4 BrLLETIN 50, UNITED STATES NATIONAL MUSEUM. 



do, concerning the position or rank of certain fox'ms, it of course 

 becomes necessary to consider carefully the points of disagreement, to 

 weigh impartiallj' the evidence and arguments adduced by the advocates 

 of such divergent decisions, and, if possible, decide independentlj^ as 

 to the seemingly better allocation of the form in question. This has 

 been attempted in the following scheme, but the author is fully con- 

 scious that his disposition of such doubtful cases may still not be final, 

 especially when different from previous decisions, as in a few cases 

 has been the result of his revision. 



It is unfortunate that those who possess the most thorough knowl- 

 edge of avian anatom}^ and morphology do not always seem to have 

 succeeded in satisfactorily diagnosing the groups which they adopt, 

 nor in clearly presenting a synthetic summary of the facts revealed 

 through their investigations. So-called diagnoses are sometimes 

 found, when carefully analyzed, to be really not diagnostic at all; 

 more often thej'' prove to be so in part only. A by no means extreme 

 example, the case of the Limicolse and Lari (suborders of the Order 

 Charadriiformes), as defined in Gadow's Classification of Vertebrata 

 (1898, p. 35), may be given for illustration. These two groups are 

 thus characterized by Dr. Gadow: 



LimicoljE. — "Nidifugous, schizognathous, without spina interna 

 sterni; hypotarsus complicated." 



Laei. — ' 'Aquatic, schizognathous, vomer complete. Without basip- 

 terygoid processes. Front toes webbed; hallux small or absent. Large 

 supraorbital glands." 



Of the characters mentioned in these two diagnoses the following are 

 common to the two groups, and therefore are not diagnostic of either: 

 (1) Nidifugous young;' (2) schizognathism; (3) complete vomer; 

 (4) absence of spina interna sterni; (5) absence of basipterygoid proc- 

 esses;' (6) webbing of front toes;' (7) small or obsolete hallux; 

 (8) aquatic habits.* 



The difl'erential characters of the two groups are thus reduced to 

 the following: 



LiMicoLiE. — Hypotarsus complicated; supraorbital glands small. 



Laei. — Hypotarsus simple; supraorbital glands large. 



With this example of so-called Suborders which are characterized 

 by a verjr small number of relatively unimportant differences may be 

 contrasted that of the several subdivisions of the Order Gruiformes. 

 to which Dr. Gadow only allows family rank, although apparently far 

 better characterized than the so-called suborders of Charadriiformes, 



' The young of the Lari are not, it is true, strictly nidifugous, but they are often 

 more nearly so than nidicolous. 



'' Basipterygoid processes are present in some Limicohv, absent in others. 

 'The front toes are webbed in some Limicolje (e. g., Himantopus). 

 * The Phalaropes are quite as truly aquatic as Gulls. 



