How do insects breathe? An outline of the tracheal system

Hexapods (including insects) have two pairs of openings on their thorax, called spiracles. These open into chitinous tubes called tracheae which then further subdivide until becoming less than 1 mm in diameter tracheoles, 2-3 µm away from metabolically-active tissues, forming a network all through the body of the insect, as seen in a larval Lestes dragonfly above (Kennedy, 1922). Oxygen diffuses from the outside through the spiracles and into this tracheal system – this is how an insect breathes, by diffusion. It’s also one of the major reasons behind the limit of insect size, and the converse is that the smallest insects don’t need a full tracheal system (e.g. the system of tiny ptiliid beetles is highly reduced) because the surface area:volume ratio doesn’t need to get expanded for diffusion to occur.

The spiracles are muscular valves in the insect body, and their opening can be controlled, mostly in order to regulate water loss, since the tracheal system is always saturated with water vapour. In fact, if control over them is lost (e.g. by nervous system malfunction), the insect will die of dehydration (Mellanby, 1935). Diagrammed above (Gullan & Cranston, 2004) is a cross-section of a typical spiracle, with the opening having hair for filtering and the valve for controlling air flow.

The tracheal system is very easy to look at by dissecting any large insect. Cockroaches are the classics, but if your students are squeamish, grasshoppers or bees should do the trick (even if they’re a bit harder to dissect for the inexperienced). Important is to keep the insect alive, anaesthetised with chloroform, because when the tracheae are filled with air, they’re silvery and easy to see. If you need a guide to such a dissection, contact me!

Development of the Tracheal System

The tracheal system derives from the invagination of the external cuticle of the embryonic insect – in effect, it’s nothing more than an extension of the exoskeleton. This is why tracheal tubes are always surrounded by cuticle. The precise details of the morphogenesis of the tracheal system is an area of intense research because of the tracheal system’s interesting branching pattern (similar to capillaries and lungs); see Affolter & Cassinus (2008) for a review. This research isn’t just pure, but has potential applications in regenerative medicine by informing us about how lungs are designed (Nichols et al., 2009).

The quick summary of the tracheal system’s development, from Drosophila larvae, is as follows: there are 10 pairs of placodes (~40 cells each) derived by budding off from the cuticle under the action of the genes trachealess and tango (Ghabrial et al., 2003), one in each embryonic segment corresponding to the adult 3nd thoracic segment to 8th abdominal segment. Each of these then undergoes the exact same process independently. They bud off sideways, to form 6 primary branches that then join together. Then, the left and the right sides connect, unifying all the branches together into the tracheal system; further branching then takes place to form tracheoles (Samakovlis et al., 1996), and the end result is what you see in the diagram above (Capinera, 2010). It’s an efficient repetitive process, one that is controlled by 200 genes (Ghabrial et al., 2011) regulated by metabolic and oxygen levels around the body (Centanin et al., 2010).

The tracheal system doesn’t scale up equally with size, as discussed in this post. The tracheal system increases in relative mass the larger an insect gets, a phenomenon called tracheal hypermetry. In tenebrionid beetles at least, this mass increase is most noticeable in the legs (Kaiser et al., 2007); whether this is a general trait for insects still needs to be researched. The size of individual tracheae depends on oxygen concentrations during development (Loudon, 1988), with lower concentrations making wider tracheae.

Gas Exchange

Carbon dioxide is excreted by pumping the thorax and/or abdomen, either up and down or by scrunching it along its length, increasing the pressure to push the CO2 out, as in a ventilation system. This also serves to increase the amount of oxygen that can enter the tissues, since it pushes the oxygen out of the tracheoles at the times when gas exchange is needed the most (movement and especially flight). When the insect is resting, this doesn’t occur (discontinuous gas exchange) – the spiracles are typically closed, opening only temporarily to let oxygen in and the built-up CO2 out (Chown et al., 2006).

More ventilatory pressure can be produced by the circulation system, which forms a “valve” at the junction between thorax and abdomen. The insect circulatory system is unique in that blood flow is occasionally reversed. When this happens, these valves contract, causing the tracheal system to get compressed and then relaxed – a ventilation effect.

Another method some insects have to increase the rate of gas exchange is called passive suction. They store CO2 as bicarbonate ions in the haemocoel (“bloodstream”), creating a vacuum when oxygen goes into the tissues but isn’t replaced by CO2, thereby drawing in more oxygen from the outside (basic laws of physics). The bicarbonate ions are occasionally converted to CO2 and released.

Westneat et al. (2003) reported a brand new mechanism of breathing in which the tracheae in the thorax and thorax compress and expand in fast cycles. This is hypothesised to be a form of active breathing, but its importance and extent is still under dispute.

Modifications of the Tracheal System

The tracheal system is modified according to the specific ecology of the insects, something most strikingly seen in aquatic larvae. Most of them have a closed tracheal system, meaning the spiracles are sealed off, with oxygen diffusing through that seal. To increase the surface area and thus increase the rate of oxygen diffusion, some have extended their tracheal system to outside the body, forming elaborate gills, best seen on the abdomen of larval mayflies, as diagrammed above (Bradley, 2009). The most unique system is that of chironomid larvae who are apneustic, getting their oxygen entirely by absorbing it through their cuticle and through gills. A similar set-up is found in Trichoptera (caddisflies), where this system is synapomorphic.

Other aquatic insects may not have special adaptations in the tracheal system, but will preserve a bubble of air around their spiracles (held up usually by specialised hairs), the plastron, and use it as a lung, maintaining it by surfacing occasionally. A tangential point: the limitations of the tracheal system are one of the reasons why there are no marine insects. These insects would have to stay near the surface, where they would get eaten to extinction by fish (whereas crustaceans migrate through the depths to avoid predation). Also, they would be too buoyant with air-filled tracheae to actually sink. A plastron would collapse from pressure at depth. Finally, pressure makes the total amount of air in the tracheal system limited, and the hypothetical marine insect wouldn’t be able to exchange gases fast enough to live.

The most interesting use of the tracheal system is for making sound, a function that evolved in the famous Madagascar hissing cockroach for use as social signals (Nelson, 1979); it’s also found in other members of the Gromphadorhina genus. The sound is achieved by the modification of the spiracle valve being enlarged, and the connected end of the trachea being similarly enlarged; in effect, it’s no different than a horn. Sounds are changed as the air flow in the horn changes (due to thorax pumping), and by how open the spiracle valve is.

Another accoustic use for the tracheal system is with the tympanal organ of those insects that can hear, e.g. grasshoppers and other orthopterans. In these insects, there is a trachea in the tympanal organ modified to become much larger or to form a tracheal sac, and it’s this that serves as the membrane which the sound waves vibrate (Heinrich et al., 1993), analogous to the three bones in our ear. A similar system exists in hemipterans, except there it’s called the tymbal organ, not the tympanal organ.

Evolution of the Tracheal System

Discussing the origin of the tracheal system requires a higher-level phylogeny of the arthropods. One of the hypotheses is that hexapods and myriapods (centi- and millipedes, and a couple of smaller groups) are sister groups forming the Tracheata clade (also known as Atelocerata, “without horns”, in reference to the lack of a secondary antenna). They’re united by their common possession of a tracheal system (among other things), meaning the tracheal system in hexapods is symplesiomorphic.

However, the validity of this taxon is under question with the Pancrustacea hypothesis gaining ground. This one states that Crustacea is paraphyletic and the Hexapoda are nested within it. This would mean that the tracheal system could be a hexapodan apomorphy. The pros and cons of each hypothesis fall outside the scope of this post.

The evolution of the tracheal system was a pivotal moment in the history of the insects, as it allowed insects to colonise both the land (by providing a way to breathe outside of water) and the air (by supporting aerobic metabolism in-flight (Komai, 2001)).

Outside of the hexapods, arthropods with a tracheal system include:

  • Myriapods: Interestingly, they also have haemocyanins (the arthropod equivalent of haemoglobin), even if they don’t transport oxygen in their haemolymph (“blood”);
  • Opilionids (harvestmen): one pair of spiracles on second opisthosomal segment;
  • Pseudoscorpions: Spiracles on first two opisthosomal segments;
  • Solifugae (camel spiders): Very extensive, insect-like tracheal system;
  • Many araneomorphs: Modify either the second or both book lungs into tracheae;
  • Ricinulei (hooded tickspiders): Spiracles at posterior of prosoma;
  • Acari (mites, ticks): Spiracles can be anywhere;
  • Onychophora (velvet worms): Same remark as myriapods, evidence from Kusche et al. (2002).

What is thus clear is that a tracheal system can emerge independently pretty easily in terrestrial arthropods, and that it is lacking in marine ones. Many arachnids have a tracheal system, and this is most likely to be a derived character originating from the book gills of a xiphosuran-like ancestor, not shared with the last common ancestor with the hexapods/tracheates (Bromhall, 1987). It’s thus obvious that it’s a key innovation enabling successful terrestrialisation in arthropods.

This is also a major hit against a possible homology between myriapodan and hexapodan tracheal systems, and thus also against the Tracheata hypothesis. However, as is often the case with such open questions in phylogeny, a homology can be reasonably argued for, as Klass & Kristensen (2001) do with a (biased) appeal to the phylogenetic value of spiracle positioning, muscularity, and innervation. I am personally against a homology, since I’m a big fan of the Pancrustacea concept, and a look at the entire tracheal system besides the spiracles reveals more differences than commonalities anyway.

And such rampant convergent evolution is interesting in and of itself. If I had a lab with lots of funding (a pipe dream at this point), one of the main projects will be to look at the tracheal system from an evo-devo perspective. How does a strikingly similar system arise independently in all arthropods that go on land? Are there any deep homologies to be found, or is it simply the easiest way to get a breathing function with an arthropodan body type? If I had to, I would use isopods as a model organism, since the Oniscoidea, a terrestrial isopod superfamily (woodlice et al.), has relatives that are all marine, and so marine and terrestrial can be easily compared to each other. But one can also look deeper in the arthropod by looking at onychophores and tardigrades – since the cuticle and its formation is essentially the same in all these groups, there is a reasonable case to be made for a similar pathway to tracheal system formation. If a reader happens to have a lab, get on it! (I’m not aware of any such research programs happening atm.)


Affolter M & Caussinus E. 2008. Tracheal branching morphogenesis in Drosophila: new insights into cell behaviour and organ architecture. Development 135, 2055-2064.

Bradley TJ. 2009. Animal Osmoregulation.

Bromhall C. 1987. Spider tracheal systems. Tissue and Cell 19, 793-807.

Capinera J. 2010. Insects and Wildlife: Arthropods and their relationships with wild vertebrate animals.

Centanin L, Gorr TA & Wappner P. 2010. Tracheal remodelling in response to hypoxia. Journal of Insect Physiology 56, 447-454.

Chown SL, Gibbs AG, Hetz SK, Klok CJ, Lighton JRB & Marais E. 2006. Discontinuous gas exchange in insects: a clarification of hypotheses and approaches. Physiological and Biochemical Zoology 2, 79.

Ghabrial AS, Luschnig S, Metzstein MM & Krasnow MA. 2003. Branching morphogenesis of the Drosophila tracheal system. Annual Review of Cell and Developmental Biology 19, 623-647.

Ghabrial AS, Levi BP & Krasnow MA. 2011. A Systematic Screen for Tube Morphogenesis and Branching Genes in the Drosophila Tracheal System. PLoS Genetics 7, e1002087.

Gullan PJ & Cranston P. 2004. The Insects: An Outline of Entomology, 3rd ed..

Heinrich R, Jatho M & Kalmring K. 1993. Acoustic transmission characteristics of the tympanal tracheae of bushcrickets (Tettigoniidae). II: Comparative studies of the tracheae of seven species. Journal of the Acoustical Society of America 93, 3481-3489.

Kaiser A, Klok CJ, Socha JJ, Lee W-K, Quinian MC & Harrison JF. 2007. Increase in tracheal investment with beetle size supports hypothesis of oxygen limitation on insect gigantism. PNAS 104, 13198-13203.

Kennedy CH. 1922. The Homologies of the Tracheal Branches in the Respiratory System of Insects. Ohio Journal of Science 22, 84-89.

Komai Y. 2001. Direct measurement of oxygen partial pressure in a flying bumblebee. Journal of Experimental Biology 204, 2999-3007.

Klass K & Kristensen NP. 2001. The ground plan and affinities of hexapods: recent progress and open problems. Annales de la Société Entomologique de France 37, 265-298.

Kusche K, Ruhberg H & Burmester T. 2002. A hemocyanin from the Onychophora and the emergence of respiratory proteins. PNAS 99, 10545-10548.

Loudon C. 1988. Development of Tenebrio molitor in low oxygen levels. Journal of Insect Physiology 34, 97-103.

Mellanby K. 1935. THE EVAPORATION OF WATER FROM INSECTS. Biological Reviews 10, 317-333.

Nelson MC. 1979. Sound production in the cockroach, Gromphadorhina portentosa: The sound-producing apparatus. Journal of Comparative Physiology A 132, 27-38.

Nichols JE, Niles JA & Cortiella J. 2009. Design and development of tissue engineered lung: Progress and challenges. Organogenesis 5, 57-61.

Samakovlis C, Manning G, Steneberg P, Hacohen N, Cantera R & Krasnow MA. 1996. Genetic control of epithelial tube fusion during Drosophila tracheal development. Development 122, 3531-3536.

Westneat MW, Betz O, Blob RW, Fezzaa K, Cooper WJ & Lee W-K. 2003. Tracheal Respiration in Insects Visualized with Synchrotron X-ray Imaging. Science 299, 558-560.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.