that connect the Orders and Families of Birds. 421 



that the tribes which meet our earliest attention are those respect- 

 ing which our information is most deficient ; and where, conse- 

 quently, we must feel most hesitation in deciding on the affinities 

 which connect them, and the subdivisions into which they branch 

 out. I shall not therefore at present dwell at much length upon 

 this order, but shall content myself with a few brief observations, 

 more particularly on such of the groups as containing European 

 species, and being of course more open to examination, may be 

 spoken of with some degree of confidence. 



There are three important groups in this order, which form 

 distinct and prominent subdivisions of it ; the families of Vultu- 

 rida, Falconida, and Strigidce, corresponding with the Linnean 

 genera Vultur, Falco, and Strix. To these may perhaps be added 

 a fourth group, the Gypogeramis of M. Illiger, which, though it 

 has sometimes been disposed in a different order, is now, I believe, 

 generally admitted to be a Bird of Frey. In this arrangement 

 I feel every disposition to acquiesce. The essential characters 

 observable in the structure of the Gypogeranus accord accurately 

 with those of the Raptores ; and it bears a resemblance to the 

 Grallatores, among which it is sometimes arranged, only in the 

 length of its tarsi. But in judging by such insulated characters, 

 however striking or important, as only indicate a similarity be- 

 tween groups not otherwise connected together by equally essen- 

 tial particulars, it becomes necessary to consider whether the 

 conformity of these characters may not be the result of a con- 

 formity in habit or situation, which may incidentally assimilate 

 the groups in which they are found, rather than of a natural affi- 

 nity that determinately unites them. In other words, we ought 

 to reflect whether an analogous mode of life or an analogous 

 place of resort may not give rise to a partially analogous confor- 

 mation. In the present case, the sandy and unsubstantial nature 

 of the plains which these birds frequent, appears to me to demand 



3 I 2 a cor- 



