14 i?. S. Lull—Dinosaicrian Distribution. 



saurus stem are derived, as side lines, the Newark dinosaurs 

 Anohisaurus and Ammosaurus of the Connecticut valley; 

 while the first known of these, Meg'adactylus polyzelus from 

 Springfield, Massachusetts, v. Huene (1906, pp. 115-118) 

 refers to the genus Thecodontosviurus itself. Tanysfrophceus, 

 ancestral to the delicate, hollow-boned Cceluridge, begins also in 

 the Muschelkalk, and while differing widely from Thecodonto- 

 saurus, its successor Ccelophysis 1 from beds in Colorado equiva- 

 lent to the Upper Keuper of Europe, converges again toward 

 the Thecodontoid phylum, so that the later representatives, 

 Coelurus on the one hand and the Compsognathoid forms on 

 the other, are closely approximated. 



In the Lower Keuper a new genus, Zanclodon, appears in 

 Europe, of greater size than its contemporaries, and from 

 which v. Huene (1909, p. 20) would derive Ceratosaurus of the 

 Morrison with no annectant forms. There are, however, among 

 the Connecticut valley footprints (Rhsetic), besides numerous 

 ones referable to the Thecodontoid types (Anchisauripus 

 Lull — 1904, p. 468), those of a large carnivore with powerful 

 anterior claws but with a relatively feeble hallux. This track 

 which Hitchcock called Gigandipus (Lull, loc. cit., p. 492) 

 because of its great size, may w T ell have been made by a mem- 

 ber of the Zanclodon phylum, the bones of which are as yet 

 unknown in these deposits. 



The Middle Keuper ushers in another genus in the form of 

 Teratosaurus ; giving rise, in the Rhsetic, according to v. 

 Huene, to two main branches, from one of which arose, through 

 Gressylosaumis and Eusk&losaurus, the great Megalosaurian 

 line, the other giving rise, through Plateosaurus, to the Sau- 

 ropoda. This seems to me, however, to place the divergence 

 of the Sauropoda somewhat too late in time ; to the implied 

 phylogeny 1 take no exception. 



The Connecticut valley forms, which had reached great 

 profusion, to judge from the abundance and variety of their 

 footprints, are contemporaneous with the European Rhsetic. 

 Footprints apparently of equivalent age and character are 

 found in New Mexico as well. 



The lower Keuper beds contain Thecodo7itosaurus and 

 possibly Massospo?idylus in India (Lydekker 1890, p. 22), 

 while the Upper Keuper entombed the former genus in 

 Australia. 



The Upper Karoo beds of Africa, referred by v. Huene to 

 the Khsetic and by Broom (1907, p. 161) to the Lower Jura 

 (Storm berg Beds), contain Thecodontosaurus, Euskelosaurus 

 and Massospjondylus, all Triassic types. 



