vermann, Ernst Stromer (b. 1871) and Otto Jaekel (b. 1863). At
Upsala in Sweden Carl Wiman has inspired a remarkably pro-
gressive group of workers, while in Switzerland Hans Georg
Stehlin (b. 1870) has continued in the great field of Riitimeyer.
For the principal contributions by palaeontologists of the third and fourth groups above described, the reader is referred to the Memoirs and Bulletins of the American Museum of Natural History, of the university of California, of the Carnegie Institution of Washington, to the Contributions from the Palaeontological Laboratory (Peabody Museum) of Yale Uni- versity, to the Memoirs and Catalogues of the British Museum (Natural History), to the Palaeontographica, and to the Me- moirs of the Societe Paleontologique Suisse. It is upon the researches of these workers in field and laboratory that the great synthetic volumes referred to earlier are chiefly founded, and that the following generalizations of modern vertebrate palaeontology are chiefly due.
ORIGINS OF THE GREAT VERTEBRATE STOCK AND ITS BRANCHES
Origin of Chordates. No discovery has thus far lessened the gap between the modern Protochordates (Amphioxus, tunicates, etc.) and any of the known phyla of invertebrates. Some of the cephalaspid ostracoderms have been cited by Patten as favouring the view that the chordates have been derived from certain arthropods, but such resemblances are ascribed to convergence by Dollo and many others. The earliest ostracodcrm remnant actually known is a dermal plate of a genus named Astraspis from the Upper Ordovician near Canyon City, Colorado; this represents a new family Astraspidae allied to the Psammosteidae of the Silurian and Devonian (C. R.Eastman, 1917). These chord- ates, heavily shielded and hence known as ostracoderms, were dom- inant in the Upper Silurian, radiating into six families and many genera, abundant in the Lower Devonian, diminishing in the Middle Devonian and becoming extinct in the Upper Devonian.
Origin of Fishes. The earliest fish remnant actually known is the fin-spined Onchus from the Upper Silurian of Scotland, which appears to represent the group of acanthodian sharks, covered with fine quadrate scales like those of ganoids and with a skull structure distinctly elasmobranch. The elasmobranchs (shark and ray types) are still the oldest known gnathostomes or true jaw-bearing vertebrates, constituting (a) one of the four primary gnathostome groups, i.e. jawed groups, the others being (b) the fringe-finned ganoids (Crossopterygii), (c) the ray-finned ganoids and teleosts collectively known as Actinopterygii and ( the lungfishes (Dipnoi). The fossil ancestors of the fringe- finned ganoids have not yet been discovered; so these animals are theoretically traced to unknown cartilaginous fishes of Silurian times. The oldest Crossopterygian actually known is the Osleolcpis macrolepidotus of the Middle Old Red Sandstone of Scotland. There were two principal periods of adaptive radia- tion among the Crossopterygii, the first in Middle and Upper Devonian times, the second in Mesozoic times which produced the family Coelacanthidae, from which may have sprung the existing fishes Polypterus and Calamoichthys as degenerate off- shoots. From the earlier Devonian radiation of the Crossoptery- gians is traced the theoretic origin of the Dipnoi or lungfishes, on the one hand, and of the oldest known amphibians on the other. The Devonian Crossopterygian skull and fins appear to be " archetypal," to the lungfish type, on the one hand, and to the amphibian type on the other. Cope's genius in separating the Actinopterygii is sustained, for there is as yet no fossil evi- dence of the connexion of this group with the Crossopterygii, other than the supposed community of origin in Silurian times. Here the reader should consult the writings of Smith Woodward, Joseph F. Whiteaves, Bashford Dean, William K. Gregory and the synthetic reviews of Osborn (1918) and Lull (1917).
Origin of Amphibians and First Tetrapods. In this epoch- making transition from the fringe-finned fish type to the tetrap- odal amphibian and terrestrial type, the prophecies of Huxley, Cope and Baur and other great anatomists of the second and third groups of palaeontologists appear to be fulfilled. The Silurian period marked the parting of the ways among the great
primary groups of fishes and the first steps towards the frame of the terrestrial amphibians. Not until the Upper Devonian of Pennsylvania do we find a footprint ( Thinopus antiquus Marsh) , which may be referred to an amphibian tetrapod. The first known actual skeletons occurred in the Coal Measures (Upper Carboniferous) of Europe and America and represented four widely radiating groups. The structural gap separating the earliest tetrapod amphibians and fishes is perhaps the greatest known in the whole range of vertebrate evolution, but all modern authorities agree that the amphibians were probably derived from a Silurian or early Devonian type of fringe-finned fish. Even as far back as the Upper Carboniferous and even in the Lower Carboniferous the Amphibia were adaptively radiating into several orders and numerous families comprising highly special- ized forms. During the Carboniferous we find numerous inde- pendent phyla of eel-like or burrowing, and of compressed, swimming, as well as of large-bodied, predatory forms. The latter culminate in the gigantic labyrinthodonts of the Triassic. The exact connexion of any of these forms with the modern Amphibia (urodcles and Anura) is doubtful. The Anura first appear in the Jurassic, and at the present time they retain many characters reminiscent of such Palaeozoic Amphibia as the branchiosaurs and the Eryops group. The urodeles are first known in the genus Hylaeobatrachus of the Lower Cretaceous of Europe. Both groups, especially the Anura, appear to have gone through a wide adaptive radiation during the Tertiary. The connexion of the modern caecilians with the ancient types is obscure. The reader is referred especially to the contributions of Williston, Case, Watson, Gregory, Broili, and the synthetic re- views of Osborn (1918) and Lull (1917).
Origin of Reptiles. The oldest-known reptiles, solid-headed Cotylosauria of Cope, are regarded as amphibians which had eliminated the aquatic stages in development, the oldest reptile actually known being the genus Eosauravus from the Coal Measures of Ohio. In other words, the cotylosaur reptiles are traceable to solid-headed stegocephalian amphibians, which, in turn, are traceable to solid-headed unknown Crossopterygians of Silurian times. The oldest and most primitive reptiles (Coty- losauria} occurring in the Upper Carboniferous and Permian, are thus structurally very close to certain contemporary stegoce- phalian amphibians. The first great adaptive radiation of the reptiles into the two grand divisions, the solid-headed (Coty- losauria} and the temporal-arched (Pelycosauria}, began in the Upper Carboniferous and still more widely diverged in Permian times. As early as the Permian, occurs a mammalian-like series of reptiles which exhibits an extensive adaptive radiation and gives off one branch, the Cynodontia, which, in turn, survives into Triassic times and clearly approaches the mammalian grade of organization. From the primary temporal-arched also appear the forerunners of the Mesozoic reptiles, the plesiosaurs, ichthyo- saurs, dinosaurs and pterosaurs, widely separated from each other in the Triassic and thus having their branches deep down in the Permian and Carboniferous, each grand division giving rise to an adaptive radiation of its own. These have been traced in detail by such authorities as Andrews, Dollo, Abel, von Huene, Williston and Osborn.
Here the reader is referred to the writings of Williston, Hatcher, Osborn, Merriam, Lambe, Lull, and especially during the past decade to those of Charles W. Gilmore of the United States National Museum, Washington, and of Dr. Robert Broom of South Africa, as well as to the synthetic reviews of Osborn (1918) and Lull (1917).
The two greatest achievements of the decade are the clearing up of the relationships of the primitive South African terrestrial Reptilia of the Perm-Trias, beginning with the solid-headed types (pareiasaurs) and ending in their highest expression, the mammal-like types known as Cynodonts and Theriodonts. The field explorations of Robert Broom and the profound compar- ative researches of D. M. S. Watson and of William K. Gregory have given us a clear comprehension of the habits and relation- ships of this first terrestrial radiation group. Williston and Case have covered the same great period in America.
