Okdbe TUBINARES, 

 the tube-nosed swimmers. 



CllAK. Swimming birds with tubular nostrils, the horny covering of the bill 

 consisting of several distinct pieces, separated by more or less marked grooves. 

 Terminal portion of maxilla produced into a strongly hooked unguis. Feet fully 

 webbed, anteriorly. Hallux rudimentary, consisting of an elevated sessile, often 

 minute, claw, sometimes wholly absent. Wings usually very long. Basipterygoids 

 usually absent ? Egg single, white. 



The number of families into which the Order Tubinares is properly divisible is an 

 unsettled question. In a "Report on the Anatomy of the Petrels (Tubinares)," 

 which forms the leading article of Vol. IV. of the Zoological Reports of H.M.S. 

 " Challenger," the late Professor W. A. Forbes divides the Tubinares into two fami- 

 lies as follows : (1) ProcellariidtP, including as sub-families Procellariinoi and Diome- 

 deinai ; and (2) Oceanitida', composed of the genera Fregetta, Pelagodrovia, Oceanites, 

 and Garrodla. According to this arrangement, the Albatrosses are held to be much 

 more nearly related to the genera Procellavia, Cymochorea, and Ilalocyptena than are 

 Oceanites and the other Oceanitidoi ^ — a proposition which, notwithstanding the 

 reasons advanced,^ we are not prepared to accept. 



The arrangement we have to propose is not supposed to be a perfectly natural one, 

 but there can be no question as to the naturalness of the groups defined below : — 



1, Diomedeidde. Wings very long and narrow, on account of the extreme development of 



the humerus and ulna. Remiges 39-50 (the largest numher in any known hird). Nasal 

 tuhes lateral, widely and completely separated by tbe intervening " culminicorn." No 

 hind toe. Size very large. 



2. Procellariidae. Wings lengthened, hut of different structure from the preceding (remiges 



20-39, usually about 30). Nasal tubes near together, laid side by side upon the cul- 

 men, the nostrils opening anteriorly. Hind toe present, though sometimes minute. Size 

 extremely variable. 



1 ". . . In spitt of the general superficial resemblance of the Occrtn2<i(?(B to the smaller forms of Procc?- 

 lariidcc, with which all ornithologists previous to Garrod had confounded them, the differences between the 

 two families are, it will be seen, numerous and important. The special points of resemblance wliich the 

 Oceanitklce have with such Procellarian genera as ProccUaria and Cymochorea — such as the general small 

 size, style of coloration, form of skull, comparative simplicity of the tensor 2Mtagii arrangement, simple 

 sternum and syrinx (the last three peculiarities being also common to Pelecanoides) — may best be 

 explained by supposing that these small Procellarian forms are on the whole less specialized than the 

 larger ones (Fulmars, Albatrosses, Shearwaters, etc.), and so retain more of the characters possessed by 

 the primitive and now extinct common form from which both the Procellariidcc and Oceanitidae must 

 have been derived" (Foubks, t. c. p. 56). 



"^ "According to modern ideas, the object of a classification is not so much to rejiresent morphological 

 facts as to indicate the phylogcnetic relations of the diiferent forms concerned " (Foubes, t. c. p. 58). 



