Skip to content

Aepyornithidae: the largest maniraptors of Madagascar

November 29, 2011

Aepyornis by Ville Sinkkonen. This picture is the one that best illustrates the absurdly large (by paleognath standards) skull of these ratites.

Elephant birds, or, as they are known in malagasy, Vorompatra (“marsh bird”) are iconic giant ratites from the island subcontinent of Madagascar. While they are very commonly present in books illustrating extinct animals, very little is actually presented of these birds; their eggs, the largest among birds and in fact the largest dinosaur eggs ever found until the discovery of even larger eggs in China, attributed to sauropods (giant oviraptorid eggs have also been found, but I’m not sure if they were truly larger or just slightly longer, given their shape), are usually the focus when they are mentioned, and if you’re lucky they might also refference how they were heavier than their taller New Zealand relatives, the moas.

In truth, while admitely hardly charismatic even to some ornithologists, are rather interesting because they heavily hint at how Madagascar’s fauna evolved during the Cenozoic, and further illustrate the way paleognath diversity was established.


Aepyornis skull. Much like in cassowaries and moas, aepyornithids fused the nares, anteorbital frenestrae and eye sockets in a single opening, although in a more extreme way, almost resembling azhdarchoids in a way. This was likely to facilitate air passage, allowing the animal to cool the head in the tropical climate of Madasgacar.

All aepyornithids are known from the Pleistocene/Holocene, a given due to the nigh-absence of earlier Cenozoic malagasy fossils (the exception being sea cows). They are represented by two genera: the famous Aepyornis and the extremely obscure Mullerornis. The number of species is not certain; four species of Aepyornis and three of Mullerornis are recognised, but it has been suggested that many so called “species” might represent animals at different growth stages. For the sake of convenience, I will focus more on the genera as a whole and not on the supposed individual species.

As to be expected from flightless birds, aepyornithids suffered several anatomical modifications. The notarium was lost, as was most of the forelimbs with the exception of the minuscule humerus, which was probably located internally, rendering the bird completly wingless. As to be expected from a ratite, the sternum was small and flat, no longer exhibiting a recognisable keel. The skeleton in general is very robust; although many bones remained pneumatic as means to control body temperature as in other dinosaurs, being free from the pressures of powered flight meant that marrow production could become more extensive, and as such the hindlimbs and pelvis were nothing short of massive. In life, the largest Aepyornis weighted over 400 kg, and were considered the biggest neornithes of all time (the largest moas and terror birds were taller, but much less heavy), until later research gave that title to australian dromornithids, which managed to be both heavier and taller.

The skulls of Aepyornis were unusually large for ratite standards; while not proportionally the largest skulls among birds, in sheer size they were among the biggest, appearently only surpassed by the terror bird Kelenken. Paleognaths, due to their lack of an ossified palate, are generally noted for having small skulls in comparation to other birds; even long billed forms like kiwis have very shallow beaks. This seems to be refuted by the very deep rostrums of aepyornithids, which superficially vaguely resembled those of azhdarchoid pterosaurs, particularly chaoyangopterids. The size of the skull was presumably an adaptation for the hot climate of Madagascar; the nares, anteorbital frenestrae and the eye sockets were fused into one large cavity, allowing the passage of air in other to keep the head cool.

Despiste the large beaks, the animals probably had a weak bite force, due to the lack of a bony palate, although, given the strong gastric acids of most birds, this probably did not impair the animal much. Aepyornithids are thought to have been high browsers, occupying on Madagascar the same ecological niche that giraffids occupy on mainland Africa. Many malagasy rainforest tree species have fruits with thick, highy sculptured endocarps, often dark blue or purple in colour, much like those consumed by modern cassowaries, so aepyornithids might have played important roles as frugivores, dispersing the seeds of the species they fed on. In the presence of the also browsing giant lemurs, aepyornithids might have had a more flexible diet, feeding on small animals in addition to leaves like modern cassowaries and rheas, but this is not certain. While they were almost certainly present in rainforests, most aepyornithid remains were found in Madagascar’s wetlands; it’s not certain if this was a preffered environment or if was just where the fossils were best preserved.

Not much is known about aepyornithid social behaviour; probably it was fairly loose, likely tending towards being solitary but without the agressive territory creation that cassowaries and kiwis engage on. The male was almost certainly  the one responsible for incubating the eggs and raising the young like in most modern ratites; it’s not clear if they formed harems like most modern ratites or if they only raised one or two eggs laid by a single female. The chicks were certainly precocial; the presence of more predators perhaps indicates that were guarded by their parents like in most modern ratites, unlike modern kiwis and maybe the extinct moas, which are essencially superprecocial. Regardless, the period of parental protection was likely short, as aepyornithid growth rates are slow like those of moas and kiwis, with the birds reaching sexual maturity after several years.

Gradma, what big eggs and bones you have!

Aepyornis eggs. The biggest among neornithes, their size probably is an evolutionary response to predation.

Aepyornithids are generally dismissed as the product of an environment without macropredators, much like pratically every flightless bird in existence. However, this view is erroneous; while many birds lost flight in the absence of predators, we now know that flight is more expendable than that, and in fact many flightless avifaunas evolved in environments ruled by many predators, such as the Cretaceous hesperornithes and patagopterygids. Even ostriches and rheas likely evolved in the company of predators; while their ancestry goes back to a time where predatory mammals were rare, sebecian and pristichampsid crocodyllians (and possibly gastornithids, if they aren’t herbivores instead) ruled the ancient forests of the Eocene, and forms like Palaeotis and Diogenornis were already cursorial.

Did aepyornithids evolve in a world without predators? A look at malagasy Holocene fauna indicates the presence of three macropredators; the giant osteolamine crocodile Voay robustus and the giant fossa Cryptoprocta spelea likely posed a threat to the juveniles, probably even the adults of Mullerornis in Voay‘s case. However, all aepyornithids probably feared the remaining malagasy macropredator, Stephanoaetus mahery, the Malagasy Crowned Eagle, also known as Mahery. The similar Harpagornis moorei from New Zealand was a moa specialist, capable of dealing with even the larger species, so it is likely that the Mahery, with the ability to fly and thus strike at the vulnerable neck and head, was a threat to even the largest Aepyornis.

For obvious reasons, its not clear if aepyornithids have been under the influence of similar predators across the Cenozoic, and if they evolved flightlessness in a time when Madagascar had no predators. However, it seems clear that modern aepyornithids do have some adaptations that can be interpreted as defenses against the contemporary predators; the sheer robust built of these birds in comparation to other ratites, specially to the also insular moas, seems to suggest that these creatures became so massive in order to defend themselves. The only other birds that are equally as robust are gastornithids and dromornithids, which also evolved in environments with terrestrial predators. In fact, these three groups evolved in the company of large crocodyllians, with pristichampsids co-existing with gastornithids, mekosuchines with dromornithids and Voay with the aepyornithids, so this might suggest that the robust built of these birds might be related to fending off crocodyllians specialised in killing terrestrial prey.

While Voay might not had been the predator that influenced aepyornithid evolution, similar crocodiles might had been present earlier, when the birds evolved.

Similarly, the eggs of aepyornithids are unusually large. While ratites in general lay very large eggs, the only ones with such proportionally large eggs are kiwis, which are superprecocial. It is possible that aepyornithid eggs became that large in other to decrease predation on the young; no known malagasy carnivore can break the shell, and the young would have been born big enough to be only threatened by the larger carnivores. This too seems indicative that aepyornithids evolved to cope with the native predators.

Like moas, aepyornithids are clearly not specialised runners. However, the metatarsals (like in all neornithes, they are fused) appear to be proportionally slightly longer than in moas, and the hallux is absent. This seems to suggest that aepyornithids were more efficient runners than moas, presumably relying on speed to escape, although fully grown Aepyornis probably didn’t not had to run.

The origins of the Vorompatra

Ratite dispersal diagram by Daniel Foidl. Aepyornithids might be most closely related to the ratites of Oceania, rather than to ostriches. If so, one of the most impressive feats of oceanic travel among terrestrial birds might had occured.

Historically, the closest relatives of the elephant birds have been considered to be the ostriches, a rather logical conclusion given that ostriches live in Africa and aepyornithids live in Madagascar; since both landmasses were part of the supercontinent Gondwanna, both bird clades would had evolved from an ancestor that became isolated in Africa and Madagascar.

In reality, things are not as simple. Ratite diversity took place in the Cenozoic or late Cretaceous, long after Gondwanna was largely split apart; Africa was already an island continent in the early Cretaceous, long before neornithes evolved. Indeed, no african paleognaths have been recognised so far with the possible exception of Eremopezus, and ostriches evolved in Asia, not in Africa. It is possibe that ostriches and aepyornithids evolved from a common ancestor when India and Madagascar were a single landmass, with both groups evolving independently once India broke apart and moved northwards towards Asia.

However, genetic studies revealed that aepyornithids were not a sister clade with ostriches, but were surprisingly more closely related to emus, cassowaries and kiwis, the ratites of Oceania. If so, their ancestors engaged in a rather impressive feat of oceanic travel; it is unlikely they crossed the Indian Ocean from Australia, likely having colonised Madagascar from Antarctica instead, but even then the distance by water was still great, only slightly less than today. The ancestors of aepyornithids were most likely very competent flyers; Paleocene/Eocene volant paleognaths, known as lithornithids (which might be not be a natural clade, as they likely are just a paraphyletic group of species representing basal ratites), are thought to have been quite aerial birds, possibly similar to modern Charadriiformes like lapwings.

However, since the closest relatives of the Aepyornithidae+Casuariidae+Apterygidae clade are the barely volant tinamous, it is also very likely that the ancestor of aepyornithids was a poorly volant fowl-like species. If so, the oceanic travel to Madagascar must have been a very lucky incident, likely via floating rafts of natural vegetation.

The process of becoming flightless would have occured more quickly if the ancestor species was tinamou like rather than “lithornithid” like, followed quickly by an increase in size. Unlike moas, aepyornithids still had remaints of the wings of their volant ancestors, but since they were basically just humeri this probably means little in terms of determining when they became flightless. It seems very likely that, by the Miocene, aepyornithids as the ones known in the Holocene were already present. If aepyornithid diversity was higher in the earlier Cenozoic, and if they co-existed with other paleognaths, is currently unknown.

Eggs atributed to aepyornithids are known from the Canary Islands, precisely on the opposite side of Africa. Needless to say, in the absence of known aepyornithid fossils from mainland Africa, they are thought to belong to other birds instead, such as the albatross like pelagornithids.


Mullerornis by Daniel Foidl. More gracile than Aepyornis, and likely a better runner, it was still a rather robust bird. The last individuals lived as recently as 1260 BP, well after humans reached Madagascar.

Aepyornithids, like the rest of the Holocene megafauna, were brought to extinction due to anthropogenic influence. The humans that colonised Madagascar likely hunted aepyornithids, but appearently not to the extent that they hunted moas; for some reason, vorompatras were appearently considered sacred, and accordingly there’s relatively few specimens that indicate death by humans; notably, nearly all of these specimens belonged to Mullerornis, and not to the larger Aepyornis. The eggs, however, were not regarded to the same standard, and they are much more common in archeological sites.

Other possible causes of extinction were the introduction of diseases transmitted by chickens brought by the human colonisers. Being of indonesian ancestry, the early colonisers of Madagascar brought with them junglefowl, that likely carried pathogens that affected the native birds of Madagascar. For obvious reasons, this hypothesis has not been tested yet.

In spite of these factors (in part thanks to their sacred status), aepyornithids might have outlived the other members of the malagasy megafauna by several hundreds of years. Mullerornis is known to have lived until at least the end of the first millenium, and Aepyornis might have lived to see europeans arrive to the island. Indeed, the Androy region of Madagascar might have been one of the last places where these birds still dealt; Étienne de Flacourt described the giant birds of the Ampatres. The name “elephant bird” was atributed due to the legend of the Roc, a bird of arabian lore capable of carrying elephants, and due to the neotenous appearence of ratites, it would had been easy to see the vorompatras as chicks of a gigantic bird. It is possible that folk memories of the Mahery helped as well.

EDIT: The article was slightly changed in order to better credit the artist behind the last two pictures, Daniel Foidl. Under his suggestion, I uploaded an up-dated version of his Mullerornis drawing. I don’t know how to thank him properly.

5 Comments leave one →
  1. November 30, 2011 5:53 am

    Fascinating! Almost Tet Zoo worthy. Any references?

    • November 30, 2011 12:39 pm

      Wikipedia, David Attenborough’s doc about these birds, conversations with Tim Morris, the recent studies on paleognath genetics and comparations with other birds.

  2. gray Stanback permalink
    March 22, 2012 2:30 pm

    it is likely that the Mahery, with the ability to fly and thus strike at the vulnerable neck and head, was a threat to even the largest Aepyornis.

    That’s possible, but bear in mind that its closest living relative–the African Crowned Eagle– is a specialized hunter of monkeys and other forest-dwelling mammals. Its relative on Madagascar was probably little different, preying on the giant lemurs that inhabited the island at the time. Even today, lemurs have an irrational fear of flying predators, even though there are none that can kill them.

    • March 22, 2012 4:53 pm

      True. And there was also an Aquila eagle on Madagascar as well, so maybe both species had a distinct niche partitioning in prey selection

  3. gray Stanback permalink
    March 22, 2012 10:39 pm

    If anything, the Aquila (if it was as big as I’m imagining it) seems more likely to have been an elephant bird killer. After all, the Haast’s Eagle was also a member of the Aquila+”Heiraeetus” complex.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: