174 PROCEEDINGS OF THE ACADEMY OP 



tnral modifications ; and as such unavailable for the truly scientific definition 

 of groups of a higher grade than families. 



In calling attention to the foregoing considerations, I wish to be understood 

 as offering no opinion upon the questions involved, and particularly as by no 

 means asserting that the Halodromes are not true Procellaridians. It is rarely 

 of use to exchange one doubtful opinion for another; and for the present I 

 shall follow the usually received classification. But it is safe to affirm that by 

 the determination of the proper aflSnitiesof these birds the exact value of the 

 character of tabulation of the rhiuotheca is to be ascertained. 



Subfamily DIOMEDEIN.E. 



In a careful study of the Albatrosses, the interesting fact becomes evident, 

 that we have an easy and convenient means of accurate diagnosis of species in 

 the characters afforded us by the bill alone. All the known species differ from 

 each other by perfectly tangible and readily appreciable variations in the size, 

 shape and color of the bill ; in the configuration of its several corneous ele- 

 ments, and in the outline of the feathers around its base. This latter feature, 

 conjointly with the shape of the corneous covering of the culmen in that por- 

 tion of its extent which is posterior to the nares, gives us such reliable data 

 that we need hardly enquire further. I shall, theiefore, in the following pages 

 confine myself chiefly to detailed descriptions of the bill ; and it will be noticed, 

 as supporting the foregoing assertions, that a synoptical table may be drawn 

 up solely upon the characters mentioned above. 



As we shall study the bill somewhat in detail, I introduce, for convenience 

 of description, several words expressive of the different corneous elements 

 which cover it; the meaning of which will be obvious. I may remark that the 

 piece interposed between the inferior mandibular rami at the lower border of 

 their symphysis (here called the " interramicorn,") is a feature which also 

 definitely characterizes this group, as it is present in no other. The presence 

 of a well defined membranous fringe on the exterior toes is also highly charac- 

 teristic. 



In the following pages I describe eleven species — one of them supposed to be 

 new — and indicate the possible existence of a twelfth. Of these one differs so 

 much from the rest that it may be properly made the type of a genus distinct 

 from Diomedea. . The remaining species have also been subdivided into several 

 genera, chiefly by Prof. Reichenbach. Such a collocation of species is cer- 

 tainly natural, regarded as simply expressive of the fact that certain of them 

 are more intimately allied to each other, than they are to the species of another 

 group ; but the differences presented seem hardly sufficient to warrant our at- 

 taching generic import to them. The following will serve to explain the point 

 alluded to. 



Group A. Comprising ea;M^aw.s, brachyura, nigripes, gibbosa. Of largest and 

 medium size. The bill is very broad, stout and heavy: and especially very 

 wide at its base, and is uniform in color. The colors of the plumage are white, 

 variegated with black, especially upon the wings ; or uniform fuliginous. Tbe 

 tail is very short. The nostrils are large, and wide. Exulans may be consid- 

 ered as typical of this group. The length of tail reaches its minimum in bra- 

 chyura, upon which character Prof. Reichenbach founds his genus Phahas- 

 Iria. 



Group B. Comprising melanophrys, GUHana, n. sp. catiia, culminata, chloro- 

 rhyncha, olivaceirontris. Of medium and rather small size. Bill shorter, weaker, 

 and considerably compressed, usnally bright or parti-colored. White, with black 

 back and wings. Tail long, slightly rounded. Melanophrys may be taken as the 

 type of this group, which constitutes the genus Thalassarche Reich. Both mela- 

 nophrys emd Gilliana differ from the other three species in the character of the 

 culminicorn, as will be hereafter more particularly elucidated. 



So varying are the characters of shape of bill, outline of frontal feathers, 

 length of tail, etc., that I think they can hardly be made typical of distinct 



[May, 



