Mourasuchus

Description

Mourasuchus is well known for its strange anatomy, its skull exceptionally dorsoventrally flattened, broad, and overall platyrostral in shape, differing greatly from any other caiman currently known, most closely resembling the enigmatic crocodyliform Stomatosuchus from the Late Cretaceous Bahariya Formation. The nares are elevated and so is the high postrostral cranium and short skull table. The orbital margins are likewise raised above the rostrum with thickened anteromedial margins formed by a knob-like swelling of the frontal and prefrontal bones. Overall, the orbits are smaller than the infratemporal fenestrae. The teeth were generally small and conical, numbering upwards of 40 teeth on each side of the upper and lower jaw and interlocking perfectly. The posterior teeth of the dentary and maxilla are laterally compressed. The premaxilla of Mourasuchus show noticeably large occlusal pits, especially noticeable in M. arendsi. However, such pits can be observed in a variety of extant and fossil taxa and are not considered to be diagnostic between species of Mourasuchus.

The mandible is U-shaped with a sharp transition between the first five teeth and the rest of the dentary teeth. The first to fifth tooth are located on the curved anterior portion of the mandible, while the bone is straight from the sixth onward. The mandibular symphysis is short and slender, only extending to the first posterior alveoli, meaning the animal most likely had a relatively weak bite. The osteoderms of Mourasuchus show conspicuous spines on their dorsal surface.

The humerus of Mourasuchus is long and slender and interpreted to possibly indicate weakened forelimbs, potentially corresponding with a predominantly aquatic lifestyle.

Size

Due to the fragmentary remains of Mourasuchus the body size is difficult to determine. Mourasuchus skulls range up to a meter (3.3 ft) in length, with the holotype skull of M. pattersoni being 108.1 centimetres (42.6 in) long while that of M. arendsi could reach a length of 108.5 centimetres (42.7 in) long. The largest skull belongs to M. amazonensis at 113.5 centimetres (44.7 in). In a 2020 study, Cidade et al. attempted to determine the body sizes of the four recognized Mourasuchus species based on the head: body ratio of modern genera like Caiman latirostris and Alligator mississippiensis. Their results recovered a mean body size of 6.3 m (21 ft) and 1.2 metric tons (1.3 short tons) for M. atopus, the smallest species, and upwards of 9.5 m (31 ft) and 4.4 metric tons (4.9 short tons) for M. amazonensis. However, a later study contradicted these results. In 2022 Paiva and colleagues argued that the dorsal cranial length was a poor basis for size estimates compared to the width of the skull. Additionally, they found that prior studies commonly included juvenile animals in their data, muddying the results. This study calculated a length between 2.5–3.1 m (8 ft 2 in – 10 ft 2 in) for Mourasuchus atopus, 3.8–4.8 m (12–16 ft) for Mourasuchus arendsi, 4.3–5.5 m (14–18 ft) for Mourasuchus pattersoni and for Mourasuchus amazoniensis, the largest species, 4.7–5.98 m (15.4–19.6 ft). These results were achieved by specifically restricting the dataset to extant species of caimans, while calculations using measurements from all of Crocodilia generally rendered greater sizes but may be overestimates. Still, the authors note that the generally smaller stature of modern caimans may have influenced the results in a similar fashion.

Phylogeny

Although Mourasuchus has been firmly placed within Caimaninae by authors, the exact relationship it has with other crocodilians of this family and even the relationship between the different species of the genus had long been unclear. Some studies have proposed that Mourasuchus was closely related to the Eocene Orthogenysuchus from North America, and more distantly to the giant caiman Purussaurus which it shared its habitat with. However, more recent papers indicate that ongoing preparation conducted on Orthogenysuchus significantly influences the scoring of this taxon's characters, leading to some authors removing the animal from analysis until further publications. The cladogram below shows the phylogenetic tree used by Bona et al. (2012).

The following tree is based on the results recovered by Cidade et al. (2017), excluding the North American Orthogenysuchus and including the then newly named Mourasuchus pattersoni while also following the synonymy of M. nativus with M. arendsi. Like Bona et al. before them, the authors recover Mourasuchus and Purussaurus as closely related clades, this time as sister genera with Centenariosuchus just outside this shared clade.

Paleobiology

Diet

Much like with the unrelated but morphologically similar Stomatosuchidae, the feeding ecology of Mourasuchus is enigmatic and poorly understood, with a variety of hypotheses having been proposed to make sense of its specialized anatomy. With slender and short mandibular rami and perfectly occluding, slender and almost homodont teeth, Mourasuchus was not built to capture and hold prey like modern crocodilians. In his 1965 publication on the crocodilians of Cenozoic Columbia, Langston proposed three ideas on how Mourasuchus may have fed. He suggests that it may have waited with opened jaws for approaching prey, fed on aquatic and floating plant material or searched for prey by sifting through substrate at the bottom of bodies of water. While his other hypothesis have gone without much attention, the later of the three has been discussed by later publications, referring to this ecology as either "filter feeding", "straining technique" or "gulp feeding". Cidade et al. follow Langston's proposed feeding hypothesis, speculating that Mourasuchus may have fed on small invertebrates, mollusks, crustaceans, and small fish, specializing in trying to ingest as much prey at once as possible. This they suggest explains the platyrostral-broad skull morphology, which differs from the usual crocodilian skull morphology that typically either favors longirostrine forms providing speediness or robust-snouted forms with increased bite force. Langston also proposed a throat pouch based on comments Franz Nopsca made regarding Stomatosuchus, however there is no evidence for such a structure in Mourasuchus nor is its presence certain in stomatosuchids. As the prey Mourasuchus would have fed on in accordance with the straining technique hypothesis would likely be found either in free water or substrate, the animal would have also captured inedible material during feeding. For this a type of filtering mechanism or behavior would be beneficial, however nothing indicating as much has been found so far. Due to the absence of evidence for actual filtering, Cidade et al. instead favors the term "gulp feeding" for Mourasuchus hypothetical hunting behavior.

In a 2020 study Cidade et al. aimed to determine Mourasuchus ability to perform the death roll, a crucial behavior observed in modern crocodiles and alligators that is commonly observed during feeding, when crocodiles rapidly spin to rip pieces of flesh from their prey. The study calculated a very low "death roll" capability indicator (DRCI) similar to that seen in the Slender Snouted Crocodile, Indian Gharial and False Gharial, which suggests that Mourasuchus was very unlikely to be able to perform the death roll.

Paleoenvironment

Mourasuchus lived during an era of great crocodilian diversity in South America, being found in environments shared with genera inhabiting a diverse set of niches. In the Peruvian Pebas Formation M. atopus was found alongside the large bodied Purussaurus neivensis, a medium-sized species of Gryposuchus, G. pachakamue and three distinct taxa of durophagous caimans. The later species M. arendsi coexisted with an even greater host of contemporary crocodilians in the Urumaco Formation, sharing its habitat with the larger Purussaurus mirandai, several large-bodied gharials, multiple durophagous caimans and the true crocodile Charactosuchus. In some areas such as the Urumaco and Solimões Formation multiple species of Mourasuchus are known, with the former being home to both M. pattersoni and M. arendsi while the later has yielded M. arendsi and M. amazonensis. The great diversity of crocodylomorphs in these Miocene-age (Tortonian stage, 8 million years ago) wetlands suggests that niche partitioning was efficient, which would have limited interspecific competition.