Spinosaurus

Annotated skull diagram

The skull had a narrow snout filled with straight conical teeth that lacked serrations. There were six or seven teeth on each side of the very front of the upper jaw, in the premaxillae, and another twelve in both maxillae behind them. The second and third teeth on each side were noticeably larger than the rest of the teeth in the premaxilla, creating a space between them and the large teeth in the anterior maxilla; large teeth in the lower jaw faced this space. The very tip of the snout holding those few large anterior teeth was expanded, and a small crest was present in front of the eyes.[4] Using the dimensions of three specimens known as MSNM V4047, UCPC-2, and BSP 1912 VIII 19, and assuming that the postorbital part of the skull of MSNM V4047 had a shape similar to the postorbital part of the skull of Irritator, Dal Sasso and colleagues (2005) estimated that the skull of Spinosaurus was 1.75 meters (5.7 ft) long,[4] but more recent estimates suggest a length of 1.6–1.68 meters (5.2–5.5 ft).[15][16] The Dal Sasso and colleagues skull length estimate is questioned because skull shapes can vary across spinosaurid species and because MSNM V4047 may not belong to Spinosaurus itself.[3][17]

Reconstruction of the holotype fossils

Very tall neural spines growing on the back vertebrae of Spinosaurus formed the basis of what is usually called the animal's "sail". The lengths of the neural spines reached over 10 times the diameters of the vertebral bodies from which they extended.[18] The neural spines were slightly longer front to back at the base than higher up, and were unlike the thin rods seen in the pelycosaur finbacks Edaphosaurus and Dimetrodon, contrasting also with the thicker spines in the iguanodontian Ouranosaurus.[18]

Spinosaurus sails were unusual, although other dinosaurs, namely the ornithopod Ouranosaurus, which lived a few million years earlier in the same general region as Spinosaurus, and the South American sauropod Amargasaurus, might have developed similar structural adaptations of their vertebrae. The sail may be an analog of the sail of the Permian synapsid Dimetrodon, which lived before the dinosaurs even appeared, produced by convergent evolution.[18]

The structure may also have been more hump-like than sail-like, as noted by Stromer in 1915 ("one might rather think of the existence of a large hump of fat [German: Fettbuckel], to which the [neural spines] gave internal support")[19] and by Jack Bowman Bailey in 1997.[18] In support of his "buffalo-back" hypothesis, Bailey argued that in Spinosaurus, Ouranosaurus, and other dinosaurs with long neural spines, the spines were relatively shorter and thicker than the spines of pelycosaurs (which were known to have sails); instead, the dinosaurs' neural spines were similar to the neural spines of extinct hump-backed mammals such as Megacerops and Bison latifrons.[18]

S. aegyptiacus skeletal reconstruction in swimming posture

Two species of Spinosaurus have been named: Spinosaurus aegyptiacus (meaning "Egyptian spine lizard") and the disputed Spinosaurus maroccanus (meaning "Moroccan spine lizard").[20][21]

The first described remains of Spinosaurus were found and described in the early 20th century. In 1912, Richard Markgraf discovered a partial skeleton of a dinosaur in the Bahariya Formation of western Egypt. In 1915, German paleontologist Ernst Stromer published an article assigning the specimen to a new genus and species Spinosaurus aegyptiacus.[19][20]

Fragmentary additional remains from Bahariya, including vertebrae and hindlimb bones, were designated by Stromer as "Spinosaurus B" in 1934.[22] Stromer considered them different enough to belong to another species, and this has been borne out. With the advantage of more expeditions and material, it appears that they pertain either to Carcharodontosaurus[23] or to Sigilmassasaurus.[21]

S. maroccanus was originally described by Dale Russell in 1996 as a new species based on the length of its neck vertebrae.[21] Specifically, Russell claimed that the ratio of the length of the centrum (body of vertebra) to the height of the posterior articular facet was 1.1 in S. aegyptiacus and 1.5 in S. maroccanus.[21] Later authors have been split on this topic. Some authors note that the length of the vertebrae can vary from individual to individual, that the holotype specimen was destroyed and thus cannot be compared directly with the S. maroccanus specimen, and that it is unknown which cervical vertebrae the S. maroccanus specimens represent. Therefore, though some have retained the species as valid without much comment,[24][25][26] most researchers regard S. maroccanus as a nomen dubium[4][27][28] or as a junior synonym of S. aegyptiacus.[23]

Six main partial specimens of Spinosaurus have been described.

Plate I in Stromer (1915) showing S. aegyptiacus holotype elements

BSP 1912 VIII 19, described by Stromer in 1915 from the Bahariya Formation, was the holotype.[19][20] The material consisted of the following items, most of which were incomplete: right and left dentaries and splenials from the lower jaw measuring 75 centimeters (30 in) long; a straight piece of the left maxilla that was described but not drawn; 20 teeth; 2 cervical vertebrae; 7 dorsal (trunk) vertebrae; 3 sacral vertebrae; 1 caudal vertebra; 4 thoracic ribs; and gastralia.[19] Of the nine neural spines whose heights are given, the longest ("i," associated with a dorsal vertebra) was 1.65 meters (5.4 ft) in length.[19] Stromer claimed that the specimen was from the early Cenomanian, about 97 million years ago.[19][20]

This specimen was destroyed in World War II, specifically "during the night of 24/25 April 1944 in a British bombing raid of Munich" that severely damaged the building housing the Paläontologisches Museum München (Bavarian State Collection of Paleontology).[20] However, detailed drawings and descriptions of the specimen remain. Stromer's son donated Stromer's archives to the Paläontologische Staatssammlung München in 1995, and Smith and colleagues analyzed two photographs of the Spinosaurus holotype specimen BSP 1912 VIII 19 discovered in the archives in 2000.[20] On the basis of a photograph of the lower jaw and a photograph of the entire specimen as mounted, Smith concluded that Stromer's original 1915 drawings were slightly inaccurate.[20] In 2003, Oliver Rauhut suggested that Stromer's Spinosaurus holotype was a chimera, composed of vertebrae and neural spines from a carcharodontosaurid similar to Acrocanthosaurus and a dentary from Baryonyx or Suchomimus.[28] This analysis was rejected in at least one subsequent paper.[4]

NMC 50791, held by the Canadian Museum of Nature, is a mid-cervical vertebra which is 19.5 centimeters (7.7 in) long from the Kem Kem Beds of Morocco.[21] It is the holotype of Spinosaurus maroccanus as described by Russell in 1996.[21] Other specimens referred to S. maroccanus in the same paper were two other mid-cervical vertebrae (NMC 41768 and NMC 50790), an anterior dentary fragment (NMC 50832), a mid-dentary fragment (NMC 50833), and an anterior dorsal neural arch (NMC 50813).[21] Russell stated that "only general locality information could be provided" for the specimen, and therefore it could be dated only "possibly" to the Albian.[21]

Specimen MNHN SAM 124 of S. maroccanus, Muséum National d'Histoire Naturelle, Paris

MNHN SAM 124, housed at the Muséum National d'Histoire Naturelle, is a snout (consisting of partial premaxillae, partial maxillae, vomers, and a dentary fragment).[29] Described by Taquet and Russell in 1998, the specimen is 13.4 to 13.6 centimeters (5.3–5.4 in) in width; no length was stated.[29] The specimen was located in Algeria, and "is of Albian age."[29] Taquet and Russell believed that this specimen along with a premaxilla fragment (SAM 125), two cervical vertebrae (SAM 126-127), and a dorsal neural arch (SAM 128), belonged to S. maroccanus.[29]

BM231 (in the collection of the Office National des Mines, Tunis) was described by Buffetaut and Ouaja in 2002.[27] It consists of a partial anterior dentary 11.5 centimetres (4.53 in) in length from an early Albian stratum of the Chenini Formation of Tunisia.[27] The dentary fragment, which included four alveoli and two partial teeth, was "extremely similar" to existing material of S. aegyptiacus.[27]

UCPC-2 in the University of Chicago Paleontological Collection consists mainly of two narrow connected nasals with a fluted (ridged) crest from the region between the eyes.[4] The specimen, which is 18.0 centimetres (7.09 in) long, was located in an early Cenomanian part of the Moroccan Kem Kem Beds in 1996 and described in the scientific literature in 2005 by Cristiano Dal Sasso of the Civic Natural History Museum in Milan and colleagues.[4]

Specimen MSNM V4047 of S. aegyptiacus in the Civic Natural History Museum in Milan

MSNM V4047 (in the Museo di Storia Naturale di Milano), described by Dal Sasso and colleagues in 2005, consists of a snout (premaxillae, partial maxillae, and partial nasals) 98.8 centimetres (38.9 in) long from the Kem Kem Beds.[4] Like UCPC-2, it is thought to have come from the early Cenomanian. Arden and colleagues in 2018 tentatively assigned this specimen to Sigilmassasaurus brevicollis given its size. In the absence of associated material, however, it is difficult to be certain which material belongs to which taxon.[17]

FSAC-KK 11888 is a partial subadult skeleton recovered from the Kem Kem beds of North Africa. Described by Ibrahim and colleagues (2014) and designated as the neotype specimen (although Evers and colleagues 2015 reject the neotype designation for FSAC-KK-11888).[30] It includes cervical vertebrae, dorsal vertebrae, neural spines, a complete sacrum, femora, tibiae, pedal phalanges, caudal vertebra, several dorsal ribs, and fragments of the skull.[5] The body proportions of this specimen have been debated as the hind limbs are disproportionately shorter in the specimen than in previous reconstructions. However, it has been demonstrated by multiple paleontologists that the specimen is not a chimaera, and is indeed a specimen of Spinosaurus that suggests that the animal had much smaller hind limbs than previously thought.[31][32][33] Other known specimens consist mainly of very fragmentary remains and scattered teeth. These include:

• A 1986 paper described prismatic structures in tooth enamel from two Spinosaurus teeth from Tunisia.[34]
• Buffetaut (1989, 1992) referred three specimens from the Institut und Museum für Geologie und Paläontologie of the University of Göttingen in Germany to Spinosaurus: a right maxilla fragment IMGP 969-1, a jaw fragment IMGP 969-2, and a tooth IMGP 969-3.[35][36] These had been found in a Lower Cenomanian or Upper Albian deposit in southeastern Morocco in 1971.[35]

Reconstructed subadult skeleton from a private collection

• Kellner and Mader (1997) described two unserrated spinosaurid teeth from Morocco (LINHM 001 and 002) that were "highly similar" to the teeth of the S. aegyptiacus holotype.[37]
• Teeth from the Chenini Formation in Tunisia which are "narrow, somewhat rounded in cross-section, and lack the anterior and posterior serrated edges characteristic of theropods and basal archosaurs" were assigned to Spinosaurus in 2000.[38]
• Teeth from the Echkar Formation of Niger were tentatively referred to Spinosaurus in 2007.[39]
• A partial tooth 8 cm long purchased at a fossil trade show, reportedly from the Kem Kem Bed of Morocco and attributed to Spinosaurus maroccanus, showed 1–5 mm wide longitudinal striations and micro-structures (irregular ridges) among the striations in a 2010 paper.[26]

MHNM.KK374 to.KK378 are five isolated quadrates (skull bones) of different sizes which were collected by locals and acquired commercially in the Kem Kem region of southeastern Morocco, provided by François Escuillié and are deposited in the collections of the Muséum d’Histoire Naturelle of Marrakech. The quadrates show two different morphologies suggesting the existence of two spinosaurines in Morocco.[15]

Material possibly belonging to Spinosaurus has been reported from the Turkana Grits of Kenya.[40]

Some scientists have considered the genus Sigilmassasaurus a junior synonym of Spinosaurus. In Ibrahim and colleagues (2014), the specimens of Sigilmassasaurus was referred to Spinosaurus aegyptiacus together with "Spinosaurus B" as the neotype and Spinosaurus maroccanus was considered as a nomen dubium following the conclusions of the other papers.[4][5][23] A 2015 re-description of Sigilmassasaurus disputed these conclusions, and considered the genus valid.[30] This conclusion was further supported in 2018 by Arden and colleagues, who consider Sigilmassasaurus to be a distinct genus, though a very close relative of Spinosaurus, the two unified in the tribe Spinosaurini, coined in the study.[17]

The subfamily Spinosaurinae was named by Sereno in 1998, and defined by Holtz and colleagues (2004) as all taxa closer to Spinosaurus aegyptiacus than to Baryonyx walkeri. And the subfamily Baryonychinae was named by Charig & Milner in 1986. They erected both the subfamily and the family Baryonychinae for the newly discovered Baryonyx, before it was referred to the Spinosauridae. Their subfamily was defined by Holtz and colleagues in 2004, as the complementary clade of all taxa closer to Baryonyx walkeri than to Spinosaurus aegyptiacus. Examinations by Marcos Sales and Cesar Schultz and colleagues (2017) indicate that the South American spinosaurids Angaturama, Irritator, and Oxalaia were intermediate between Baronychinae and Spinosaurinae based on their craniodental features and cladistic analysis. This indicates that Baryonychinae may in fact be non-monophyletic. Their cladogram can be seen below.[41]

The cladogram below depicts the findings of Arden and colleagues (2018):[17]

1915 illustration of S. aegyptiacus dorsal vertebrae

The function of the dinosaur's sail or hump is uncertain; scientists have proposed several hypotheses including heat regulation and display. In addition, such a prominent feature on its back could make it appear even larger than it was, intimidating other animals.[18]

The structure may have been used for thermoregulation. If the structure contained abundant blood vessels, the animal could have used the sail's large surface area to absorb heat. This would imply that the animal was only partly warm-blooded at best and lived in climates where nighttime temperatures were cool or low and the sky usually not cloudy. It is also possible that the structure was used to radiate excess heat from the body, rather than to collect it. Large animals, due to the relatively small ratio of surface area of their body compared to the overall volume (Haldane's principle), face far greater problems of dissipating excess heat at higher temperatures than gaining it at lower. Sails of large dinosaurs added considerably to the skin area of their bodies, with minimum increase of volume. Furthermore, if the sail was turned away from the sun, or positioned at a 90 degree angle towards a cooling wind, the animal would quite effectively cool itself in the warm climate of Cretaceous Africa.[42] However, Bailey (1997) was of the opinion that a sail could have absorbed more heat than it radiated.[18] Bailey proposed instead that Spinosaurus and other dinosaurs with long neural spines had fatty humps on their backs for energy storage, insulation, and shielding from heat.[18]

Many elaborate body structures of modern-day animals serve to attract members of the opposite sex during mating. It is possible that the sail of Spinosaurus was used for courtship, in a way similar to a peacock's tail. Stromer speculated that the size of the neural spines may have differed between males and females.[19]

Gimsa and colleagues (2015) suggest that the dorsal sail of Spinosaurus was analogous to the dorsal fins of sailfish and served a hydrodynamic purpose.[43] Gimsa and others point out that more basal, long-legged spinosaurids had otherwise round or crescent-shaped dorsal sails, whereas in Spinosaurus, the dorsal neural spines formed a shape that was roughly rectangular, similar in shape to the dorsal fins of sailfish. They therefore argue that Spinosaurus used its dorsal neural sail in the same manner as sailfish, and that it also employed its long narrow tail to stun prey like a modern thresher shark. Sailfish employ their dorsal fins for herding schools of fish into a "bait ball" where they cooperate to trap the fish into a certain area where the sailfish can snatch the fish with their bills. The sail could have possibly reduced yaw rotation by counteracting the lateral force in the direction opposite to the slash as suggested by Gimsa and colleagues (2015). Gimsa and colleagues specifically wrote :

Restoration

Spinosaurus anatomy exhibits another feature that may have a modern analogy: its long tail resembled that of the thresher shark, employed to slap the water to herd and stun shoals of fish before devouring them (Oliver and colleagues, 2013). The strategies that sailfish and thresher sharks employ against shoaling fish are more effective when the shoal is first concentrated into a ‘bait ball’ (Helfman, Collette & Facey, 1997; Oliver and colleagues, 2013; Domenici and colleagues, 2014). Since this is difficult for individual predators to achieve, they cooperate in this effort. When herding a shoal of fish or squid, sailfish also raise their sails to make themselves appear larger. When they slash or wipe their bills through shoaling fish by turning their heads, their dorsal sail and fins are outstretched to stabilize their bodies hydrodynamically (Lauder & Drucker, 2004). Domenici and colleagues (2014) postulate that these fin extensions enhance the accuracy of tapping and slashing. The sail can reduce yaw rotation by counteracting the lateral force in the direction opposite to the slash. This means that prey is less likely to recognize the massive trunk as being part of an approaching predator (Marras and colleagues, 2015; Webb & Weihs 2015).

Spinosaurus exhibited the anatomical features required to combine all three hunting strategies: a sail for herding prey more efficiently, as well as flexible tail and neck to slap the water for stunning, injuring or killing prey. The submerged dorsal sail would have provided a strong centreboard-like counterforce for powerful sidewards movements of the strong neck and long tail, as performed by sailfish (Domenici and colleagues, 2014) or thresher sharks (Oliver and colleagues, 2013). While smaller dorsal sails or fins make the dorsal water volume better accessible for slashing, it can be speculated that their smaller stabilization effect makes lateral slashing less efficient (e.g. for thresher sharks). Forming a hydrodynamic fulcrum and hydrodynamically stabilizing the trunk along the dorsoventral axis, Spinosaurus’ sail would also have compensated for the inertia of the lateral neck by tail movements and vice versa not only for predation but also for accelerated swimming. This behaviour might also have been one reason for Spinosaurus’ muscular chest and neck reported by Ibrahim and colleagues (2014).

Reconstructed skull and neck

It is unclear whether Spinosaurus was primarily a terrestrial predator or a piscivore, as indicated by its elongated jaws, conical teeth and raised nostrils. The hypothesis of spinosaurs as specialized fish eaters has been suggested before by A. J. Charig and A. C. Milner for Baryonyx. They base this on the anatomical similarity with crocodilians and the presence of digestive acid-etched fish scales in the rib cage of the type specimen.[44] Large fish are known from the faunas containing other spinosaurids, including the Mawsonia, in the mid-Cretaceous of northern Africa and Brazil. Direct evidence for spinosaur diet comes from related European and South American taxa. Baryonyx was found with fish scales and bones from juvenile Iguanodon in its stomach, while a tooth embedded in a South American pterosaur bone suggests that spinosaurs occasionally preyed on pterosaurs,[45] but Spinosaurus was likely to have been a generalized and opportunistic predator, possibly a Cretaceous equivalent of large grizzly bears, being biased toward fishing, though it undoubtedly scavenged and took many kinds of small or medium-sized prey.[11] A study by Cuff and Rayfield (2013) concluded that bio-mechanical data suggests that Spinosaurus was not an obligate piscivore and that its diet was more closely associated with each individual's size.[46] The characteristic rostral morphology of Spinosaurus allowed its jaws to resist bending in the vertical direction, however its jaws were poorly adapted with respect to resisting lateral bending.[46]

In 2009, Dal Sasso and colleagues. reported the results of X-ray computed tomography of the MSNM V4047 snout.[47] As the foramina on the outside all communicated with a space on the inside of the snout, the authors speculated that Spinosaurus had pressure receptors inside the space that allowed it to hold its snout at the surface of the water to detect swimming prey species without seeing them.[47]

A 2010 isotope analysis by Romain Amiot and colleagues found that oxygen isotope ratios of spinosaurid teeth, including teeth of Spinosaurus, indicate semiaquatic lifestyles.[48] Isotope ratios from tooth enamel and from other parts of Spinosaurus (found in Morocco and Tunisia) and of other predators from the same area such as Carcharodontosaurus were compared with isotopic compositions from contemporaneous theropods, turtles, and crocodilians.[48] The study found that Spinosaurus teeth from five of six sampled localities had oxygen isotope ratios closer to those of turtles and crocodilians when compared with other theropod teeth from the same localities.[48] The authors postulated that Spinosaurus switched between terrestrial and aquatic habitats to compete for food with large crocodilians and other large theropods respectively.[48] A 2013 study performed by scientists Andrew R. Cuff and Emily J. Rayfield showed that Spinosaurids like Spinosaurus had relatively poor resistance in their skulls for torsion compared to other members of this group (Baryonyx) and modern alligators, thus showing Spinosaurus preyed more regularly on fish than it did on land animals, although considered predators of the former too.[46]

A 2018 study by Donald Henderson, however, refutes the claim that Spinosaurus was semiaquatic. By studying the buoyancy in lungs of crocodilians and comparing it to the lung placement in Spinosaurus, it was discovered that Spinosaurus could not sink or dive below the water surface. It was also capable of keeping its entire head above the water surface while floating, much like other non-aquatic theropods. Furthermore, the study found that Spinosaurus had to continually paddle its hind legs to prevent itself from tipping over onto its side, something that extant semiaquatic animals do not need to perform. Henderson therefore theorized that Spinosaurus probably did not hunt completely submerged in water as previously hypothesized, but instead would have spent much of its time on land or in shallow water.[49][6]

Hip region of an older reconstruction

Hip region of the new reconstruction

Although traditionally depicted as a biped, it has been suggested since the mid-1970s that Spinosaurus was at least an occasional quadruped.[10][42] This was bolstered by the discovery of Baryonyx, a relative with robust arms.[50] Because of the mass of the hypothesized fatty dorsal humps of Spinosaurus, Bailey (1997) was open to the possibility of a quadrupedal posture,[18] leading to new restorations of it as such.[50] Theropods, including spinosaurids, could not pronate their hands (rotate the forearm so the palm faced the ground),[51] but a resting position on the side of the hand was possible, as shown by fossil prints from an Early Jurassic theropod.[52] The hypothesis that Spinosaurus had a typical quadrupedal gait since fell out of favor, however it was still believed that spinosaurids may have crouched in a quadrupedal posture, due to biological and physiological constraints.[44][51]

The possibility of a quadrupedal Spinosaurus was revived by a 2014 paper by Ibrahim and colleagues that described new material of the animal. The paper found that the hind limbs of Spinosaurus were much shorter than previously believed, and that its center of mass was located in the midpoint of the trunk region, as opposed to near the hip as in typical bipedal theropods. It was therefore proposed that Spinosaurus was poorly adapted for bipedal terrestrial locomotion, and must have been an obligate quadruped on land. The reconstruction used in the study was an extrapolation based on different sized individuals, scaled to what were assumed to be the correct proportions.[5] Palaeontologist John Hutchinson of the Royal Veterinary College of the University of London has expressed scepticism to the new reconstruction, and cautioned that using different specimens can result in inaccurate chimaeras.[53] Scott Hartman also expressed criticism because he believed the legs and the pelvis were inaccurately scaled (27% too short) and didn't match the published lengths.[54] However, Mark Witton expressed agreement with the proportions reported in the paper.[55] In their 2015 re-description of Sigilmassasaurus, Evers and colleagues argued that Sigilmassasaurus was in fact a distinct genus from Spinosaurus, and therefore doubted whether the material assigned to Spinosaurus by Ibrahim et. al. should be assigned to Spinosaurus or Sigilmassasaurus.[30] In 2018, an analysis by Henderson found that Spinosaurus probably was competent at bipedal terrestrial locomotion; the center of mass was instead found to be close to the hips, allowing Spinosaurus to stand upright like other bipedal theropods.[6]

An ungual phalanx measuring 21 mm (0.83 in) belonging to a very young juvenile Spinosaurus indicates that the theropod developed its semiaquatic adaptations at a very young age or at birth and maintained them throughout its life. The specimen, found in 1999 and described by Simone Maganuco and Cristiano Dal Sasso and colleagues, is believed to have come from an animal measuring 1.78 m (5.8 ft) (assuming it resembled a smaller version of the adult), making it the smallest specimen of Spinosaurus currently known.[56]