696 MISS B. LINDSAY ON THE AVIAN STERNUM. [JuDC 16, 



(Spheniscidfe), partly overlying the pectoralis medius, a muscle 

 described by Watson and named by him the Dermo-humeralis, which 

 takes its rise from the liead of the humerus, and forms in tlie 

 abdominal region a thin band of longitudinal fibres lying near the 

 median line : this is, in other words, a ventral part of the rectus 

 continuous wich a lateral part differentiated from the pectoralis. In 

 these birds also the rectus is comj^lex, consisting of two sheets 

 (Watson opud Bronn). 



B. Middle sheet, most massive of the three, and attached to the 

 posterior-lateral borders of the sternum, which apparently corre- 

 sponds with the rectus of other birds. 



C. Thin inner sheet, which has the transversus attached to its 

 edges. This takes attachment after passing inside the sternum for 

 nearly one third tbe whole length of the latter. In the later stages 

 its attachment travels backwards towards the end of the sternum, till 

 in the embryo of 17 days it is completely united with the main body 

 of the rectus, while the transversus has acquired its usual position 

 with regard to the latter. This late ontogenetic cliange in the rectus 

 seems to represent the posterior translation of the attachment of the 

 rectus, during the late phylogenetic development of the long meta- 

 sternum characteristic of this type, which apparently splits the muscle 

 in two, and carries the upper part backwards as it grows. 



3. In the earliest stage of the Guillemot, as in the Ostrich, there 

 are seen transverse bands of muscle ; these are afterwards lost, 

 and fused into a uniform supercostal sheet continuous with the 

 obHquus externus. This supercostal sheet eventually, in the latest 

 stages examined, acquires attachment to individual ribs, while at the 

 same time the obliquus externus acquires an attachment to the ribs 

 and lateral borders of the sternum ; the phylogenetic meaning of 

 these clianges is obvious. 



4. There are indications of a slight posterior shortening of the 

 costal sternum. There are 7 sternal ribs in tlie adult, but in the early 

 embryo 9 ; of these the 7th, 8th, and 9th have no uncinate processes. 

 Fig. I. (p. 690) shows an adult specimen in which the 9th rib is 

 not only free at its sternal end, but atrophied also towards the 

 dorsal region of its spinal parts. This mode of atrophy at both ends, 

 which leaves a rudiment of the middle part of the rib, occurs again 

 in the Chick. In the Ostrich, as has been seen, atrophy begins at or 

 near the sternal end, and travels uniformly towards the spinal region. 

 Neither in the Chick nor in the Ostrich do we find that marked 

 division between spinal and sternal portions of the rib which in the 

 case of the rudimentary 9th rib of the Guillemot divides it into two 

 pieces. 



The allied species, TJ. brunnichii, may be compared with this ; 

 the adult has also 7 sternal ribs, and 2 posterior ribs without uncinate 

 processes, which do not reach the sternum. 



There is also some indication of anterior shortening of the sternum, 

 for one of the early embryos shows the attachment of a rib, which in 

 the adult {cf. Plate XLIY. fig. 9) falls short of the sternum, 



5. The metasternum is remarkable for the great reduction of the 



