BIRDS or NORTH AND MIDDLE AMERICA. 15 



jj. Rictal region feathered, postocular region partly naked 

 .smaller forms (total length abont 100 mm.) with coloration 



much variegated Hypocnemis (extralimital).« 



a. Frontal or loral regions, or both, scantily feathered (sometimes 

 quite bare). 

 j. Loral and suborbital regions (sometimes forehead and crown 

 also) bare or but scantily bristled; adult males uniform black 

 (with white markings on wing-coverts). 



Gymnocichia (p. 97). 

 jj. Loral and suborbital regions feathered (only the postocular 

 and rictal regions naked); adult males not black (or else 

 without white markings on wing-coverts). 



I. Tail less than four-fifths as long as wing; bill stouter; 



nostrils smaller, more rounded; forehead more thinly 

 feathered, the feathers semi-decomposed, eemi-erect. 



Myrnieciza (p. 103). & 



II. Tail more than five-sixths as long as wing (sometimes 



longer than wing); bill more slender; nostrils larger, 

 more longitudinal; forehead more densely feathered, the 

 feathers more compactly webbed, decumbent. 



Myrmoderas (extralimital).*' 

 dd. Plumage harder, very dense and compact, only the tail-coverts semi- 

 decomposed or loose webbed. <^ (Formicariese.) 

 e. Tail less than two-thirds as long as wing; plumage of rump not abnormally 

 long and dense; feathers of latero-frontal antiae short and dense, not 

 antrorse, not extending above nostril; bill relatively longer and less 

 depressed basally, the exposed culmen as long as middle toe without 

 claw; under parts neither streaked nor barred. 



Formicarius (p. 115). 

 ee. Tail more than two-thirds as long as wing; plumage of rump very long 

 and dense; feathers of latero-frontal antia? longer, antrorse. extending 

 anteriorly, above the nostril, to anterior end of nasal fossse; bill rela- 

 tively shorter and more depressed basally, the exposed culmen much 

 shorter than middle toe without claw Chamaeza (extrahmital).* 



o Hypocnemis Cabanis, in Wiegmann's Archiv fiir Naturg., xiii, pt. i, 1847, 212. 

 Type, Turdtis tintinnabulatus GTaelm = For7nicarius cantator Boddaert. (Guiana 

 and upper Amazon Valley; two species.) 



^ Including Myrmclastes Sclater, which I am unable to separate generically. 



c Myrmoderas Ridgway, Proc. Biol. Soc. Wash., xxii, April 17, 1909, 70. Type 

 Myiothcra loricata Lichtenstein. 



Besides the type, the following species (placed in Myrmeciza by Dr. Sclater) also 

 belong here: M. cinnamomea (Gmelin), M. ruficauda (Maximilian), M. squamosa 

 (Pelzeln), and M. atrothorax (Boddaert); prol)ably also Myrmeciza pelzelni Sclater and 

 M. heviimelscna Sclater, which, however, I have not seen. The group ranges from the 

 Guianas and Amazon Valley to southeastern Brazil. 



d I am not at all satisfied with these characters, but am unable to find better ones 

 for separating these two exceedingly distinct genera from the rest, collectively. They 

 doubtless possess marked anatomical differences, for they certainly stand clearly apart 

 from all the rest of the family. 



« Chamxza Vigors, Zool. Journ., ii, 1825, 395. Type, C. meruloides Yigors= Turdus 

 brevicaudus Gmelin. — Chamxzosa (emendation) Cabanis, in Wiegmann's Archiv fiir 

 Naturg., i, 1847, 218. (Colombia to southeastern Brazil; six species.). 



