Saturday, February 21, 2015

Reports of gut contents in herbivorous dinosaurs

I generally avoid writing "nuts and bolts" posts about dinosaurs, because there's a lot of people already doing that who know what they're doing. There is, however, one topic on which I'm uniquely qualified to pontificate. In a previous lifetime I was a fresh-faced graduate student who wrote his thesis on possible gut contents in the "Leonardo" specimen of Brachylophosaurus. As such, I spent a fair amount of time considering the history of possible gut contents in herbivorous dinosaurs, which consists of an exclusive club populated by four hadrosaurs, one ankylosaur, and one sauropod. In retrospect, the Leonardo project was somewhat beyond my powers, but I battled on against overwhelming oddities and came out of it eventually. (For you folks playing along at home, by "herbivorous dinosaurs" I mean more or less classic dinosaurs that aren't theropods. I've got nothing against birds, which include many a herbivorous dinosaur, but they aren't why I got interested in these things, and I don't feel sentimental about them being dinosaurs, because "my" dinosaurs are almost all ornithischians and have all been dead for 66 million years.)

The heartbreak of taphonomy

One of the things I took with me from the Leonardo project is that possible gut contents of herbivorous dinosaurs are a lot of work, especially for the return on investment. Perhaps I am somewhat gut-shy now, but it's always helpful to be extra-cautious when dealing with a topic that invites speculation. Unless you have somehow found a specimen where integument impressions clearly cover the entire torso, with no possible avenue for the inside to get out or the outside environment to get in except those originally engineered for that function, you cannot and should not rule out the possibility of the original gut contents escaping. (I leave you to consult the literature on cadavers at your leisure.) You also probably don't want to cut into this hypothetical specimen for sampling, but that's another matter. For all other specimens, you can come up with all kinds of circumstantial evidence to support gut contents if you are being very thorough or just really, really want to see gut contents, but it hardly takes the dark skills of a taphonomy expert to poke holes. (Taphonomy serves two invaluable functions in paleontology: it describes what happened to a fossil after the original owner expired, and it's often an important check on sensational proposals, acting as a "dope slap" that ideally reminds us "no, we can't know that, and if we could, this would not be evidence of it.") The sad fact of the matter is that carcasses have an alarming tendency to open up. You want rapid burial for the best chance of preserving a carcass at least somewhat intact, but putting a hadrosaur-sized carcass in a setting where it can be rapidly buried, like, say, a large river moving a lot of sediment, is also a great way to expose it to currents that can empty it, and bring in plant crud from the environment. A feature the size of a large herbivorous dinosaur is going to be a place where things like masses of vegetation get caught.

Furthermore, if the gut contents were trapped inside the carcass, they are probably not going to be in good shape. The digestive tract is going to break down quickly and leave the gut contents in a chemical stew of various substances. If we interpret herbivorous dinosaurs as primarily reargut fermenters, you've also got a GI tract's worth of plant-digesting bacteria to contend with as well. As noted by Carpenter (1987), there's no guarantee that the stuff approximates a normal diet; because the animal obviously died shortly after eating, gut contents could represent material eaten in response to starvation or other physiological distress. Finally, even under the best of conditions you've got a grand total of one data point. The lure of gut contents is undeniable and obvious: actual physical proof of what that animal was eating. The reality is for fans of ambiguity, at least in terms of herbivores.

The "Trachodon mummies"

The "Trachodon mummies" are a pair of famous hadrosaur specimens collected by the Sternberg family shortly after the turn of the 20th century. One of them, the more famous in the English-speaking world and sometimes known as "the" mummy, is on display at the American Museum of Natural History in New York (AMNH 5060). The other is on display at the Naturmuseum Senckenberg in Frankfurt am Main, Germany (NMS R4036). Regarding AMNH 5060, Charles H. Sternberg (1909) reported that carbonized gut contents were present, but these have never been described and, I believe, were removed during preparation. Although the taphonomy was not documented or reported as thoroughly as we could do today, there is enough to paint a picture of a carcass that dried out during a drought, which protected it from consumption and decay, and later was caught in a flood and rapidly buried in a point bar (Carpenter 2007). A drought is one way to halt microbial action and thin out the crowd of potential scavengers, which can keep the gut contents around in the first place, but we come back again to the process of burial as potentially interfering.

AMNH 5060, from Osborn (1912).

The Senckenberg specimen included a brown concentration of plant fragments, mostly needles of the conifer Cunninghamites elegans, with branch fragments of conifers and angiosperms, and small seeds or fruits (Kräusel 1922). Kräusel (1922) was quite comfortable with the material being gut contents; the main point of interest at the time was the implication that the animal was feeding on land instead of sucking pond scum. This is the specimen whose guts are cited in various "Dinosaur Renaissance"-era works that brought hadrosaurs out of the swamps, going back, actually, to about a decade beforehand, in Ostrom (1964), one of the most important publications on duckbills (the source of about 90% to 95% of the modern consensus). However, the "in-place or washed-in" debate reared its head almost instantly in Abel (1922), just a few pages after Kräusel's note.
AMNH 5060 under glass: "Mother of Mercy, is this the end of Rico?"

Stokes' sauropod

William Stokes (1964) reported on a sauropod specimen from the Morrison Formation of Utah that he interpreted as including gut contents. The description is in a two-page Nature article, so don't expect huge amounts of detail, but the proverbial red flags and sirens go up when it turns out that the skeleton was weathering out of the ground and the bones scattered: there is no way to know about the integrity of the carcass. The fact that the material was found within the skeleton is not what I'd consider "strong" evidence: it is the minimum I'd want to see. The plant material is "vaguely stratified in one plane", which is indicative of having been at least stirred up by a current. The fossils are unlike the woody fragments seen throughout the Morrison, but then again crud being washed into the carcass of a sauropod is probably subject to a rather different depositional setting than plant matter in the rest of the Morrison. Finally, the detail that usually bothers other paleontologists is that among the woody fragments is a fair amount of bone and an Allosaurus tooth. Stokes interpreted this as showing that sauropods were mobile trash cans that included organic-rich muds and sediments in their diets. Whether or not he was right, the taphonomy of the specimen is inadequate to support this claim. To me, it reads more like a selection of environmental debris got caught in the carcass during burial

Minmi sp. [named Kunbarrasaurus ieversi 12-2015]

Like the "Trachodon mummies", the specimen in question here is much better known for its other qualities: it's one of the finest ankylosaur specimens known, and is essentially responsible for our ideas of what Minmi looked like, which makes it a bit funny that all these years later it's still just Minmi sp. It is in the collections of the Queensland Museum, and is designated QM F18101. Molnar and Clifford (2001) published a detailed description of a mass of plant material found in the abdomen. This is the one I like the best in terms of gut contents. It certainly has auspicious taphonomy: it's one of the small but noteworthy number of armored dinosaur specimens that found their way into marine sediments, where plant fragments are at a shortage. (Your typical classic dinosaur was not particularly fitted for marine exploration, but we do have a handful of specimens found in marine sediments. Most are armored dinosaurs; perhaps the armor helped keep the carcass together while the voluminous gut cavity served as a float, or, more prosaically, there were just a lot of armored dinosaurs living near the seashore. Dinosaurs in marine sediments are usually ascribed to the phenomenon known fondly as "bloat and float", although I suppose that the heaps of well-cured dinosaur jerky we call "mummies" could have also rafted out a fair distance.)

The material in the Minmi is all in the posterior part of the abdomen, and consists of 0.6-2.7-mm-long fragments of fibrous or vascular plant tissue, fruiting bodies about 4.5 mm across, small spherical seeds about 0.3 mm across, and possible fern sporangia. The plant fragments are interpreted as vascular bundles of leaves. The fragments are broken perpendicular to the fibers, while the seeds and fruiting bodies are unbroken (Molnar and Clifford 2001). For me, the most interesting feature of this material is the implication that the leaves were cut into very small pieces in the mouth of the ankylosaur, either by nibbling or chopping in the mouth. The feeding capabilities of ankylosaurs have historically been given little respect; they didn't have impressive dental batteries like hadrosaurs or horned dinosaurs, and their teeth are tiny compared to the teeth of most other dinosaurs. The fine size of the fragments, and the fine size of fragments in a coprolite attributed to an ankylosaur from the Morrison, suggest that ankylosaurs had a few tricks up their sleeves.

The RTMP Corythosaurus

Most of the other specimens listed here are fairly well-known, at least to dinosaur fans (with the exception of the sauropod). Our next guest has slipped through the cracks of fame, although you can see it displayed in all its glory at the Royal Tyrrell Museum of Palaeontology in Drumheller, Alberta. It is a Corythosaurus, RTMP 80.40.1, consisting of a skeleton lacking the hips and tail, but including skin impressions. Like most of the other specimens discussed here, it was found articulated and three-dimensional, and even has the classic hadrosaur death pose (pulled-back head, arms out, folded legs). Within the specimen is an assortment of plant fragments, mostly woody material or bark 1-4 cm at longest, generally twiggy bits of three to nine years' growth, with conifers outnumbering angiosperms 2 to 1. Seeds and seed pods are also present, as well as tiny charcoal bits (Currie et al. 1995).

Stereo RTMP 80.40.1! Points if you can spot the champsosaur.

As Currie et al. (1995) noted, there is nothing about the circumstances that would make the plants material unlikely gut contents. However, the bits of charcoal and diverse palynomorphs (organic microfossils, mostly pollen and spores), particularly reworked Early Cretaceous dinoflagellate cysts, indicate that at least some of the internal material was introduced (Currie et al. 1995). (That, and the inside was filled with sand. Most creatures are not made of sand.) The carcass could have functioned as a sort of trap, slowing down water flow in the fluvial burial setting, and causing plant matter to be dumped within. Currie et al. (1995) came down squarely on the ambiguous side. When I saw the specimen and the blocks that had been removed from the internals, I saw nothing that would cause me to side more strongly on the gut contents side than they did.

They left most of the stuffing in, but a few blocks are stored separately, in the collections.

"Leonardo" the Brachylophosaurus

If you've never seen a dinosaur "mummy" in person before, the photos don't really do them justice. There you are, in front of a sedimentary simulacrum of an actual dead dinosaur, not just a pile of bones. The first thing I thought upon seeing Leo was that it looked uncannily like an animal ready for dissection in an anatomy class. Leonardo is one of several articulated brachylophosaur skeletons found in a small area near Malta, Montana, along with a brachylophosaur bonebed, but it's the only one to date to have these nifty integument impressions. The right shoulder and the neck are particularly impressive, showing three-dimensional structures. In hindsight, looking over Carpenter (2007), I suspect that Leo's postmortem history was something similar to AMNH 5060's.

They should have sent a poet.

The gut region material is exposed in three areas of the specimen: mid-abdomen beneath the ribs; behind the knee; and along the posterior edge of the specimen, near where the ischial rods were before they broke off. The material is in a clayey matrix, and consists of mm-scale dark fragments of plants. It's kind of sepia-colored, in contrast to the orange-brown of the bones and the tan sandstone of the exterior. The plant-rich material is at least a few cm thick, so it's not just a single layer of plant bits. Incidentally, the side we see of Leo is actually the side it was buried on; after extraction, it was flipped over, and has remained flipped over to this day.

Leonardo's leg and abdomen, with two areas where the integument impressions are missing, exposing plant-rich finer sediment (one beneath the ribs, one behind the knee).

How does the evidence for gut contents stack up? Well, we're in the right part of the body, and there's nothing eye-opening about the possible gut contents (like, say, chewed tyrannosaur fingers), so we've got that going for us. Two interesting points: the plant material is not particularly diagnostic, and it retains this anonymity throughout the body. If you were looking for a clear menu, or some kind of differentiation between areas of the body, you're out of luck. It's basically fingernail-sized chunks of carbonized plant material wherever you look, and judging by the thinness of the fragments and occasional traces of surface patterning, leaves are the bulk of it (Tweet et al. 2008). Again, this is consistent with an animal that may have stripped leaves (Ostrom 1964) and processed them heavily in the mouth before digestion, but still circumstantial. One thing it does have going for it is a more uniform size range than the Corythosaurus fragments. The internal sedimentary grain size is much finer than the external grain size, so at least the carcass wasn't filled up as an empty rib cage (doesn't rule out deposition of fines in a cul-de-sac or other restricted setting, but at least flow wasn't trucking right on through). In the end, we came down on the side of probable gut contents, which had been influenced by flowing water (Tweet et al. 2008).

Leo does its thing while I work. Some of us can't get by on our looks, you know!


Abel, O. 1922. Diskussion zu den Vorträgen R. Kräusel and F. Versluys. Paläontologische Zeitschrift 4:87. [in German]

Carpenter, K. 1987. Paleoecological significance of droughts during the Late Cretaceous of the Western Interior. Pages 42–47 in P. Currie and E. Koster, editors. Fourth symposium on Mesozoic terrestrial ecosystems, short papers. Occasional Papers of the Tyrrell Museum of Palaeontology 3.

Carpenter, K. 2007. How to make a fossil: part 2 – Dinosaur mummies and other soft tissue. The Journal of Paleontological Sciences.

Currie, P. J., E. B. Koppelhus, and A. F. Muhammad. 1995. "Stomach" contents of a hadrosaurid from the Dinosaur Park Formation (Campanian, Upper Cretaceous) of Alberta, Canada. Pages 111–114 in Sun Ailing and Wang Yuangqing, editors. Sixth symposium on Mesozoic terrestrial ecosystems and biota, short papers. China Ocean Press, Beijing, China.

Kräusel, R. 1922. Die Nahrung von Trachodon. Paläontologische Zeitschrift 4:80. [in German]

Molnar, R. E., and H. T. Clifford. 2001. An ankylosaurian cololite from Queensland, Australia. Pages 399–412 in K. Carpenter, editor. The armored dinosaurs. Indiana University Press, Bloomington, Indiana.

Osborn, H. F. 1912. Integument of the iguanodont dinosaur Trachodon. [be warned: it's a 76 mb pdf] Memoirs of the American Museum of Natural History 1:33–54.

Ostrom, J. H. 1964. A reconsideration of the paleoecology of the hadrosaurian dinosaurs. American Journal of Science 262(8):975–997.

Sternberg, C. H. 1909. A new Trachodon from the Laramie Beds of Converse County, Wyoming. Science 29(749):753–54.

Tweet, J. S., K. Chin, D. R. Braman, and N. L. Murphy. 2008. Probable gut contents within a specimen of Brachylophosaurus canadensis (Dinosauria: Hadrosauridae) from the Upper Cretaceous Judith River Formation of Montana. Palaios 23(9):624–635.


  1. Great summary! I was playing catch up after the recent Borealopelta paper and this was a big help :)