Shoulder-girdle in Cretaceous Ornithosauria. 439 



surface from which an anterior part has come away. And if 

 this specimen is compared with the neural arch of the sacral' 

 vertebra (pL x. figs. 8, 9) it will be found to correspond en- 

 tirely." The vertebra referred to is singular in the circum- 

 stance that the transverse process has a higher position 

 relatively to the neural canal than is usual, the infra-neural 

 articulation preserves the concavo-convex articulations, while 

 the neural spine shows a vertical suture with the arch next it. 

 The neural spine of this vertebra is not absolutely the same in 

 character with the bones already referred to. It is a remarkably 

 thick wedge, forming more than half the height of the vertebra. 

 Its sides are smooth, vertical, and flattened, with a defining 

 ridge above the transverse neural platform. I believe this 

 vertebra to be one of three which I suppose to have been anchy- 

 losed by their neural arches in OrnitJiocheirus^ to form the 

 articulation for what is usually the fx-ee extremity of the 

 scapula. And it follows that the other specimens to which I 

 have referred are portions of separated neural arches of this 

 structure, in which three consecutive neural spines are blended 

 together. 



This difference of condition from an ordinary pelvic sacrum 

 is exactly what might have been anticipated, for the ordinary 

 transverse processes evidently carried costal ribs, though they 

 are not shown in English specimens of dorsal vertebrae of 

 Ornithocheirits, and therefore the mechanical stimulus to ossi- 

 fication was necessarily absent from the infra-neural parts 

 of the vertebrae, which in tiie pelvic sacrum has blended the 

 vertebrse together. All these bones are in the Woodvvardian 

 Museum of the University of Cambridge. 



If the interpretation of them which I offer is legitimate, 

 it is probable that the portion of the neural arch which 

 is impressed with the vertical, parallel, transversely ovate 

 facets, concave from front to back, for the scapular arti- 

 culation, is a distinct ossification imbedded in the neura- 

 pophyses, comparable to the neural spines of lower Vertebrates, 

 which have not usually a separate existence in the Ornitho- 

 saurian skeleton. It has every appearance of being a separate 

 bone, but it is manifest that the evidence of its relation to 

 the vcrtebrge is imperfect. I have published figures of all 

 the materials, and offer a restoration of the scapular ossifi- 

 cation on which I have drawn the outline of the most 

 complete supra-neural bone which has been found separate 

 (fig. 1). My reason for including three vertebras is 

 based upon a comparison of my figures in the ' Ornitho- 

 sauria,' pi. X. figs. 8, 9, and pi. xii. fig. 17, with pi. xii. 

 figs. 15, Ifi, and the figure of 1859 already quoted in Owen'.s 



