176 FOKMICARliniE. 



Family II. FORMICARIIDiE. 



The Formicarians were associated by the older authors partly 

 ■with the Shrikes (Laniido'), and partly with the Timeliiue Thrushes. 

 If more obvious external characters are taken only as a p:uide, this 

 is by no means an unnatural arrangement. The formidable hooked 

 bill of one of the larger ThtmnophUi would well become a typical 

 Lanius ; while the general conformation of the ordinary Formicarian, 

 with its thickened rump-feathers, when compared with that of some 

 of the Timeliines of the Eastern Archipelago, is by no means dis- 

 similar. Dr. Cabanis, in his well known " Ornithologische iS^otizen," 

 published in 1847 *, following up the anatomical investigations of 

 Johannes Miiller, first showed that the Formicariicla were distin- 

 guishable by external as well as internal characters, if these were 

 carefully looked for. Nitzseh in his ' Pterylographie,' and SundevaU 

 in his celebrated ' Tentamen,' advanced our knowledge of the subject ; 

 and I do not presume to say that the present arrangement of the 

 Formicariida', which is mainly based upon Dr. Cabanis's plan, is any 

 more than an improved edition of the results which he arrived at. 



The division of the Formica ri'idoi into subfamilies is a matter of 

 much difficulty. As in many other cases, the diversities of the 

 extreme forms (such as Thamnophilus and Grallaria) are sufficientlj' 

 obvious. But there is a long series of intermediate stages, and the 

 question is where can we best draw a line, or lines, between the 

 groups. Upon reconsideration I have come to the conclusion that 

 I cannot maintain the three subfamilies exactly as adopted in my 

 Synopsis of the Ant-Thrushes, published in 1858 f. 



The simplest plan ajipears to be to cut off first the long-legged 

 Grallarice and their allies to form the subfamily Grallariimp. The 

 remainder may then be separated into two subfamilies — oneembracing 

 the Shrike-like Thamnophilince, the other the typical Formicariince, 

 which imitate rather the Thrushes and Sylvians. When the trachefe 

 of these numerous forms have been carefully examined, it is probable 

 that a more satisfactory way of division will be arrived at. 



The Formicariidcf are a very numerous group in the middle part 

 of the Neotropical Eegion. They thin out very much in the Trans- 

 panamic Subi-egion on the north, and still more so in the Chilian 

 Subregion on the South. Only about 20 species occur in Central 

 America ; and Mr. Hudson and 1 have been able to include only 4 

 species in the Ornithology of the Argentine Eepublic, most of these 

 being, moreover, found only near the northern limits of that country. 



* Wiegm. Arch. f. Niit. xiii. Bd. i. pp. 186, 308. 



t " Synopsis of tlie American Ant-birds {Formicarik^a)" P. Z. S. 1858, pp. 202, 

 232, 272, also separately paged. 



