. In insects, the oxygen does not enter the body fluid at all before reaching the tissues, but stays in the air throughout the tracheal system. The advantage is that oxygen can diffuse much more quickly, approximately 10,000 times, in air than when dissolved in water. The tracheal system essentially works by passive diffusion, driven by the difference in oxygen concentration between the outside and the tissues inside the insect consuming the oxygen. This puts an upper limit on how large an insect can grow and still maintain efficient gas exchange. When the oxygen concentration in the atmosphere is high as it was in the late Carboniferous/early Permian, the speed of diffusion in the tracheal system increases. This in turn means that insects could grow bigger and still be able to transport oxygen efficiently to their tissues. The mass extinctions at the end of the Paleozoic at the Permo-Triassic boundary also affected the terrestrial insect communities. Eight out 27 insect orders recognized from this time went extinct. Among these were the Paleodictyoptera or "super-suckers". Other prominent insects that disappeared were the gigantic griffin flies, but their close relatives the dragonflies survived. Interestingly, by the end of the Permian the oxygen concentration had dropped to less than half of the peak levels in the early Permian and was lower than it is today. This probably put large insects at a disadvantage because of the reduced diffusion rates in the tracheal system. After the end Permian extinctions and throughout the Mesozoic different insect groups rose to dominance and the insect fauna gradually evolved to resemble the modern fauna. We will explore this in the next video.
