Giant Prehistoric Bird Crushed Seeds, Not Little Horses

If the huge bird were still alive, Gastornis would be an ornithophobe’s nightmare. Equipped with an extraordinarily deep beak, this six-foot-tall bird was among the largest creatures to roam the forests of prehistoric Europe and North America. And it’s exactly that imposing avian aspect which spawned a long-running fossil meme about the ancient bird.

While not so huge as the largest non-avian dinosaurs, Gastornis was nevertheless a giant in its Paleocene and Eocene heyday between 55 and 40 million years ago. In Europe the bird towered over the mammals who inhabited the same forests – the largest herbivores and carnivores of the day were about the size of a German shepherd, with many being considerably smaller. (In North America, where Gastornis fossils were previously labeled “Diatryma, some of the contemporary herbivorous mammals grew to bigger sizes, but there were still many smaller beasts running about.) So it seemed only natural that the monstrous bird would have preyed on the scurrying mammals, pouncing on “dawn horses” and cleaving lemur-like primates in two with it’s powerful beak. In museums and documentaries, Gastornis marked the last gasp of dinosaur dominance before mammals took over the world.

[In this clip from Walking With Beasts, Gastornis reprises the classic apex predator role.]

But recent research has found that Gastornis wasn’t so terrifying, after all. While a 1991 paper concluded that the bird’s beak could have made short work of many small mammals, other publications pointed out that such a beak would have been just as well-suited to cracking seeds and crunching tough fruit. More recently, tracks of Gastornis – née “Diatryma” – found in Washington show that the bird had blunted toes rather than vicious talons, and a preliminary study of dietary clues preserved in the bones of a German specimen of the bird suggested a menu of plants rather than flesh. And now paleontologist Delphine Angst and colleagues have added another line of evidence that Gastornis probably wasn’t a rapacious mammal-muncher.

Chemical signatures in the extinct bird’s bones are at the center of the Naturwissenschaften study. Angst and coauthors studied carbon isotope (δ13C) traces in the bones of Gastornis, small herbivorous mammals the bird lived alongside, and modern hawks and ostriches. This isotope acts as a proxy for diet. Generated inside plants, the carbon isotope becomes preserved in the tissues of herbivores that eat those greens and, further down the line, in the tissues of the carnivores that consume those herbivores. Locked in bones and teeth, this carbon isotope allows paleontologists to outline what individual animals were consuming and how they may have split up resources in the same habitat.

By themselves, these chemical traces don’t automatically label an individual as an herbivore, carnivore, or omnivore. What the carbon isotope values mean rely on comparison and analysis. In the case of Gastornis, Angst and colleagues used the data to investigate the opposing views of the bird as carnivore or herbivore.

Part of what makes carbon isotopes useful in paleontology is that they can be tied back to different types of plants that photosynthesize in different ways. This detail is  what led  Angst and coauthors to throw out the idea that Gastornis sliced small mammals. If the big bird was a carnivore, the researchers found, then the carbon isotope signatures inside it’s bones would indicate that it ripped open prey that, in turn, relied upon C4 plants – grasses and other plants that rely on a distinct form of carbon fixation. The snag is that plants didn’t evolve that C4 method of photosynthesis until about 14 million years after Gastornis lived. The chemical trace didn’t match up with the ecology of the time.

The skeleton of Gastornis on display at the National Museum of Natural History. Photo by Brian Switek.
The skeleton of Gastornis on display at the National Museum of Natural History. Photo by Brian Switek.

When Angst and coauthors looked at the carbon isotope through the lens of Gastornis being an herbivore, however, the signature was a better match and was comparable to those of herbivorous mammals living at the same time. The bird’s carbon isotope profile was that of an avian that crushed seeds and crunched thick-skinned fruit.

And the researchers went a step further. Through dissections of modern birds ranging from Darwin’s finches to Eurasian sparrowhawks, Angst and colleagues studied the anatomy and connection points of the external adductor muscle in modern herbivorous and carnivorous birds. This is a major muscle that powers bird bites, and the herbivorous, seed-cracking birds typically had wider muscles with increased space for attachment on the lower jaw. That fits they way they feed. Much more power is needed to bust open hard fruits than to tear soft flesh.

The actual muscles of Gastornis rotted away over 40 million years ago, but the bird’s lower jaw shows a wide space for the external adductor muscle to attach. Taken together, the bird’s beak, feet, reconstructed musculature, and chemical signature best fit a large herbivore that snacked on plants rather than the mammals that lived underfoot.

Does this mean that Gastornis never took a swipe at little Eohippus? Of course not. Animals are not so strict about a herbivorous or carnivorous diet as our species can be. Gastornis may very well have snagged the occasional unwary mammal or tried carrion. But the bird’s anatomy did not evolve for a life of predation like that of the famous “terror birds”, and the traces in bone suggest a life primarily fueled by vegetation.

Visions of Gastornis as an herbivore towering over the heads of Paleogene mammals is stronger than ever before. But there is still one line of evidence that could finally cement this shift in paleo imagery. If you want to know about a prehistoric animal’s diet, there’s no better source of information than the food that actually passed through their mouth – gut contents, mashed-up food preserved in the intestines (cololites), and fossil feces (coprolite). As far as I’m aware, no one has yet described such trace fossils with Gastornis, but such finds are not outside the realm of possibility. With stomach scraps or a prehistoric pellet in hand, paleontologists could finally outline Gastornis grub.


Angst, D., Lécuyer, C., Amiot, R., Buffetaut, E., Fourel, F., Martineau, F., Legendre, S., Abourachid, A., Herrel, A., 2014. Isotopic and anatomical evidence of an herbivorous diet in the Early Tertiary giant bird Gastornis: Implications for the structure of Paleocene terrestrial ecosystems. Naturwissenschaften. doi: 10.1007/s00114-014-1158-2

10 thoughts on “Giant Prehistoric Bird Crushed Seeds, Not Little Horses

  1. I’ve seen several other examples of surprise vegetarians, or vegetarians with traits usually associated with carnivores, such as Tiarajudens eccentricus, which had canines like those of a saber-toothed cat, but its other teeth are clearly those of a herbivore. There are some monitor lizards which eat fruit, and also an extinct crocodilian or similar reptile, and some living alligators were found to include fruit in their diet. The Pacu are large, frugivorous cousin of the piranha. A unique case is that of a wounded nurse shark that was taken into captivity and found to prefer eating the vegetarian feed intended for other denizens of its aquarium.
    It makes me wonder, are there cases of surprise meat-eating in herbivores? We once viewed chimpanzees as peaceful herbivores, but now know they can be violent and they do hunt animals for food. I’ve heard of a few isolated cases of farm animals exhibiting fluke carnivory, the one-time observation of a horse eating a chicken, for example. Not much to go on. I’d be interested in other examples of surprises both ways.

  2. I’ve worked with horses most of my life and knew a pony when I was younger that would willingly take and eat hot dogs, hamburgers and deli meats. As I’ve gotten older and mentioned this to other horse people I’ve had others confirm they knew horses and ponies that also did this and heard rumors of horses that would take bites of carrion. Even more rare but nonetheless passed about are the rumors of horses that were witnessed stalking and killing small mammals. I know some of this may be dismissed as hyperbole or attention seeking type behavior, but I find it hard to believe that every story I’ve heard of this is false. It stands to reason that more animals are opportunistic omnivores than we know, it surely can’t hurt in a survival type situation.

  3. I think it might be useful to point out that phorusrhacids were different birds from gastornithids and were actually carnivorous. I could easily imagine people confusing the two and then proclaiming phorusrhacids were herbivores too.

  4. On that topic, David: Captive parrots certainly seem to love eating chicken flesh, so I’d be curious to know whether they’ve ever been observed eating meat in the wild. (Other than the Kea, of course, which is a known omnivore.)

  5. Very interesting article, Brian, and I’m glad that you didn’t completely discard the idea that Gastornis occasionally ate small mammals. The evidence from the anatomy and the carbon-13 isotopes suggests this was the avian equivalent of a grizzly bear – a generalized omnivore that took in meat to supplement a heavily vegetarian diet. Which, to be honest, is not unlike some ratites (especially cassowaries and emus).

  6. Meat-eating herbivores? Sheep have been known to eat ground-nesting seabirds. White-tailed Deer have taken small birds out of mist nets, much to the distress of the ornithologists intending to band the birds. I think most birds and mammals aren’t as constrained into simple dietary categories as we assume.

  7. This seems slightly flawed. They don’t seem to have compared any fossil carnivores from the same deposits, just herbivores and the bird. Given that diagenesis can radically alter isotope composition in bone, it is pretty critical to compare an unknown (Gastornis) to modern herbivores, modern carnivores, fossil herbivores AND fossil carnivores. Dropping that fourth test makes this study unsatisfying and inconclusive to my mind.

  8. I must agree with you that this is somewhat “disappointing” to childhood concepts, but science is science whether we like it or not.

    What I’d like to see is some comparative anatomy. A grosbeak has a large, seed-crunching beak, but he’s small and flies. A large flightless bird like an ostrich or emu has a tiny skull. What can we use that is alive today to help our study?

    So here’s a question. Is this merely applicable to Gastornis, or all Phorusrhacids? And what about the mihirungs?

  9. It’s interesting that this paper doesn’t mention a 2012 journal article where North American researchers described Gastornithids as gentle herbivores, based on anatomical evidence. The reference: Mustoe, G., Tucker, D., Kemplin, K. 2012. Giant Eocene bird footprints from northwest Washington, USA. Palaeontology 55, 6: 1293-1305. BBC Nature presented a nice online summary, which can be accessed at

Leave a Reply

Your email address will not be published. Required fields are marked *