472 



BAROX FRANCIS NOPCSA ON THE 



If we wish to identify the furcula of birds with the clavicle of 

 other vertebrata, we must assume that, first of all, the clavicle 

 managed to slip along the anterior border of the scapula, then, to 

 pass the glenoid fossa without, however, affecting the motion 

 of the humerus by so doing, and finally to attach to the coracoid. 

 Worse than that, we must also suppose that in the middle of the 

 body a very curious, almost mysterious, and unique reduction 

 managed to wipe out even the last ontogenetical trace of the 

 interclavicle without afl^ecting the clavicles. Surely it is some- 

 wliat risky to evoke a hypothesis of this sort. 



Suppo.smg now, instead of these changes, that the furcula is an 

 ossified tendon that was developed when birds began to glide to 

 strengthen a reduced scapular arch, it is easy to explain its 

 origin on the coracoid, the primary fusion of the median ends, its 



Text-figure 7. 



Scapula and coracoid in Dinosaurs and birds. 

 (A) Gorgosa-urus ; {W) Archceopteri/x ; {0} Struthio ; (D) Nothura. 



successive fixing on to the sternum, and the absence of clavicles 

 and interclavicle. Of course it is necessary in this case to fall 

 back on the hypothesis of a terrestrial and bipedal Proavis in 

 which the shoulder-girdle had suft'ered reduction. 



In accordance with some old zoologists, I consider the furcula 

 to be an ossified tendon, and not to be homologous with the 

 clavicles in reptiles. tSince this supposition agrees with the 

 results hitherto obtained, I thinJc this hypothesis is the right 

 one. 



All evidence brought forward up to the present shows that the 

 fiying state of birds was preceded by a state of bipedal running. 

 Having established this, of course it becomes necessary to investi- 

 gate when and where pi-imitive feathers and passive flight began. 



If passive flight and feathers arose after cursorial reptiles had 



