1885.] MISS B. LINDSAY ON THE AVIAN STERNUM. 697 



lateral processes, and for a crescent-shaped ridge at the end. The 

 former character is easily understood if we regard these processes as 

 homologous with those of the Ratitse, and consider the metasternum 

 to have received special elongation at a late date. Plate XLIV. 

 fig. 9 shows a condition which supports this view ; the median part 

 of the nietasternum is here not longer than the process, while in 

 tlie adult it is considerahly longer ; also the process is continued 

 anteriorly into a ridge forming part of the undoubtedly costal 

 sternum, which may be distinguished from the thinner cartilage of 

 the metasternum adjoining it towards the median line. 



The crescent-shaped ridge marks the posterior insertion of the 

 pectoralis minor, which thus lies as it were in a depression of the 

 sternum ; this muscle is remarkable for its strength and its great 

 elongation, taken in comparison with those of the pectoralis major. 



(iii.) The Gull. 

 (II specimens, from about 16 days' to 5 days' embryos.) 



1. In several individuals of about 14 days there was traced a thin 

 ventral portion of the rectus continuous with the pectoralis major. 



2. Two specimens of about 11 or 12 days exhibited traces, con- 

 sisting in greater thickne-^s and greater strength of the fibres in the 

 regions indicated, of a division of the rectus into tliree bauds, a 

 median, and two lateral attached to the posterior-lateral processes. In 

 the adult the lateral part of the rectus is, on the contrary, very thin. 

 The embryonic condition recalls that of the Ostrich ; possibly the 

 width of the sternum, which is somewhat broad for its length, may 

 be associated with the early lateral thickness of the rectus. 



3. The intercostales externi exhibit in the later stages the changes 

 already described in the Guillemot ; the supercostal sheet at the 

 stage in which it is continuous sends a thin continuation under the 

 pectoralis major, which passes over the sternum and is attached to it 

 near the borders of the pectoralis minor. 



A 5 days' embryo atibrds the clue to the intercostal nature of 

 the primitive transverse bands seen in the previous types ; bands 

 precisely similar are seen alternating with and attached to the ribs, 

 not overlying them as in the former cases ; while part of their dorsal 

 extent is already fused to form the continuous supercostal sheet 

 referred to above. The connecting link between the supercostal and 

 intercostal state of these bands is seen in the Chick, ■where the bands 

 overlie the ribs, but where they are found on dissection to alternate 

 with them in the cervical region. 



4. The metasternum has two pairs of processes: of these the 

 outer is identified as homologous with the posterior-hiteral process in 

 the Guillemot and Ostrich by the fact that its outer border affords the 

 sternal attachment of the obliquus externus. The inner one is there- 

 fore of later origin ; perhaps it owes its existence to the peculiarity 

 of the posterior end of the pectoralis major, of which the lateral ])art 

 is unusually strong and longer than the median, and is attached to 



