In part 1 we looked at two competing hypotheses of pterosaur origins: the more popular and traditional archosaur hypothesis and the typically avoided and heretical fenestrasaur hypothesis.
In part 2 we took a closer look at the character traits used by Senter (2003) and Nesbitt (2011) to connect pterosaurs to archosaurs.
Here, in part 3, you’ll meet a series of taxa that document a gradual increase in pterosaurian traits: Lacertulus, Meyasaurus, Huehuecuetzpalli and the fenestrasaurs. An asterisk (*) marks the pterosaurian characters. Links will take you to reptileevolution.com for additional imagery and data. References follow the descriptions.
Click to enlarge. Squamates, tritosaurs and fenestrasaurs in the phylogenetic lineage preceding the origin of the Pterosauria.
Lacertulus bipes (Carroll and Thompson 1982) Late Permian, ~5 cm snout/vent length. Lacertulus was originally described as a facultatively bipedal* primitive lizard or eosuchian representing a distinct lineage of one or the other. Here Lacertulus was derived from a sister to Homoeosaurus and Gephyrosaurus and was a phylogenetic predecessor to a new clade of lizards, the Tritosauria (“third lizards”) that includes Meyasaurus and its descendants, including pterosaurs. The Lacertulus fossil is complete and articulated, but poorly preserved, especially anteriorly.
Overall, Lacertulus was about half the size of Homoeosaurus. The nasal was notched* above the naris slightly expanding the size of the opening. The maxilla formed the majority of the ventral margin of the naris.* The pterygoid extended anteriorly as far as the palatine.* The lateral process of the palatine was set further posteriorly.* Distinct from most lizards, the teeth were subthecodont in implantation.*
There were 24 presacral vertebrae, perhaps one fewer than in Gephyrosaurus. The neural arches were low and not fused to the centra. The neural spines were low to absent.
The forelimbs were much shorter than the hind limbs. The radius and ulna had little to no end expansion, so they were aligned closer to one another*. The olecranon was ossified in a pattern distinct from living lizards. The carpals were tiny but ossified and rounded, rather than locked in a tight mosaic as in many living lizards.
The pelvis was taller than long with a distinct anterior process on the ilium.* Such a process in living lizards capable of bipedal locomotion helps raise the front of the body off the ground (Snyder 1954). The pubis was narrower, producing a larger thyroid fenestra. The femur was longer than half the glenoid-acetabulum length.* The tibia was more robust than the fibula.* The astragalus, calcaneum and centrale remained unfused,* unlike most lizards.* The pes was twice the length of the manus. Metacarpals 1-4 were bound together. The foot was large with a tendril-like digit 4 ideal for arboreal (tree branch) locomotion. The phalanges of pedal digit 5 were likewise elongated.*
Meyasaurus faurai (Vidal 1915, Evans and Barbadillo 1996) Early Cretaceous, ~10 cm skull length. Despite being a large Late Cretaceous relic, Meyasaurus was derived from a sister to the Late Permian Lacertulus and phylogenetically preceded Huehuecuetzpalli.
Distinct from Lacertulus the skull of Meyasaurus had a more sharply angled rostrum* and a longer temporal area. The postorbital did not descend as far. The premaxillary dorsal process extended beyond the naris.* The parietal was larger. The pineal opening was a mere pinprick and located further forward*. The frontal-parietal suture was straight and much wider than the frontal-nasal suture.* The frontal was narrower between the orbits.* The teeth were tiny and the posterior teeth had twin cusps (bicuspid), a character sometimes retained in Gekko. Multiple cusps are seen in Sharovipteryx and Longisquama as well as basal pterosaurs.
The cervicals of Meyasaurus were more robust* with large paddle-shaped ribs. The scapulocoracoid was fenestrated anteriorly,* as in other lizards. The clavicle was more robust. The humerus was expanded distally. Metacarpal 4 was shorter than 3. The pelvis, tail and hind limbs are unknown in the adult, only in a referred juvenile.
Huehuecuetzpalli mixtecus (Reynoso 1998) Middle to Late Albian, Early Cretaceous ~110 mya, ~9.5 cm snout to vent length. Huehuecuetzpalli nested at the base of all lizards in the original analysis of Reynoso (1998). Here Huehuecuetzpalli was derived from a sister to Meyasaurus and was a phylogenetic predecessor to Cosesaurus, and Macrocnemus,
Distinct from Meyasaurus, the skull of Huehuecuetzpalli was narrower* (in dorsal view) with a posteriorly displaced and enlarged naris*. The premaxilla was extended anteriorly*. The premaxillary ascending process extended nearly to the orbit*. The nasal was further split by the posterodorsal expansion of the naris*. The pineal foramen was on the frontal/parietal suture*. The postfrontal was reduced*. It lost contact with the upper temporal fenestra, replaced with a lateral extension of the parietal and a medial extension of the postorbital*. Rather than a broad and solid palate (as in Lacertulus), the pterygoid was shorter, narrower and reduced to just the transverse processes and the quadrate processes*. The vomers and palatines were also reduced to struts*.
Transverse processes and chevrons were smaller.* Three sacrals were present.* The caudals were attenuate.* The chevrons paralleled the centra.*
An ossified sternum* covered the posterior tip of the T-shaped interclavicle. The scapula and coracoid were further fenestrated anteriorly,* as in most living lizards. The scapula was very short, but likely was made taller by a cartilage extension, as in many living lizards. The radius and ulna were expanded at their ends.* Metacarpals 3 and 4 were subequal.* The carpus was unossified, even in the adult. This had important ramifications in the lineage of a successor, Cosesaurus with regard to the appearance of the pteroid.
The ilium developed a long posterior process.* The fibula was less than half the width of the tibia.* The calcaneum was half the size of the astragalus. Metatarsal 5 was shorter and torsioned.* Metatarsals 3 and 4 were subequal. Pedal digit V was further elongated with p5.1 more than half the length of mt 4.*
The robust hind limb and three sacral vertebrae might suggest a facultative bipedal capability in the manner of similar living lizards and Lacertulus, despite the lack of an anterior ilial process.
Mammals, crocs, birds, turtles and other lizards all practice allometric growth in which changes occur with maturity. In contrast, a juvenile Huehuecuetzpalli is known and it is virtually identical in proportion to the adult (Reynoso 1998). More data is needed, but from what little is known, this clade experienced isometric growth.* Juveniles and hatchlings were virtually identical to adults. They lacked a short snout and large eyes.
(Nopcsa 1931) Ladinian, Middle Triassic ~220mya
The smallest specimen of Macrocnemus, BES SC 111 (Renesto and Avanzini 2002), was originally considered a juvenile, but it was morphologically distinct from the other larger specimens. Here it is the most primitive Macrocnemus now known. BES SC 111 was derived from a sister to Huehuecuetzpalli and it documents a further elongation and narrowing of the naris,* a shorter temporal region,* a longer neck of eight vertebrae,* an incipient strap-like scapula,* a broader ventral pelvis* and a more pterosaur-like palate,* but pedal phalanx 5.1 is reduced, which is an autapomorphy (a unique trait within this clade). None of the tail vertebrae show any sign of caudal autotomy,* which in several living lizards permits the tail to breakoff during an attack only to regrow later.
Langobardisaurus tonneloi was a sister to Macrocnemus that retained an elongated pedal 5.1 phalanx.
Cosesaurus aviceps (Ellenberger and DeVillalta 1974) Middle Triassic ~225 mya, ~16cm long was derived from a sister to Macrocnemus. Cosesaurus phylogenetically preceded Sharovipteryx and Longisquama.
Distinct from Macrocnemus, the rostrum of Cosesaurus had a more deeply concave dorsal profile, which further reduced the naris to a long slit. The rostrum was shorter. By contrast, the postorbital region was expanded to include a larger cranium.* The maxilla had three openings, together representing the origin of the single antorbital fenestra without a fossa.* The palatine had no lateral process.* The parasphenoid was relatively enlarged. The occiput was steeply angled as if the cervicals were typically held further beneath the skull.*
The cervicals were midway in length between those of Macronemus and Huehuecuetzpalli.* Four sacrals were present.*
The scapula was reduced to a slender strap-like shape oriented posterodorsally.* The coracoid was reduced to a stem,* resulting from further coracoid fenestration expanded posteriorly until only the posterior quadrant-shaped rim remained. The coracoid stem inserted into a socket* created by the anterior migration of the sternum* to the transverse processes of the interclavicle.
The sternum was broader* and anteriorly displaced* such that the anterior rim was coincident with the transverse processes of the interclavicle and clavicles. The clavicles were shorter and only transverse in orientation.* They did not extend along the anterior rims of the coracoids and scapula, as in most tetrapods including Macrocnemus. Instead the clavicles wrapped around the coincident interclavicle and sternum anterior rims creating the sternal complex* otherwise found in Longisquama and pterosaurs (Wild 1993). The interclavicle developed an anterior process making it cruciform.* The two centralia, best seen in Sphenodon, had migrated to the medial edge of the wrist where they assumed new identities as the pteroid* and preaxial carpal* (Peters 2009). The manus was much larger and longer than the forearm,* with metacarpals and digits less disparate in size* except digit V, which was reduced to a vestige.*
The pelvis was considerably smaller overall, but with a longer anterior process of the ilium* and a much smaller pubis and ischium. A new bone, the prepubis,* extended beyond the pubis increasing its effective length. The proximal tarsals were the same width*. Metatarsals I-IV were less disparate in length.* Pedal 5.1 was longer than the rest of the digit.*
Soft tissue impressions of a sagittal crest and plumes, plus extradermal membranes trailing the limbs (primitive patagium and uropatagium)* were preserved. “Hairs” emanated from the caudal vertebrae.*
In Cosesaurus the coracoids were socketed, as in birds and pterosaurs. If similar in function, Cosesaurus was flapping its forelimbs long before the advent of a wing-like morphology. This was likely a secondary sexual characteristic enabled by facultative bipedal locomotion.
Digitigrade, occasionally bipedal footprints attributed to the ichnogenus Rotodactylus (Peabody 1948) were matched to the pes of Cosesaurus (Peters 2000a). Digit 5 made a small circular impression far behind the other four toes*, of which digit 1 only impressed the ungual. The proximal phalanges were all held elevated,* because the metatarsophalangeal joint was a simple butt joint*, incapable of much movement. A flexed digit V* impressed its dorsal surface far behind the other four toes, a configuration unknown outside of the Fenestrasauria.
Sharovipteryx miribilis (Sharov 1971) Norian, Late Triassic, ~210 mya was originally considered a pseudosuchian, then a prolacertiform (Peters 2000). Here Sharovipteryx nests as a tritosaurid lizard derived from a sister to Cosesaurus. Sharovipteryx is a sister to Longisquama and pterosaurs.
Distinct from Cosesaurus, the skull of Sharovipteryx had an upturned premaxilla with procumbent teeth.* The naris was enlarged* and a smaller antorbital fenestra was present. The postorbital was higher relative to the orbit.* The rostrum was straighter.* The ventral mandible was more convex. The teeth were more varied in shape, with rear teeth having several cusps.* Long hyoids emerged from the base of the throat. These were moved laterally to create aerodynamic strakes from the extended neck skin.
The cervicals were elongated. The dorsals were shortened.* 24 presacrals were present, but three of them were located between the ilia.* Several more caudal vertebrae joined the sacral series.* This was probably in response to the increase in stress at the fulcrum of the obligate bipedal configuration. The caudals were lengthened.* The dorsal ribs extended horizontally creating a wide but shallow torso, yet another aerodynamic surface.
The coracoid was straighter. The entire forelimb was reduced. The humerus was short but robust with a large deltopectoral crest.* The ulna and radius were also robust, but shorter than the metacarpus. Digit IV was discovered by Sharov (1971) extending back to the pelvis*. The other shorter digits, long considered missing, follow the pattern of sister taxa.
The ilium was hyper-elongated, both anteriorly and posteriorly.* The puboischium was deeper than in Cosesaurus.* The prepubis was straighter. The hind limbs were extremely long* with a femur longer than the torso. The distal femur had a short anterior extension to prevent overextension of the tibia. The tibia was longer than the femur.* The metatarsals spread apart, unlike other fenestrasaurs. Digit V was further elongated such that pedal 5.1 was longer than metatarsal 4.
Various extradermal membranes surrounded Sharovipteryx. The neck skin was six times wider than the cervicals and able to be spread even wider by lateral extension of the hyoids. Fiber-supported uropatagia extended from the hind limbs,* from digit 5 to the base of the tail. Smaller membranes extended anterior to the femur and at the base of the tibia. Webbing was present between the toes.* Elongated fibers tipped the tail*, a likely precursor to the vane in pterosaurs. Fiber-embedded membranes were also preserved in the damaged areas around the forelimbs.
With its diminutive forelimbs, Sharovipteryx would seem to be dissimilar to pterosaurs, but the rest of its traits find no better match. With such long hind limbs, Sharovipteryx was unable to walk quadrupedally, but leaping was greatly improved. It is widely held that pterosaurs were ALL quadrupedal, but Cosesaurus and Sharovipteryx were facultative and obligate bipeds respectively. Hone and Benton (2007) objected when Peters (2000) presented fenestrasaurs as bipeds despite the fact that matching footprints confirm that configuration and less well-endowed living lizards are able to stand and run bipedally (Snyder 1954). They offered no competing reconstructions.
The origin of the Pterosauria from basal Fenestrasauria
Longisquama insignis (Sharov 1970) Norian, Late Triassic, ~210 mya was derived from a sister to Cosesaurus and Sharovipteryx and was a phylogenetic predecessor to the basal pterosaur, MPUM6009.
Distinct from Cosesaurus, the skull of Longsiquama had a more constricted (in dorsal view) snout, which enhanced binocular vision. The orbits were larger. The posterior teeth had larger cusps*. The quadrate leaned posteriorly.*
The cervicals were shorter* and the dorsal series was longer, especially so near the hips and between the ilia. Even so the presacral count was reduced to 21.* The sacrum curved dorsally 90 degrees,* which elevated the attenuated tail.* These vertebral modifications made Longsiquama somewhat similar to a lemur, which also leaps from tree to tree. Such a long torso provided more room for plumes, gave the backbone great flexibility, and provided more room for egg production.
The pectoral girdle was little changed from Cosesaurus, but the clavicles curved around the sternal complex* and the sternal keel was deeper.* Now fused together the interclavicle, clavicles and sternum formed a sternal complex.*
It is difficult to determine if metacarpal IV was torsioned* or not and difficult to ascertain the ability and degree of finger 4 to fold. The fingers were all greatly enlarged, especially finger 4.* The loss of continuity in the PILs (parallel interphalangeal lines) indicates that finger 4 no longer worked in consort with digits 1-3.*
The posterior ilium was angled dorsally* matching the dorsally curved sacrum. The pubis and ischium were much deeper,* which provided a much larger pelvic aperture to pass a much larger egg. The hind limb was more robust. The foot was relatively large with digits of increasing length laterally.* The metatarsals were compressed,* as in Cosesaurus. Pedal digit V had a curved proximal phalanx.
Longisquama was named for its extradermal dorsal plumes. Another set of plumes arose from its skull and neck. Former caudal hairs (in Cosesaurus) formed a tail vane*. Pterosaur-like patagia trailed the forelimbs*. Longisquama was overloaded with secondary sexual characteristics. From plumes to flapping arms, Longisquama was all about creating an exciting presentation unrivaled in the Triassic. Longisquama had everything Cosesaurus had, only wildly exaggerated. With increased bipedalism and active flapping, Longiquama probably experienced the genesis of aerobic metabolism.*
Pterosaurs did not develop from Longisquama. They shared a common ancestor. The pterosaurs, once they developed larger wings, did not continue to elaborate the plumes as those would have hampered flight. Pterosaurs did not have more robust hind limbs or a long dorsal vertebral series.
Eudimorphodon? Carniadactylus? MPUM 6009 (Wild 1978) Norian, Late Triassic, ~210 mya was originally considered a juvenile Eudimorphodon ranzii. Later Dalla Vecchia (2009) considered this specimen congeneric with Carniadactylus, but the morphological differences are too great. Often called “the Milan specimen” MPUM 6009 is a distinct taxon and the most primitive known pterosaur following the analysis of (Peters 2007) with the resulting tree seen here.
Distinct from Longisquama, the skull of the Milan specimen was proportionately larger. The antorbital fenestra was larger, reaching the top of the skull. The mandible was more gracile.
Atypical for most pterosaurs, but similar to Longisquama and anurognathid pterosaurs, the cervicals were relatively short, but much more robust. The dorsal series was shorter. The caudals were more slender.
The sternal complex had posterior indentions marking the contributions of the now fused clavicles and sternum. The coracoid retained its quadrant curve, but was more robust. The hand was relatively smaller, but manual digit IV was hyper-elongated to form the spar for a wing membrane. Complete folding of digit IV against the ulna was possible due to the torsion of metacarpal IV. Manual digit V became an even smaller vestige, but still retained three phalanges including a sharp ungual.
The ilium was slightly shorter. The pubis and ischium were ventrally separated. The femoral head was inturned but no distinct neck was visible. The prepubis extended halfway to the knee. The hind limb was relatively the longest among pterosaurs, similar to that of Longisquama. The digitigrade pes was relatively smaller. Pedal 5.2 and p5.3 were fused and preserved folded against p5.1, which was straight. The pedal digit V ungual was retained.
Cranial and dorsal plumes were similar to those in Cosesaurus, but smaller than in Longisquama. The wing membranes extended behind the elongated wing finger. A better flyer than Longisquama, MPUM 6009 had wing proportions more like those of other pterosaurs. It is doubtful that MPUM 6009 was ever a quadruped owing to the relative limb lengths.
Many of the evolutionary changes from Longisquama to MPUM 6009 (i.e. larger skull, shorter torso) could have been the result of paedomorphosis or they could have been the result of natural selection in a flyer, rather than a leaper, with certain traits (long legs, elevated tail, short neck) retained. As in Longisquama, the pelvic aperture and wide posterior sacrals indicate a relatively larger egg could have been delivered.
This series of taxa documents the origin of the pterosaur wing without having to imagine wing pronation, loss of digit V, loss of ungual 4 and migration of metacarpals I-III to the anterior face of metacarpal IV, as Bennett (2008) postulated.
Bennett SC 1996. The phylogenetic position of the Pterosauria within the Archosauromorpha. Zoolological Journal of the Linnean Society 118: 261–308.
Bennett SC 2008. Morphological evolution of the forelimb of pterosaurs: myology and function. Pp. 127–141 in E. Buffetaut & D.W.E. Hone (eds.), Flugsaurier: pterosaur papers in honour of Peter Wellnhofer. Zitteliana, B28.
Carroll and Thompson 1982. A bipedal lizardlike reptile fro the Karroo. Journal of Palaeontology 56:1-10.
Dalla Vecchia FM 2009. Anatomy and systematics of the pterosaur Carniadactylus (gen. n.) rosenfeldi (Dalla Vecchia, 1995). Rivista Italiana de Paleontologia e Stratigrafia 115 (2): 159-188.
Ellenberger P and de Villalta JF 1974. Sur la presence d’un ancêtre probable des oiseaux dans le Muschelkalk supérieure de Catalogne (Espagne). Note preliminaire. Acta Geologica Hispanica 9, 162-168.
Hone DWE and Benton MJ 2007. An evaluation of the phylogenetic relationships of the pterosaurs to the archosauromorph reptiles. Journal of Systematic Palaeontology 5:465-469. PDF online
Hone DWE and Benton MJ 2008. Contrasting supertree and total-evidence methods: the origin of the pterosaurs. In: Hone DWE, Buffetaut E, editors. Flugsaurier: pterosaur papers in honour of Peter Wellnhofer. Vol. 28. Munich: Zittel B. p. 35-60.
Nesbitt SJ 2011. The early evolution of archosaurs: relationships and the origin of major clades. Bulletin of the American Museum of Natural History 352: 292 pp.
Nopcsa F 1931. Macrocnemus nicht Macrochemus. Centralblatt fur Mineralogie. Geologic und Palaeontologie; Stuttgart. 1931 Abt B 655–656.
Peabody FE 1948. Reptile and amphibian trackways from the Lower Triassic Moenkopi formation of Arizona and Utah. University of California Publications, Bulletin of the Department of Geological Sciences 27: 295-468.
Peters D 2000a. Description and Interpretation of Interphalangeal Lines in Tetrapods. Ichnos 7:11-41.
Peters D 2000b. A reexamination of four prolacertiforms with implications for pterosaur phylogenesis. Rivista Italiana di Paleontologia e Stratigrafia 106: 293–336.
Peters D 2007. The origin and radiation of the Pterosauria. In D. Hone ed. Flugsaurier. The Wellnhofer pterosaur meeting, 2007, Munich, Germany. p. 27.
Peters D 2009. A reinterpretation of pteroid articulation in pterosaurs. Journal of Vertebrate Paleontology 29: 1327-1330
Renesto S and Avanzini M 2002. Skin remains in a juvenile Macrocnemus bassanii Nopsca (Reptilia, Prolacertiformes) from the Middle Triassic of Northern Italy. Jahrbuch Geologie und Paläontologie, Abhandlung 224(1):31-48.
Sharov AG 1971. New flying reptiles from the Mesozoic of Kazakhstan and Kirghizia. – Transactions of the Paleontological Institute, Akademia Nauk, USSR, Moscow, 130: 104–113.
Snyder RC 1954. The anatomy and function of the pelvic girdle and hind limb in lizard locomotion. American Journal of Anatomy 95:1-46
Vidal LM 1915. Nota geologica y paleontologica sobre el Jurásico superior de la provincia de Lérida. Bolletino del Instituto de Geologica Minerales España, 36: 1-43.
Wild R 1978. Die Flugsaurier (Reptilia, Pterosauria) aus der Oberen Trias von Cene bei Bergamo, Italien. Bolletino della Societa Paleontologica Italiana 17(2): 176–256.
Wild R 1993. A juvenile specimen of Eudimorphodon ranzii Zambelli (Reptilia, Pterosauria) from the upper Triassic (Norian) of Bergamo. Rivisita Museo Civico di Scienze Naturali “E. Caffi” Bergamo 16: 95-120.
We’ll look at soft tissue in pterosaurs in the next two posts.