1888.1 PROCEEDINGS OF UNITED STATES NATIONAL MUSEUM. 277 



terior portion of the united niesetknioid and rostrum. Their inner 

 laminae are produced forward to anchylose with the limbs of the vomer. 



This latter boue is quite an extraordinary structure in an Albatross, 

 and differs not a little in the various species. To get at its exact shape 

 and relations to the surrounding structures I found that I was obliged 

 to cut away certain portions of a spare skull and remove it, together 

 with the pterygoids and mutilated parts of the palatines. From this 

 specimen I made the drawings presented in Figs. 1G and 17. 



Viewing this from above, we find that all three bones contribute cer- 

 tain of their parts to form a deeply excavated, longitudinal groove that 

 extends the entire length of the structure. During life the spear-shaped 

 rostrum rides in this, occupying, however, but the hinder two- thirds of 

 the channel. Seen from the side, we find that the vomerine portion of 

 this rostral bed is continued downward and forward as a median cari- 

 nation, which anteriorly curves down between the maxillo-palatines, to 

 have its apex finished off in a little foot-like process which appears, as 

 above described, in the interpalatine cleft. 



Fig. 15. Posterior view of the skull of Diomedea hrachyura; mandible removed; life size, 

 author from same specimen as Figs. 12 et seq. 



By the 



In the undiagnosed skull of an Albatross (No. 16738) the pterygoids 

 and palatines behave in the same way as in the Short-tailed variety, 

 but the vomer shows no mid channel beyond the end of the sphenoidal 

 rostrum, is fully double the width of the other, and rather reminds us 

 of the extraordinary vomer of Eodgers' Fulmar. Anteriorly, however, 

 its tip is carried down to appear in the inter-palatine cleft, as in I). 

 albatrus. Forbes figures a vomer similar to this for D. exulans in his 

 Challenger memoir. 



The maxillo-palatines are large, compact, elliptical plates. They stand 

 but a few degrees removed from the vertical plane, each facing outward 

 and slightly downward. Upon a lateral view the mandibular side 

 nearly shuts this bone out of sight, and it is only in certain positions 

 that we can secure a good look at it. The surface next the median 



