784 BULLETIN 50, UNITED STATES NATIONAL MUSEUM. 



Rufous phase. 



Adults {sexes alike). — ^Above deep cinnamon-rufous or hazel, 

 darkest on back and scapulars; pileum and hindneck spotted with 

 Hghter cinnamon-rufous; lower hindneck crossed by a collar of black 

 (more or less interrupted in middle portion), immediately below which 

 the color is lighter cinnamon-rufous or rufous- tawny; back, rump, 

 scapulars, and wing-coverts indistinctly barred or transversely spotted 

 with lighter cinnamon-rufous, but sometimes nearly immactilate, 

 especially the scapulars; tail banded with light cinnamon-rufous and 

 dusky, the bands varying in relative width, the paler ones sometimes 

 continuous but sometimes interrupted at shafts of rectrices; whole 

 face light cinnamon-rufous, rarely approaching dull rusty white on 

 "eyebrow" aud lores; chin and malar region white (usually tinged, 

 more or less, with rusty) ; band across throat deep cinnamon-rufous, 

 usually more or less intermixed with black laterally; upper chest, 

 median f oreneck, and median line (sometimes very narrow) of breast 

 and abdomen white; sides of breast, and sides, imiform cinnamon- 

 rufous, the flanks similar but broken into more or less distinct stripes 

 of cinnamon-rufous and white or buffy white, the latter, however, 

 sometimes indistinct; thighs, tarsi, and under tail-coverts tawny- 

 ochraceous, the last indistinctly edged with whitish; under wing- 

 coverts and spots on inner webs of remiges ochraceous-buffy; bill, 

 etc., as in brown phase. 



Young. — ^Essentially like adults, but pileum, hindneck, and back 

 unspotted. 



Adult male.— Length (skin), 150;" wing, 89-95.5 (92.1); tail, 

 49.5-63.5 (56.5); cuhnen (from cere), 11.5-12.2 (11.8).» 



o One specimen. 



b Two specimens; one (the smaller) from Send, Panami; the other from mountains 

 south of Cartago, Costa Rica. 



Eight specimens of undetermined sex measure as follows: Length (skins), 145-165 

 (153.1); wing, 95-104.5 (99.1); tail, 53-64.5 (58.6); culmen (from cere), 10.5-11.5 

 (10.9). 



The very considelrable variation in pattern and colors in the series examined indi- 

 cates the possibility of geographic differentiation, but the number of specimens is 

 much too small to warrant any conclusion in the matter. 



