26 Proceedings of the Biological Society of Washington. 



especially; upper tail-coverts and tail brownish black; throat deep colonial 

 buff, the breast more or less strongly shaded with citrine, passing into 

 Martius yellow posteriorly; tibiae buffy citrine; crissum old gold; under 

 wing-coverts buffy yellow; inner margins of remiges buffy below; "iris 

 brown; feet dark plumbeous; bill black, flesh-color below." Female 

 similar, but the vertical spot wanting or merely indicated in buffy or 

 ochraceous. 



Measurements.— M&\e: wing, 64-69 (66); tail, 53-61 (57); bill, 10-11 

 (10.4); tarsus, 15.5-18 (16.4). Female (seven specimens); wing, 58-61 

 (60); tail, 51-54 (52); bill, 10-10.5 (10.3); tarsus, 13.5-17 (14.8). 



Range. — Guiana, west through central and southern Venezuela to Colom- 

 bia (east of the Andes), and south to the north bank of the Amazon and 

 through upper "Amazonia" to eastern Peru. 



Remarks. — The "Barbichon de Cayenne," male, of Buffon and D'Au- 

 benton is the basis of Gmelin's name for this species, the figure being easily 

 recognizable. Latham and Stephens, the next authors to notice the species, 

 merely copied Buffon's account. So far as we know Swainson had no 

 specimens of this form before him when he made it the type of his new genus 

 T]irannula, but only of the Brazilian bird (mastacalis). Cabanis, who 

 in due course had access to examples collected by Schomburgk in British 

 Guiana, is the first author to clearly recognize the distinctness of the 

 present bird. Whitely also secured specimens in British Guiana, and in 

 more recent years numerous other specimens have been taken in French 

 and Dutch Guiana, Venezuela, and thence westward to the foot of the 

 Andes in Colombia. Without having examined the specimens on which 

 the records are based it is impossible to be sure, of course, but we are 

 inclined to place the records for northeastern Peru here rather than under 

 M. atricaudus, mainly on geographical grounds. The measurements for 

 one of the specimens given by Taczanowski indicate a bird of the present 

 form, and not atricaudus, which averages smaller. The wing-measurement 

 of the other specimen, given as "59" millimeters, may be an error for 69. 

 It is extremely unlikely that atricaudus would be found on both sides of 

 the Andes in Peru, judging by analogy. For Brazil, Mr. Hellmayr finds 

 that Natterer's specimens from Marabitanas on the Rio Negro and from 

 Borba on the lower Rio Madeira belong here. A specimen, easily referable 

 to this form, from Avojutuba, on the Rio Negro, appears to confirm the 

 Marabitanas record, but we may be excused for venturing to doubt the 

 Borba record in view of the range of variation known to obtain in M. 

 Tnastacalis. It is more likely that the Rio Madeira separates the respec- 

 tive ranges of mastacalis and barbatus, but further field-work alone can 

 demonstrate this. Miss Snethlage has recorded two specimens from Bom 

 Lugar, on the Rio Puriis, which from the description are almost certainly 

 barbatus. The indications are, therefore, that this form has an extensive 

 range in northern South America, from Guiana on the east to Colombia and 

 Peru on the west. Individual variation is very much less than in )nasta- 

 calis, from which the present form may always be told by its much brighter 

 yellow rump and under surface, which latter is shaded with citrine, not 

 buffy or old gold. 



