154 fringillidjE. 



sides of body, and flanks pale yellowish brown, streaked with blackish 

 brown, more narrowly on the latter ; abdomen dull whitish ; thighs 

 and under tail-coverts pale yellowish brown, with darker brown 

 shaft-streaks ; under wing-coverts and axillaries buffy white, the 

 edge of the wing mottled with dark-brown bases to the feathers ; 

 quills below dusky, ashy whitish along the inner edge : " bill 

 brownish horn-colour ; feet horn-colour ; iris dark brown '"' (Jelski). 

 Total length 4-3 inches, culmen 0-4, wing 1*9, tail 1*5, tarsus 

 0-6. 



The adult female in seasonal plumage is more olive-brown than in 

 summer, and has more of an olive-yellowish tinge on the breast and 

 sides of the body, the dark streaks being very distinct and the abdo- 

 men pure white. 



The young female resembles the adult female, but is umber-brown, 

 with a rufescent tinge on the fore part of the crown and on the 

 edges of the greater wing-coverts and inner secondaries ; the under 

 surface deep ochraceous buff, almost uniform, with only a few dusky 

 centres to the feathers of the fore neck and breast. 



Peruvian and Brazilian specimens have wdiite axillaries, white 

 under wing-coverts and quill-lining ; the white shoulder-spot is very 

 large and distinct. 



Central-American specimens have black axillaries and under wing- 

 coverts and black quill-lining ; the white shoulder-spot is very small 

 and difficult to distinguish. Specimens from Trinidad and Tobago, 

 also some from British Guiana, Rio Negro, and Bogota, have the 

 same characters. 



It seems certain, from an examination of a large series, that some 

 specimens from the above localities occasionally show an admixture of 

 white on the axillaries, and also have the quill-lining more or less 

 white ; and as these will generally be found to be more or less in 

 seasonal plumage, it would appear that the male, in its first winter, 

 still retains sufficient of the characters of the adult female (which 

 it at first greatly resembles) to account for a certain amount of 

 white under the wing. This seems the more probable, as these 

 intermediate specimens do not agree in the amount of white, which 

 bears every indication of passing into a completely black under- 

 wing. 



If, therefore, it could be proved that the above reasoning were 

 correct and the characters constant, we should have two species 

 with well-defined ranges ; but some specimens from Colombia and 

 Para, fully adult males, appear to be intermediate as regards 

 the amount of white ; and therefore it seems better to recognize 

 but one species, varying to the south in the direction of a white 

 quill-lining, and to the north in the direction of a perfectly black 

 one. The specimens obtained by Mr. "Whitely on the Carimang 

 River in different months also confirm this last suggestion. 

 Bab. The same as that of the genus. 



