| Acerosodontosaurus | |
|---|---|
 | |
| Skull diagram | |
Scientific classification | |
| Kingdom: | Animalia |
| Phylum: | Chordata |
| Class: | Reptilia |
| Family: | †Tangasauridae |
| Genus: | †Acerosodontosaurus Currie 1980 |
| Type species | |
| †Acerosodontosaurus piveteaui | |
Acerosodontosaurus is anextinctgenus ofneodiapsidreptiles that lived during theLate Permian ofMadagascar.[1][2] The only species ofAcerosodontosaurus,A. piveteaui, is known from a naturalmold of a single partial skeleton including a crushed skull and part of the body and limbs. The fossil was discovered in deposits of theLower Sakamena Formation. Based on skeletal characteristics, it has been suggested thatAcerosodontosaurus individuals were at least partiallyaquatic.[2]
Acerosodontosaurus has generally been considered a "younginiform", part of aparaphyleticgrade of Permiandiapsids which linked the most basal ("primitive") diapsids (araeoscelidians such asPetrolacosaurus) to more derived ("advanced") diapsids, including the earliest ancestors of modern reptiles such ascrocodilians andlizards (Sauria).[2] However, its position within the grade is controversial. Initially considered a specimen of the contemporaneousTangasaurus,Acerosodontosaurus was later described as ayounginid in 1980[1] and atangasaurid in 2009.[2] More recent studies have even supported the hypothesis that it was not a "younginiform" at all, but rather that it was an early member ofPantestudines, the reptile lineage that would lead to modernturtles.[3][4] However, this theory is still controversial,[5] and other studies have continued to support the "younginiform" interpretation.[6]
Acerosodontosaurus is known from a single specimen stored at theMuseum National d'Histoire Naturalle inParis,France. This specimen, MNHN 1908-32-57, was referred toTangasaurus and stayed undescribed untilPhillip J. Currie recognized it as a new genus. He described it asAcerosodontosaurus piveteaui in 1980, a name which roughly translates to "Piveteau's needle-toothed lizard". The specific name referencesJean Piveteau, a paleontologist who pioneered the study of Permian reptiles of Madagascar.[1]
The specimen was extracted from an unknown locality in the Lower Sakamena Formation, an Upper Permian geological formation inMadagascar. The Sakamena Formation was formed by coastalrift valleys, and can be characterized by a diverse assortment of aquatic, semiaquatic, and terrestrial animals, including many early diapsids. TheAcerosodontosaurus specimen is a partial skeleton preserved as acompression fossil within asandstonenodule. The skeleton is mostly but not entirely articulated, preserving most of the trunk, right forelimb, and small portions of the left forelimb and hindlimbs. Bones on the left side of the skull were displaced as the animal decayed, but the majority of the right side was able to fossilize.[1]
Most of the actual fossilized bones had been eroded away once the specimen was found, although well-preservedmolds were left behind in the sandstone. Currie's description was based onlatex casts made from the molds.[1] During a 2009 redescription by Constanze Bickelmann, Johannes Mueller, &Robert R. Reisz, new high-fidelity latex molds were created using improved techniques.[2]
The preserved skeleton was about 30 centimeters (12 inches) in length from the tip of the snout to the base of the tail. The tail is unknown, but Currie (1980) estimated that the total length of the animal from snout to tail tip would have been 60 to 70 centimeters (24 to 28 inches). The body was wide and the limbs were of medium length. The overall impression was that of a medium-sizedlizard, althoughAcerosodontosaurus is completely unrelated to true modern lizards.[1] Although the tail is unknown, it could have had a fin-like structure similar to that ofHovasaurus, which has been considered a close relative ofAcerosodontosaurus.[2]
The tip and upper side of the snout were not preserved, but most of the other skull bones are known. Theorbits (eye holes) were large, and the skull was longer in front of the eyes than behind them. The lower edge of each orbit was formed by a long forward branch of thejugal bone, which overlies an equally elongated rear branch of the tooth-bearingmaxillary bone. The skull was somewhat broader than that of most other "younginiforms", as indicated by the width and curvature of thefrontal andprefrontal bones. Nevertheless, it was proportionally similar to that ofHovasaurus.[7] The skull had an estimated total length of 55 millimeters (2.2 inches).[1]
Themandible (lower jaw) was long and slender. The rear part of the mandible, which was formed by thearticular bone, bears a facet which connects to thequadrate bone of the cranium to form the jaw joint. Both the quadrate and the corresponding facet on the mandible are very large and strongly built.[2]
The numerous teeth ofAcerosodontosaurus are conical, sharply pointed, and somewhat recurved. They are slightly longer towards the front of the skull, but otherwise are similar in size and shape throughout the skull and jaw, in contrast to the condition in earlier diapsids likePetrolacosaurus. Currie (1980) estimated that 37 teeth were present in the maxilla and 32 were in the preserved portion of the mandible, based on both preserved teeth and empty tooth sockets.[1] Under the same criteria but using different latex casts, Bickelmann, Mueller, & Reisz (2009) estimated 36 and 34, respectively.[2] The mandible would have had many more teeth than these estimations, since the front half of the jaw was not preserved.Acerosodontosaurus translates to "Needle-tooth lizard", referring to its high number of needle-like teeth which differ from those of other "younginiforms" in both shape and abundance.[1]
As a diapsid,Acerosodontosaurus had two holes known as temporal fenestrae on each side of the skull, behind the eyes. However, thelower temporal fenestra was not completely enclosed from below. In most early diapsids, such asPetrolacosaurus andYoungina, the lower edge of the lower temporal fenestra is formed by a bony bar composed of the rear branch of the jugal and the forward branch of thequadratojugal bone. YetAcerosodontosaurus has a short rear branch of the jugal, and is completely missing a quadratojugal, rendering the lower temporal fenestra incompletely enclosed. While many later diapsids also have an incomplete lower temporal fenestra, the only other "younginiform" with such a trait isHovasaurus.[2]
At least 21vertebrae were present, all from the neck, back, and hip; the tail is missing. Thecervical (neck) vertebrae had shorter centra (main components) than the dorsal (back) vertebrae. A small pit (subcentral foramen) was present on both the left and right sides of each vertebral centrum. The neural spines (spinous processes) of the dorsal vertebrae are characteristically tall and rectangular compared to most "younginiforms", though not quite as tall as those ofHovasaurus. The centra are at their maximum length and the neural spines are at their maximum height at the vertebrae directly in front of the hip. The transverse processes (rib facets) were quite short. Although many ribs are incomplete, those that were mostly complete were long and curved, indicating thatAcerosodontosaurus was a rather wide-bodied animal. As with most early diapsids, small wedge-like bones known as intercentra fill in the gaps between vertebrae. Numerous small, slendergastralia (belly ribs) were also present.[1][2]
The forelimbs ofAcerosodontosaurus are among the most diagnostic parts of the body. The only preserved portion of the pectoral girdle is a thin, slightly curved bone with tapered tips. This is likely acleithrum, a bone which is absent in practically all diapsids. The only other neodiapsid known with certainty to retain a cleithrum isHovasaurus.[2]
Thehumerus (forearm bone) is incomplete, but theentepicondyle and its correspondingforamen are rather well-developed. Thecapitulum andtrochlea, on the other hand, are poorly differentiated.[2] Theradius is twisted and curved, unusually similar to that of thechoristodereChampsosaurus. Theulna is hourglass-shaped, straighter and characteristically slightly longer than the radius. Unlike in terrestrial diapsids, theolecranon process of the ulna was absent. Both the radius and ulna were flatter near the wrist and more blocky near the elbow.[1]
As with other early diapsids,Acerosodontosaurus had 11carpal (wrist) bones. All of the carpals are well-separated. Most are similar in size and shape to those ofHovasaurus, with the exception of the medial centrale, a bone between theradiale anddistal carpal I which is much smaller inAcerosodontosaurus. Thefirst metacarpal (the hand bone which connects to the thumb) is small and stout, while thesecond,third, andfourth (which connect to the index, middle, and ring fingers) are incrementally longer. This contrasts with most other "younginiforms", in which the fourth metacarpal is significantly longer than the third.Hovasaurus is an exception, but evenAcerosodontosaurus differs from that genus in that its metacarpals are not as short. Thefifth metacarpal (which connects to the little finger) is as short as the first, but has a more "pinched" appearance. Somephalanges (finger bones) are obscured by the overlain skull, but those that are visible are short, with sharp and curved claws.[1][2]
Theilium (upper blade of the hip) is fairly similar to that ofHovasaurus and other reptiles, though seemingly slightly thicker. Thepubis (forward blade of the hip) is a heavily built bone, strongly curving inwards. As a result, the hips would have been quite wide in life. The outer surface of each pubis has a large, diagonally-oriented tubercule (knob- or ridge-like growth). The plate-likeischium (rear-blade of the hip) is incomplete and was initially overlooked. Thefemur (thigh bone) is long, curved, and robust, while thetibia andfibula (shin bones) are only preserved near the knee so their form is difficult to determine. No foot bones are preserved.[1][2]
Initially considered a specimen ofTangasaurus, Currie (1980) subsequently assignedAcerosodontosaurus to the familyYounginidae.[1] The next year, his study ofHovasaurus noted thatAcerosodontosaurus shared similarities with bothtangasaurids and younginids, and he considered it ancestral to both families.[7] Other studies found a similar result, placing it as the most basal member of "Younginiformes".[8] The 2009 redescription of the genus partially deconstructed the concept of amonophyletic "Younginiformes" and assigned it to Tangasauridae.[2] The 2011 description of thebasalneodiapsidOrovenator considered tangasaurids to bestemward (further from modern reptiles) compared to younginids,[9] while a 2016 study ofarchosauromorphs placedAcerosodontosauruscrownward (closer to modern reptiles) compared toYoungina.[10]
Beveret al. (2015) analyzedAcerosodontosaurus alongside other neodiapsids via bothmaximum parsimony andBayesian analyses, and cranial-only vs full body data. Their full specimen parsimony analysis found that it was apantestudinelepidosauromorph related tosauropterygians,Eunotosaurus,Odontochelys, andPappochelys. This result argues thatAcerosodontosaurus was more closely related to turtles than to any other modern reptile, and that turtles were more closely related to lizards and kin (lepidosaurs) than to crocodilians and birds (archosaurs). Their cranial-only parsimony analysis matrix retainedAcerosodontosaurus's identity as a basal pantestudine but removed Pantestudines fromSauria (the lepidosauromorph +archosauromorph clade). The link betweenAcerosodontosaurus and turtles was more tenuous in theBayesian analyses, where it was placed as a neodiapsid closely related to Sauria but independent from Pantestudines.[3]
During their redescription ofPappochelys, Schoch & Sues (2018) revised the matrix of Beveret al. (2015) but failed to resolve the relations of most reptile lineages near the base of Sauria, includingAcerosodontosaurus.[11] Liet al. (2018) described the new pantestudineEorhynchochelys and revised Schoch & Sues (2018)'s data matrix further. They were able to find more stability, recoveringAcerosodontosaurus as the sister taxon ofClaudiosaurus and not closely related toHovasaurus.Acerosodontosaurus andClaudiosaurus were still indeterminate neodiapsids if all taxa were included, but they were found to be the earliestpantestudines under a specific set of parameters which excluded 21 poorly-described taxa.[4]
Several skeletal features support the hypothesis thatAcerosodontosaurus was aquatic. Theolecranon process of theulna was absent (or not developed into bone), a trait in common with the aquaticnothosaurs,ichthyosaurs, andthalattosaurs, but in contrast to terrestrial reptiles such asThadeosaurus. Moreover, theradius is twisted (similarly to that of the aquaticchampsosaurs), and the wrist has a "relaxed" structure, with thecarpals (wrist bones) separated by gaps rather than tightly connected.[2]
The Lower Sakamena Formation was deposited in a wetland environment situated within a North-South orientatedrift valley, perhaps similar toLake Tanganyika. The climate at the time of deposition was temperate, warm, and humid, with seasonal rainfall and possible monsoons.[12] Flora from the formation includes theequisetaleanSchizoneura, theglossopterid gymnospermGlossopteris, and seed fernLepidopteris. Other vertebrates known from the Lower Sakamena Formation include thepalaeoniscoid fishAtherstonia, theprocolophonid parareptileBarasaurus, the gliding reptileCoelurosauravus, the neodiapsidsHovasaurus,Claudiosaurus,Thadeosaurus fragments ofrhinesuchidtemnospondyls, an indeterminatetheriodonttherapsid and thedicynodontOudenodon.[13]
| |||||||||||||||||||||||
| |||||||||||||||||||||||
| |||||||||||||||||||||||
