Under the Roof of Stegosaurus

How have we not talked about Stegosaurus here yet? Simply unreasonable. It goes without saying that Stegosaurus is one of the most recognizable dinosaurs ever. Every kid who knows what a dinosaur is knows about Stegosaurus. That said, many people don’t know Stegosaurus as well as they think they do. There’s a lot of pop culture myths and misunderstandings around this strange beast.

Stegosaurus was discovered at just the right time to earn its status as a true icon of palaeontology. The first described fossils of this dinosaur, fully dubbed Stegosaurus armatus, came from Colorado in the 1870’s, and were named by the oil baron of palaeontology, Othniel Charles Marsh (Marsh, 1877). This was during that ultra-competitive time in American palaeontology known as the Bone Wars, when Marsh and his rival E. D. Cope scoured the west for undescribed fossil creatures new to the eyes of science*. Based on those big bony plates that were found with the incomplete specimen, Marsh initially thought that Stegosaurus was some sort of big, aquatic, turtle-like reptile, believing the plates lay flat across the back and formed a protective dorsal shell. This shelled dinosaur idea inspired the name Stegosaurus, which means ‘roofed lizard’.

*This history of the Bone Wars has been covered and summarized extensively in many different places. Nonetheless, please do comment if want to see it done here as well.

Over the following decades, more complete specimens provided better insight into what Stegosaurus really looked like (and allowed Marsh to name several new species which, typical to form, only a few of are still considered valid). This included fossils with the dorsal plates in generally the right position (sticking up out of the back, but the exact arrangement wouldn’t be figured out for a while) as well as those famous tail spikes. Eventually Marsh had enough material to publish a restoration of what he thought the skeleton of Stegosaurus looked like (Marsh, 1891). This restoration by Marsh bears most of the features of Stegosaurus that became so recognizable throughout the 20th century- the dramatically arched back with its row of enormous plates, the comparatively tiny, narrow head with beaked jaws, the short, sprawling front legs and column-like back legs, and the long tail tipped with fearsome spikes.

Marsh’s 1891 skeletal reconstruction of Stegosaurus. Like many dinosaurs found during this era, we’ve refined a lot of the details known about these creatures. Credit: Public domain

This restoration, though, saw some refinement in the early 20th century. Marsh had mistakenly doubled the number of tail spikes (also famously known as thagomizers thanks to that classic Far Side comic) from four to eight, and had arranged the plates in a single row along the back. Palaeontologists who came after Marsh realized that Stegosaurus really had two rows of plates along the back, but there was some debate over the arrangement of these rows. Charles Gilmore argued, based on incomplete yet partially articulated specimens, that the plates of Stegosaurus were in arranged in staggered as opposed to paired rows, which has been largely accepted in palaeontology ever since (Gilmore, 1914). One thing Marsh did include in 1891 though, which many restorations in the past have overlooked, is the gular armor of Stegosaurus– small bony ossicles in the skin that formed a protective layer over the animal’s throat.

There were actually a few other Stegosaurus species named after the description of S. armatus and, true to form for Marsh, not all of them are still valid. The type species was so fragmentary that it’s not diagnostic, and so Stegosaurus stenops was eventually established as a replacement holotype for this genus. S. stenops is far better known from more complete specimens, and has shaped much of the public’s general idea of a ‘typical’ Stegosaurus with its broad, blunt-tipped plates and tail spikes that were round in cross-section. There’s also the larger Stegosaurus ungulatus, which had narrower, more pointed plates. There’s also the partially complete Stegosaurus sulcatus, distinguished by its tail spikes which bore noticeably enlarged bases.

Silhouettes showing the broad differences between two species of Stegosaurus, S. stenops and S. ungulatus. Credit: KoprX, CC BY-SA 4.0

The low-fronted, droopy-postured, lethargic Stegosaurus became the standard look for this dinosaur for decades. This image, and many other myths, quickly became ingrained in dinosaur pop culture, and we’re still combating the effects to this day. Have you ever heard that Stegosaurus went extinct because it was so stupid, ungainly, and sluggish that it just couldn’t survive? Or that we have no idea what those plates were used for? Or that it needed a secondary brain in its posterior just to be able to function? It’s time to do Stegosarus some justice.

The more species within the larger Stegosauria group that were discovered, the more palaeontologists realized how weird Stegosaurus was (or seemed to be) even among members of its own group. Compared to other stegosaurs, it seemed to have gone to more of an extreme with features like the short front limbs, arched back, short neck, and giant plates. While true that Stegosaurs was indeed one of the largest and more stegosaur-y of the stegosaurs, the extreme proportions we once believed in were never completely accurate. This came to light in 2015 when a relatively complete Stegosaurus stenops specimen from Wyoming was thoroughly described (Maidment et al., 2015). This specimen, nicknamed ‘Sophie’ and housed at the Natural History Museum in London, provided a lot of insight into the true postcranial anatomy of Stegosaurus.

Since the time of Marsh, diagrams and skeletal mounts of Stegosaurus had been composites- skeletons put together with bones from several individuals, as no complete skeleton was known. This led to individuals of differing ages, preservation quality, and size being shown to represent a generalized form of Stegosaurus. While our knowledge of this dinosaur’s posture (and that of most other species) changed during the Dinosaur Renaissance to show it as a more upright, horizontally-built animal, the general proportions stayed the same.  The ‘Sophie’ specimen, although not quite fully grown, allowed palaeontologists to refine their knowledge of the skeleton of Stegosaurus. As summed up and illustrated nicely here, we now know that Stegosaurus wasn’t quite as bizarrely proportioned as we once thought. It had a longer neck, shorter torso, and the hind legs were slightly shorter compared to the front limbs than we’d previously thought. All in all, ‘Sophie’ showed that Stegosaurus was a more horizontally even animal than past reconstructions showed, and more in line with what we knew about other stegosaurs.

The wonderful skeleton of ‘Sophie’, a very informative Stegosaurus specimen. Credit: Susannah Maidment et al. & Natural History Museum, London, CC BY 4.0

Okay, so there’s some neat insights into the posture and proportions of Stegosaurus. What about the rest of the animal? After all, it’s still pretty strange. Culturally, I think we’ve become quite used to Stegosaurus and its weirdness. I also think that, if we never knew about Stegosaurus or any of its relatives for all this time, and then suddenly a complete Stegosaurus was excavated and published on tomorrow, the media and dinosaur fans alike would have a field day over this bizarre hallucination of an animal dug out of the prairie badlands. The dinosaur was indeed considered wondrously strange by at least some people closer to the time of its initial discovery. In The Lost World, Arthur Conan Doyle described Stegosaurus, thorough the eyes of the novel’s protagonist, as: “the most extraordinary creature that I had ever seen. It was the wild dream of an opium smoker, a vision of delirium” (pg. 35). Clearly, he thought it was pretty weird.

Let’s start at the front, with the skull of Stegosaurus. This body part doesn’t seem to get much attention, outside of mentions that it was remarkably small compared to the dinosaur’s rhino to elephant-sized body. The skull was long and narrow, almost like an elongated turtle skull, but with teeth and more openings. Like other thyreophorans (the ‘armored’ dinosaurs), Stegosaurus had relatively weak jaws tipped with a keratinous beak, with small, leaf-shaped teeth in the back of its mouth.

It’s the braincase, of course, that often the subject of discussion here. Many of us grew up learning that the brain of Stegosaurus was the size of a golf ball, or a walnut. The striking ratio between its brain and body size was certainly not lost on early palaeontologists including Marsh himself (Marsh, 1880), who was amazed that such a big animal could have such a large brain. And while the brain of Steogsaurus was indeed relatively small, there’s a bit of mammalian bias showing here. Not that Stegosaurus was likely capable of any sort of advanced reasoning or complex social behaviors, of course. But I’ve literally been asked by people how Stegosaurus could have possibly come into existence without the first of its kind blundering off cliffs, into trees, or into the jaws of carnivores. Here’s the thing to know- intelligence is all relative, and humans tend to value out own form of intelligence as objectively the most valuable. You don’t actually need to be a super genius to survive in nature. Human-level intelligence helps if you want to build complicated tools, form sophisticated social relationships, or solve tricky puzzles. But the Morrison ecosystem demanded none of that out of Stegosaurus. The requirements to survive long enough to reproduce in that world meant that animals had to be tough, know where and how to find food and water, and be able to deal with other creatures they coexisted with. For about five million years Stegosaurus could do all of that, before environmental changes eventually drove it to extinction. Would you and I fare so well in the late Jurassic of western North America?

We’ll continue on with the weird anatomy of Stegosaurus before getting back to how it survived in its demanding environment later. Now, about those dorsal plates… These structures are actually osteoderms, literally pieces of bone that grown within an animal’s skin, and are not directly connected to the main skeleton. Osteoderms take on many different shapes in different species. The armor of mammals like armadillos and glyptodonts, the round bony texture texture on Gila monster skulls, the knobby surface of crocodiles, and the spikes and scutes of ankylosaurian dinosaurs all come from osteoderms. Along the back Stegosaurus, they evolved into broad, somewhat triangular plates which were likely covered in a sheath of keratin.

While we’ve been able to figure out the arrangement of these plates, what’s less certain is what their function was. Marsh believed they may have been for defense from larger-brained predators (Marsh, 1880). Bakker agreed with this in the 1980’s and argued that the plates may have even been somewhat moveable in order to present predators with a continuous surface of sharp, pointed bony structures (Bakker, 1986), but this mobility and function of the plates has been challenged in other research (Buffrenil, 1986). A 2010 publication on scaly skin impressions from the related stegosaur Hesperosaurus confirmed a keratinous covering on stegosaur plates, which may have given them a relatively sharp edge that, even if not strictly for protection, may have been hazardous for a predator to bite into (Christiansen & Tschopp, 2010).

A more popular theory for plate function in Stegosaurus is thermoregulation. The bone surface of Stegosaurus plate show they would’ve had a good deal of blood vessels running along each side beneath the keratin surface (Buffrenil, 1986). Having blood vessels close to the surface of large body parts is a good way to get rid of excess heat. As cool wind moves across the surface of the heat radiator, it draws heat energy out of the animal’s blood and away from heat-sensitive parts like the brain. Several types of mammals like live in warm places today, such as elephants, hares, and foxes, use their blood vessel-rich ears for exactly this. Could Stegosaurus have done the same? Some argue it could have. The internal vascularity of Stegosaurus plates has been shown to be similar to the osteoderms of crocodilians, which use these little bones to help shed excess heat, meaning Stegosaurus may have done something similar (Farlow et al., 2010).

Others argue that we should treat the thermoregulatory idea with caution, and see little to no direct evidence of it (Main et al., 2005). It’s worth noting that, while Stegosaurus plates were broad and flat, making them a decent shape for conducting heat exchange, many if not most other stegosaurs had much smaller, narrower plates which may have been much less suited for this. However, before moving on to the last idea, there may be some compromise here. Stegosaurus was on the large side for a stegosaur, with perhaps only the European Dacenturus outweighing it (Paul, 2016). It inhabited the hot, seasonally arid lowlands of late Jurassic North America where, presumably, an animal of its size may have been prone to overheating. I’m only speculating, but it might be that the plates of Stegosaurus served some other sort of primary function, while secondarily acting as somewhat decent heat radiators as a perk that gave the dinosaur a bit of an edge in this environment. Going back to elephants as an example, they use their ears not just for shedding heat but also for communication, flaring their ears out when alarmed or angry.

This brings me to the final plate use theory: visual communication. Big, elaborate, weird features on animals, including dinosaurs, often tend to be use for display purposes. Horns, antlers, spikes, crests, frills, wattles, pouches, those sorts of things. Was visual display, perhaps to attract mates and ward off predators and rivals, the reason why the plates of Stegosaurus evolved? It could be. A study on the plate shapes of close relative Hesperosaurus argued for sexual dimorphism in the plate shapes for use as a continuous display structure in males (Saitta, 2015). The keratinous plate coverings, which may have greatly enlarged the outer size of each plate, might have even been brightly coloured in certain sexes or age categories, but it was unlikely that they could change colours at will as some popular media has depicted (Christiansen & Tschopp, 2010). Perhaps this display function also indirectly helped in defence against predators, making Stegosaurus look bigger and more imposing, but who knows for sure. We may eventually find evidence for some more conclusive use for these big flat osteoderms but, until then, I’d say display with a possible secondary thermoregulation

Hypothetical signs of sexual dimorphism and associated plate function in the stegosaur Hesperosaurus, from Saitta (2015). Credit: © 2015 Evan Thomas Saitta, CC BY 4.0

Now for those tail spikes, or thagomizers. Like the dorsal plates, these too are simply highly modified osteoderms, and other stegosaur genera had spikes not just at the tip of the tail but along the top of it, the back, or even the shoulders (check out Gigantspinosaurus for some really extreme spikes). You’ll often see the thagomizers, especially on older dinosaur toys, sticking upwards from the top of the tail. This orientation would make the thagomizers fairly useless as defensive tools, and we now know that they actually stuck out horizontally from the sides of the tail, which is much more conducive to lashing out at hungry Allosaurus’ and such (Carpenter, 1998). This is almost certainly what thagomizers were used for. While there was some early display-or-defence arguments (Gilmore, 1914) (Bakker, 1986), fossil evidence including tail spikes with trauma-induced damage (McWhinney et al., 2001) and thagomizer-shaped holes in Allosaurus bones (Carpenter et al., 2005) have largely settled the debate.

This tail-focused defence strategy brings us to a Stegosaurus myth that just won’t completely go away no matter how hard palaeontologists try to make it. You may have heard from some popular dinosaur media source in the past that Stegosaurus, as well as some or most other dinosaurs, had a second brain located elsewhere in their body (often said to be in the hip region). The idea is that the comically-sized brain of Stegosaurus was so tiny, the dinosaur needed a second one just to be able to perform the most basic of functions. This is not true though, and even with its relatively small brain-to-body size ratio, there’s no reason to think that Stegosaurus had or even needed a ‘second brain’. The source of this myth stems from the size of the neural canal in the dinosaur’s sacral vertebrae, which could accommodate a larger volume than the braincase (Marsh, 1880). There was, of course, no secondary brain in this space. Birds also have an enlarged sacral canal region which houses a glycogen body, an organ that may serve to supply glycogen to the animal’s nervous system, and it’s thought that this too is what was going on in the hips of Stegosaurus (Buchholz, 1990).

The diet and ecology of Stegosaurus – how it lived in and interacted with its environment – is a good place to wrap things up. The dinosaur was clearly a plant eater, with its short beak and peg-shaped teeth, but exactly what types of plants Stegosaurus was capable of processing isn’t totally understood by palaeontologists. Different publications have argued for either a weak (Reichel, 2010) or relatively strong bite force (Lautenschlager et al., 2016) in Stegosaurus.

Some experts see Stegosaurus as mostly a low browser, feeding on plants less than a meter off the ground (Weishampel, 1984). One reoccurring idea with Stegosaurus is its ability to, hypothetically, rear up onto its back legs in order to reach higher food. Marsh himself thought this was likely, suggesting that the tail was used for balance, giving the dinosaur a tripod stance (Marsh, 1891). This stemmed from the now outdated view of Stegosaurus hind limbs being radically longer than its front limbs, as an animal with proportions like that theoretically could rear up. Robert Bakker, ever the fan of active, nimble dinosaurs, supported the idea of a rearing Stegosaurus (Bakker, 1978). It has since been challenged (Carpenter, 1998) and tentatively supported (Mallison, 2010) in subsequent publications. Depending on all these different bite force and rearing ability interpretations, Stegosaurus may have been limited to low-growing ferns, horsetails, and mosses on one end of the scale, or able to process tougher leaves and twigs of cycads and conifers on the other.

It seems like there’s no overwhelming consensus one way or the other as to whether Stegosaurus physically could rear up onto its hind legs, at least not yet. That said, assuming it could have done this, is it likely it would have? Hard to say, but perhaps occasionally. There’s an important difference between behaviours that animals do on the regular versus behaviours they do very rarely. The ornithopods and sauropods that Stegosaurus coexisted with would’ve made for some real competition for higher plant resources. I’d imagine it would’ve reared up for access to higher foods, such as tree leaves, only when there was nothing to eat within easy reach closer to the ground, perhaps during seasonal droughts and such.

Periodic dry spells were something that Stegosaurus would have certainly had to deal with. The Morrison Formation, which Stegosaurus shared with a greatest-hits list of Jurassic dinosaurs, was a vast basin of open fern savannahs lined with lush, forested river systems and wetlands during the wet seasons, which became brutally hot and arid during the long dry seasons (Turner & Peterson, 2004). It seems like many dinosaurs preferred to stick to these more hospitable riparian forests, but may have been drawn out to shrinking waterholes during times of food and water shortage. The Morrison is an extensive and well-studied formation with an amazing palaeobiodiversity, and would be impossible to sum up adequately here. It was also a demanding environment, home to giant sauropods, carnivorous theropods both big and small, shielded ankylosaurs and herbivorous ornithopods as well as pterosaurs, mammals, small reptiles, and much more.

As you can see, for such a popular dinosaur, there’s still a lot we don’t know for sure yet about Stegosaurus. This icon of the prehistoric past is still worth revisiting, studying, and searching for out there in the badlands. Hopefully this has helped some folks look on Stegosaurus with new eyes. This bizarre creature straight out of dreams of an opium den actually existed, once walked the arid Jurassic plains of the American West millions of years ago, and is now buried beneath the dusty prairie badlands. Isn’t it wonderful that life on Earth turned out this way?

By Nicholas Carter


Bakker RT (1978). “Dinosaur feeding behavior and the origin of flowering plants”. Nature274 (5672): 661–63.

Bakker, R (1986). The Dinosaur Heresies. Penguin Books. pp. 229–34.

Buchholz (née Giffin) EB (1990). “Gross Spinal Anatomy and Limb Use in Living and Fossil Reptiles”. Paleobiology16 (4): 448–58.

Buffrénil (1986). “Growth and Function of Stegosaurus Plates”. Paleobiology12 (4): 459–73.

Carpenter, Kenneth (1998). “Armor of Stegosaurus stenops, and the taphonomic history of a new specimen from Garden Park, Colorado”. Modern Geology23: 127–44.

Carpenter K, Sanders F, McWhinney L, Wood L (2005). “Evidence for predator-prey relationships: Examples for Allosaurus and Stegosaurus.”. In Carpenter K (ed.). The Carnivorous Dinosaurs. Indiana University Press. pp. 325–50.

Christiansen, N. A.; Tschopp, E. (2010). “Exceptional stegosaur integument impressions from the Upper Jurassic Morrison Formation of Wyoming”. Journal of Geosciences103 (2): 163–171. doi:10.1007/s00015-010-0026-0

Farlow, James O.; Hayashi, Shoji; Tattersall, Glenn J. (2010). “Internal vascularity of the dermal plates of Stegosaurus (Ornithischia, Thyreophora)” (PDF). Swiss J Geoscia103 (2): 173–85.

Gilmore CW (1914). “Osteology of the armored Dinosauria in the United States National Museum, with special reference to the genus Stegosaurus“. Series: Smithsonian Institution. United States National Museum. Bulletin 89 (89).

Lautenschlager, Stephen; Brassey, Charlotte A.; Button, David J.; Barret, Paul Mm. (2016). “Decoupled form and function in disparate herbivorous dinosaur clades”Scientific Reports6: 26495. Bibcode:2016NatSR…626495Ldoi:10.1038/srep26495.

Maidment SCR, Brassey C, Barrett PM (2015) The Postcranial Skeleton of an Exceptionally Complete Individual of the Plated Dinosaur Stegosaurus stenops (Dinosauria: Thyreophora) from the Upper Jurassic Morrison Formation of Wyoming, U.S.A. PLoS ONE 10(10): e0138352. https://doi.org/10.1371/journal.pone.0138352

Main R, de Ricqlès A, Horner JR, Padian K (2005). “The evolution and function of thyreophoran dinosaur scutes: implications for plate function in stegosaurs”. Paleobiology31 (2): 291–314.

Mallison, H. (2010). “CAD assessment of the posture and range of motion of Kentrosaurus aethiopicus HENNIG 1915″. Swiss Journal of Geosciences103 (2): 211–233.

Marsh, Othniel Charles (1880). “Principal characters of American Jurassic dinosaurs, part III”American Journal of Science3 (19): 253–59.

Marsh, O. C. (1877). “A new order of extinct Reptilia (Stegosauria) from the Jurassic of the Rocky Mountains”American Journal of Science3 (14): 513–514.

Marsh, Othniel Charles (1891). “Restoration of Stegosaurus“. American Journal of Science3 (42): 179–81.

McWhinney LA, Rothschild BM, Carpenter K (2001). “Posttraumatic Chronic Osteomyelitis in Stegosaurus dermal spikes”. In Carpenter K (ed.). The Armored Dinosaurs. Indiana University Press. pp. 141–56

Paul, Gregory S. (2016). The Princeton Field Guide to Dinosaurs: 2nd Edition. United States of America: Princeton University Press.

Reichel, Miriam (2010). “A model for the bite mechanics in the herbivorous dinosaur Stegosaurus (Ornithischia, Stegosauridae)”. Swiss Journal of Geosciences103 (2): 235–240.

Saitta E.T. (2015). “Evidence for Sexual Dimorphism in the Plated Dinosaur Stegosaurus mjosi (Ornithischia, Stegosauria) from the Morrison Formation (Upper Jurassic) of Western USA”PLOS ONE10 (4): e0123503.

Turner, Christine E.; Peterson, Fred (May 2004). “Reconstruction of the Upper Jurassic Morrison Formation extinct ecosystem—a synthesis”. Sedimentary Geology167 (3–4): 309–355.

Weishampel, DB (January 1984). “Interactions between Mesozoic Plants and Vertebrates:Fructifications and seed predation”. N. Jb. Geol. Paläontol. Abhandl167 (2): 224–50.