ANATOMY OF THE SUOE-BILL. 697 



Uameras. — A good deal of attention has been paid to the 

 anterior surface of the proximal end of the humerus in birds, and 

 I have compared the conditions presented by Balo'iiiceps^ Scopus, 

 Storks and Herons. Bakeniceps is markedly different in this 

 respect from the others and I'esembles much more closely the 

 structure shown by the same bone of the Pelican, which has 

 been adequately figured by Pycraft (35, fig. 6, p. 90). The 

 expanded end of the humerus is verj^fiat a,nd I'ather symmetrical. 

 The coraco-humeral groove is extremely faint, rather better 

 marked than in the Pelican but entirely different from the deep 

 transverse groove shown by Sco'pns. Storks and Herons. The 

 crista inferior is hardly raised above the surface ; only the 

 faintest distal groove separates it from the shaft of the bone. 

 In the Pelican it is much more salient ; Py craft's figure rather 

 under-represents its prominence. In Herons the crista is rather 

 like that of Bakeniceps, but more prominent ; in Scopus it is still 

 more prominent, and in Storks, including Taidcdiis, it is even 

 more prominent. The pectoral crest is still longer and better 

 marked in Bakuniceps than in the Pelican, but is generally 

 similar, and in both the oval impression for the insertion of the 

 second division of the pectoral is very clear. There is no trace 

 of this in the Herons. In the Storks, including Tantalus, it is 

 well-marked. This region of the humerus in Scopus is charac- 

 terized by the enormous development of the pectoral crest, which 

 is very much larger than in any other of the birds with which I 

 am dealing here, and there is a faint oval impression which 

 certainly seems to imply the existence of a divided pectoral in 

 that bird, as in Storks and Bakeniceps (see supra, p. 676). 



Hand. — The hand of birds is usually described as containing 

 the representatives of three digits, of which that corresponding 

 with the index finger (on the usually accepted supposition that 

 the pollex is present) is the largest. The proximal phalanx of 

 this digit has usually a broad flange on the ulnar side on which 

 one or more of the primary quills rest. In dissecting this region 

 in Bakeniceiys I was struck by the way in which the ai-rangement 

 of muscles suggested that this phalanx represented the phalanges 

 of two adjacent digits united by a narrow sheet of bone. On 

 examining the skeleton in Bakeniceps itself, Scopus, Arcka and 

 a number of Storks, the suggestion is still more striking. The 

 broadened phalanx is obviously thickened along the ulnar and 

 radial bordei's, and if these bordei'S were the phalanges of two 

 digits united by an ossified sheet of fibre the structure would be 

 intelligible. The point requires investigation, both by com- 

 parative anatomy aiid embi'yology, and I mention it here, only 

 to direct attention to it, but I shall be surprised if it does not 

 turn out that the hands of these and many other birds show four 

 not three digits. 



Tibial bridge. — The bony canal for the tendon of the extensor 

 muscle of the digits is complete in Bakeniceps as in Scopus, Storks 

 and Herons. 



