Pterodactylus

Holotype specimen of the species P. spectabilis, now considered a juvenile specimen of P. antiquus

Like other pterosaurs (most notably Rhamphorhynchus), Pterodactylus specimens can vary considerably based on age or level of maturity. Both the proportions of the limb bones, size and shape of the skull, and size and number of teeth changed as the animals grew. Historically, this has led to various growth stages (including growth stages of related pterosaurs) being mistaken for new species of Pterodactylus. Several detailed studies using various methods to measure growth curves among known specimens have suggested that there is actually only one valid species of Pterodactylus, P. antiquus.[25]

The youngest immature specimens of Pterodactylus antiquus (alternately interpreted as young specimens of the distinct species P. kochi) have a small number of teeth, as few as 15 in some, and the teeth have a relatively broad base.[23] The teeth of other P. antiquus specimens are both narrower and more numerous (up to 90 teeth are present in several specimens).[25]

Pterodactylus specimens can be divided into two distinct year classes. In the first year class, the skulls are only 15 to 45 millimeters (0.59 to 1.77 in) in length. The second year class is characterized by skulls of around 55 to 95 millimeters (2.2 to 3.7 in) long, but are still immature however. These first two size groups were once classified as juveniles and adults of the species P. kochi, until further study showed that even the supposed "adults" were immature, and possibly belong to a distinct genus. A third year class is represented by specimens of the "traditional" P. antiquus, as well as a few isolated, large specimens once assigned to P. kochi that overlap P. antiquus in size. However, all specimens in this third year class also show sign of immaturity. Fully mature Pterodactylus specimens remain unknown, or may have been mistakenly classified as a different genus.[23]

Fossil pelvis of a large specimen, now referred to the dubious species P. grandipelvis

The distinct year classes of Pterodactylus antiquus specimens show that this species, like the contemporary Rhamphorhynchus muensteri, likely bred seasonally and grew consistently during its lifetime. A new generation of 1st year class P. antiquus would have been produced seasonally, and reached 2nd-year size by the time the next generation hatched, creating distinct 'clumps' of similarly-sized and aged individuals in the fossil record. The smallest size class probably consisted of individuals that had just begun to fly and were less than one year old.[23][27] The second year class represents individuals one to two years old, and the rare third year class is composed of specimens over two years old. This growth pattern is similar to modern crocodilians, rather than the rapid growth of modern birds.[23]

Comparisons between the scleral rings of Pterodactylus antiquus and modern birds and reptiles suggest that it may have been diurnal. This may also indicate niche partitioning with contemporary pterosaurs inferred to be nocturnal, such as Ctenochasma and Rhamphorhynchus.[28]

Based on the shape, size, and arrangement of its teeth, Pterodactylus has long been recognized as a carnivore specializing in small animals. A 2020 study of pterosaur tooth wear supported the hypothesis that Pterodactylus preyed mainly on invertebrates and had a generalist feeding strategy, indicated by a relatively high bite force.[29]

Fossil specimen of the species Rhamphorhynchus muensteri, which was previously assigned as the species Pterodactylus münsteri

Numerous species have been assigned to Pterodactylus in the years since its discovery. In the first half of the 19th century any new pterosaur species would be named Pterodactylus, which thus became a "wastebasket taxon".[32] Even after clearly different forms had later been given their own generic name, new species would be created from the very productive sites, throughout Europe and North America, often based on only slightly different material.[48]

The earliest reassignments of pterosaur species to Pterodactylus started in 1825, with the description of Rhamphorhynchus; fossil collector Georg Graf zu Münster alerted the German paleontologist Samuel Thomas von Sömmerring about several distinct fossil specimens, Sömmerring thought that they belonged to an ancient bird.[49] Further fossil preparations had uncovered teeth, to which Graf zu Münster created a skull cast. He later sent the cast to Professor Georg August Goldfuss, who recognized it as a pterosaur, specifically a species of Pterodactylus. At the time however, most paleontologists incorrectly consider the genus Ornithocephalus (lit. 'bird-head') to be the valid name for Pterodactylus, and therefore the specimen found was named as Ornithocephalus Münsteri, which was first mentioned by Graf zu Münster himself.[50] Another specimen was found and described by Graf zu Münster in 1839, he assigned this specimen to a new separate species called Ornithocephalus longicaudus; the specific name means 'long tail', in reference to the animal's tail size.[51] German paleontologist Hermann von Meyer in 1845 officially emended that the genus Pterodactylus had priority over Ornithocephalus, so he reassigned the species O. münsteri and O. longicaudus into Pterodactylus münsteri and Pterodactylus longicaudus.[52] In 1846, von Meyer created the new species Pterodactylus gemmingi based on long-tailed remains; the specific name honors the fossil collector Carl Eming von Gemming.[53] Later, in 1847, von Meyer finally erected the generic name Rhamphorhynchus (lit. 'beak snout') due to the distinctively long tails seen in the specimens found, which are much longer than those seen in Pterodactylus. He assigned the species P. longicaudus as the type species of Rhamphorhynchus, which resulted in a new combination called Rhamphorhynchus longicaudus.[54] The species R. münsteri was later changed to R. muensteri by Lydekker in 1888, due to the ICZN rule that prohibits non-standard Latin characters, such as ü, in scientific names.[22]

Holotype jaw fragments and teeth of Cimoliopterus, which was previously known as Pterodactylus cuvieri

Beginning in 1846, many pterosaur specimens were found near the village of Burham in Kent, England by British paleontologists James Scott Bowerbank and Sir Richard Owen. Bowerbank had assigned fossil remains to two new species; the first was named in 1846 as Pterodactylus giganteus;[55] the specific name means 'the gigantic one' in Latin, in reference to the large size of the remains, and the second species was named in 1851 as Pterodactylus cuvieri, in honor of the French scientist Georges Cuvier.[56] Later in 1851, Owen named and described new pterosaur specimens that have been found yet again in England. He assigned these specimens to a new species called Pterodactylus compressirostris.[57] In 1914 however, paleontologist Reginald Hooley redescribed P. compressirostris, to which he erected the genus Lonchodectes (lit. 'lance biter'), and therefore made P. compressirostris the type species, and created the new combination L. compressirostris.[58] In a 2013 review, P. giganteus and P. cuvieri were reassigned to new genera; P. giganteus was reassigned to a genus called Lonchodraco ('lance dragon'), which resulted in a new combination called L. giganteus, and P. cuvieri was reassigned to the new genus Cimoliopterus ('chalk wing'), creating C. cuvieri.[59] Back in 1859, Owen had found remains the front part of a snout in the Cambridge Greensand, and assigned it into the species Pterodactylus segwickii; in honor of Adam Sedgwick, a British geologist.[60] This species however, was reassigned to the genus Camposipterus in 2013, therefore creating the new combination Camposipterus segwickii.[59] Later, in 1861, Owen had uncovered multiple distinctively looking fossil remains yet again in the Cambridge Greensand, these were assigned to a new species named Pterodactylus simus,[61] though the British paleontologist Harry Govier Seeley had created a separate generic name called Ornithocheirus, and reassigned P. simus as the type species, which created the combination Ornithocheirus simus.[62] Between the years 1869 and 1870, Seeley had reassigned many pterosaur species into Ornithocheirus, while also creating several new species.[62][63] Many of these species however, are now reclassified to other genera, or considered nomina dubia.[59] In 1874, further specimens were found in England, again by Owen, these ones were assigned to a new species called Pterodactylus sagittirostris,[64] this species however, was reassigned to the genus Lonchodectes in 1914 by Hooley, which resulted in an L. sagittirostris.[58] This conclusion was revised by Rigal et al. in 2017, who disagreed with Hooley's reassignment, and therefore created the genus Serradraco, which afterwards resulted in a new combination called S. sagittirostris.[65]

Specimen YPM1177, the first uncovered skull of Pteranodon, which was back then assigned as a species of Pterodactylus

Assigning new pterosaur species to Pterodactylus was not only common in Europe, but also in North America; paleontologists such as Othniel Charles Marsh in 1871 for example, described several toothless pterosaur specimens, which were accompanied by teeth that belonged to the fish Xiphactinus, which Marsh assumed that these teeth belonged to the pterosaur specimens he found, since all pterosaurs discovered at the time had teeth. He then assigned these specimens to a new species called "Pterodactylus oweni", but this was changed to Pterodactylus occidentalis because "P. oweni" was found to have been preoccupied by a pterosaur species described with the same name back in 1864 by Seeley.[66][67] In 1872, American paleontologist Edward Drinker Cope also found various pterosaur specimens in North America, he assigned these to two new species known as Ornithochirus umbrosus and Ornithochirus harpyia, Cope attempted to assign the specimens he found to the genus Ornithocheirus, but misspelled forgetting the 'e'.[68] In 1875 however, Cope reassigned the species O. umbrosus and O. harpyia into Pterodactylus umbrosus and Pterodactylus harpyia, though these species had been considered nomina dubia ever since.[69][67] Paleontologist Samuel Wendell Williston unearthed the first skull of the pterosaur, and found that the animal was toothless,[67] this made Marsh create the genus Pteranodon (lit. 'toothless wing'), and therefore reassigned all the American pterosaur species, including the ones that he named, from Pterodactylus to Pteranodon.[70]

Later, in the 1980s, subsequent revisions by Peter Wellnhofer had reduced the number of recognized species to about half a dozen. Many species assigned to Pterodactylus had been based on juvenile specimens, and subsequently been recognized as immature individuals of other species or genera. By the 1990s it was understood that this was even true for part of the remaining species. P. elegans, for example, was found by numerous studies to be an immature Ctenochasma.[25] Another species of Pterodactylus originally based on small, immature specimens was P. micronyx. However, it has been difficult to determine exactly of what genus and species P. micronyx might be the juvenile form. Stéphane Jouve, Christopher Bennett and others had once suggested that it probably belonged either to Gnathosaurus subulatus or one of the species belonging to Ctenochasma,[24][25] though after additional research Bennett assigned it to the genus Aurorazhdarcho.[17] Another species with a complex history is P. longicollum, named by von Meyer in 1854, based on a large specimen with a long neck and fewer teeth. Many researchers, including David Unwin, have found P. longicollum to be distinct from P. kochi and P. antiquus. Unwin found P. longicollum to be closer to Germanodactylus and therefore requiring a new genus name.[4] It has sometimes been placed in the genus Diopecephalus because Harry Govier Seeley based this genus partly on the P. longicollum material. However, it was shown by Bennett that the type specimen later designated for Diopecephalus was a fossil belonging to P. kochi, and no longer thought to be separate from Pterodactylus. Diopecephalus is therefore a synonym of Pterodactylus, and as such is unavailable for use as a new genus for "P." longicollum.[71] "P." longicollum was eventually made the type species of a separate genus Ardeadactylus.[17]

Juvenile type specimen of Pterodactylus kochi, now reassigned as Diopecephalus kochi

The only well-known and well-supported species left by the first decades of the 21st century were P. antiquus and P. kochi. However, most studies between 1995 and 2010 found little reason to separate even these two species, and treated them as synonymous.[4] More recent studies of pterosaur relationships have found anurognathids and pterodactyloids to be sister groups, which would limit the more inclusive group Caelidracones to just two clades.[72][71] In 1996, Bennett suggested that the differences between specimens of P. kochi and P. antiquus could be explained by differences in age, with P. kochi (including specimens alternately classified in the species P. scolopaciceps) representing an immature growth stage of P. antiquus. In a 2004 paper, Jouve used a different method of analysis and recovered the same result, showing that the "distinctive" features of P. kochi were age-related, and using mathematical comparison to show that the two forms are different growth stages of the same species.[25] An additional review of the specimens published in 2013 demonstrated that some of the supposed differences between P. kochi and P. antiquus were due to measurement errors, further supporting their synonymy.[17]

By the 2010s, a large body of research had been developed based on the idea that P. kochi and P. scolopaciceps were early growth stages of P. antiquus. However, in 2014, two scientists began publishing research that challenged this paradigm. Steven Vidovic and David Martill concluded that differences between specimens of P. kochi, P. scolopaciceps, and P. antiquus, such as different lengths of neck vertebrae, thinner or thicker teeth, more rounded skulls, and how far the teeth extended back in the jaws, were significant enough to separate them into three distinct species. Vidovic and Martill also performed a phylogenetic analysis which treated all relevant specimens as distinct units, and found that the P. kochi type specimen did not form a natural group with that of P. antiquus. They concluded that the genus Diopecephalus could be returned to use to distinguish "P". kochi from P. antiquus. They named the new genus Aerodactylus for P. scolopaciceps as well. So, what Bennett considered early growth stages of one species, Vidovic and Martill considered representatives of new species.[48][32]

In 2017, Bennett challenged this hypothesis, he claimed that while Vidovic and Martill had identified real differences between these three groups of specimens, they had not provided any rationale that the differences were enough to distinguish them as species, rather than just individual variation, growth changes, or simply due to crushing and distortion during the fossilization process. Bennett pointed in particular to the data used to distinguish Aerodactylus, which was so different from the data for related species, it might be due to an unnatural assemblage of specimens. As a result, Bennett continued to consider Diopecephalus and Aerodactylus simply as year-classes of immature Pterodactylus antiquus.[73]

During its over-200-year history, the various species of Pterodactylus have gone through a number of changes in classification and thus have acquired a large number of synonyms. Additionally, a number of species assigned to Pterodactylus are based on poor remains that have proven difficult to assign to one species or another and are therefore considered nomina dubia (lit. 'doubtful names'). The following list includes names that were used to identify new pterosaur species that now have been reclassified, or until recently thought to be pertaining to Pterodactylus proper, and names based on other material that has as yet not been assigned to other genera. This list also includes species that are nomina nuda ('naked names'), which are species that were not published formally. Species that are nomina oblita ('forgotten names') are the ones that have been disused, and species that are nomina rejecta ('rejected names') are the ones that have been rejected because a more preferable name had been accepted instead.[22][74]