Bones of dinosaurs and other reptiles from the Triassic-Jurassic of the Connecticut Valley: Over 200 years of published history

The bones from the Connecticut Valley in eastern USA came from the Newark Supergroup with the Lower Jurassic remains coming from the uppermost red sandstone bed (Portland Formation, Middle Hettangian) in Connecticut, and the Longmeadow Formation in Massachusetts. The dinosaurs mostly represent herbivorous or omnivorous facultatively bipedal early sauropodomorph non-eusauropod (“prosauropod”) dinosaurs. Two fragmentary records of animals with a body length of ~1.5 m were discovered during blasting for a water well. The first (YPM VP 2125), in 1818 from East Windsor, Conn., is the earliest published record of a North American dinosaur. The second (AMC 41104), in 1855 from Springfield, Mass., is the holotype of Megadactylus polyzelus Hitchcock, 1865, which was the type species of Anchisaurus Marsh, 1885. In the 1880s, three articulated skeletons came from a quarry in Manchester, Conn. YPM VP 209, the type of Anchisaurus solus Marsh, 1892, is a poorly preserved juvenile skeleton (body length ~1 m) with a fragmentary skull. YPM VP 1883, an almost complete well-preserved skeleton (body length 1.6 m) with skull but no tail, is the type of Anchisaurus colurus Marsh, 1891. YPM VP 208, the largest specimen (body length 2.5 m), consisting mostly of the hip region and hindlimbs, is Anchisaurus major Marsh, 1889, the type species of Ammosaurus Marsh, 1891. Long identified as an early carnivorous coelurosaurian theropod dinosaur, Ammosaurus is now regarded as a basal sauropodomorph dinosaur and usually as a subjective junior synonym of Anchisaurus. However, based on differences in the proximal end of the pubis, with a plesiomorphic closed obturator foramen in YPM VP 208 and an autapomorphic open one in YPM VP 1883 with a triangular sutural surface for the ischium, these are valid sister taxa that constitute the Anchisauridae Marsh, 1885. Other autapomorphies or unique characters for Anchisaurus are a postorbital with a transversely expanded ventral process, a lateral pit on the ventral part of the quadrate, and three small foramina in a vertical row ventrally on the proximal part of sacral rib 2. Autapomorphies for Ammosaurus include a depression ventrally on the proximal part of sacral rib 2, and no connection beyond the bases of the transverse process and the sacral rib of sacral vertebra 3. YPM VP 2125 may represent a second specimen of Ammosaurus based on the hatchet-shaped form of the deltoid process of the humerus and the robust form of distal tarsal 1 and carpal 1. Because of the inadequate nature of the holotype (AMC 41104), YPM VP 1883 was designated by the ICZN (2015) as the neotype of Megadactylus polyzelus Hitchcock, 1865, the type species of Anchisaurus Marsh, 1885. Three different dispersal events brought the Lower Jurassic basal sauropodomorphs (Anchisauridae, Seitaard, Sarahsaurus) to North America. Unlike footprints, the bones of carnivorous theropod dinosaurs are very rare. There are two coelophysoid grade skeletons, that of Podokesaurus holyokensis Talbot, 1911 from near South Hadley, Mass., a fairly complete skeleton (body length ~1.15 m) but no skull, and a natural cast of the pubis, tibia and ribs of a larger individual (~2.5 to 3 m) from Middletown, Conn. After burial, the bones of the latter were removed by dissolution and the cavities filled with sand. The difficulty distinguishing such sandstone casts of bones from the surrounding sandstone is one reason why skeletal remains are so rare in the region compared to those of footprints. Two isolated, Dilophosaurus-like theropod teeth, from the lower Shuttle Meadow Formation in North Guildford, Conn., represent an animal about 4 m long. These teeth are the earliest dinosaur remains from the Connecticut Valley, occurring a little above the End-Triassic Extinction. McMenamin (2021a) describes the distal humerus (width 123 mm) of a large 9 m long supposed theropod dinosaur from the Portland Formation near Amherst, Mass. However, the EDS spectrum shows mostly silica and oxygen, with only trace amounts of phosphorus and calcium, so it is not bone but some version of silica. McMenamin (2021b) also describes the wrist region of a supposed pterosaur from South Hadley, Mass. but clear labelled drawings and photos are needed to establish the identity of this specimen. Hitchcock (1836, 1858, 1865) described the abundant three toed footprints from the Connecticut Valley as “ornithichnites” that were presumed to be the tracks of large ground living flightless birds. The genus Anchisauripus Lull, 1904a was erected for footprints supposedly made by the prosauropod Anchisaurus, but they were probably made by a small Coelophysis-like theropod dinosaur. However, trackways of Otozoum were made by a prosauropod walking with a more plantigrade posture than the prosauropod Plateosaurus. The identification of trackways of the brontozoid ichnotaxa Anchisauripus-Eubrontes-Gigantipus continuum as those of prosauropods requires a nearly complete lack of correspondence between pad structure and osteology. However, there is an excellent fit between the pes of the theropod dinosaur Dilophosaurus wetherilli from Arizona (Lower Jurassic, Kayenta Formation) and the pes reconstructed for Eubrontes. The predominance of footprints of carnivorous theropod dinosaurs from the early Early Jurassic of Eastern (and Western) North America may indicate an extremely unusual, inverted food pyramid, with the terrestrial carnivorous theropods probably feeding mostly on fish. By the late Early Jurassic to Middle Jurassic, herbivorous dinosaurs became much more numerous (as indicated by both skeletal fossils and footprints around the world) and normal terrestrial food pyramids were widely established. The only non-dinosaurian record from the Lower Jurassic of the Connecticut Valley is a partial skeleton from the Longmeadow Sandstone near Longmeadow, Mass. This is Stegomus longiceps Emerson & Loomis, 1904, the type species of the protosuchian crocodylomorph Stegomosuchus Huene, 1922. This small (body length 25 to 30 cm, a juvenile) agile, cursorial, carnivorous animal is correlated with the ichnotaxon Batrachopus gracilis. Several different non-dinosaurian remains have been discovered in the Upper Triassic (pre-latest Rhaetian) from the New Haven Formation of the Hartford basin. Conn. A partial right scapula (estimated animal’s body length ~225 cm) was found in 1888 in Simsbury. Described as Rutiodon validus (Marsh, 1893), it is an indeterminate phytosaur, a group of aquatic predators resembling living crocodiles. A natural mold of the underside of 16 plates of the dorsal part of the dermal armor (body length ~150 cm) was found in 1895 in Fair Haven. Stegomus arcuatus Marsh, 1896b is referred to the Aetosauria, a group of quadrupedal herbivorous reptiles. The pseudosuchian Erpetosuchus is represented by a skull found in Cheshire in 1995. It was a small (body length ~61 cm), lightly built, cursorial predator. The parareptilian procolophonid Hypsognathus fennert is represented by a mold of the anterior part of the body with an almost complete skull as bone, was found in 1967 in Meriden. The molar-like teeth, with the deep cheek region and deep lower jaws, indicate that it fed on tough high fiber plant material. A small partial skull (body length >1 m), found in 1965 in central Connecticut, was named Colopops noviportensis Pritchard et al., 2018 as the most basal rhynchosaur, However, Scheyer et al. (2020) note that the dorso-ventrally compressed and disarticulated bones are those of the skull of a normal rhynchocephalian lepidosauromorph, not a rhynchosaur, and that the supposed reinforced snout and exceptionally enlarged temporal region (indicative of very large jaw closing muscles) are artifacts of preservation.