500 PROCEEDINGS OF THE ACADEMY OP [Oct., 



the rusty tint being mainly restricted to the outer vanes of the feathers 

 and the black subterminal bands scarcely perceptible. 



This race is the brightest of the 'bay-backed ' series and approaches 

 insulana, the dullest of the 'ferruginous-backed' series, in the color of 

 the upper parts. 



Type No. 71,581, Amer. Mus. Nat. Hist. Rio Cauca, Colombia, 

 May 27, 1898. 6". J. H. Batty. Wing, 5.55 ins. ; tail, 12.10 ins. 

 Fiaya oayaua nigriorissa (Sclater). 



Piaya nigricrissa Sclater, P. Z. S., 1860, p. 285 and 297 [New Grenada and 

 Peru]. 



Length of wing, 5.45; tail, 10.20. 



Above bay with a strong wine purple gloss on wings and tail, thighs 

 dark gray, crissum blacker, less sooty than in mehleri; under side of 

 rectrices as in cayana, uniform dull black with the exception of the 

 white tips. 



Specimens examined from Colombia — Bogota ; Ecuador — Napo River, 

 Archidona, Guayaquil; Peru — ^headwaters of the Huallaga River, 

 Pebas. 



Piaya cayana pallescens (Cab. and Heine). 



Pyrrhocorax pallescens Cabanis and Heine, Mus. Hein., IV,^p. 86, 1862 [North 



Brazil]. 

 Piaya cayana cabanisi Allen, Bull. Amer. Mus. Nat. Hist., V, p. 136, 1893 



[Chapada, Matta Grosso, JBrazil]. 



Length of wing, 5.95; tail, 11.75. 



Upper parts similar to columbiana but paler, the bright ferruginous 

 modified by a tone of ochre; crissum and thighs paler than in colom- 

 biana, and abdomen paler than in any other race, under side of rectrices 

 nearly uniform rusty brown except for the white tips; the pale rusty 

 tint usually pervades the whole dark area instead of forming definite 

 patches, in some lights, however, the dusky subterminal bands are 

 clearly discernible. 



.g[I have studied Dr. Allen's series of cabanisi from Chapada, Matto 

 Grosso, and cannot find sufficient difference between them and birds 

 from more northern Brazilian localities to w^arrant separation from 

 pallescens. Five specimens of the latter give average length of wing 

 5.90 and tail 11.30, while ten Matto Grosso birds give wing 6, tail 12. 

 These differences are less than those shown by series of mehleri from 

 different parts of its range and since, as Dr. Allen admits, there are 

 practically no color differences, there seems to be no ground for recog- 

 nizing cabanisi as distinct. 



Dr. Allen was misled by the general recognition accorded to the more 



