Dilophosaurus

Name and Etymology

The genus name Dilophosaurus was coined by Samuel P. Welles in 1970. In 1942, Welles collected the first specimens from a site near Tuba City, Arizona, that had been discovered in 1940 by Jesse Williams, a Navajo man. At the time, the crests were not clearly recognized, and Welles described the material in 1954 as Megalosaurus wetherilli. It was only after his 1964 discovery of a larger, better-preserved specimen (UCMP 77270) with clearly defined crests that he realized the animal warranted a new genus, which he formally named Dilophosaurus in a 1970 note in the Journal of Paleontology. The species epithet wetherilli honors John Wetherill, a Navajo councilor and noted explorer in the Four Corners region.

Taxonomic Status

Only one species is currently recognized: Dilophosaurus wetherilli (Welles, 1954) Welles, 1970. A Chinese species, Dilophosaurus sinensis, was named in 1993 from the Lufeng Formation of Yunnan, but subsequent studies reassigned it to Sinosaurus triassicus (Carrano et al., 2012), removing it from Dilophosaurus. The informally named Dilophosaurus "breedorum" (based on UCMP 77270, proposed by Pickering in 1995) is considered a nomen nudum with no significant differences from D. wetherilli (Gay, 2005; Marsh & Rowe, 2020).

Scientific Significance

Dilophosaurus is a pivotal taxon for understanding theropod body-size increase and ecological diversification across the Triassic–Jurassic boundary. Its axial skeleton exhibits several derived features relative to Late Triassic theropods, which may be linked to the evolution of macropredatory habits (Marsh & Rowe, 2020). It also provides critical data on the early evolution of cranial ornamentation in neotheropods, demonstrating that nasolacrimal crests evolved independently in multiple early theropod lineages.

Geological Age

Dilophosaurus inhabited the Early Jurassic, with the Kayenta Formation spanning approximately the Sinemurian through Toarcian stages. U-Pb detrital zircon dating by Marsh et al. (2014) yielded a maximum depositional age of 183.7 ± 2.7 Ma from the 'silty facies' in northern Arizona, indicating a late Pliensbachian to early Toarcian age for at least part of the formation. Additional analyses by Marsh (2018) recovered ages of 197.0 ± 1.5 to 195.2 ± 5.5 Ma (mid-Sinemurian) from other horizons, suggesting the Kayenta Formation was deposited over a substantial time span. A 2025 revised chronostratigraphy by Rasmussen et al. further refined the age constraints.

The holotype and paratype come from the lower Kayenta Formation, approximately 10 m above the contact with the underlying Moenave Formation, while the TMM specimens from Gold Spring derive from the middle Kayenta Formation ('Silty Facies').

Formation and Lithology

The Kayenta Formation is part of the Glen Canyon Group and is distributed across the Colorado Plateau region of northern Arizona, northwest Colorado, Nevada, and Utah. It ranges from approximately 100 to 120 m in thickness and consists primarily of reddish-brown to pink sandstone interbedded with varicolored mudstone and siltstone, deposited in fluvial (stream and river) and floodplain environments.

Paleoenvironment

The Kayenta Formation is interpreted as recording a semi-arid to arid climate with seasonally variable precipitation. Sedimentary structures including lenticular bedding, desiccation cracks, and ripple marks indicate streams with fluctuating discharge and associated floodplains. Some limestone horizons and freshwater gastropod and bivalve fossils indicate intermittent ponds or lakes.

Paleomagnetic studies (Steiner & Tanner, 2014) indicate that the Colorado Plateau occupied a near-equatorial paleolatitude during the Early Jurassic, placing Dilophosaurus in a tropical to subtropical arid belt. The associated fauna included freshwater fishes (hybodont sharks, lungfish), crocodylomorphs (Protosuchus, Kayentasuchus), tritylodontids (Kayentatherium, Oligokyphus), the thyreophoran Scutellosaurus, the sauropodomorph Sarahsaurus, the small theropod Coelophysis kayentakatae (= Megapnosaurus kayentakatae), and early turtles (Kayentachelys).

Holotype and Key Specimens

Diagnostic Features

Marsh & Rowe (2020) identified 29 autapomorphies for Dilophosaurus wetherilli. Key diagnostic characters include: a small ridge on the narial process of the premaxilla; dorsally expanded nasals and tall lacrimals that combine to form the distinctive paired parasagittal nasolacrimal crests; the posterior centrodiapophyseal lamina of the cervical vertebrae bifurcates and then reconverges along the neck in a continuous series; and the mandibular fenestra is anteroposteriorly compressed.

Specimen Limitations

The holotype skull is laterally compressed, making interpretation of some structures difficult. Welles (1984) once suggested that the 1964 specimen (UCMP 77270) might represent a different genus, but Marsh & Rowe's (2020) phylogenetic analysis recovered all specimens as a monophyletic group, confirming a single valid species. Differences among specimens are attributable to ontogenetic stage and preservational variation.

Body Plan and Size

Dilophosaurus was a slender, lightly built, medium-to-large theropod. The largest specimen (UCMP 77270) is estimated at approximately 7 m in total length with a skull length of 590 mm and a body mass of about 400 kg. The holotype (UCMP 37302) is estimated at approximately 6 m in length, 283 kg in mass, and a hip height of about 1.36 m, with a skull length of 523 mm (Paul, 2010; Marsh & Rowe, 2020). This made it the largest known terrestrial animal in Early Jurassic North America.

Cranial Crests

The most distinctive feature of Dilophosaurus is the pair of parasagittal crests atop the skull. These begin as a low ridge on the premaxilla and expand dorsally through the nasals and tall lacrimals. In UCMP 77270, the crests reach their greatest height above the midpoint of the antorbital fenestra and project upward and laterally at approximately 80° when viewed anteriorly.

The exact shape of the crests is incompletely preserved, but Marsh & Rowe (2020) compared them to the casque of the helmeted guineafowl (Numida meleagris), suggesting the bony crests were covered in keratin or keratinized skin and may have been considerably larger in life than the fossil bone alone indicates. CT scans reveal that the crests contained pneumatic sinuses connected to the paranasal air sac system surrounding the brain, indicating they were very lightweight. Their probable function was species recognition, sexual display, or visual signaling, though direct evidence is limited (Knell & Sampson, 2011).

Skull and Dentition

The skull is large relative to the body but delicately constructed. The snout is narrow in anterior view and tapers dorsally. A subnarial gap between the premaxilla and maxilla creates a diastema (toothless region) in the tooth row, a feature shared with some other early theropods.

The teeth are generally long, thin, and posteriorly recurved, with serrations on both anterior and posterior carinae. The maxilla bears 12 teeth and the dentary 17, with the largest teeth occurring near the fourth maxillary alveolus. The maxillary teeth are larger than the dentary teeth.

Forelimbs and Hands

The forelimbs are robust with a long, slender humerus. The hand has four digits: the first is short but powerful with a large claw; the second and third are longer and more slender with smaller claws; the fourth is vestigial. Given the range of motion and proportional length of the arms, the mouth likely made first contact with prey during hunting rather than the hands (Senter & Robins, 2015).

Hind Limbs and Locomotion

The hind limbs are long and powerful, with a robust femur and strong, clawed toes. Dilophosaurus was an obligate biped, with the long tail serving as a counterbalance. No direct speed estimates exist, but the limb proportions and body plan indicate an active predator capable of sustained locomotion.

Diet and Prey

Dilophosaurus was carnivorous and was almost certainly the apex predator in its ecosystem. Earlier interpretations (Welles, 1984) suggested that the subnarial gap and apparently delicate skull construction limited it to small prey or fish. However, Marsh & Rowe (2020) demonstrated that the skull was considerably more robust than previously understood, noting that while Dilophosaurus could have consumed fish and small prey in its riverine habitat, claims that it could only eat small, weak animals were overstated. Its sharp, serrated teeth, powerful forelimbs, and large body size indicate it was capable of hunting a variety of prey, including small to medium-sized herbivorous dinosaurs such as Scutellosaurus and Sarahsaurus.

Ecological Niche

Within the Kayenta Formation fauna, Dilophosaurus was the largest predator by a substantial margin. The contemporary theropod Coelophysis kayentakatae was much smaller (2–3 m in length) and would not have competed for the same prey resources. Potential prey items included the basal thyreophoran Scutellosaurus lawleri, the sauropodomorph Sarahsaurus aurifontanalis, and various small vertebrates.

Behavior and Sociality

The 1942 quarry yielded three individuals in close proximity, raising the possibility of gregarious behavior. However, this could equally represent time-averaged accumulation of carcasses at a common water source or other taphonomic bias rather than true social behavior. The crests, if used for visual signaling, may have facilitated social interactions, but direct evidence remains limited.

Geographic Range

All confirmed Dilophosaurus body fossils are restricted to the Kayenta Formation of northern Arizona. Major collecting sites include the Tuba City/Moenave area (UCMP V4214 quarry, holotype and paratype locality), the Gold Spring area (TMM specimens), and Rock Head Quadrangle (additional specimens reported by Gay, 2001, including juvenile material approximately 190 km from the type locality).

Trace Fossils

Ichnofossils (trackways) attributed to Dilophosaurus or similarly sized large theropods have been reported from the Kayenta Formation and other Glen Canyon Group units. The ichnotaxa Dilophosauripus williamsi, Kayentapus hopii, and Eubrontes giganteus have all been proposed as potential Dilophosaurus trackmaker products (Lucas et al., 2006). Similar large theropod tracks are known from eastern North America, Europe, Africa, and China in Triassic–Jurassic boundary strata.

Paleogeographic Setting

During the Early Jurassic, western North America was positioned at near-equatorial latitudes, considerably closer to the equator than its present location. Paleomagnetic studies of the Glen Canyon Group (Steiner & Tanner, 2014) indicate the Colorado Plateau was at approximately 5–15°N, placing Dilophosaurus in a tropical to subtropical arid belt at the southern margin of the expanding Pangaean desert system.

History of Classification

The phylogenetic placement of Dilophosaurus has been contentious since its discovery. Welles (1954) originally classified it within Megalosaurus. By 1970, he recognized it as a distinct genus, and subsequent authors placed it within Coelophysoidea (Rowe & Gauthier, 1990; Holtz, 1994; Tykoski & Rowe, 2004) or Ceratosauria.

Current Phylogenetic Position

Marsh & Rowe's (2020) comprehensive phylogenetic analysis recovered Dilophosaurus as a non-averostran neotheropod positioned on the stem lineage leading to Averostra (= Ceratosauria + Tetanurae). It is not a coelophysoid, contrary to many earlier analyses. In this framework, Dilophosaurus forms a grade with other stem-averostrans including Cryolophosaurus ellioti (Antarctica) and Zupaysaurus rougieri (Argentina).

The previously proposed family 'Dilophosauridae' is not supported as monophyletic. The development of nasolacrimal crests appears to have occurred independently in several early theropod lineages and is likely a broadly distributed plesiomorphic tendency rather than a synapomorphy uniting a specific clade.

Intrageneric Taxonomy

Dilophosaurus sinensis (Hu, 1993) from the Lufeng Formation of China was reassigned to Sinosaurus triassicus by Carrano et al. (2012), removing it from the genus. The only valid species is therefore D. wetherilli.

Established Facts

Anatomical features confirmed across multiple specimens include the paired parasagittal nasolacrimal crests, the subnarial gap and corresponding dental diastema, sharp serrated teeth, four manual digits (with the fourth vestigial), obligate bipedalism, and a body length of 6–7 m.

Well-Supported Hypotheses

The phylogenetic position as a non-averostran neotheropod (Marsh & Rowe, 2020), the function of crests for display rather than combat (based on their fragility), and the semi-arid fluvial paleoenvironment of the Kayenta Formation are all supported by robust evidence.

Uncertain Hypotheses

The precise extent and shape of the keratinous crest covering, body coloration and patterning, the presence or absence of protofeathers, the degree of sexual dimorphism, and the specific nature of social behavior all lack direct fossil evidence and remain speculative. Some trace fossils from the Early Jurassic have been interpreted as feather impressions associated with theropod resting traces (Gierlinski, 1996; Lockley et al., 2003), but whether these represent genuine integumentary evidence or sedimentological artifacts remains debated.

Popular Media vs. Science

The 1993 film Jurassic Park depicted Dilophosaurus as a dog-sized animal that spits blinding venom and fans a colorful neck frill, like a frilled lizard. None of these features have any basis in the fossil record. The real Dilophosaurus was a 6–7 m apex predator, roughly the height of a human at the hip. No anatomical structures supporting venom glands, a venom-delivery mechanism, or a neck frill have ever been found. Director Steven Spielberg reportedly reduced its size deliberately to avoid confusion with the film's Velociraptor.