284 BULLETIN 50, UNITED STATES NATIONAL MUSEUM. 



about half its length, the rectrices (12*^) abruptly and conspicuously 

 acuminate, with their very rigid shafts twisted and strongly decurved 

 subterminally. Tarsus about as long as exposed culmen, less than 

 one-fourth as long as wing, slender, distinctly scutellate (enda- 

 spidean) ; middle toe, with claw, shorter than tarsus ; outer toe, with 

 or without claw, equal to or very slightly longer than middle toe; 

 inner toe, without claw, reaching to a little beyond subterminal 

 articulation of middle toe; hallux much shorter than inner toe, not 

 stouter; middle toe united to outer toe by whole of its first and about 

 half of its second phalanx, to inner toe by greater part of its first 

 phalanx; anterior claws rather large, very strongly curved and 

 acute, that of the hallux slightly curved, as long as or longer than 

 the digit. 



Coloration. — Tail-coverts, tail, and remiges chestnut or rufous- 

 chestnut; pileum and back olive-brown, the former with narrow 

 indistinct streaks of paler; beneath light olive-brown, throat brown- 

 ish buff or clay color, the chest sometimes spotted with the same. 



Nidijication. — Nests in holes; eggs glossy wliite. 



Range. — Costa Rica to British Guiana and upper Amazon Valley. 

 (Four species. ^) 



This genus is distinctly intermediate between Sittasomus and Dendrodncla, and 

 I quite agree with Hellmayr that it is more nearly related to the latter than to the 

 former; but I can not believe that it is "perhaps barely separable " from Dendrodncla 

 (see Hellmayr, Bull. Brit. Orn. Club, xiv, 1904, 52), the difference in the character 

 of the tips of the rectrices (in which character Dcconychura is precisely like GZ?//3^o- 

 ryncMis and Sittasomus and very unlike Dendrodnda), even apart from the differ- 

 ence in the form of the bill, being, in my opinion, quite sufhcient to justify generic 

 separation. I have not, however, seen Dendrodnda longicauda Pelzeln, which 

 Hellmayr refers to Deconychura, and which may possibly bridge the gap apparently 

 separating the two genera as understood by me. 



KEY TO THE SPECIES OF DECONYCHURA. C 



a. Rump brown like back (only the upper tail-coverts chestnut-rufous); bend of 

 wing washed with cinnamon-rufous. 

 6. Smaller (wing 95-102 in male, 86-90 in female; bill, from rictus, 24-25.5); 

 breast spotted or streaked with buff. (Southwestern Costa Rica and Panamd.) 



Deconychura typica (p. 285). 



66. Larger (wing 107-111 in male, 102 in female; bUl, from rictus, 27-29 in male, 



25 in female); breast plain, like abdomen. (British Guiana to Rio Negro 



and Rio Madeira.) Deconychura longicauda (extralimital).'* 



a Hellmayr was the first to call attention to the fact that this genus possesses twelve 

 rectrices, not ten, as stated by its describer. (See Bull. Brit. Orn. Club, xiv, 

 1904, 52.) 



& According to Hellmayr (Novit. Zool., xiv, 1907, 368). I have, however, seen 

 only D. typica^ from which alone the above diagnosis and description are taken. 



c Adapted from Hellmayr, Novit. Zool., xiv, 1907, 368. 



^Dendrodnda longicauda Pelzeln, Orn. Bras., i Abth., 1868, 42, 60 (Borba; Mara- 

 bitanas; Barra da Rio Negro); Sclater, Cat. Birds Brit. Mus., xv, 1890, 165. — Deco- 

 nychura longicauda Hellmayr, Novit. Zool., xiv, no. 2, Nov., 1907, 367 (crit.). 



