GREGORY, THEORIES OF THE ORIGIN OF BIRDS 37 



saurs, and especially by the "Pseudosuchians/' such as Ornithosuchus, 

 EuparJceria. Some of these small reptiles may have reared np on their 

 hind legs in running, as certain lizards do ; indeed the hind feet of Ewpar- 

 herm, according to Broom, exhibit incipient adaptations to bipedal pro- 

 gression, while others were in all probability actively hopping types. But 

 none of these animals show any very pronounced bird-like characters in 

 the skeleton. Their structure, however, was, on the whole, of so general- 

 ized a type that the diverse peculiarities of the birds, pterosaurs and other 

 groups could readily be derived from this source. In some of the Pseu- 

 dosuchians the body is kno^vn to have been covered with horny plates, but 

 others may have been clothed with the overlapping scales which must 

 have j)receded the evolution of feathers. 



Thus the paleontological record as to the immediate ancestry of the 

 birds is regrettably indecisive, but the principles of comparative anatomy 

 appear to lead to some pretty safe inferences, as follows : 



The transformation of long overlapping scales into feathers, whenever 

 it did occur, was one of the critical steps in the evolution of birds. This 

 transformation took place apparently over the whole surface of the body 

 and along the legs and arms. No doubt it was correlated with the higher 

 vitality and improved circulation of the pro-avis as compared with the 

 ancestral reptile; no doubt also the air -retaining mesh of the feathery 

 covering not only prevented undue radiation of heat from within during 

 cold periods, but also screened off some of the burning heat of the sun. 

 Possibly this feather-armor would also protect the ambitious aviator in 

 his many falls from the branches. The pro-Aves were surely quick run- 

 ners, both on the ground and in the trees, but it is not yet clear whether 

 the upright position was first attained upon the ground or in the trees. 

 They very early acquired the habit of perching upright on the branches, 

 as shown by the consolidated instep bones, grasping first digit and strong 

 claws of Arcliceopteryx. Their slender arms ended in three long fingers 

 provided with large claws which were at first doubtless used in climbing. 



These active pro-Aves contrasted widely in habits with their sluggish 

 remote reptilian forebears. In the pursuit of their prey they jumped 

 lightly from branch to branch and finally from tree to tree, partly sus- 

 tained by the folds of skin on their arms and legs and later by the long 

 scale-feathers of the pectoral and pelvic "wings" and tail. That they 

 held the arms perfectly still throughout the gliding leap still appears 

 doubtful, for all recent animals that do that have never attained true 

 fiight. I cannot avoid the impression that a vigorous downward flap of 

 the arms, even before they became efficient wings, would assist in the 

 "take off'' for the leap, and that another flap just before landing would 

 check the speed and assist in landing. 



