Inspired by comments made following the emu dissection article from Monday, I got thinking about elongate tracheae in birds. As we’ll see, this subject is fertile ground if you like serious weirdness and spectacular extremes [Trumpet manucodes shown here, from wikipedia.. all will become clear]. Before we get to that serious weirdness and those spectacular extremes, a bit of basic anatomy…
The trachea – or windpipe – is, as I’m sure you already know given that you’ve got one, a tube that extends along the ventral surface of the neck from the larynx to the lungs [singular: trachea, plural: tracheae. Thanks David]. While the tetrapod larynx normally contains vocal chords used in sound production, it does not contain vocal chords in birds and hence has long been thought to have no role in vocalisation at all. Recent work has shown that this is not true, and that a significant amount of laryngeal movement in birds helps them to control their vocalisations. In fact some birds (like crowing roosters) descend the larynx during vocalisation: huh, once upon a time, laryngeal descent was thought unique to humans [for more on laryngeal descent in non-humans, see the Fallow deer article]. Incidentally, the role of laryngeal movement in avian vocalisation was initially predicted on the basis of dissection data (Homberger 1986, 1999) and, about 20 years later, confirmed by cineradiography.
While the larynx does play a role in vocalisation after all, the syrinx (located at the base of the trachea and deep within the chest) is more important. It’s a pretty complicated structure, involving muscles, membranes and internal foramina. There’s a lot that could be said about it (it’s tremendously variable), so I’ll leave well alone for now. Incidentally, syringeal musculature is absent in some birds (namely some storks), so the only noises they make are ‘mechanical’ ones like bill-clattering. Posterior to the syrinx, the trachea bifurcates into the two primary bronchi, and these then lead into the lungs. Cartilaginous tracheal rings are arranged along the trachea’s length and keep it stiff and open all the time [Trumpeter swan trachea and sternum shown here, from Banko (1960). Read on...].
Hypertrophied trachea in swans and cranes, and ‘problems’ of dead space and storage
In the vast majority of birds, the trachea is, apart from the syrinx, pretty simple. However, several groups have evolved incredibly elongate tracheae where extra loops, hoops and other deviations increase its length, sometimes to an utterly ridiculous degree. The most familiar birds with tracheal looping are swans: tracheal length is variable in the four species that exhibit it, but in all of them the trachea forms an S-shaped loop at its base. In species without an elongate tracheal loop, like the Black swan Cygnus atratus, the base of the trachea fits into the space between the furcula and the anterior edge of the sternum. In contrast, the extremely long loop of the Trumpeter swan C. buccinator invades the chest cavity and is attached to the medial (inner) surface of the sternum, forming an obvious tube on the sternal surface (Banko 1960). You can see it in the photo below: the trachea invades the sternum at its anterior end and then forms the bulbous mass you can see on the upper surface.
The Trumpeter swan trachea is more than three times larger in volume than expected for a bird of its size (Hinds & Calder 1971), and having a trachea of this length means that the birds have to cope with a tremendous dead space (dead space = the amount of air that has to be shunted out of the way before the animal can take in a new breath). Here it would be all too easy to go off on a tangent and talk about air-sacs and avian respiration, and I regret I can’t do that… I will say, however, that dead space is probably not as much of a problem for birds as it is for mammals and it’s probably relatively easy for birds (with their extensive air sac system) to move large volumes of air around (Clench 1978, p. 428). Must resist mention of sauropod dinosaurs – – err, dammit.
Very long tracheal loops are also present in cranes. Having said that the condition in swans is most familiar, I should note that this could instead be said for cranes: Kaiser Friedrich II described crane tracheae in 1250, the French naturalist Pierre Bollonius was writing about tracheal elongation in cranes again during the 1500s, and crane tracheal morphology was illustrated by V. Coiter in 1575 (Johnsgard 1983, Fitch 1999). In some cranes, the tracheal loop comes into contact with the anterior margin of the sternum and the sternum’s leading edge is strongly recessed to receive it [see adjacent diagrams, from Johnsgard (1983)]. In forms with longer tracheal loops, the trachea properly invades the body of the sternum, and in the species with the very longest tracheae (like Eurasian cranes Grus grus and Whooping cranes G. americana) the entire length of the sternum is occupied by tracheal looping: the loops double-up on themselves, forming a complex coil. In the Japanese crane G. japonicus, the trachea can be 1.6 m long in total (Johnsgard 1983). Crane sterna, and their associated tracheae, are shown in the adjacent image [from Johnsgard (1983)]. The more ‘extreme’ European crane and Whooping crane are shown at the bottom.
Elongate tracheae are also present in magpie geese, in chachalacas and various other cracids, in ptarmigans, capercaillies and some guineafowl, in some spoonbills, in painted snipe, and in limpkins. The presence of an elongate trachea in the Capercaillie Tetrao urogallus is particularly interesting given the claim that this is one of very few birds that can produce infrasound (Moss & Lockie 1978). However, later testing showed that capercaillie calls did not include an infrasonic component after all: however, the ‘flutter jumps’ that these large birds use as part of their displays do create infrasound (Lieser et al. 2005). At the moment the only confirmed infrasonic birds are cassowaries (and they don’t have elongate tracheae). Thanks to exceptionally preserved specimens, one of which was discovered with an in-situ trachea [shown here, after Worthy], we know that two moa – Euryapteryx and Emeus – had very long, looping tracheae (Worthy & Holdaway 2002).
Where do birds store their elongate tracheae? We’ve already seen that the tracheae of swans and cranes are either coiled up against, or even upon or within, the sternum. In the guineafowl with looped tracheae (two members of the genus Guttera, most notably the Crested guineafowl G. pucherani), the loop is fitted away inside the hollow furcula (Chapin 1932, Frith 1994). The European spoonbill Platalea leucorodia and Yellow-billed stork Mycteria ibis (apparently) are unusual in that the coiled tracheal loop is within the thoracic cavity: that is, medial to the clavicles and sternum, though not in close association with the sternum as is the case in swans. In a bizarre and fascinating case of convergence, the elongate moa trachea was also intrathoracic (Worthy & Holdaway 2003). It’s more normal, however, for the tracheal loop or loops to be positioned ventral to the sternum, and be tucked up against the pectoral muscles. This is the case in Magpie geese Anseranas semipalmata, cracids, Painted snipe Rostratula benghalensis, and in the birds-of-paradise that have long tracheae (Frith 1994). And I’ve left birds-of-paradise to last. Why? Because their tracheal loops are perhaps the most extreme of all. They. Are. Ridiculous.
Most ridiculous of all: birds-of-paradise
Two kinds of birds-of-paradise have looped tracheae: the Trumpet bird or Trumpet manucode Phonygammus keraudrenii and the manucodes proper (Manucodia). It is probable that, among birds-of-paradise, the long tracheae possessed by these two taxa evolved only once, as Phonygammus and Manucodia appear to be sister-taxa (in fact, some authors include Phonygammus within Manucodia). The elongate Trumpet bird trachea was first reported in 1826, figured and discussed a few times in the late 1800s, and reviewed comprehensively by Clench (1978): it really has to be seen to be believed. Aaaand, here it is… [as represented by three specimens, from Clench (1978)]…
It is extremely variable ontogenetically, sexually, and between populations, but – at its most extreme – consists of five coils that form a spiral on the animal’s ventral surface. The trachea descends posteroventrally on the bird’s left side, extends posteriorly all the way to the cloaca, then curves to the right before travelling anteriorly to form the next loop.
The tracheal looping is similarly ridiculous in the other manucodes, but there aren’t as many spirals. However, the trachea extends even further posteroventrally in the manucodes than it does in the Trumpet bird: Clench (1978) figures one specimen [shown here] where the trachea loops upwards near the cloaca, and then extends on the side of the animal’s right thigh before looping back to the underside!
Why have super-long trachea?
So – – why do these birds have these super-long, looping or coiled tracheae? While it has been suggested that elongate tracheae might have a role in respiration or physiology (see Fitch 1999 for review), it has generally been thought that elongate, coiled tracheae allow their owners to make especially loud, resonant noises, and it’s notable that the birds we’ve just been looking at are the ones that create some of the loudest, most powerful, furthest-carrying noises within Aves. Trumpeter swans, Whooping cranes and Trumpet birds are the noisiest members of their respective groups, and exhibit the most complicated and elongate tracheae of their respective groups. I think we can safely infer that extinct birds with long, looping tracheae – like those moa – made loud, striking calls too.
Fitch (1999) argued that the calls made by birds with elongate tracheae are disproportionately loud compared to the size of the bird: therefore, elongate tracheae do not just permit particularly deep or loud calls, but also lower the formant frequencies and hence function in acoustic size exaggeration. This is a neat trick, as larger body size (or, the impression of large body size) is advantageous for sexual competition and advertisement, and may also play a role in intimidating predators. Also notable is that birds with long, looped trachea are all relatively large, and at least some of them nest in areas with dense vegetation where visibility is restricted (Fitch 1999).
Like a trombone – or like a violin?
Do the tracheal loops act in the same manner as the tubes of a trombone? This is what most authors have assumed, but Gaunt et al. (1987) argued that – in cranes – tracheal coiling made no difference to call volume. Instead, they proposed the novel idea that the integration of the tracheal loops with the sternum allows the entire apparatus to function like a stringed instrument. Perhaps, they proposed, the entire surface of the large sternum resonates during a call. Its vibrations are then, they proposed, amplified by the internal air sac system. I have no idea whether this hypothesis is reasonable or not and am unaware of any follow-up work. The idea that big birds might use their enormous bony sterna, super-long tracheae and extensive air sac systems to make particularly loud noises is pretty exciting, and must be one of the most incredible cases of exaptation among Tetrapoda… if correct. Gaunt et al. (1987) noted that, if they are correct, then birds like cranes function more like violins than trombones [Eurasian crane shown here, from wikipedia].
Incidentally, there is a possible cost to having large, complex tracheae: they might be especially prone to fungal infections (Souza & Degernes 2005).
While writing this piece I worked hard to resist all the diversions that came along: the avian larynx and syrinx, the syringeal bullae present in some waterfowl, the expanded (but not long) tracheae of umbrellabirds and other cotingas, the dead space problem and how it might link to pneumaticity, and so much more. In part I succeeded, but in part I didn’t. This is a very neat area and it’s really got me inspired, and I hope you enjoyed it too.
Refs – –
Banko, W. E. 1960. The Trumpeter Swan. North American Fauna Series No. 63. US Fish and Wildlife Service, Washington, D.C.
Chapin, J. P. 1932. The birds of the Belgian Congo. Volume 1. Bulletin of the American Museum of Natural History 65, 1-756.
Clench, M. H. 1978. Tracheal elongation in birds-of-paradise. The Condor 80, 423-430.
Fitch, W. T. 1999. Acoustic exaggeration of size in birds via tracheal elongation: comparative and theoretical analyses. Journal of Zoology 248, 31-48.
Frith, C. B. 1994. Adaptive significance of tracheal elongation in manucodes (Paradisaeidae). The Condor 96, 552-555.
Gaunt, A. S., Gaunt, S. L. L., Prange, H. D. & Wasser, J. S. 1987. The effects of tracheal coiling on the vocalizations of cranes (Aves; Gruidae). Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology 161, 43-58.
Hinds, D. S. & Calder, W. A. 1971. Tracheal dead space in the respiration of birds. Evolution 25, 429-440.
Homberger, D. G. 1986. The lingual apparatus of the African grey parrot Psittacus erithacusLinné (Aves: Psittacidae): description and theoretical mechanical analysis. Ornithological Monographs 39, 1-233.
– . 1999. The avian tongue and larynx: multiple functions in nutrition and vocalization. In Adams, N. & Slotow, R. (eds) Proceedings of the 22nd International Ornithological Congress,University of Natal, Durban, South Africa. BirdLife (Johannesburg), pp. 94-113.
Johnsgard, P. A. 1983. Cranes of the World. Croom Helm, London.
Lieser, M., Berthold, P. & Manley, G. A. 2005. Infrasound in the capercaillie Tetrao urogallus. Journal of Ornithology 146, 395-398.
Moss, R. & Lockie, I. 1978. Infrasonic components in the song of the capercaillie Tetrao urogallus. Ibis 121, 95-97.
Souza, M. J & Degernes, L. A. 2005. Mortality due to aspergillosis in wild swans in northwest Washington State, 2000-02. Journal of Avian Medicine and Surgery 19, 98-106.
Worthy, T. H., Holdaway, R. N. 2002. The Lost World of the Moa. Indiana University Press, Bloomington, Indiana.