Better late than never… what was the identity of that unusual string of vertebrae I featured here however-many-days-ago? Most of you realised – correctly – that it was the neck of a bird, and several of you guessed moa. This wasn’t a bad guess, but it wasn’t the right one. The correct answer was given three guesses in by Adam Yates of Dracovenator. Yes, it’s the neck of a kiwi (Apteryx). Well done Adam, your Tet Zoo-ing skills serve you well (I’ll not mention the fact that Adam’s skills also extend to fishes and molluscs, gack). Why post a picture of a kiwi neck? Because it’s so frikkin’ weird, that’s why. The neck vertebrae of kiwi look very broad and robust for a bird, and I was hoping that these characteristics might have fooled you into thinking that you were looking at a baby apatosaur neck or something…
Mivart (1877) provided a very good description of ratite vertebral anatomy and described how Apteryx ‘differs from all [other ratites] in the greater relative stoutness of the neck and production of its processes’ (p. 34). Like all ‘ordinary’ tetrapod vertebrae, the cervical vertebrae of Apteryx have prezygapophyses, postzygapophyses and neural spines on the neural arch, and parapophyses and diapophyses on the centrum [for help with the terminology see Tutorial 2 at SV-POW!]. Adjacent to the posterior part of the neural spine and sitting atop the postzygapophyses are additional structures, the hyperapophyses (singular: hyperapophysis. Not to be confused with hypapophysis [plural: hypapophyses]. Hypapophyses are keel-like structures that project from the ventral surface of the centrum: in the kiwi neck, they’re small on the atlas, absent on the axis, and then present from the 12th cervical to the third dorsal). I’m not entirely sure what hyperapophyses are (homology-wise): it’s tempting to think that they might be homologous with the epipophyses of non-avian theropods and basal birds, but epipophyses are typically placed more posteriorly on the postzygapophyses. I could say a lot more about kiwi cervical vertebrae, but I think I should move on.
Indeed, it’s worth briefly discussing the other weirdness present in the kiwi skeleton. I’ll try and keep comments on the skull as brief as possible as we all know how easy it is to find loads to say about skulls. In marked contrast to most other maniraptorans, the orbits of kiwi are small and, in contrast to other crown-group birds, the nostrils are located right down at the tip of the bill. Clustered around the tips of both the upper and lower jaws are numerous small pits that house specialised cells called mechanoreceptors (specifically, the sort of mechanoreceptors present in bird bills are called Herbst corpuscles and Grandry corpuscles). These provide kiwis with tactile jaw tips (Martin et al. 2007), so they don’t rely on their well-developed olfaction entirely as used to be thought. We previously looked at avian bill mechanoreceptors back when Mark Witton and I paid some attention to them in our work on azhdarchids. The kiwi bill is long, but what’s surprising is how variable its length is within the group, both between sexes and between species (Sales 2005). Having mentioned species, I should note that kiwi taxonomy has recently been revised (or, at least, is in the process of being revised): genetics indicates that there are several cryptic species within what used to be known as Apteryx australis [adjacent image, showing kiwi bill tips, from Martin et al. (2007). Image of kiwi skull below from Digimorph].
Moving to the thoracic skeleton, the neural spines on the dorsal vertebrae are tall and – in some kiwi taxa – so long anteroposteriorly that they butt up against one another to form a continuous neural lamina. They aren’t fused together, however, as they are in some other neornithines (when the neural spines fuse together, they create a laterally compressed structure termed a notarium). The ribs are also weird in being strikingly broad.
What the hell is going on with those weird uncinate processes?
The large, flattened uncinate processes are also broad: in most birds (and other maniraptorans) they’re slender and spur-like. Uncinate process have been shown to play a role in anchoring muscles that facilitate flaring of the ribcage when movement of the sternum is prevented during sitting (Codd et al. 2005). Kiwi sit down when, for example, brooding their gigantic eggs. At such times they are, like other birds, unable to use the sternum to ventilate the posterior thoracic and abdominal air sacs (which, like those of other ratites, are small and reduced compared to those of other neornithines). However, kiwi have a particularly short sternum [more on the sternum below] that does not extend further posteriorly than the fourth thoracic rib (in most other neornithines it extends at least as far posteriorly as the sixth rib).
Might it be, therefore, that the big uncinate processes are needed in order to compensate for the small sternum? This idea requires testing and I’m not aware of data that might have a bearing on it. However, it doesn’t look like it’s the case, because other maniraptorans with similarly small sterna (including oviraptorosaurs, dromaeosaurs, elephant birds and emus) don’t have the weird, broad uncinate processes so characteristic of kiwi. Could the uncinate processes help stabilise the rib cage somehow? Maybe, but I can’t think why kiwi would need special lateral stabilisation like this. For now, the evolution of those broad ribs and uncinate processes is mysterious.
Weird arms, hands, legs and feet
The kiwi scapulocoracoid is small with subparallel margins that make the whole structure look strap-like. The wing is strange: all the bones are tiny and gracile, the humerus is strongly curved and very slender, and the hand is monodactyl with a small claw (though this is not always present). The number of phalanges seems to vary between two and four. The kiwi taxa differ in how stiff their wing quills are.
The sternum is anteroposteriorly short and very broad with a convex ventral surface that usually lacks a keel (a vestigial keel is occasional present). It’s also thin at its lateral margins. The very short coracoidal sulci are widely separated on anterolateral processes by a large U-shaped border that lacks any kind of anterior process. Unlike that of other ratites, the kiwi sternum has two lateral trabeculae, both of which have incisions on their medial sides. Some sterna are perforated by one or two foramina. There are five prominent articular facets for sternal ribs: the ribs themselves increase in length posteriorly.
The kiwi pelvis is interesting because, if you know non-avian dinosaurs, it has a more familiar look to it and lacks some of the weirdness of the neornithine pelvis that often confuses people [kiwi pelves below from McGowan (1982)]. Ordinarily in neornithines, the dorsal surface of the elongate ischium is fused for much of its length to the ventral surface of the post-acetabular part of the ilium, leaving a large foramen (the ilioischiadic foramen) ventral to the anterior part of the post-acetabular part of the ilium. At the same time, the strongly reduced, rod-like pubis usually extends posteroventrally in close contact with the ischium. In contrast, the pubis and ischium of kiwi are clearly distinct and project posteroventrally beneath the ilium, being separate both from each other and from the ilium (thought the pubis and ischium may occasionally unite distally). The pelves of moa and elephant birds are similar in these respects. In case you haven’t noticed, the pelves of dromaeosaurs, basal birds, palaeognaths and neognaths form a neat and obvious ‘sequence’. But then, you could say the same about any other part of theropod anatomy when looked at within a phylogenetic context.
The pelvis in kiwi is remarkable for its breadth, which exceeds that of any other ratite. In dorsal view the flaring blades of the ilia produce a heart-like shape. The tail is long and a pygostyle is generally stated to be absent (and, in the kiwi material I’ve looked at, it doesn’t look like any caudal vertebrae are fused together to form a pygostyle). However, Mivart (1877) thought that a pygostyle is indeed present and that it consisted of ‘three or four successively smaller and smaller vertebrae ankylosed together’ (p. 39). Pygostyles seem to be generally absent in ratites, but ostriches have a short one (De Beer 1956). Mivart’s comments make me think that further work is needed in order to determine the distribution of pygostyles within the ratites.
Finally, there is also some interesting stuff in the hindlimb. The middle trochlea on the tarsometatarsus projects markedly distally and there are two short hypotarsal ridges on the posterior surface of the tarsometatarsus, as there are in moa. Cracraft (1974) used this character (among a few others) to support the monophyly of a kiwi + moa clade, but since then some studies have rejected the existence of such a clade. The most unusual thing about the kiwi hindlimb is that the hallux is not reversed as it is in other neornithines (that is, metatarsal I does not project posteriorly from the tarsometatarsus): instead, it projects anteromedially (forwards and inwards). In this respect it is similar in orientation to that of non-avian theropods (McGowan 1979*, Paul 2002). This is unlikely to be a retained primitive character, and is almost certainly a reversal.
* Note that some of the observations made in this paper were corrected by Vanden Berge (1982).
That all went on for much longer than I’d planned. It’s obvious that kiwi are highly specialised for a nocturnal lifestyle in which flight and good eyesight have been replaced with cursoriality and a reliance on olfaction and tactile abilities. These features, combined with the hair-like feathers, whiskers, and cryptic habits of kiwi have led many biologists to regard these bizarre birds as ‘honorary mammals’. What makes them doubly interesting is that, in several aspects of their morphology, they are reminiscent of Mesozoic maniraptorans and might have been similar to them in physiology and soft-tissue anatomy [adjacent image from wikipedia].
There’s tons more that could be said about kiwi. While doing research for some of the text above, I discovered some old material on the group that I’ll recycle here at some stage.
Refs – -
Codd, J. R., Boggs, D. F., Perry, S. F. & Carrier, D. R. 2005. Activity of three muscles associated with the uncinate processes of the giant Canada goose Branta canadensis maximus. The Journal of Experimental Biology 208, 849-857.
Cracraft, J. 1974. Phylogeny and evolution of the ratite birds. Ibis 116, 494-521
De Beer, G. 1956. The evolution of ratites. Bulletin of the British Museum of Natural History (Zoology) 4, 59-70.
Martin, G. R., Wilson, K.-J., Wild, J. M., Parsons, S., Kubke, M. F. & Corfield, J. 2007. Kiwi forego vision in the guidance of their nocturnal activities. PLoS
One 2 (2): e198. doi:10.1371/journal.pone.0000198
McGowan, C. 1979. The hind limb musculature of the brown kiwi Apteryx australis mantelli. Journal of Morphology 160, 33-74.
- . 1982. The wing musculature of the the brown kiwi Apteryx australis mantelli and its bearing on ratite affinities. Journal of Zoology 197, 173-219.
Mivart, S. G. 1877. On the axial skeleton of the Struthionidae. Transactions of the Zoological Society 10 (1), 1-52.
Paul, G. S. 2002. Dinosaurs of the Air: the Evolution and Loss of Flight in Dinosaurs and Birds. Johns Hopkins University Press (Baltimore).
Sales, J. 2005. The endagered kiwi: a review. Folia Zoologica 54, 1-20.
Vanden Berge, J. C. 1982. Notes on the myology of the pelvic limb in kiwi (Apteryx) and in other birds. The Auk 99, 309-315.