1907.] ORIGIN OF FLIGHT. 229 



even during ontogenetical evolution, a backward rotation of the 

 ischium, and in consequence a flattening of the pelvic girdle, is 

 to be met with. Pteromys and Anomalurus, according to Dollo, 

 have to be tei-med passive flyers : the first partially active flyei- 

 seems to be Galeojyithecus, for, according to Wallace, this animal 

 is not only capable of sailing downward, but at the end of its 

 downward oblique glide to rise a little upward. 



Galeojyithecus, however, is a long-tailed, comparatively short- 

 armed patagial animal, in which nevertheless the patagium ex- 

 tends even to the tips of the digits and to the end of the tail ; 

 while when we turn from this to the specialised actively flying 

 Chiroptera, we are impressed firstly by the elongation of the 

 wing, and secondly by the frequent partial or total loss of the 

 tail. 



Both in Pterosaurs and Bats the main movement during flight 

 seems to have been, and still is, dependent on the humei'al articu- 

 lation. The similarity of the patagial structure in Rhampho- 

 rhynchus and Bats, as remarked by Zittel, is also to be noted. 

 The hairless condition of the patagium in Chiroptera compared 

 with Galeopithecus is likewise a more specialised feature ; while 

 PteropvjS vulgaris (more specialised than any Bat in regai-d to the 

 caudal i-egion, " chevauchement de specialisation ") shows, by 

 possessing some hair on the interior surface of the patagial 

 membrane, an intei-mediate stage. 



In Pterosaurs, as also in Pterojnis, the number of sacrals is 

 augmented, and in the latter they even form, by co-ossifying with 

 the ischium, a pseudosacrum. 



The more or less perfect reduction of tibia and ulna is another 

 character that is noteworthy in all patagium-bearing Mammals. 



In an analogous manner to the Cretaceous Ornithocheb'us, also 

 in some Bats a rigid thorax is attained, though in this case the 

 ribs and not the vertebrae co-ossify. 



Since we may safely assume that Bats descended from Mammals 

 which possessed a well-developed neural spine, the reduction of 

 this process, noticeable also in the Flying Lizai^d iJDraco volans), 

 has also to be considered as a sign of specialisation 



The thin and dense skull -bones also unite in specialised Bats, 

 very much as in Birds and Pterosaurs ; and as to the brain, 

 there exists a great amount of resemblance between the cast of 

 the brain-cavity in some Eocene Bats, in Hesperornis^ and in 

 Scaphognalhus. 



Only in one point is there a pronounced difference between the 

 Pterosaurs and the Bats, and this is in the transformation of the 

 phalanges of the pes and manus. While in the Pterosaui-s a i-educ- 

 tion of the claws takes place in the pes, and they remain present 

 in the manus, in Chii'optera exactly the ojaposite happens ; but 

 this divergence is easily understood when we consider that the 

 Chiroptera had, in consequence of adapting four fingers to flight, 

 only their hind feet at their disposition, for resting and suspend- 

 ing on branches, while the Pterosaurs, which developed only one 



