Insect Flight: Phylogenetic Intro

Of all the insects, 98% of them have wings (the Pterygota). That statistic alone should emphasise how important the wing is to the success of the insects – remember that they are the most diverse group of animals. Flight has enabled them to invade just about every ecosystem there is by allowing them to use resources scattered all over the landscape, as well as permitting extremely effective ways to escape from predators (of course, the predators reacted, most notably the spiders with whom they coevolved). They also allow the critters to spread all over the world and colonise new habitats. Also, as a small trophy, they were the very first animals to fly.

The Pterygota are a monophyletic group – the wings more or less seal the deal, although two other characters can be mustered up:

  • Loss of an aorta-derived ring around the oesophagus (present in the other hexapods).
  • Diaphragms in the legs separating the distal- and ventral-directed haemolymph streams.

There are other characters, such as the fused tentorium (cephalic endoskeleton), but they don’t count as true apomorphies since they can also be found in some (not all) basal hexapods. The fact that all flying insects have direct copulation instead of a female picking up a sperm packet is also not a reliable character, since Carboniferous odonates had no copulatory organs that we know of, so it is not an ancestral freature. Nevertheless, as I said, the wings on the meso- and metathorax are the key apomorphy of the Pterygota – Carboniferous fossils even hint that the prothorax may also have had ‘wings’ once, although these are nothing more than projections instead of fully-developed wings.

When talking of the insect wing, we’re not just talking of a simple geometrical change and reduction in weight, as it is in vertebrates. Insect wings are, for all intents and purposes, novel structures, even if they do end up just being derived limbs (see later post). Not only is the wing incredibly complex, how the wing attaches to the thorax is also important, and let’s not forget that wing flapping is controlled by the thorax, so the whole musculature there had to be overhauled.

The winged insects have been classically (since Martynov (1925)) divided into the Neoptera and the Palaeoptera. The latter is proposed as being the most basal, and comprises the odonates and the mayflies – they are characterised by an inability to fold their wings backwards, and this is then supposed to be the ancestral condition. The Neoptera are definitely a monophyletic group due to their ability to fold their wings back over their abdomen, but the phylogeny of the Palaeoptera has been a niggling issue since Martynov’s publication – the aptly-named “Palaeoptera Problem”. The picture below shows the three main concurrent hypotheses.

Palaeoptera Problem. Simon et al. (2009).

I will not throw my weight behind any of them for the sake of fairness; let it be known that I strongly support the Metapterygota hypothesis, so if any bias does sneak through, I apologise.

The Metapterygota hypothesis is supported by many morphological characters, most of which are not even involved with flight. For example, both odonates and neopterans have the same type of mandible that allows only limited movement, but much more powerful force; the way it works has to do with the joint, and a related trait that is seen in the metapterygotes is the loss of many ancestrally-present muscles associated with the mandible. Another telling character is the lack of a subimago: the pre-final moult of a mayfly and other hexapods has all the morphological features of the adult, but is sexually immature; this life cycle is not found in the metapterygotes, where only the final imago (adult) emerges with wings and the complete set of features. Another uniting characteristic of the metapterygotes is that the wings and the meso- and metathoracic legs are each supplied by two trachaeas, whereas in the others, they each have just one trachaea. As always, there are other characters, but they are sometimes found in other lineages and have not been proven to be secondarily derived there.

The Palaeoptera hypothesis is somewhat older, especially gaining popularity because of Hennig’s support in his 1969 book on the phylogeny of the insects, and has recently experienced an upsurge in popularity. It’s mostly based on wing venation characters, which in this case are not very reliable, as we’ll see later.

As for the Chiastomyaria hypothesis, it depends on how the thoracic and flight musculature is interpreted. While it is true that the odonates have quite remarkable musculature, it is not hard to imagine it being derived from the neopteran ground plan; the Chiastomyaria hypothesis demands that it is completely unique.

I only mentioned the morphological arguments here; that’s because as usual, molecular analyses (which are the ones referenced in the picture) are all over the place and suffer from their usual taxon and genetic sampling problems.

The picture below summarises the situation for the Palaeoptera and Metapteraygota hypotheses. Ask for more info – I only include it so you can see how easy it is to gather up characters, and how hard it is to actually tell which are useful or not.

Palaeoptera Problem (Resolutions). Willkommen & Hörnschemeyer (2007).

We won’t go into neopteran phylogeny, as it is an even more notoriously difficult issue than the Palaeoptera Problem and should be the topic of its own post (if requested!). Suffice it to say that the Neoptera comprises 11 taxa: Plecoptera, Dictyoptera, Grylloblattodea, Dermaptera, Orthoptera, Phasmatodea, Embioptera, Zoraptera, Acercaria, Mantophasmatodea and, of course, Endopterygota/Holometabola. Most, if not all, the insects/hexapods you commonly recognise fall into one of those groups, from grasshoppers (Orthoptera) to cockroaches (Dictyoptera) to butterflies and beetles (Holometabola). You can group those taxa in any way you like to make a phylogenetic hypothesis – it’s probably been published aleady with reasonable evidence.

The figure below shows the standard, classical view of pterygotan phylogeny. The various annotations are apomorphic character states in the wing base – ask if you want more info. The main problem with trying to figure out the relationships between the pterygotes (a task that’s been attempted since the times of Aristotle, mind you) is that we have no way to root the tree – without concrete knowledge of the ground plan (i.e. fossils), there’s no way to make the reliably objective comparisons needed for proper phylogenetics.

Standard pterygote phylogeny. Hörnschemeyer & Willkommen (2007).


Hörnschemeyer, T. & Willkommen, J. 2007. The Contribution of Flight System Characters to the Reconstruction of the Phylogeny of the Pterygota. Arthropod Systematic & Phylogeny 65, 15-23. [Warning: PDF!]

Simon, S., Strauss, S., von Haeseler, A. & Hadrys, H. 2009. A Phylogenomic Approach to Resolve the Basal Pterygote Divergence. Molecular Biology and Evolution 26, 2719-2730.

Willkommen, J. & Hörnschemeyer, T. 2007. The homology of wing base sclerites and flight muscles in Ephemeroptera and Neoptera and the morphology of the pterothorax of Habroleptoides confusa (Insecta: Ephemeroptera: Leptophlebiidae). Arthropod Structure & Development 36, 253-269.

Research Blogging necessities :)

Simon, S., Strauss, S., von Haeseler, A., & Hadrys, H. (2009). A Phylogenomic Approach to Resolve the Basal Pterygote Divergence Molecular Biology and Evolution, 26 (12), 2719-2730 DOI: 10.1093/molbev/msp191

Leave a Reply