1895. THE STRUCTURE OF ARCH&OPTERYX . 115 



the London specimen, though it may be that the supposed cast of the 

 brain (!) shown at b, in Fig. 2, is a portion of the skull. 



The ribs, both vertebral and ventral, are very slender. There 

 are no uncinate processes visible. 



Of the sternum nothing is known, though much has been 

 written. In the Berlin specimen it probably lies still hidden in the 

 matrix. The position of the ventral ribs shows that it must have 

 been small. 



The scapulas in the Berlin specimen were broken in exposing 

 the specimen. The right one is easily recognisable in the plate. 

 They are flat curved bones, not unlike those of a modern bird. 

 Their length is 43 mm. or thereabouts, according to Dames. In the 

 photograph only a portion is seen. 



The coracoids are in the Berlin specimen largely hidden. I 

 have not specially examined what portion is exposed in the London 

 specimen. The dorsal ends are exposed in the Berlin specimen 

 and possess a furcular tuberosity as in other birds. 



Fig. 1. — The Skull of ArchaopUryx. After Dames. (Natural size.) 



Of the furcula, a small portion is seen at the left shoulder of 

 the Berlin specimen. It was, however, imperfectly exposed at the 

 time when the photograph was taken. A larger portion is seen in 

 the London specimen. It is a characteristically avian furcula, 

 U-shaped ventrally, and articulating with the furcular tuberosity of 

 the coracoid at each shoulder. 



The humerus is a well-developed bone in each wing. Its form 

 and dimensions may be seen in the plate of my former paper (vol. 

 iii., p. 275), as also in the one accompanying the present communi- 

 cation. In the London specimen the inner surface of the right 

 humerus (called "left" by Owen) is seen on the right side of the 

 specimen (see h' in Fig. 2), while both are seen from the outer 

 surface in the Berlin specimen. It differs from that of other 

 birds in being devoid of the pectoral crest or ridge for the 

 insertion of the great pectoral muscle. As Dames points out, this 

 confirms his view that the sternum must have been small, as must 

 also the great pectoral muscle. In the plate illustrating my previous 

 article the proximal end of the humerus is covered by a portion of 

 the matrix, which has since been removed (at 11 in Plate I. of the 

 present article), and that plate consequently gives an impression of a 



