28 Proceedings of the Biological Society of Washington. 



the color may be in part attributable to the age of a large part of our 

 specimens (collected at Chapada, in 1882-85), though it is exhibited in 

 a degree by specimens collected at Puerto Suarez in 1908, and at Urucum 

 in 1913. However, I should prefer seeing a large series of freshly collected 

 birds before commenting further on their general colorations. These 

 birds further differ from our other specimens in being smaller, with shorter 

 bills, presenting, indeed, in these respects, the minimum measurements 

 of our entire series. 



We are, however, here concerned chiefly with the extent to which this 

 series of birds shows the "pied" character which distinguishes our series 

 from Bolivia. Thus, seventeen of eighteen males, and nine of the fifteen 

 females are marked with feathers in whole or in part yellow. The feathers 

 are never as numerous as they are in our birds from the Andean region 

 of Bolivia, and it is evident that the pied character is disappearing. This, 

 it seems to me, is less surprising than that it should be present in so large 

 a percentage of the specimens in a region over four hundred miles from 

 what appears to be the region of its greatest development. 



Amazon River (Solimoes, near Manaos, 1 cf ; Santarem, 3 <? d" ; Rio 

 •Tocantins, 1 9 ; Marajo, 1 o 71 ). — These specimens are obvious inter- 

 grades between what may be loosely called the northern and southern 

 forms. In general black tone of color they are nearer the former, in 

 size they are fairly intermediate, while four of the six birds exhibit traces 

 of the pied markings found in most of our southern specimens. The 

 Solimoes bird has a single breast-feather broadly tipped with yellow, 

 and of the three Santarem birds, one has three breast-feathers, the other, 

 one narrowly fringed with yellow. The Marajo bird has one yellow 

 feather and one broadly tipped with yellow on the breast. On the whole, 

 these Amazon birds are to be referred to decumanns rather than to macn- 

 losus. 



North of the Amazon (Dutch Guiana, 3 cTc 71 ; British Guiana, 1 d\ 

 1 9 ; Tobago, 1 d" ; Venezuela, 6 d" d" ; Ecuador, 1 d\ 1 9 ; Colombia, 

 6- o"d\ 11 9 9; Panama, 10 d 1 d\ 3 9 9).— As before stated, lack of 

 adequate topotypical material prohibits a report on the variation of these 

 specimens inter se and I consider them here only with regard to the pied 

 marking which characterizes the southern form. 



Of the forty-four specimens here listed only one shows any evidence of 

 this marking, a male from Cristobal Colon, Paria Peninsula, Venezuela, 

 having one breast-feather lightly fringed and one almost imperceptibly 

 margined with paler yellow. It is therefore the practical absence of these 

 yellow feathers north of the Amazon as well as their presence south of the 

 Amazon, which indicates that they constitute a character of racial value. 



Variation in Size. 



Examination of the measurements of a considerable number of speci- 

 mens, shows that while females from the same locality present a com- 

 paratively small range of variation in size, the males vary widely. Fur- 



