Clubs, spurs, spikes and claws on the hands of birds (part I)

All too few people seem to realise that birds have hands*; it's just that these parts of the body are - normally - mostly obscured from view by the feathers. While the main role of the bird hand is to support remiges (the big wing feathers), less well known is that many birds possess claws, spurs, spikes and knobs on their hands and wrists that they use in offence or defence [avian hand skeletons below will be identified and discussed in part II].


* Two groups of birds atrophied and eventually lost their hands during evolution. Hesperornithines - a group of toothed, foot-propelled diving birds from the Cretaceous - reduced their wings over time and some members of the group ended up with just a rod-shaped humerus (all the bones distal to it having been lost). Moa lost their hands as well as their entire wings, and in fact even went as far as losing the socket for the humerus on the shoulder girdle.


Before we start looking at those claws, spurs, spikes and knobs, a very brief tutorial on avian hand anatomy [adjacent skeleton photo from here. Do you have any idea how hard it is to find good figures of the avian wing skeleton online?].

While birds do still possess the bones of three individual fingers, the metacarpal and distal carpal bones have united to form a single structure called the carpometacarpus. The proximal end of the carpometacarpus is a large, trochleated, convex lump that is very obviously homologous with the semi-lunate carpal bone of non-avian maniraptoran theropods. Properly called the carpal trochlea (or trochlea carpalis, if you prefer), this structure allows the hand to fold up tightly close to the ulna. Of the three digits, the one on the leading or anterior surface of the hand (the one that looks like it should be a thumb, or digit I) is often called the alular digit. The adjacent, 'middle' digit (the one that looks like it should be digit II) is often called the major digit, while the third digit is typically termed the minor digit.


A projecting structure termed the extensor process (or extensor attachment) is present between the carpal trochlea and the alular digit; it belongs to metacarpal I (also called the alular metacarpal), but this is hard to appreciate in modern birds because metacarpal I is fused imperceptibly into the carpometacarpus as mentioned above [in the adjacent diagram, from Campbell (2008), the extensor process is the site of origin of the collateral ligament marked K]. Various tendons attach to this process and it plays a key role in the movement of the hand.

If you know anything about bird hands you'll know that there's long been a debate on digit homology in the avian wing: palaeontological evidence indicates that the three digits of the bird hand are digits I-III whereas embryological evidence is supposed to show that digit I forms in the embryo but is later resorbed, and that the three fingers of the adult are actually digits II-IV. Some workers have argued that both views can be correct, and that an embryological event termed a frame shift occurred during theropod evolution... and this in turn has led some people to reidentify the manual digits of fossil theropods. I covered some of this stuff in the Limusaurus article from 2009. I think that the embryological evidence has been misinterpreted and that the fingers of bird hands really are I-III (Vargas & Fallon 2005). It's a complicated subject that I'd rather avoid for now.

Hand claws are normal for birds

Given that birds are amniotes, it follows that they have hand claws within their ancestry. While it's well known that the fossil birds of the Mesozoic often had clawed hands, it doesn't seem to be widely known that hand claws are widespread and in fact wholly normal in extant birds. They're not an anachronism unique to the Hoatzin Opisthocomus hoazin [embryo (above) and adult Hoatzin hands shown below].


In fact, digit I claws are generally present in ratites, gamebirds, waterfowl, divers, storks and kin, finfoots, owls, New World vutures, the Secretary bird Sagittarius serpentarius, waders and many others (Jefferies 1881, Fisher 1940) - a distribution which strongly suggests that they've been retained throughout neornithine history but lost selectively here and there, especially among so-called 'higher landbirds' [ostrich hand shown below; note obvious claws].


Digit II claws are rarer but are present in some neornithines (like some ratites, juvenile waterfowl and flamingos), so might also be primitive for the clade. It's sometimes said that cassowaries have a particularly large claw on digit II that they use as a weapon (Stettenheim 2000)*. Indeed, like other ratites, they do have such a claw, but I've never otherwise heard of them using the claw as a weapon and would be interested to know if this is true. Cassowary wings also possess stiff, barbless quills that can be as much as 20 cm long, but I don't think these are used in combat either.

* I realised after writing this that it's a confused reference to digit II in the foot. The especially large claw on this digit is used as a weapon. More on cassowary combat here.


So, retaining claws on the hands isn't much of a big deal - if anything's a big deal, it's that so few people know this stuff (I'm always confounded by the fact that people don't even seem to look at the dead animals they consume on a regular basis - chickens have obvious hand claws, yet the average person is confused by the mere mention of hands in birds) [adjacent image, from Fisher (1940), shows assorted avian hand claws: A. Pagophila. B. Cygnus. C. Rallus. D. Lophortyx. E. Cathartes. F. Archaeopteryx].

Spurs, spikes and clubs in waterfowl (again, sorry)

Rather more remarkable is the fact that birds belonging to several distinct lineages have evolved entirely new structures - namely, bony spurs, knobs and clubs - on their wrists, arm bones or carpometacarpi. Technically, these aren't really 'new' structures: instead they're all modifications of existing bits and pieces, especially the extensor process. An excellent review of these structures was provided by Rand (1954), and I've relied heavily on this paper when preparing this article (note that Sclater (1886) also looked briefly at avian carpal spurs as well as claws).

We'll begin with waterfowl (by which I mean: the whole of Anseriformes). Screamers (Anhimidae) are, in terms of hand anatomy, among the most incredible of birds because they possess two large spikes on the carpometacarpus. The larger one emerges from the extensor process, and a more distally located one grows from near the end of the major metacarpal. In the Black-necked screamer Chauna chavaria, the spikes are oval in cross-section, but in the Southern screamer C. torquata there's a sharp-edged keel along the spike's proximal edge, and in the peculiar Horned screamer Anhima cornuta the spikes are triangular in cross-section and thus equipped with three sharp keels. I've only ever seen figures of the Black-necked screamer's spikes, but it turns out that those of the Horned screamer are larger: in the former, the proximal spike can be 45-47 mm long, but in the latter it can reach 61 mm in males (Rand 1954). Screamer wing spikes were previously covered here.


Moving now to anatids (ducks, geese and swans), we find that several species possess spurs and other structures. In Plectropterus gambensis, the Spur-winged goose, the spur is an extension of the radiale: one of the two distal carpal bones [Plectropterus hand skeleton shown here; from Rand (1954)]. It therefore has a completely novel origin compared to the spurs and spikes seen in screamers and other waterfowl. It's a stout, conical structure that is typically oval in cross-section, but sharp keels are present on the spur in some individuals. The Spur-winged goose was looked at fairly recently on Tet Zoo: go here.


Carpal spurs are also present in the Torrent duck Merganetta armata [male shown here; from wikipedia]: these are completely different from the spurs of Plectropterus as they're borne on the extensor process. The Torrent duck spur has a very stout base and tapers rapidly to a point. The spurs are typically bigger and more sharply pointed in males than females (9-17 mm vs 6-13 mm). The species name for this South American duck, coined by Gould in 1842, means 'armed', so I assume that it was named specifically for its hand spurs.

Horny wrist knobs are present in several anatids, including steamer ducks (Tachyeres), the Comb duck Sarkidiornis melanotos and kelp geese (Chloephaga). These structures (which have hardly ever been illustrated or depicted in the literature) are, I believe, inflated extensor processes. They can be used in particularly brutal fashion: for an account of their role in steamer duck behaviour see Attack of the flying steamer ducks.

And, in the interests of giving up here because I want to publish something before I go away to the Royal Society pterosaur exhibition (many photos to follow), I'll deal with the remaining birds of interest in part II. Plovers, jacanas, sheathbills, giant pigeons and ibises to follow.

Bird hand anatomy has been mentioned or discussed a few times previously on Tet Zoo. See...

Refs - -

Campbell, K. E. 2008. The manus of archaeopterygians: implications for avian ancestry. Oryctos 7, 13-26.

Fisher, H. I. 1940. The occurence of vestigial claws on the wings of birds. American Midland Naturalist 23, 234-243.

Jefferies, J. A. 1881. On the claws and spurs on birds' wings. Proceedings of the Boston Society of Natural History 21, 301-306.

Rand, A. L. (1954). On the spurs on birds' wings. The Wilson Bulletin, 66, 127-134

Sclater, P. L. 1886. On the claws and spurs of birds. Ibis 28, 147-151.

Stettenheim, P. R. 2000. The integumentary morphology of modern birds - an overview. American Zoologist 40, 461-477.

Vargas, A. O. & Fallon, J. F. 2005. The digits of the wing of birds are 1, 2, and 3. A review. Journal of Experimental Zoology (Mol Dev Evol) 304B, 206-219.

More like this

Time to finish one of those long-running series of Tet Zoo articles: at last, the long-awaited, much anticipated third and final instalment in the series on the clubs, spurs, spikes and claws present on the hands of numerous neornithine bird species. If you haven't done so already, do check out…
A little while back we looked at the claws, bony knobs and other structures present on the hands of certain palaeognaths, waterfowl and other birds. Time to look at more of this sort of stuff - I kind of got distracted by lapwing taxonomy, so this is all going on for a bit longer than expected,…
Another waterfowl fact. This one is gonna be brief. In the previous article we looked at the wing spurs of Plectropterus. They're pretty cool, but they're far from unique, and even more incredible are the much larger, dagger-like spikes seen in screamers... Screamers (Anhimidae) are a small group…
Yesterday we looked briefly at goose digestion. Pretty incredible stuff, I'm sure you'll agree. Hey: wouldn't it be weird if some waterfowl were poisonous? Yeah, wouldn't it. Well... guess what? One of the most dangerous birds in the world - I'm not kidding here - is the African spur-winged goose…

It´s really strange that it is so generally unknown that many birds possess real tallons. Actually I have to admit that I discovered this also comparably late.
Matt Wedel has some excellent photos at his blog which show the wing-claws of emus and rheas during a dissection:
I have never seen really prominent claws at chicken wings, but I remember that I have seen them at the wings of a roasted duck. And the stiff quills of cassowaries look really strange, as well as so much other aspects of their anatomy:…

Do all these birds use their spurs and clubs for fighting? Do any use them for, say, climbing?

By Andreas Johansson (not verified) on 30 Jun 2010 #permalink

You think that that "the embryological evidence has been misinterpreted and that the fingers of bird hands really are I-III (Vargas & Fallon 2005)."
That sounds reasonable.
Concerning the phalanges would it be
1-2-1-x-x as mentioned by David MarjanoviÄ in another thread?

It's sometimes said that cassowaries have a particularly large claw on digit II that they use as a weapon (Stettenheim 2000).

You forgot the ref, but never mind -- that claw is on the 2nd toe, not the 2nd or any finger.

I'm always confounded by the fact that people don't even seem to look at the dead animals they consume on a regular basis - chickens have obvious hand claws

Indeed, and so do turkeys. Size, shape, and distribution (1st, 2nd, or both fingers) vary in chickens and turkeys.

By David MarjanoviÄ (not verified) on 30 Jun 2010 #permalink

Moa lost their hands as well as their entire wings, and in fact even went as far as losing the socket for the humerus on the shoulder girdle.

does that mean they aren't tetrapods anymore? ;p

As promised, now delivered...
You say birds have hands (exceptions noted), but calling these structures hands is a bit generous. When an anatomical feature is re-purposed that radically, I think it's fair to give it a different name. My earbones aren't jawbones, though they used to be.
The hoatzin is losing prestige all over the place! Nothing special in his claws, nested comfortably in Neoaves. Pathetic! Let's just start calling them "stinkbirds" and get it over with.
Still, I'd rather know the facts and have a more complete view of the big picture, even if that means losing out on a few romantic notions. And this is all pretty cool stuff.

By CS Shelton (not verified) on 30 Jun 2010 #permalink

Is that a complete list of waterfowl wing-weaponry? I remember being told (or maybe reading) as a child that swans (some swans, male swans, some male swans...) have knobs on their wings which makes them particularly dangerous when they attack.

By Allen Hazen (not verified) on 30 Jun 2010 #permalink

Sorry, just looked again and noted "B. Cygnus" in the assortment of wing claws.

By Allen Hazen (not verified) on 30 Jun 2010 #permalink

O, thanks, waiting for more!

Question 1: Why cassowaries have these five quills? Having seen them, they are useless for fighting. The only function I can imagine is that they can produce sound when cassowary shakes its vestigal wings.

Question 2: Can a swan, with or without wing knob, break man's leg? I once estimated that a blow of swan wing is about equal to dropping 10 to 15 kg weight narrow edge on your outstretched leg.

It is strange that this isn't more widely known, especially when certain examples grace the family dinner table on a regular basis. Maybe we should let our children play with their food a little more often...

A pair of Australian Shelduck lived uncaged at the Melbourne Zoo (Victoria, Australia). They were at the 'inciting' stage of pairing-up for spring. The female rushed up to two teenage boys as if about to peck them, then 'got a fright' and rushed back to her drake in apparent agitation. The drake then rushed at one boy, seized his jeans just below the knee, and inflicted really savage blows on his shin with the wing carpal joint. When shaken off and shooed away, the Shelducks repeated the whole performance. The boys showed great restraint in not kicking the birds -- they said 'That really hurts!' They were very interested in this behaviour, having decided for themselves that the duck wanted the drake to defend her, and so to convince her that he was a suitable mate.
I have never seen a Shelduck skeleton, but I wonder if the male's carpal knob is better developed than the female's.

@Anthea Fleming: Oh I know exactly what you're talking about. I've seen human females do this too.