Introduction

The Morrison Formation of western North America is one of the most well-studied geological formations within the world. These rock layers, which extend from Montana to New Mexico and from Utah to South Dakota, are dated to the late Jurassic Period approximately 155-148 million years ago (MYA). Tens of thousands of dinosaur bones have been uncovered within the rocks of the Morrison Formation, many of which come from the most famous and recognizable dinosaurs that every elementary school child knows such as Allosaurus, Brachiosaurus, Stegosaurus, and Diplodocus. However, the Morrison Formation was home to many more species other than the well-known names that are seen in children’s dinosaur books. One such species is the theropod dinosaur Marshosaurus, an enigmatic species known from fragmentary remains, and its position within the dinosaur family tree is unclear.

Discovery

In the 1970s, fragmentary remains of a theropod dinosaur were discovered in the Cleveland-Lloyd Dinosaur Quarry in east-central Utah. The stratum where the fossils were found was described as “150 feet (45.75 meters) below the top of the Brushy Basin Member of the Morrison Formation” (Madsen 1976, page 52), which would date the bones to the latest part of the Kimmeridgian Stage of the late Jurassic Period, approximately 150 MYA. Several meat-eating dinosaurs had already been found within the Morrison Formation such as Allosaurus and Ceratosaurus, but this was something new. In 1976, an article was published by James H. Madsen Jr. in which this animal was officially named Marshosaurus bicentesimus. The animal was named in honor of Professor Othniel Charles Marsh of Yale University, the famous paleontologist who was one of the major figures in “the Bone Wars” of the 19th Century. Its species name is in reference to it being named in 1976, the bicentennial date of the signing of the Declaration of Independence. The identification of this dinosaur as a new genus and species was based primarily based upon the unique structure of its hip, specifically its left ilium (the bone that forms the top of the hip); the posterior part of the ilium is strongly triangular and comes to a definite point at the back (Madsen 1976, pages 51-52).

Marshosaurus bicentesimus holotype (originally UUVP 2826, now UMNH VP 6373). Left ilium in lateral (A) and medial (B) views. Abbreviations: act- =  acetabulum; anb- = anterior blade; isc- = ischiadic peduncle; meb- = medial blade; pob- = posterior blade; pu- = pubic peduncle. Madsen 1976, page 53.

James H. Madsen Jr’s diagnostic description (an analysis of what parts of the animal’s anatomy make this animal unique and distinguishable from other animals) of the Marshosaurus holotype reads as follows:

“Marshosaurus bicentesimus…is distinct from other Morrison theropods…in the unusual character of the type specimen, a left ilium, and in the referred materials, which include the tooth-bearing elements of the skull and jaw and the complete pelvic girdle…Ilium heavy with long, low posterior blade; pubic peduncle stout, its articular surface rugose and divided into anterior and posterior concavities; ischiadic peduncle moderately long with a rugose, rounded surface truncated at apex; acetabulum deep and nearly symmetrical. And from referred specimens: interdental plates separated to their base; premaxilla loosely attached to maxilla; maxilla with a divided sinus having two openings in the usual position of the maxillary fenestra. Number of teeth in premaxilla, 4; maxilla, 16; and dentary, 22. Maxillary teeth with slightly expanded roots. Pubis long, slender, and anteriorly bowed with a rugose proximal end doubly convex for articulation with the pubic peduncle; the rugose ischiadic articulation crescent shaped and deeply notched on the underside; the bowed shaft of the pubis terminating in a posteriorly expanded foot or pedicle. Ischium shorter than pubis, its shaft relatively straight, grooved medially at midlength, and expanded distally; the pubic articulation crescentic and rugose; the iliac articulation concave and rugose; the proximoventral end expanded, striated laterally and medially to margin of medial contact” (Madsen 1976, pages 51-52). Although the holotype specimen was found in the 1970s, in truth, the first specimens of this animal were actually discovered in 1912, when a partial skeleton of an unidentified meat-eating dinosaur was found at Dinosaur National Monument near the Utah/Colorado border. The fossils consisted of the back half of the upper jaw, the back half of the right lower jaw, a series of neck vertebrae and anterior dorsal vertebrae, a scapula, a humerus, and a rib showing evidence of pathology. However, this specimen wasn’t officially described until 1997 – a full eighty-five years after it was found! – and it was only then that scientists realized that it belonged to Marshosaurus due to the dorsal vertebrae’s neural spines possessing notched craniodorsal margins (Chure et al. 1997, page 38).

Marshosaurus fossils have so far been found at the Cleveland-Lloyd Dinosaur Quarry in east-central Utah, the Dry Mesa Quarry in western Colorado, and Dinosaur National Monument on the Utah/Colorado border. In all localities, the fossils date to the late Kimmeridgian and early Tithonian Stages of the late Jurassic Period approximately 150 MYA, within Zones 2 and 3 of the Morrison Formation (Madsen 1976, pages 51-60; Britt 1991, pages 59-60; Chure et al. 1993, page 28; Turner and Peterson 1999, page 92; Carrano et al. 2012, page 235; Paleofile. “Marshosaurus”; Paleofile. “Marshosaurus sp”). This date will be important, as I’ll discuss later.

Locations where Marshosaurus fossils have been found as of 2025.

Description

Marshosaurus shared its environment with other meat-eating dinosaurs like Allosaurus, Ceratosaurus, Torvosaurus, Coelurus, and Ornitholestes. Many of these genera are known from numerous skeletons and are well-known to the general public. By contrast, very little is known about Marshosaurus largely because the remains from this dinosaur are maddenlingly sparse, amounting to bits and pieces from four skeletons. Although a complete skeleton of Marshosaurus has never been found, paleontologists have guessed that the animal’s size was somewhere around 15-20 feet long (Madsen 1976, page 52; Lambert 1990, page 75; Lessem and Glut 1992, page 294; Holtz 2007, page 371; Paul 2010, page 91).

Most Marshosaurus specimens are housed in the collections of the Natural History Museum of Utah (UMNH). Other specimens are housed within Brigham Young University (BYU), the Carnegie Museum of Natural History (CMNH), the Denver Museum of Nature and Science (DMNH), and the Yale Peabody Museum (YPM). Known Marshosaurus specimens, aside from isolated teeth, consist of the following fossil bones:

• Right premaxilla (paratype) (UMNH VP 7820 (it used to be classified as UUVP 3266))
• Left maxilla (paratype) (UMNH VP 7824 (it used to be classified as UUVP 1864))
• Right maxilla (paratype) (UMNH VP 7824 (it used to be classified as UUVP 1846))
• Right maxilla (paratype) (UMNH VP 7825 (it used to be classified as UUVP 4695))
• Left dentary (paratype) (UMNH 6364 (it used to be classified as UUVP 40-555))
• Left dentary (paratype) (UMNH 6367 (it used to be classified as UUVP 3454))
• Right dentary (paratype) (UMNH VP 6368 (it used to be classified as UUVP 3502))
• Dorsal vertebra (UUVP coll.)
• Left ilium (UMNH VP 6372 (it used to be classified as UUVP 1182))
• Left ilium (paratype) (UMNH VP 6372 (it used to be classified as UUVP 1845))
• Left ilium (holotype specimen) (UMNH VP 6373 (it used to be classified as UUVP 2826))
• Right ilium (paratype) (UUVP 1882)
• Right ilium with a pathology (paratype) (UMNH VP 6374 (it used to be classified as UUVP 2742))
• Left ischium (paratype) (UMNH VP 380 (it used to be classified as UUVP 2878))
• Right ischium (paratype) (UMNH VP 6379 (it used to be classified as UUVP 2832))
• Left pubis (paratype) (UMNH VP 6384 (it used to be classified as UUVP 40-295))
• Left pubis (paratype) (UMNH VP 6386 (it used to be classified as UUVP 1867))
• Right pubis (paratype) (UMNH VP 6387 (it used to be classified as UUVP 4736))
• A series of anterior tail vertebrae (BYUVP 5201)
• Tail vertebra (UUVP 99)
• Tail vertebra (UUVP 441)
• Tail vertebra (UUVP 5247)
• Tail vertebra (UUVP 5780)
• Two partial skeletons of juvenile Marshosaurus (CMNH 21704 (formerly classified as DINO 16455b and DMN 343)). Consists of posterior skull elements, posterior mandible, atlas, axis, cervical vertebrae, dorsal vertebrae, dorsal rib, scapula, and humerus.
• Skull elements and vertebrae (DMNH 3718). Consists of 1 left premaxilla, 1 right maxilla, and 4 caudal vertebrae.
• Partial pelvis (YPM-PU 72-1)

James H. Madsen Jr. made the following observations concerning Marshosaurus’ skull and teeth: “The premaxilla of Marshosaurus in comparison with that of Allosaurus (Madsen, in press) is unusual in several respects: (1) the maxillary suture is weak and restricted to a slightly roughened area on the convex, posterolateral surface of the premaxilla, while in Allosaurus the contact is continuous along most of the posterior edge; (2) the teeth are laterally compressed, but in Allosaurus they are nearly circular in cross section; (3) interdental plates are present and separate, but they are fused in Allosaurus; and (4) there is little indication of the subnarial foramen, which is evident in Allosaurus…The skull of the diminutive theropod Velociraptor mongoliensis, described by Osborn (1924) from the Protoceratops zone of central Mongolia, exhibits at least two characteristics important in a comparison with Marshosaurus: (1) a similar broad excavation of the ex ternal surface of the maxilla anterior to the maxillary fenestra(e) is bordered below by a sharp shelf that sweeps posteriorly under the antorbital fenestra; and (2) deep nutrient foramina are in conspicuous, linear arrangement, parallel to the tooth margins of the maxilla and dentary. However, the difference in size and the significant stratigraphic separation indicate that Marshosaurus and Velociraptor are probably not closely related…The maxilla of Marshosaurus is also similar to that of Deinonychus (Ostrom, 1969, p. 18) in having two exits from the maxillary sinus(es). Colbert and Russell (1969 , p. 7) also note the occurrence of two openings in the maxilla of Dromaeosaurus. Variation in the development of chambers or sinuses in the superior rami of theropod maxillae, which may be of phyletic importance, warrants further study…The dentary of Marshosaurus is long and narrow with a noticeable, downward bow in contrast to the straight dentary of Deinonychus (Ostrom , 1969, p. 29), but similarly it has a rounded anterior end terminating medially in a relatively weak symphysis. Additionally, the dentary lacks the marked lateral bow of Allosaurus, which indicates a relatively narrow, pointed lower jaw. The significance of the diastema in one of the Marshosaurus dentaries (UUVP 40-555) is unknown, but it is curious that a similar abnormality is present on a dentary of Ceratosaurus (UUVP 158), on two dentaries of Allosaurus (UUVP 5289 and 5748) in the Cleveland-Lloyd Dinosaur Collection, and on the type of Labrosaurus (USNM 3215)” (Madsen 1976, pages 56-57).

No complete skull, or even a complete upper jaw or lower jaw, has been found which belongs to Marshosaurus. However, we have enough fragments of the skull from the four partial skeletons which have been found that we can make a composite image of what the whole head would have looked like. I looked at as many photographs of Marshosaurus skull bones as I could. Then I deleted the background in each picture so that the skull bones were laid atop a blank white background. Afterwards, I re-sized the pictures so that all of them were to the same scale. Finally, I fitted the pictures of the different bones together like a puzzle. I noticed the following features:

1. The teeth are strongly angled backwards.
2. Large nares located at the front of the cranium.
3. Large pits on the front of the premaxillia.
4. A large notch between the premaxilla and the maxilla.
5. A large beefy upper jaw, but a comparatively thin and slender lower jaw. However, it’s possible that the lower jaw comes from a smaller individual.

I should state that the two partial skeletons found at Dinosaur National Monument are believed to belong to juveniles (Chure et al. 1993, page 30; Chure et al. 1997, page 38), and examination of other meat-eating dinosaurs such as Allosaurus, Albertosaurus, Tarbosaurus, and Tyrannosaurus has shown that juveniles had a markedly different appearance to adults. Therefore, the composite reconstruction of what a fully-grown Marshosaurus looked like is probably not accurate. Joe Sertich believes that these elements do not belong to Marshosaurus bicentesimus, but instead belong to another species of Marshosaurus or possibly to an entirely new genus (“Science Division Live: Jurassic Giants of DMNS”). However, I am skeptical of this claim. Numerous specimens of creatures like Coelophysis, Allosaurus, and Tyrannosaurus show considerable anatomical variation between individuals, so I do not believe that the slight anatomical differences warrant a new species or new genus name for this specimen.

Although most Marshosaurus remains are elusive, Marshosaurus teeth are abundant, and thankfully its teeth are easy to identify because they’re very distinctive. Like many theropod dinosaurs, Marshosaurus teeth had serrations on both the front and back edges. However, the serrations on the teeth of Marshosaurus only extend from the tip down two-thirds of the front edge; the lower third of the front edge doesn’t have any serrations. Also, the serrations on Marshosaurus’ teeth are extremely small. Each anterior serration measures only one-fourth of one millimeter in size, while the posterior tooth serrations measure one-third of a millimeter. The serrations are so small that the teeth might as well be serration-less (“Madsen 1976, page 56; “Marshosaurus teeth – Can you identify?”; “Morrison Fm: Theropod Serrations Comparison”). The teeth of Marshosaurus cannot be confused with the teeth of Ceratosaurus because Ceratosaurus teeth have serrations extending along both the anterior and posterior edges from the tip to the gumline. Ceratosaurus teeth also have “lingual grooves” – a series of thin shallow grooves that run up the length of the tooth’s side – on the tooth’s lingual side (“Identification of Ceratosaurus teeth”). However, it was noted that the teeth of Marshosaurus are similar to those of Piatnitzkysaurus  (Hendrickx et al. 2015, page 635).

Marshosaurus is commonly stated to be a member of the super-family Megalosauroidea, which is divided into the families Piatnitzkysauridae, Megalosauridae, and Spinosauridae. Megalosauroid vertebrae possess prominent rims around the anterior articular surfaces of cervical centra. Features of vertebrae in piatnitzkysaurids include flat anterior articular surface on the centra of the neck vertebrae, the presence of a pneumatic foramen or fossa on the posterior of some neck vertebrae, and bifurcated anterior neural spines on the dorsal vertebrae. However, the neck vertebrae of Marshosaurus have a subtle convex front surface on the centra, which is seen in the neck vertebrae of Leshansaurus, Monolophosaurus, and Sinraptor. The neck vertebrae which are closest to the skull possess a ridge along the vertebrae’s underside. This is a feature which absent in most tetanuran theropods, but it is present in some non-tetanurans. Subtle bifurcated neural spines are also present in the anterior dorsal vertebrae in Condoraptor, Marshosaurus, and Piatnitzkysaurus (Carrano et al. 2012, page 235; Dai et al. 2020). All of these features imply that Marshosaurus is either an advanced non-tetanuran or a primitive tetanuran. Either way, it rests close to the boundary which separates these two groups.

Marshosaurus’ hip structure was one of the most distinctive aspects of its anatomy, so much so that it was listed as an autopomorphy when the time came to officially name it in 1976. James H. Madsen Jr. makes the following statements concerning Marshosaurus’ hip structure: “The pelvic structure of Marshosaurus is a generalized theropod design. The ilium of Marshosaurus is grossly similar in morphology to those of Allosaurus, Ornitholestes, Ceratosaurus, Stokesosaurus, Deinonychus, and Tyrannosaurus in having elongate, laterally compressed posterior and anterior blades that project beyond the ischiadic and pubic peduncles. The major characteristics of the ilium, which set it apart from that of other theropods, are the faceted ischiadic peduncle and the ventral concavities of the pubic peduncle…The pubis of Marshosaurus represents a slight departure in design from other theropods, but superficially it resembles Marsh’s Coelurus agilis (1884), although it is approximately twice the length and, therefore, from a much larger individual. It differs from other Late Jurassic species in the unusual shape of the foot, the anterior bow of the shaft, and the knobs on the iliac articulation. The anterodistal expansion of the foot extends slightly beyond the anterior surface of the shaft. Additionally, the posterior extension of the pubic foot of Marshosaurus is also restricted, especially so in comparison with that of the tyrannosaurids and the Lower Cretaceous theropod Microvenator (Ostrom, 1970, plate 12)…The ischium of Marshosaurus is unusual primarily because of its relative shortness, which is approximately two-thirds the length of the pubis. The distal end of the ischium in Marshosaurus is asymmetrically expanded, more so than in Allosaurus. It is in sharp contrast with the thin shaft of the tyrannosaurid ischium, which is without distal expansion (Russell, 1970, p. 2), and the ischium of Elaphrosaurus (Janensch, 1925, figure 8), which is irregularly expanded…Regrettably, the distal ends of the ischium and pubis of Marshosaurus cannot be compared in detail with those of Ceratosaurus, since as noted by Gilmore (1920, p. 108) they are missing in the type and may not have been available when Marsh (1884) made his original description” (Madsen 1976 pages 57-59). However, it must be noted that the pelvis of the mid-Jurassic megalosaur Eustreptospondylus oxoniensis and the late Jurassic megalosaur Torvosaurus tanneri also have ilia which come to a sharp triangular posterior point, so this feature is not exclusive to Marshosaurus (Galton and Jensen 1979, pages 3, 9-11).

Left pelvic girdle of Marshosaurus bicentesimus in  lateral view. A = left ilium; B = pubis; C = ischium. D = acetabulum in ventral view. Abbreviations: act- = acetabulum; amb- = probable origin of ambiens muscle; isc- = ischiadic peduncle; on- = obturator notch; op- = obturator process; pu- = pubic peduncle. Madsen 1976, page 58.

Phylogeny

Ever since its discovery, Marshosaurus has been of uncertain placement within Theropoda. In 1979, James Madsen and Wade E. Miller said that Marshosaurus was a megalosaurid (Madsen and Miller 1979; Mindat. “Marshosaurus”). This was repeated in 1984 by Dale Russell (Carrano et al. 2012, page 215). In 1988, Gregory Paul stated that Marshosaurus was a megalosaurid, and even more specifically was a close relative of Eustreptospondylus, but didn’t provide any evidence (Paul 1988, pages 286-287, 289). In 1989, Sergei M. Kurzanov said Marshosaurus was an allosaurid, but didn’t provide any evidence (Kurzanov 1989; Carrano et al. 2012, page 215)Mindat. “Marshosaurus”). In 1990, David Lambert said that Marshosaurus was possibly an allosaurid, again with no evidence (Lambert 1990, page 75). In 1991, Brooks Britt stated that Marshosaurus might be a dromaeosaur (Britt 1991, pages 2-3), a claim that was repeated that same year by Wade E. Miller and colleagues (Miller et al. 1991; Mindat. “Marshosaurus”). In 1993, Dan Chure and colleagues said that Marshosaurus was a carnosaur (Chure et al. 1993, page 28), and updated this slightly in 1997 stating that it was a basal carnosaur, possibly a megalosaur (Chure et al. 1997, page 38). In 2004, Thomas Holtz and colleagues listed Marshosaurus as a basal tetanuran (Holtz et al. 2004, page 77), and elaborated further on this in 2007 stating that Marshosaurus possessed a medly of anatomical features that are similar to spinosaurs, carnosaurs, and coelurosaurs (Holtz 2007, page 371). In 2010, Gregory Paul said that it was a “miscellaneous carnosaur” (Paul 2010, page 91). In 2010, Roger Benson stated that Marshosaurus was grouped together in a clade including Piatnitzkysaurus, Xuanhanosaurus, and Condorraptor. Despite possessing some physical features that seem advanced, Benson’s 2010 classification of Marshosaurus shows it and its close relatives to be very archaic members of the megalosauroid super-family (Benson 2010, pages 921-922, 926). Although this clade was not given a name in Benson’s 2010 report, it has since been coined as Piatnitzkysauridae, which was erected in 2012 by Matthew Carrano and colleagues, who stated that Piatnitzkysauridae was located as the base of the megalosauroid tree (Carrano et al. 2012, pages 218, 235, 251). The piatnitzkysaurids were medium-sized theropod dinosaurs which lived mostly during the middle Jurassic Period; Marshosaurus might have been the last surviving genus of this family, a hold-over from an earlier time. In recent years, it was stated that Piatnitzkysauridae was not a family of Megalosauroidea, but was instead be the earliest diverging family of Allosauroidea (Rauhut and Pol 2019: 18826; Pradelli et al. 2025: 176). However, since Marshosaurus’ placement in the theropod tree has jumped around a lot since its discovery, and since the majority of people still think of it as a megalosaur, I think that these claims need more consensus amongst paleontologists.

One might wonder if Marshosaurus is just a juvenile Torvosaurus, since both inhabited the Morrison Formation. According to phylogenics, the answer is no. Marshosaurus was a primitive megalosaur while Torvosaurus was an advanced megalosaur. Therefore, they cannot be juvenile and adult specimens of the same species. Another possibility is that Marshosaurus was a late-surviving dilophosaur. Dilophosaurus is positioned as an advanced member of Neotheropoda, in between primitive theropods like Coelophyis and the more advanced theropods in the clade Averostra, which includes the megalosaurs, carnosaurs, and coelurosaurs. Also, both Marshosaurus and Dilophosaurus share a few features of their anatomy, such as a premaxillary notch in the upper jaw, enlarged nares, and thin blade-like teeth with extremely tiny serrations. Placing Marshosaurus in the same clade as Dilophosaurus would account for the anatomical similarities as well as the fact that Marshosaurus’ skeleton appears to have a mixture of features from several theropod families. However, Marshosaurus’ hip bones, which are the most diagnostic features, are clearly similar to a megalosaur. Therefore, while there are some aspects to its skull and tooth structure that are similar to Dilophosaurus, it’s probably safe to say that Marshosaurus is a megalosaur, not a dilophosaur. It would therefore appear that the similarities between the two are either purely coincidental, or perhaps are the result of “convergent evolution” – when species from two different groups evolve into more-or-less the same appearance due to habitat or lifestyle.

Conclusions

Marshosaurus is believed to be a megalosauroid, which includes the megalosaurs and the spinosaurs, the latter of which is known for being semi-aquatic and relying on fish for a substantial part of their diet. There is a notch in Marshosaurus’ upper jaw located between the premaxilla and maxilla bones, similar to Dilophosaurus, the megalosaurid Eustreptospondylus, and spinosaurid theropods; animals with such features are supposed to have placed an emphasis on fish-eating. The skull of Marshosaurus had enormous temporal fenestrae, inferring that this animal could rapidly snap its jaws shut – perfect for grabbing fast-moving prey. The teeth of Marshosaurus possess extremely tiny serrations, and the front edge of the tooth only has serrations for two-thirds of its length; spinosaurids have absolutely no serrations on their teeth. Based upon the information that I have obtained and the observations which I have made, I would like to offer the following hypothesis: I believe that Marshosaurus bicentesimus may represent a transitional intermediate form between Megalosauridae and Spinosauridae (if you will, a “proto-spinosaur”), in which a large portion of its diet consisted of fish, and its teeth were gradually losing their serrations and transforming themselves into serrationless teeth seen in spinosaurs.

Marshosaurus bicentesimus. © Jason R. Abdale (August 19, 2025).

However, there are many more questions that need to be answered, especially if this hypothesis is to be proven.

1. Check the isotopic analysis of the bones and teeth to see if there is evidence that these animals spent a lot of time in or near water, or if they had a diet which was either highly or exclusively fish-based. Do a compare-and-contrast isotopic analysis of the teeth of Marshosaurus with those of Dilophosaurus, Baryonyx, and Spinosaurus. Do any match?
2. Check muscle attachments on the back of the skull and on the neck vertebrae. Animals that need to catch fish need to quickly snap their neck downwards in order to grab them. Was Marshosaurus’ neck properly constructed for such activity?
3. How flexible are the neck vertebrae of Marshosaurus? Compare neck vertebra structure of Marshosaurus, Allosaurus, Torvosaurus, and Baryonyx.
4. Did Marshosaurus possess paroccipital processes on the back of the skull? These knobs of bone serve as attachment points for muscles which pull the head and neck downwards. Did Marshosaurus have these features, and were they unusually enlarged?
5. Do a comprehensive study on the taphonomy of the sites where all Marshosaurus fossils have been found. Do these localities have anything in common?

As you can see, there is lot of research left to do. Hopefully, this article will encourage paleontologists to do more research on Marshosaurus in order to confirm or reject the various points which I proposed.

A talk given in Liverpool, UK in 2023 by Dino Chak Hei Chu, Jordan Bestwick, and Richard Butler from the University of Birmingham stated that analysis was done on the teeth of Allosaurus fragilis, Ceratosaurus nasicornis, and Marshosaurus bicentissimus to see if there was any evidence of dietary partitioning, and indeed there was. However, no further information was given in the report (Chu et al. 2023).

Concerning the three localities where Marshosaurus remains have been found so far, Dry Mesa Quarry is believed to be the product of a drought followed by a flash flood (Richmond and Morris 1998, pages 121-143), Dinosaur National Monument is thought of as a flash flood deposit along a river bank (When Dinosaur Roamed America), and the Cleveland-Lloyd Quarry, while frequently thought of as a predator trap in mud or quicksand (When Dinosaurs Ruled, episode 2), is probably more likely to be an extreme drought event at an ephemeral pond (“Allosaurus: Your Dinosaurs Are Wrong #20”). All three of these sites involve bodies of water, which might mean that Marshosaurus made its living around water bodies.

During the 1990s, it was stated in paleo-media that the Morrison Formation was a semi-arid savannah (Dinosaur!, episode 2). However, this is only partially accurate. As stated earlier, the Morrison Formation encompasses a large geographic area and also spans ten million years. Different regions of the Morrison Formation had different environments, and the same region’s environment could change over its ten million year lifespan. What appears to have happened in prior years is that paleontologists looked at small slices of this vast geological pie, and assumed that’s how the entire place looked. Then in 2004, the Morrison Formation was looked at in more detail, and it became clear that our previous understanding of what the Morrison Formation was like was incorrect. Far from being a dry dusty savannah throughout its existence, the Morrison Formation changed over time. As I mentioned near the beginning of this article, all Marshosaurus fossils found so far are dated to the late Kimmeridgian and early Tithonian Stages of the late Jurassic Period circa 150 MYA. During the Kimmeridgian Stage, there were numerous rivers flowing west-to-east out of the hill country west of the Morrison Formation, and there was a series of lakes and marshlands throughout the eastern side of the Morrison Formation. Clearly, the Morrison Formation was a lot wetter and a lot greener during the Kimmeridgian Stage. However, during the later Tithonian Stage (149-145 MYA), the story changed. The climate got noticeably drier, the majority of the rivers and lakes disappeared, and the lush verdant landscape changed to a semi-arid savannah (Parrish et al. 2004, pages 137-162; Turner and Peterson 2004, pages 309-355; Demko et al. 2005, pages 4-7, 9).

A series of maps showing environmental changes to the Morrison Formation from the early Kimmeridgian to the early Tithonian. Turner and Peterson 2004, pages 336, 338, 340, and 342.

The fact that Marshosaurus’ fossils aren’t found in the upper layers of the Morrison Formation where the landscape was parched and desiccated for most of the year, but are instead found in the middle layers where the landscape was green and went suggests that Marshosaurus was very much at home in such a water-filled environment. The spinosaurs, too, favored areas that had lots of water. This is one more piece of evidence to suggest that Marshosaurus was taking the first steps to evolving into a spinosaur.

In 1993, Jim Farlow attempted to calculate population ratios of meat-eating dinosaurs within the Morrison Formation. He stated that for every 100 square kilomerers, there would be 2-6 large theropods like Allosaurus or Torvosaurus, 12 medium-sized Ceratosaurus, as many as 28 slightly smaller theropods like Marshosaurus, and nearly 500 small theropods like Ornitholestes (Foster 2020, page 216). Although such statistics are largely conjectural, the numbers for supposed Marshosaurus population levels are intriguing. If correct, then we ought to have found many more Marshosaurus fossils within the Morrison Formation than we actually did. So where are they? One explanation is that, quite simply, we’re looking in the wrong places. Go out into those areas which were likely to be along riverbanks and lakeshores within the Morrison Formation and see what you can find. In addition to the Morrison Formation, there are several other geologic formations which are situated along North America’s Pacific coast dating to the late Jurassic Period which Marshosaurus fossils have be found in. These include the Mariposa Formation and the Franciscan Formation (also called the Franciscan Complex) of central California, the Galice Formation of Oregon, and the Fernie Formation, Morrissey Formation, Mist Mountain Formation, Elk Formation, Nikanassin Formation, and Monteith Formation of British Columbia. All of these are marine formations, and if Marshosaurus really was a dinosaur that had a taste for fish and made its living by the water’s edge, then these formations might be worth looking into. As an aside, these formations really should be examined in more detail anyway if for no other reason than they have been rarely examined up to now. Heavily-studied geologic formations like the Morrison Formation are all well and good, but there’s also a lot of potential in geologic formations which have been barely looked at. For example, fragmentary remains of ichthyosaurus and plesiosaurs have been found within the Mariposa, Franciscan, and Galice Formations but their identity is mysterious. Are these examples of species found elsewhere in the world, or are they entirely new species? Who knows?

Marshosaurus might have even inhabited Europe as well. In addition to the remains found in the Morrison Formation, fossils which were found in Portugal might belong to Marshosaurus (Malafaia et al. 2017). Furthermore,  at least one tooth found in northern Germany which are dated to the late Jurassic had been classified as “Megalosauroidea cf. Marshosaurus” (collection ID code: NLMH106235b). However, this identification was only tentative (Gerke and Wings 2016: e0158334).

So was Marshosaurus really a very primitive spinosaur or a megalosaur which was showing the first signs of evolving into a spinosaur? At the moment, this is just an intriguing hypothesis which requires more fossils and more analysis to see if it’s true. Bluntly speaking, we need more bones, so get out there and start digging!

I truly enjoy writing my articles and drawing my art, but it’s increasingly clear that I can’t keep this up without your gracious financial assistance. Kindly check out my pages on Redbubble and Fine Art America if you want to purchase merch of my artwork. Consider buying my ancient Roman history books Four Days in September: The Battle of Teutoburg and The Great Illyrian Revolt if you or someone that you know loves that topic, or my ancient Egyptian novel Servant of a Living God if you enjoy action and adventure.

Please consider becoming a patron on my Patreon page so that I can afford to purchase the art supplies and research materials that I need to keep posting art and articles onto this website. Professional art supplies are pricey, and many research articles are “pay to read”, and some academic journals are rather expensive. Patreon donations are just $1 per month – that’s it. If everyone gave just $1 per month, it would go a long way to purchase the stuff that I need to keep my blog “Dinosaurs and Barbarians” running.

Take care, and, as always, keep your pencils sharp.

Bibliography

Books

Foster, John. Jurassic West: The Dinosaurs of the Morrison Formation and their World, Second Edition. Bloomington: Indiana University Press, 2020.

Holtz Jr., Thomas R. Dinosaurs: The Most Complete Up-to-Date Encyclopedia for Dinosaur Loves of All Ages. New York: Random House, 2007.

Lambert, David. The Dinosaur Data Book. New York: Avon Books, 1990.

Lessem, Don; Glut, Donald F. The Dinosaur Society Dinosaur Encyclopedia. New York: Random House Inc., 1993.

Paul, Gregory S. Predatory Dinosaurs of the World. New York: Simon & Schuster, 1988.

Paul, Gregory S. The Princeton Field Guide to Dinosaurs. Princeton: Princeton University Press, 2010.

Articles

Benson, Roger B. J. (2010). “A description of Megalosaurus bucklandii (Dinosauria: Theropoda) from the Bathonian of the UK and the relationships of Middle Jurassic theropods”. Zoological Journal of the Linnean Society, volume 158, issue 4 (April 2010). Pages 882-935.

Britt, Brooks (1991). “Theropods of Dry Mesa Quarry (Morrison Formation, Late Jurassic), Colorado, with emphasis on the osteology of Torvosaurus tanneri”. Brigham Young University Geology Studies, volume 37 (1991). Pages 1-72.
https://geology.byu.edu/0000017c-f2a9-dd9b-a7ff-f2af65580000/geol-stud-vol-37-britt-pdf.

Carrano, Matthew T.; Benson, Roger, B. J.; Sampson, Scott D. (2012). “The phylogeny of Tetanurae (Dinosauria: Theropoda)”. Journal of Systematic Palaeontology, volume 10, issue 2 (June 2012). Pages 211-300.
https://www.researchgate.net/publication/230808558_The_phylogeny_of_Tetanurae_Dinosauria_Theropoda.

Chu, Dino Chak Hei; Bestwick, Jordan; Butler, Richard (2023). “Unveiling Late Jurassic Theropod Niche Partitioning and Allosaurus Ontogeny using 3D Dental Microwear Textural Analysis”. Progressive Palaeontology 2023 (June 2023).
https://www.researchgate.net/publication/379989161_ProgPal_2023_Liverpool_Lightning_Talk_abstract_on_my_MSci_project_-_Unveiling_Late_Jurassic_Theropod_Niche_Partitioning_and_Allosaurus_Ontogeny_using_3D_Dental_Microwear_Textural_Analysis.

Chure, Dan; Britt, Brooks; Madsen, Jim (1993). “New data on the theropod Marshosaurus from the Morrison Formation (Upper Jurassic: Kimmeridgian-Tithonian) of Dinosaur National Monument”. In Santucci, Vincent L., ed. National Park Service Paleontology Research, Volume 1. Technical Report NPS/NRPEFO/NRTR 93/11 (June 1993). Page 23.
https://npshistory.com/publications/paleontology/nrtr-93-11.pdf.

Chure, D. J.; Madsen, J. H.; Britt, B. B. (1993). “New data on theropod dinosaurs from the Late Jurassic Morrison Formation”. Journal of Vertebrate Paleontology, volume 13, supplement to #3: 53rd Annual Meeting of the Society of Vertebrate Paleontology, Albuquerque, New Mexico (October 13-16, 1993). Page 30.
https://www.tandfonline.com/doi/epdf/10.1080/02724634.1993.10011533.

Chure, Daniel J.; Britt, Brooks B., Madsen Jr., James H. (1997). “A new specimen of Marshosaurus bicentesimus (Theropoda) from the Morrison Formation (Late Jurassic) of Dinosaur National Monument”. Journal of Vertebrate Paleontology, volume 17 supplement to #3 (September 1997). Page 38.
https://www.tandfonline.com/doi/10.1080/02724634.1997.10011028.

Dai, Hui; Benson, Roger; Hu, Xufeng; Ma, Qingyu; Tan, Chao; Li, Ning; Xiao, Ming; Hu, Haiqian; Zhou, Yuxuan; Wei, Zhaoying; Zhang, Feng; Jiang, Shan; Li, Deliang; Peng, Guangzhao; Yu, Yilun; Xu, Xing (2020). “A new possible megalosauroid theropod from the Middle Jurassic Xintiangou Formation of Chongqing, People’s Republic of China and its implication for early tetanuran evolution”. Scientific Reports, volume 10 (January 10, 2020). Article #139.
https://www.nature.com/articles/s41598-019-56959-x.

Demko, Timothy M.; Nicoll, Kathleen; Beer, Joseph J.; Hasiotis, Stephen T.; Park, Lisa E. (2005). “Mesozoic Lakes of the Colorado Plateau”. Geological Society of America Field Guide 6: Interior Western United States (2005). Pages 329-356.
https://www.researchgate.net/publication/260000500_Mesozoic_Lakes_of_the_Colorado_Plateau.

Galton, Peter M.; Jensen, James A. (1979). “A new large theropod dinosaur from the Upper Jurassic of Colorado”. Brigham Young University Geology Studies, volume 26, issue 2 (July 1979). Pages 1-12.
https://geology.byu.edu/0000017c-ec08-d1c3-a7fd-fd199f1c0001/geo-stud-vol-26-part-2-galton-jensen-pdf.

Gerke, Oliver; Wings, Oliver (2016). “Multivariate and Cladistic Analyses of Isolated Teeth Reveal Sympatry of Theropod Dinosaurs in the Late Jurassic of Northern Germany” PLOS One, volume 11, issue 7: e0158334 (July 6, 2016).
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0158334.

Hendrickx, Christophe; Mateus, Octavio; and Araújo, Ricardo. (2015). “The dentition of megalosaurid theropods”. Acta Palaeontologica Polonica, volume 60, issue 3 (2015). Pages 627-642.
https://app.pan.pl/archive/published/app60/app000562013.pdf.

Holtz Jr., Thomas R.; Molnar, Ralph E.; Currie, Philip J. (2004). “Basal Tetanurae”. In Weishampel, David B.; Dodson, Peter; Osmolska, Halszka, eds. The Dinosauria, Second Edition. Berkeley: University of California Press, 2004. Pages 71-110.

Kurzanov, Sergei Mikhailovich (1989). “O proiskhozhdenii i evolyutsii infraotryada dinozavrov Carnosauria”. Paleontologicheskiy Zhurnal, volume 4 (1989). Pages 3-14.

Madsen Jr., James H. (1976). “A second new theropod dinosaur from the late Jurassic of east central Utah”. Utah Geology, volume 3, issue 1 (1976). Pages 51-60.
https://ugspub.nr.utah.gov/publications/utah_geology/UtGeol3-1.pdf.

Madsen Jr., James H.; Miller, Wade E. (1979). “The fossil vertebrates of Utah, an annotated bibliography”. Brigham Young University Geology Studies, volume 26, part 4 (1979). Pages 1-147.

Malafaia, Elizabete; Escaso, Fernando; Mocho, Pedro; Serrano-Martínez, Alejandro; Torices, Angelica; Cachão, Mário; Ortega, Francisco (2017). “Analysis of diversity, stratigraphic and geographical distribution of isolated theropod teeth from the Upper Jurassic of the Lusitanian Basin, Portugal”. Journal of Iberian Geology, volume 43, issue (June 2017). Pages 257-291.
https://link.springer.com/article/10.1007/s41513-017-0021-7.

Miller, Wade E.; Baer, James L.; Stadtman, Kenneth L.; Britt, Brooks B. (1991). “The Dry Mesa Dinosaur Quarry, Mesa County, Colorado”. Guidebook for Dinosaur Quarries and Tracksites Tour, Western Colorado and Eastern Utah. Pages 31-46.
https://archives.datapages.com/data/grand-junction-geo-soc/data/012/012001/31_gjgs-sp0120031.htm.

Parrish, Judith Totman; Peterson, Fred; Turner, Christine E. (2004). “Jurassic ‘savannah’—plant taphonomy and climate of the Morrison Formation (Upper Jurassic, Western USA)”. Sedimentary Geology, volume 167, issues 3-4 (May 15, 2004). Pages 137-162.
https://www.academia.edu/48459168/Jurassic_savannah_plant_taphonomy_and_climate_of_the_Morrison_Formation_Upper_Jurassic_Western_USA_.

Pradelli, Luciano Agustin; Pol, Diego; Ezcurra, Martin (2025). “The appendicular osteology of the Early Jurassic theropod Piatnitzkysaurus floresi and its implications on the morphological disparity of non-coelurosaurian tetanurans”. Zoological Journal of the Linnean Society, volume 203 issue 1 (January 2025): 176.

Rauhut, Oliver W. M.; Pol, Diego (2019). “Probable basal allosauroid from the early Middle Jurassic Cañadón Asfalto Formation of Argentina highlights phylogenetic uncertainty in tetanuran theropod dinosaurs”. Scientific Reports, volume 9, issue 1 (December 11, 2019): 18826.
https://www.nature.com/articles/s41598-019-53672-7.

Richmond, Dean R.; Morris, Thomas H. (1998). “Stratigraphy and Cataclysmic Deposition of the Dry Mesa Dinosaur Quarry, Mesa County, Colorado”. Modern Geology, volume 22 (1998). Pages 121-143.
https://www.researchgate.net/publication/313750845_Stratigraphy_and_cataclysmic_deposition_of_the_Dry_Mesa_Dinosaur_Quarry_Mesa_County_Colorado.

Turner, Christine E.; Peterson, Fred (1999). “Biostratigraphy of Dinosaurs in the Upper Jurassic Morrison Formation of the Western Interior, U.S.A.”. Utah Geological Survey Miscellaneous Publication 99-1: Vertebrate Paleontology in Utah (1999). Pages 77-106.
https://www.google.com/books/edition/Vertebrate_Paleontology_in_Utah/qeRM16ndBx4C?hl=en&gbpv=1&dq=Marshosaurus&pg=PA92&printsec=frontcover.

Turner, Christine E.; Peterson, Fred (2004). “Reconstruction of the Upper Jurassic Morrison Formation extinct ecosystem—a synthesis”. Sedimentary Geology, volume 167, issues 3-4 (May 15, 2004). Pages 309-355.
https://doc.rero.ch/record/14577/files/PAL_E1793.pdf.

Websites

Laelaps. “Meet Marshosaurus, the Jurassic’s Forgotten Predator”, by Riley Black (November 30, 2015).
http://phenomena.nationalgeographic.com/2015/11/30/meet-marshosaurus-the-jurassics-forgotten-predator/. Accessed on July 18, 2025.

Mindat. “Marshosaurus”.
https://www.mindat.org/taxon-4823527.html. Accessed on July 18, 2025.

Paleofile. “Marshosaurus”.
http://www.paleofile.com/Dinosaurs/Theropods/Marshosaurus.asp. Accessed on August 19, 2025.

Paleofile. “Marshosaurus sp”.
http://www.paleofile.com/Dinosaurs/Theropods/Marshosaurussp.asp. Accessed on August 19, 2025.

The Fossil Forum. “Identification of Ceratosaurus teeth” (August 24, 2015).
https://www.thefossilforum.com/topic/57022-identification-of-ceratosaurus-teeth/. Accessed on August 21, 2025.

The Fossil Forum. “Marshosaurus teeth – Can you identify?” (August 27, 2015).
http://www.thefossilforum.com/index.php?/topic/57075-marshosaurus-teeth-can-you-identify/. Accessed on July 18, 2025.

The Fossil Forum. “Morrison Fm: Theropod Serrations Comparison” (August 25, 2015).
http://www.thefossilforum.com/index.php?/topic/57037-morrison-fm-theropod-serration-comparison/. Accessed on July 18, 2025.

The Natural World. “Meet Marshosaurus, Morrison’s Mysterious Meat Muncher!”, by Zach Neher (July 24, 2013).
http://thenaturalworld1.blogspot.com/2013/07/meet-marshosaurus-morrisons-mysterious.html. Accessed on July 18, 2025.

Videos

Dinosaur! Episode 2 – “The Tale of a Bone”. A&E, 1991.

When Dinosaur Roamed America. The Discovery Channel, 2001.

When Dinosaurs Ruled. Episode 2 – “The Real Jurassic Park”. TLC, 1999.

YouTube. Chaotic Good. “Allosaurus: Your Dinosaurs are Wrong #20” (February 14, 2017).
https://www.youtube.com/watch?v=j6E-mGjgk8g. Accessed on August 21, 2025.

YouTube. Denver Museum of Nature & Science. “Science Division Live: Jurassic Giants of DMNS” (July 8, 2021).
https://www.youtube.com/watch?v=PuGktWu43ko. Accessed on August 20, 2025.

Categories: Paleontology, Uncategorized

Tags: allosaurid, Allosauridae, America, art, dinosaur, Jurassic Period, Late Jurassic, Marshosaurus, megalosaurid, Megalosauridae, Megalosauroidea, Mesozoic, Morrison Formation, North America, paleo-art, paleontology, Piatnitzkysauridae, Prehistoric, reptile, spinosaur, Spinosauridae, theropod