Pico-Passerine Group of Birds. 33 



This character excludes the Tubinares. 



Each of these six characters appears in every Pico-Passerine 

 bird ; each of them is also found in some bird which is not 

 one of the Pico-Passeres ; but the combination of the six 

 characters is never found in any bird which is outside the 

 limits of the Pico-Passeres. 



The order to which we have restricted the name Pico- 

 Passeres may be easily divided into five suborders, which 

 appear to be natural groups. 



Passeres. 

 There has been some difference of opinion as to the exact 

 boundaries of this group. If the Passeres be diagnosed as 

 agithognathous birds, with a cacum and a nude oil-gland, the 

 Eurylsemidse will be admitted and the Upupidae will be ex- 

 cluded. If they be diagnosed as birds with free plantar 

 tendons, and a spinal feather-tract uninterrupted between the 

 crown and the upper back, the Upupiclte will be admitted 

 and the Eurylaemidse excluded. If they be diagnosed as 

 agithognathous birds with free plantars, both the Eurylsemidae 

 and the Upupidse will be excluded. The last-named diag- 

 nosis is so incomparably the best, the two characters accepted 

 being so very much more exclusively Passerine than the two 

 rejected, that it seems wisest to adopt it, with all its con- 

 sequences. The Passeres may therefore be diagnosed as 

 follows : — 



1. They are segithognathous. 



2. The hind plantar is free from the front plantar. 



The division of the Passeres into families is a complicated 

 question, which must be deferred to a future paper. 



Eurylcemi. 

 If the Broadbills be regarded as too aberrant a family to 

 be admitted into the suborder Passeres, they must be allowed 

 to constitute a suborder by themselves. They are unques- 

 tionably very nearly allied to the Passeres ; they possess 

 caeca, the oil-gland is nude, and in their pterylosis they 

 agree with Hirundo and other aberrant Passerine birds. 



SER. VI. — VOL. II. i) 



