14 THE VOYAGE OF H.M.S. CHALLENGER. 



oesophagus, instead of passing into the thorax directly in front of the vertebral column, 

 are so to speak pushed away to the right side of the latter at the lower part of the neck, 

 and at their entry into the thorax lie between the right side of the vertebral column and 

 the right limb of the furculum. In every species of Penguin which I have had an 

 opportunity of examining this arrangement holds good, although there seems to be no 

 reason why the tubes in question should not pass into the thorax in contact with the left side 

 of the column. The strongly-developed flexures of the neck in the Penguins are associated 

 with the maintenance by these birds, when on land, of the characteristically erect attitude, 

 and serve to bring the centre of gravity of the head and neck over the base of support 

 formed by the feet. In the genera which include the smaller species {Eudijptes and 

 SphenisGUs) the cervical curves are more strongly developed than in those which include the 

 larger {Pygosceles and Aptenodijtes), but the arrangement is essentially the same in all. 



Dorsal VertebrcB. 



The dorsal vertebrae, understanding by this term all those which possess moveable ribs, 

 are nine in number in every species which I have examined. The last dorsal vertebra 

 is immovably fused by means of its body, arch, spinous, and transverse processes with 

 the first lumbo-sacral vertebra. 



The bodies of the dorsal vertebrae in Euchjptes chri/socome from Tristan d'Acunha 

 are much compressed from side to side. The articular surfaces of the bodies of the 

 first and second,^ and the anterior articular surface of the body of the third dorsal 

 vertebrae are saddle-shaped, and resemble the corresponding surfaces of the cervical 

 vertebrae. The succeeding dorsal vertebrae difi"er from these inasmuch as they are 

 opisthocoelous, the anterior surfaces of their bodies being rounded and globular, while 

 their posterior surfaces are deeply concave, so that the convexity of any given vertebra 

 is received into the concavity of the body of the vertebra next preceding. In this respect, 

 as pointed out by Owen," and Gervais and Alix,^ this portion of the vertebral column of 

 the Penguin bears a remarkable resemblance to that of many reptiles. The body of the 

 last dorsal is immovably fused with that of the first lumbo-sacral vertebra. On either 

 side of the body of each dorsal vertebra, close to the anterior extremity, is a facet for the 

 reception of the head of a rib. 



The lower surfaces of the bodies of all the dorsal vertebras, with the exception of the 

 first and ninth, are provided with well-developed hypapophyses. In the first and ninth 

 these processes are absent. Those connected with the second, third, and fourth dorsal 

 vertebrae have bifid extremities which afi'ord additional surface of attachment to the 



' According to Gervais and Alix (Osteologie et Myologie des Manchots, p. 3), the body of the second dorsal vertebra 

 differs from those of the cervical vertebra in being opisthocffilous. Such is not the case in any of my specimens. Owen 

 (Cyclopsed. of Anatomy, Art. " Aves.," vol. i. p. 270) correctly remarks that the opislhoccelous character of the vertebrae 

 shows itself for the first time in the third dorsal vertebra. 



^ Cyclopjedia of Anatomy, Art. "Aves," vol. i. p. 270. 



^ Loc. cit., p. 3. 



