230 ^ 



which, in the inaia, supports the one founded on the osteological struc - 

 tare. 



A. — Dorsal portion of the spinal tract deeply divided, and each branch 

 dilated exteriorly Falcones 



B. — Dorsal portion of the spinal tract enlarged on all sides, undivided, 

 and sparsely feathered. 



a. — Eyelids with distinct lashes ; lumbar tract present ; dorsal por- 

 tion of the spinal tract sparsely feathered to the caudal pit, 

 thence diminished and continued as a narrow band along 

 the caudal vertebrae to the oil-gland Herpetotheres. 



J). — Eyelids without distinct lashes; lumbar tract absentj dorsal por- 

 tion of the spinal tract in the form of an elongated ellipse, 

 only contracted into a band immediately in front of the oil- 

 gland (but even there still broad), consisting exclusively of 

 scattered feathers, which become stronger posteriorly. 



Micrastures. 



The only close relationship between any two groups of this subfamily 

 is seen in the generalized forms of the Falcones and Polyhori {Sieracidea 

 and lfiZ«;apo), whose specialized forms {Falco and Polyborus) are so ex- 

 tremely dissimilar in appearance. In the two genera mentioned, the aj)- 

 proach is so very close as to almost form a transition between the two 

 groups.* There is a wonderful similarity in the general form and relative 

 proportions of all the parts, the arrangement of the scutellae of the tarsi 

 and toes, the character of the plumage, and the size of the species. But 

 notwithstanding this apparent correspondence of external characters, 

 they are found to differ in all those osteological characters diagnostic 

 of their respective groups, and, when their external structure is ex- 

 amined closely, agree severally with the other members of the groups to 

 which they belong in an apparently trivial yet really pertinent character, 

 viz, the cutting of the inner ioel)s of the outer primaries, which is always 

 essentially different in the two groups. The external diagnostic charac- 

 ters (associated with osteological, indicated on p. 229) are thus reduced 

 to the structure of the i)rimaries, as follows : 



Falcones. — Two, or less, outer primaries with their inner webs cut! 

 this always an abrupt emargination on the first, and situated near its 

 end. Second or third quill longest ; first longer than the fifth. 



PoLYBOEi. — Three, or more, outer primaries with their inner webs cut ; 

 this an oblique sinuation on the first, and near its middle. Third or 

 fourth quill longest ; first shorter than the fifth. 



The Falcones comprise very few genera in proportion to the number 

 of species, which is very considerable ; but it is outside the purpose of 

 this paper to discuss the subject of what the genera of the group are. 

 Suffice it to say that neither Baza nor Rarpagus belong here, as many 

 have supposed, being members of different groups {Femes and Ictinice) 

 of the subfamily Buteoninw. The genus Spiziapteryx, however, which 

 many consider a synonym of Harpagus, is a true Falcon. 



The groups Micrastures and Rerpetotheres have but one genus each, so 

 these are passed by without further notice ; but the Polyborine genera 

 are numerous, and, as they have never been satisfactorily defined, the 

 following diagnoses are presented : 



* The typical Hieracidea will, upoQ examiuation, probably be found to have the super- 

 ciliary process of the lachrymal much shorter than that of the typical Falcones, if it does 

 not approach the extreme brevity of this bone which characterizes the Polyborine group. 



