Velociraptor

Maxillae of V. osmolskae and V. mongoliensis compared

Maxillae and a lacrimal (the main tooth-bearing bones of the upper jaw, and the bone that forms the anterior margin of the eye socket, respectively) recovered from the Bayan Mandahu Formation in 1999 by the Sino-Belgian Dinosaur Expeditions were found to pertain to Velociraptor, but not to the type species V. mongoliensis. Pascal Godefroit and colleagues named these bones V. osmolskae (for Polish paleontologist Halszka Osmólska) in 2008.[13] However, the 2013 study noted that while "the elongate shape of the maxilla in V. osmolskae is similar to that of V. mongoliensis," phylogenetic analysis found it to be closer to Linheraptor, making the genus paraphyletic; thus, V. osmolskae might not actually belong to the genus Velociraptor and requires reassessment.[14]

Paleontologists Mark A. Norell and Peter J. Makovicky in 1997 described new and abundantly preserved specimens of V. mongoliensis, namely MPC-D 100/985 collected from the Tugrik Shireh locality in 1993, and MPC-D 100/986 collected in 1993 from the Chimney Buttes locality. The team briefly mentioned another specimen, MPC-D 100/982, which by the time of this publication remained undescribed.[10] In 1999 Norell and Makovicky provided more insights into the anatomy of Velociraptor with additional specimens. Among these, MPC-D 100/982 was partially described and figured, and referred to V. mongoliensis mainly based on cranial similarities with the holotype skull, although they stated that differences were present between the pelvic region of this specimen and other Velociraptor specimens. This relatively well-preserved specimen including the skull was discovered and collected in 1995 at the Bayn Dzak locality (more especifically at the "Volcano" sub-locality).[9] Martin Kundrát in a 2004 abstract compared the neurocranium of MPC-D 100/982 to another Velociraptor specimen, MPC-D 100/976. He concluded that the overall morphology of the former was more derived (advanced) than the latter, suggesting that they could represent distinct taxa.[15]

Velociraptor specimen MPC-D 100/982, possibly a new species

Mark J. Powers in his 2020 master thesis fully described MPC-D 100/982, which he concluded to represent a new and third species of Velociraptor. This species, which he named "V. vadarostrum", was stated to mainly differ from other Velociraptor species in having a shallow maxilla morphology.[16] Powers and colleagues also in 2020 used morphometric analyses to compare several dromaeosaurid maxillae, and found the maxilla of MPC-D 100/982 to strongly differ from specimens referred to Velociraptor. They indicated that this specimen, based on these results, represents a different species.[17] In 2021 Powers with team used Principal Component Analysis to separate dromaeosaurid maxillae, most notably finding that MPC-D 100/982 falls outside the instraspecific variability of V. mongoliensis, arguing for a distinct species. They considered that both V. mongoliensis and this new species were ecologically separated based on their skull anatomy.[18] The team in another 2021 abstract reinforced again the species-level separation, noting that additional differences can be found in the hindlimbs.[19]

V. mongoliensis holotype skull (right), rostrum (left), and dentary (bottom)

The skull of Velociraptor was rather elongated and grew up to 23 cm (9.1 in) long. It was uniquely up-curved at the snout region, concave on the upper surface, and convex on the lower surface. The snout, which occupied about 60% of the entire skull length, was notably narrow and mainly formed by the nasal, premaxilla, and maxilla bones. The premaxilla was the anteriormost bone in the skull, and it was longer than taller. While its posterior end joined the nasal, the main body of the premaxilla touched the maxilla. The maxilla was nearly triangular in shape and the largest element of the snout. On its center or main body, there was a depression developing a small oval to circular-shaped hole, called maxillary fenestra. Though in front of this fenestra were two small openings, referred to as promaxillary fenestrae. The posterior border of the maxilla formed (predominantly) the antorbital fenestra, one of the several large holes in the skull. Both premaxilla and maxilla had several alveoli (tooth sockets) on their bottom surfaces. Above the maxilla and making contact with the premaxilla, there was the nasal bone. It was a thin/narrow and elongated bone contributing to the top surface of the snout. Together, both premaxilla and nasal bones gave form to the naris or narial fenestra (nostril opening), which was relatively large and circular. The posterior end of the nasal was joined by the frontal and lacrimal bones.[27][28]

Skull of MPC-D 100/25 (Fighting Dinosaurs individual), in lateral (A-C), top (B), bottom (D-E), and posterior (E) views

The back or anterior region of the skull was built by the frontal, lacrimal, postorbital, jugal, parietal, quadrate, and quadratojugal bones. The frontal was large element, having a vaguely rectangular shape when seen from above. On its posterior end, this bone was in contact with the parietal, and such elements were the main bodies of the skull roof. The lacrimal was a T-shaped bone and its main body was thin and delicated. Its lower end meet the jugal (often called cheek bone), which was a large, sub-triangular-shaped element. Its lower border was notably straight/horizontal. The postorbital was located just above the jugal: a stocky and strongly T-shaped bone. As a whole, the orbit or orbital fenestra (eye socket)—formed by the lacrimal, jugal, frontal, and postorbital—was large and near circular in shape, being longer than taller. When seen from above, a pair of large and markedly rounded holes were present near the rear of the skull (the temporal fenestrae), whose main components were the postorbital and squamosal. Behind the jugal, an inverted T-shaped bone (also seen in other dromaeosaurids), known as the quadratojugal, was developed. While the upper end of the quadratojugal joined the squamosal, an irregularly-shaped element, its inner side meet the quadrate. The latter was of great importance for the articulation with the lower jaw. The posteriormost bone was the occipital bone and its projection the occipital condyle: a rounded and bulbous protuberance that meet the first vertebra of the neck.[27][28]

Isolated tooth of ZPAL MgD-I/97a

The lower jaw of Velociraptor comprised mainly the dentary, splenial, angular, surangular, and articular bones. The dentary was a very long, weakly curved, and narrow element that developed several alveoli on its top surface. On its posterior end, it meet the surangular. It had a small hole near its posterior end, called surangular foramen or fenestra. Both bones were the largest elements of the lower jaw of Velociraptor, contributing to virtually its entire length. Below them were the smaller splenial and angular, closely articulated to each other. The articular, located on the inner side of the surangular, was a small element that joined the quadrate of the upper skull, enabling the articulation with the lower jaw. An elongated, near oval-shaped hole was developed in the center of the lower jaw (the mandibular fenestra), and it was produced by the joint of the dentary, surangular, and angular bones.[27][28]

The teeth of Velociraptor were fairly homodont (equal in shape) and had several denticles (serrations), each more strongly serrated on the back edge than the front. The premaxilla had 4 alveoli (meaning that 4 teeth were developed), and the maxilla had 11 alveoli. At the dentary, between 14-15 alveoli were present. All teeth present at the premaxilla were poorly curved, and the two first teeth were the longest, with the second having a characteristic large size. The maxillary teeth were more slender, recurved, and most notably, the lower end was strongly more serrated than the upper one.[27][28]

Line diagram of the pes of V. mongoliensis (MPC-D 100/985)

The arm of Velociraptor was formed by the humerus (upper arm bone), radius and ulna (forearm bones), and manus (hand). Velociraptor, like other dromaeosaurids, had a large manus with three elongated digits (fingers), which ended up in strongly curved unguals (claw bones) that were similar in construction and flexibility to the wing bones of modern birds. The second digit was the longest of the three digits present, while the first was shortest. The structure of the carpal (wrist) bones prevented pronation of the wrist and forced the manus to be held with the palmar surface facing inward (medially), not downward. The pes (foot) anatomy of Velociraptor consisted of the metatarsus—a large element composed of three metatarsals of which the first one was extremely reduced in size—and four digits that developed large unguals. The first digit, as in other theropods, was a small dewclaw. The second digit, for which Velociraptor is most famous, was highly modified and held retracted off the ground, which caused Velociraptor and other dromaeosaurids to walk on only their third and fourth digits. It bore a relatively large, sickle-shaped claw, typical of dromaeosaurid and troodontid dinosaurs. This enlarged claw, which could grow to over 6.5 cm (2.6 in) long around its outer edge, was most likely a predatory device used to restrain struggling prey.[10][9]

Skeletal reconstruction of V. mongoliensis (MPC-D 100/25)

As in other dromaeosaurs, Velociraptor tails had prezygapophyses (long bony projections) on the upper surfaces of the vertebrae, as well as ossified tendons underneath. The prezygapophyses began on the tenth tail (caudal) vertebra and extended forward to brace four to ten additional vertebrae, depending on position in the tail. These were once thought to fully stiffen the tail, forcing the entire tail to act as a single rod-like unit. However, at least one specimen has preserved a series of intact tail vertebrae curved sideways into an S-shape, suggesting that there was considerably more horizontal flexibility than once thought.[10][9][29]

Life restoration of Velociraptor, showing large wing feathers as evidenced by the discovery of quill knobs

In 2007 Alan H. Turner and colleagues reported the presence of six quill knobs in the ulna of a referred Velociraptor specimen (IGM 100/981) from the Ukhaa Tolgod locality of the Djadochta Formation. Turner and colleagues interpreted the presence of feathers on Velociraptor as evidence against the idea that the larger, flightless maniraptorans lost their feathers secondarily due to larger body size. Furthermore, they noted that quill knobs are almost never found in flightless bird species today, and that their presence in Velociraptor (presumed to have been flightless due to its relatively large size and short forelimbs) is evidence that the ancestors of dromaeosaurids could fly, making Velociraptor and other large members of this family secondarily flightless, though it is possible the large wing feathers inferred in the ancestors of Velociraptor had a purpose other than flight. The feathers of the flightless Velociraptor may have been used for display, for covering their nests while brooding, or for added speed and thrust when running up inclined slopes.[23] Because of the presence of another dromaeosaurid in Ukhaa Tolgod, Tsaagan, Napoli and team have noted that the referral of this specimen to Velociraptor is currently subject to rexamination.[24]

Examinations of the endocranium of Velociraptor indicate that it was able to detect and hear a wide range of sound frequencies (2,368–3,965 Hz) and could track prey with ease as a result. The endocranium examinations also further cemented the theory that the dromaeosaur was an agile, swift predator. Fossil evidence suggesting Velociraptor scavenged also indicates that it was an opportunistic and actively predatory animal, feeding on carrion during times of drought or famine, if in poor health, or depending on the animal's age.[38]

In 2020, Powers and colleagues re-examined the maxillae of several eudromaeosaur taxa concluding that most Asian and North American eudromaeosaurs were separated by snout morphology and ecological strategies. They found the maxilla to be a reliable reference when inferring the shape of the premaxilla and overall snout. For instance, most Asian species have elongated snouts based on the maxilla (namely velociraptorines), indicating a selective feeding in Velociraptor and relatives, such as picking up small, fast prey. In contrast, most North American eudromaeosaurs, mostly dromaeosaurines, feature a robust and deep maxillar morphology. However, the large dromaeosurine Achillobator is a unique exception to Asian taxa with its deep maxilla.[39]

The "Fighting Dinosaurs" specimen of V. mongoliensis and Protoceratops andrewsi and restoration of same

The "Fighting Dinosaurs" specimen, found in 1971, preserves a Velociraptor mongoliensis and Protoceratops andrewsi in combat and provides direct evidence of predatory behavior. When originally reported, it was hypothesized that the two animals drowned.[6] However, as the animals were preserved in ancient sand dune deposits, it is now thought that the animals were buried in sand, either from a collapsing dune or in a sandstorm. Burial must have been extremely fast, judging from the lifelike poses in which the animals were preserved. Parts of the Protoceratops are missing, which has been seen as evidence of scavenging by other animals.[40] Comparisons between the scleral rings of Velociraptor, Protoceratops, and modern birds and reptiles indicates that Velociraptor may have been nocturnal, while Protoceratops may have been cathemeral, active throughout the day during short intervals, suggesting that the fight may have occurred at twilight or during low-light conditions.[41]

Size comparison of the Fighting Dinosaurs

The distinctive claw, on the second digit of dromaeosaurids, has traditionally been depicted as a slashing weapon; its assumed use being to cut and disembowel prey.[42] In the "Fighting Dinosaurs" specimen, the Velociraptor lies underneath, with one of its sickle claws apparently embedded in the throat of its prey, while the beak of Protoceratops is clamped down upon the right forelimb of its attacker. This suggests Velociraptor may have used its sickle claw to pierce vital organs of the throat, such as the jugular vein, carotid artery, or trachea (windpipe), rather than slashing the abdomen. The inside edge of the claw was rounded and not unusually sharp, which may have precluded any sort of cutting or slashing action, although only the bony core of the claw is preserved. The thick abdominal wall of skin and muscle of large prey species would have been difficult to slash without a specialized cutting surface.[40] The slashing hypothesis was tested during a 2005 BBC documentary, The Truth About Killer Dinosaurs. The producers of the program created an artificial Velociraptor leg with a sickle claw and used a pork belly to simulate the dinosaur's prey. Though the sickle claw did penetrate the abdominal wall, it was unable to tear it open, indicating that the claw was not used to disembowel prey.[43]

Remains of Deinonychus, a closely related dromaeosaurid, have commonly been found in aggregations of several individuals. Deinonychus has also been found in association with the large ornithopod Tenontosaurus, which has been cited as evidence of cooperative (pack) hunting.[44][45] However, the only solid evidence for social behavior of any kind among dromaeosaurids comes from a Chinese trackway which shows six individuals of a large species moving as a group.[46] Although many isolated fossils of Velociraptor have been found in Mongolia, none were closely associated with other individuals.[35] Therefore, while Velociraptor is commonly depicted as a pack hunter, as in Jurassic Park, there is only limited fossil evidence to support this theory for dromaeosaurids in general and none specific to Velociraptor itself. Dromeosaur footprints in China suggest that a few other raptor genera may have hunted in packs, but there have been no conclusive examples of pack behavior found.[47][48]

V. mongoliensis restraining an oviraptorosaur with its sickle claws

In 2011, Denver Fowler and colleagues suggested a new method by which dromaeosaurs like Velociraptor and similar dromaeosaurs may have captured and restrained prey. This model, known as the "raptor prey restraint" (RPR) model of predation, proposes that dromaeosaurs killed their prey in a manner very similar to extant accipitrid birds of prey: by leaping onto their quarry, pinning it under their body weight, and gripping it tightly with the large, sickle-shaped claws. These researchers proposed that, like accipitrids, the dromaeosaur would then begin to feed on the animal while it was still alive, and prey death would eventually result from blood loss and organ failure. This proposal is based primarily on comparisons between the morphology and proportions of the feet and legs of dromaeosaurs to several groups of extant birds of prey with known predatory behaviors. Fowler found that the feet and legs of dromaeosaurs most closely resemble those of eagles and hawks, especially in terms of having an enlarged second claw and a similar range of grasping motion. The short metatarsus and foot strength, however, would have been more similar to that of owls. The RPR method of predation would be consistent with other aspects of Velociraptor's anatomy, such as their unusual jaw and arm morphology. The arms, which could exert a lot of force but were likely covered in long feathers, may have been used as flapping stabilizers for balance while atop a struggling prey animal, along with the stiff counterbalancing tail. The jaws, thought by Fowler and colleagues to be comparatively weak, would have been useful for row saw motion bites like the modern day Komodo dragon, which also has a weak bite, to finish off its prey if the kicks weren't powerful enough. These predatory adaptations working together may also have implications for the origin of flapping in paravians.[29]

In 2010, Hone and colleagues published a paper on their 2008 discovery of shed teeth of what they believed to be a Velociraptor near a tooth-marked jaw bone of what they believed to be a Protoceratops in the Bayan Mandahu Formation. The authors concluded that the find represented "late-stage carcass consumption by Velociraptor" as the predator would have eaten other parts of a freshly killed Protoceratops before biting in the jaw area.[49] The evidence was seen as supporting the inference from the "Fighting Dinosaurs" fossil that Protoceratops was part of the diet of Velociraptor.[50] In 2012, Hone and colleagues published a paper that described a Velociraptor specimen with a long bone of an azhdarchid pterosaur in its gut. This was interpreted as showing scavenging behaviour.[51]

Velociraptor was warm-blooded to some degree, as it required a significant amount of energy to hunt. Modern animals that possess feathery or furry coats, like Velociraptor did, tend to be warm-blooded, since these coverings function as insulation. However, bone growth rates in dromaeosaurids and some early birds suggest a more moderate metabolism, compared with most modern warm-blooded mammals and birds. The kiwi is similar to dromaeosaurids in anatomy, feather type, bone structure and even the narrow anatomy of the nasal passages (usually a key indicator of metabolism). The kiwi is a highly active, if specialized, flightless bird, with a stable body temperature and a fairly low resting metabolic rate, making it a good model for the metabolism of primitive birds and dromaeosaurids.[34]

Norell with colleagues in 1995 reported one V. mongoliensis skull bearing two parallel rows of small punctures on its frontal bones that, upon closer examination, match the spacing and size of Velociraptor teeth. They suggested that the wound was likely inflicted by another Velociraptor during a fight between the species. Because its bone structure shows no sign of healing near the bite wounds and the overall specimen was not scavenged, this individual was likely killed by this fatal wound.[52] In 2001 Molnar and team noted that this specimen is MPC-D 100/976 hailing from the Tugrik Shireh locality, which has also yielded the Fighting Dinosaurs specimen.[53]

In 2012 David Hone and team reported another injured Velociraptor specimen (MPC-D 100/54, roughly a sub-adult individual) found with the bones of an azhdarchid pterosaur within its stomach cavity, was carrying or recovering from an injury sustained to one broken rib. From evidence on the pterosaur bones, which were devoid of pitting or deformations from digestion, the Velociraptor died shortly after, possibly from the earlier injury. Nevertheless, the team noted that this broken ribs shows signs of bone healing.[51]

Restoration of related Linheraptor in paleoenvironment

In both Bayan Mandahu and Djadochta formations many of the same genera were present, though they varied at the species level. These differences in species composition may be due a natural barrier separating the two formations, which are relatively close to each other geographically.[13] However, given the lack of any known barrier which would cause the specific faunal compositions found in these areas, it is more likely that those differences indicate a slight time difference.[54]

V. osmolskae lived alongside the ankylosaurid Pinacosaurus mephistocephalus; alvarezsaurid Linhenykus; closely related dromaeosaurid Linheraptor; oviraptorids Machairasaurus and Wulatelong; protoceratopsids Bagaceratops and Protoceratops hellenikorhinus; and troodontids Linhevenator, Papiliovenator, and Philovenator.[54] Sediments across the formation indicate a similar depositional environment to that of the Djadochta Formation.[55]

Restoration of V. mongoliensis in the arid Djadochta settings

Known specimens of Velociraptor mongoliensis have been recovered from the Djadochta Formation (also spelled Djadokhta), in the Mongolian province of Ömnögovi. This geological formation is estimated to date back to the Campanian stage (between 75 million and 71 million years ago) of the Late Cretaceous epoch.[56] The abundant sediments—sands, sandstones, or caliche—of the Djadochta Formation were deposited by eolian (wind) processes in arid settings with fields of sand dunes and only intermittent streams, as indicated by very sparse fluvial (river-deposited) sedimentation, under a semi-arid climate.[57][56][58]

The Djadochta Formation is separated into a lower Bayn Dzak Member and upper Turgrugyin Member. V. mongoliensis is known from both members, represented by numerous specimens.[56] The Bayn Dzak Member (mainly Bayn Dzak locality) has yielded the oviraptorid Oviraptor; ankylosaurid Pinacosaurus grangeri; protoceratopsid Protoceratops andrewsi; and troodontid Saurornithoides.[2][56] The younger Turgrugyin Member (mainly Tugriken Shireh locality) has produced the bird Elsornis; dromaeosaurid Mahakala: ornithomimid Aepyornithomimus; and protoceratopsid Protoceratops andrewsi.[59][60]

V. mongoliensis has been found at many of the most famous and prolific Djadochta localities. The type specimen was discovered at the Flaming Cliffs site (sublocality of the larger Bayn Dzak locality/region),[2] while the "Fighting Dinosaurs" were found at the Tugrik Shire locality (also known as Tugrugeen Shireh and many other spellings).[5] The latter is notorious for its exceptional in situ fossil preservation. Based on deposits (such as structureless sandstones), it has been concluded that a large number of specimens were buried alive during powerful sand-bearing events, common to the these paleoenvironments.[61]