264 OSTEOLOGY OF TDBINARES AND STEGANOPODES. 



The angles are truncate, slightly produced below, and the articular 



cups show on their upper sides at the usual sites the large pneumatic 

 foramen on either one. Much of the skull proper is likewise permeated 

 by air. 



Of the vertebral column and the rest of axial skeleton. — What I have 

 said of the vertebral column of Oceanodroma applies almost exactly to 

 the column in this Fulmar; the twenty first vertebra, however, in both 

 the common Fulmarus and F. rodgersii anchyloses as the anterior one 

 of that series which goes to form the sacrum between the pelvic bones. 



Its ribs in consequence meet a pair of costal ribs below, that in their 

 turn articulate with the sternum. 



Rodgers' Fulmar lias the hyapophysial canal of the cervical series 

 passing through the sixth to the tenth vertebra, inclusive. 



Moreover, in the dorsal region we fiud this bird differs from the 

 Petrel in having fully developed metapophyses linking the outer ends 

 of the transverse processes together. 



The eleventh, twelfth, and thirteenth vertebrae all have a large, sin- 

 gle hypapophysis. This is not so long, and has lateral, basic processes 

 in the next two, while through the dorsal series it becomes gradually 

 longer, then shortens again, to appear for the last time as a minute point 

 on the first sacral. From the second to the fourth, inclusive, the cerv- 

 ical vertebrae not only have these hypapophyses, but equally well- 

 developed neural spines. The latter gradually disappear in the next 

 two, and the arterial canal supplants the former. 



The skeleton of the trunk of Fulmarus glacialis, from the specimen 

 collected in the North Atlantic by Ludwig Kumlien in 1877, has been 

 allowed to remain nicely articulated, with all the bones in their normal 

 positions. In it the sternum is very short and concave, while the six 

 vertebral ribs descend almost directly to reach the costal ones, and 

 other particulars are observed in which it agrees with F. rodgersii. 



Epipleural appendages which belong to the ribs of this Fulmar au- 

 chylos with their borders, and never overlap more than the next suc- 

 ceeding rib behind them. 



Occasionally among the vertebra we will fiud one that shows a pneu- 

 matic foramen, but I believe the ribs are solid, and air does not gain 

 access to the interior of any of them. Sometimes in the last pair of 

 sacral ribs one or the other may be but feebly developed and not have 

 any free haemapophysis to meet it below. Such is the case in the skele- 

 ton of Rodgers' Fulmar before me. 



Kinship with the Albatross unmistakably crops out in the sternum 

 of this bird. A glance at the figures is enough to satisfy oue of this 

 fact. 



In outline the bone is quite- square, and although in some specimens 

 the xiphoidal border is, like in the oue I figure, jagged to an extent that 

 leads us to believe it to be without any regular pattern, I have, never- 

 theless, sufficient material before me to prove that the tendency of the 

 bone is to become doubly notched. 



