The two genera Eryops and Cricotus of the North-American Permian formation had until recently[[120]] been relegated to the Stegocephali. By grouping them and their nearest allies together as Proreptilia it is intended to indicate that they are the lowest known Reptiles and that they probably link this class to the Amphibia. The superficial resemblance of their tri- or bi-partite vertebrae, and their occurrence in the Lower Permian, have caused the error of classing them with the Stegocephali, but the composition of their typically gastrocentrous vertebrae leaves no doubt as to their affinities. After all, we feel certain that Reptiles have arisen from Stegocephalous Amphibia, and it is in the Lower Permian, exactly where these debatable creatures lived side by side with Stegocephali, undoubtedly likewise temnospondylous, that the change from Amphibia into Reptiles seems to have taken place. Both are referable to Amphibia with quadripartite vertebrae. The condition of the occipital condyles determines nothing. This greatly exaggerated character has lost in importance since we have known the condylar modifications of the Theromorpha; moreover, Cricotus itself seems to have possessed a single condyle. We should even expect the Proreptilia to present many Stegocephalous inheritances, for instance the condition of the skull roofed in by dermal bones, a ventral dermal armour, a very complete pectoral arch still without a sternum, and only one sacral vertebra.
Until more genera are better known than they are now, it is premature to divide the present sub-class into orders.
Fig. 57.–Trunk vertebrae of Eryops (cf. Fig. 56, 4, p. [283]). Cp, Articular facet of the capitulum of a rib.
Eryops, with several species in Texas and New Mexico. E. megacephalus is the most abundant and the largest species, its broad and flattened skull measuring more than 18 inches in length and 12 in width. With the exception of the nostrils and the small orbits, the skull is entirely encased in bone, with a rough, pitted surface, but without any distinguishable sutures. The absence of mucous canals, so common in the Stegocephali, is worthy of note. The quadrates extend obliquely outwards and backwards, so that the joint with the mandible lies in a plane behind the occiput. The mandibles are devoid of a projecting angular process. The teeth are numerous, small, and pointed. The vertebrae are typically temnospondylous, consisting each of three pairs of separately ossified pieces, which, although closely packed together, are not suturally connected. The neural arches possess high spinous processes, they articulate by short and broad zygapophyses and are, with their triangular bases, wedged in between the two ventral pieces, the posterior of which (the united interventralia) is in broader contact with the neural arch and lies behind it; the anterior piece (the united basiventrals) appear as typical, but large, intercentra, and bear on their posterior, dorsal margin the facets for the ribs. The latter are short, but are broad at their proximal ends, which are not bifurcated; they extend their articulation from the "intercentra" upon the short lateral processes of the neural arches. The tail is short and ends in a pointed coccyx, owing to fusion of the last vertebrae.
The pubes and ischia are heavy, the former flattened and broadened out. The limbs are of an almost ideal pentadactyloid type; strongly developed for terrestrial locomotion. The ulna possesses a large olecranon. The carpus consists of ten separate pieces, ulnare, intermedium, radiale, two centralia and five distal carpalia. The latter support only four metacarpals and fingers, the second finger being completely abolished, an explanation suggested by Cope and corroborated by Emery.[[121]]
Cricotus, with several species in Texas and Illinois. C. heteroclitus was perhaps 10 feet long and probably aquatic. The skull has a long, narrow, depressed snout, the margins overhanging those of the lower jaw; its surface is encased in dermal bones, most of which still show sutures, so that for instance postfrontals, postorbitals, supratemporals and squamosals can be distinguished; all these are in contact with the long parietals and with the quadrato-jugal arch, covering the temporal region; but the supratemporals have a free projecting border, like the squamosals of the crocodiles. According to Cope's description the basioccipital is connected with the first vertebra by an undivided discoid "intercentrum," probably the true centrum, while the first basiventral mass, which would be, if independent, the first true intercentrum, is more probably connected with the first neural arch, thus constituting the ring of the atlas.
The vertebrae are still temnospondylous, but no longer tripartite. The neural arch is fused with the interventralia into one mass, which carries the capitula and tubercula of the ribs, while the united basiventrals still remain as separate intercentral wedges. In the tail these wedges carry chevron-bones, and are enlarged into thick almost complete discs, or rather rings, while the whole vertebral column is still perforated, as also in Eryops, by the chorda dorsalis. The tail is long. The digits are devoid of claws.
Remains of dermal armour exist on the throat in the shape of several large gular plates, while the whole belly is covered with many closely packed bony scales, which are arranged in chevron-shaped transverse rows.
Probably several other genera of American Permian and also of European Permian strata will, when better known and critically examined, have to be referred to the Proreptilia. Thus for instance the European Melosaurus may have affinities with Eryops, while Diplovertebron of Bohemia seems to be allied to Cricotus. The difficulty of division will lie with those Lower Permian Amphibia which, like Archegosaurus, Euchirosaurus, Actinodon, possess tripartite vertebrae, which at first sight are strikingly like those of Eryops. But the tail-vertebrae permit of no mistake, and since these are quadripartite in Archegosaurus, Chelydosaurus, and Sphenosaurus, these genera are safely to be classed with the Amphibia, unless, indeed, for mere argument's sake, it be assumed that the intercentral discs of Diplovertebron and Cricotus are formed by the fusion of Amphibian interdorsals with interventrals. Anyhow, simply to state that the tripartite vertebrae of Eryops are the same as those of Actinodon, would be as convincing as saying that the English and French flags are essentially the same, both containing the same colours, but one is white, red, and blue, the other blue, white, and red. Tripartite Amphibian vertebrae are composed of basidorsals + basiventrals + interdorsals, those of Reptiles are made up of basidorsals + basiventrals + interventrals. (Cf. Fig. 56, p. [283], and Fig. 1, p. [13].)