Styracosaurus (/stɪˌrækəˈsɔːrəs/ sti-RAK-ə-SOR-əs; meaning "spiked lizard" from the Ancient Greek styrax/στύραξ "spike at the butt-end of a spear-shaft" and sauros/σαῦρος "lizard") is an extinct genus of herbivorous ceratopsian dinosaur from the Late Cretaceous (Campanian stage) of North America. It had four to six long parietal spikes extending from its neck frill, a smaller jugal horn on each of its cheeks, and a single horn protruding from its nose, which may have been up to 60 centimeters (2 feet) long and 15 centimeters (6 inches) wide. The function or functions of the horns and frills have been debated for many years.

Styracosaurus was a relatively large dinosaur, reaching lengths of 5–5.5 metres (16–18 ft) and weighing about 1.8–2.7 metric tons (2.0–3.0 short tons). It stood about 1.8 meters (5.9 feet) tall. Styracosaurus possessed four short legs and a bulky body. Its tail was rather short. The skull had a beak and shearing cheek teeth arranged in continuous dental batteries, suggesting that the animal sliced up plants. Like other ceratopsians, this dinosaur may have been a herd animal, travelling in large groups, as suggested by bone beds.

Named by Lawrence Lambe in 1913, Styracosaurus is a member of the Centrosaurinae. One species, S. albertensis, is currently assigned to Styracosaurus. Another species, S. ovatus, named in 1930 by Charles Gilmore was reassigned to a new genus, Rubeosaurus, by Andrew McDonald and Jack Horner in 2010, but it has been considered either its own genus or a species of Styracosaurus (or even a specimen of S. albertensis) again, since 2020.

The first fossil remains of Styracosaurus were collected in Alberta, Canada by C. M. Sternberg (from an area now known as Dinosaur Provincial Park, in a formation now called the Dinosaur Park Formation) and named by Lawrence Lambe in 1913. This quarry was revisited in 1935 by a Royal Ontario Museum crew who found the missing lower jaws and most of the skeleton. These fossils indicate that S. albertensis was around 5.5–5.8 metres (18–19 ft) in length and stood about 1.65 metres (5.4 ft) high at the hips. An unusual feature of this first skull is that the smallest frill spike on the left side is partially overlapped at its base by the next spike. It appears that the frill suffered a break at this point in life and was shortened by about 6 centimeters (2.4 inches). The normal shape of this area is unknown because the corresponding area of the right side of the frill was not recovered.

Barnum Brown and crew, working for the American Museum of Natural History in New York, collected a nearly complete articulated skeleton with a partial skull in 1915. These fossils were also found in the Dinosaur Park Formation, near Steveville, Alberta. Brown and Erich Maren Schlaikjer compared the finds, and, though they allowed that both specimens were from the same general locality and geological formation, they considered the specimen sufficiently distinct from the holotype to warrant erecting a new species, and described the fossils as Styracosaurus parksi, named in honor of William Parks. Among the differences between the specimens cited by Brown and Schlaikjer were a cheekbone quite different from that of S. albertensis, and smaller tail vertebrae. S. parksi also had a more robust jaw, a shorter dentary, and the frill differed in shape from that of the type species. However, much of the skull consisted of plaster reconstruction, and the original 1937 paper did not illustrate the actual skull bones. It is now accepted as a specimen of S. albertensis.

In the summer of 2006, Darren Tanke of the Royal Tyrrell Museum of Palaeontology in Drumheller, Alberta relocated the long lost S. parksi site. Pieces of the skull, evidently abandoned by the 1915 crew, were found in the quarry. These were collected and it is hoped more pieces will be found, perhaps enough to warrant a redescription of the skull and test whether S. albertensis and S. parksi are the same. The Tyrrell Museum has also collected several partial Styracosaurus skulls. At least one confirmed bone bed (bonebed 42) in Dinosaur Provincial Park has also been explored (other proposed Styracosaurus bone beds instead have fossils from a mix of animals, and nondiagnostic ceratopsian remains). Bonebed 42 is known to contain numerous pieces of skulls such as horncores, jaws and frill pieces.

Several other species which were assigned to Styracosaurus have since been assigned to other genera. S. sphenocerus, described by Edward Drinker Cope in 1890 as a species of Monoclonius and based on a nasal bone with a broken Styracosaurus-like straight nose horn, was attributed to Styracosaurus in 1915. "S. makeli", mentioned informally by amateur paleontologists Stephen and Sylvia Czerkas in 1990 in a caption to an illustration, is an early name for Einiosaurus. "S. borealis" is an early informal name for S. parksi.

A species, Styracosaurus ovatus, from the Two Medicine Formation of Montana, was described by Gilmore in 1930, named for a partial parietal under the accession number USNM 11869. Unlike S. albertensis, the longest parietal spikes converge towards their tips, instead of projecting parallel behind the frill. There also may only have been two sets of spikes on each side of the frill, instead of three. As estimated from the preserved material, the spikes are much shorter than in S. albertensis, with the longest only 295 millimeters (11.6 inches) long. An additional specimen from the Two Medicine Formation was referred to Styracosaurus ovatus in 2010 by Andrew McDonald and John Horner, having been found earlier in 1986 but not described until that year. Known from a premaxilla, the nasal bones and their horncore, a postorbital bone and a parietal, the specimen Museum of the Rockies 492 was considered to share the medially-converging parietal spikes with the only other specimen of S. ovatus, the holotype. Following this additional material, the species was added to a phylogenetic analysis where it was found to group not with Styracosaurus albertensis, but in a clade including Pachyrhinosaurus, Einiosaurus and Achelousaurus, and therefore McDonald and Horner gave the species the new genus name Rubeosaurus. Another specimen, the partial immature skull USNM 14768, which was earlier referred to the undiagnostic genus Brachyceratops, was also referred to Rubeosaurus ovatus by McDonald and colleagues in 2011. While the medial spikes of USNM 14768 were too incomplete to show if it shared the convergence seen in other R. ovatus specimens, it was considered to be the same species as it was also found in the older deposits of the Two Medicine Formation, and had a unique combination of parietal features only shared completely with the other specimens of the species.