FAMILY TROCHILIDAE 329 



The subspecific name niveoventer was given by Gould to birds 

 collected by Warszewicz near David, but in western Chiriqui this well- 

 marked form is known mainly from the Subtropical Zone near El 

 Volcan, and on the eastern side of the mountain in the Boquete region. 

 At San Felix in the lowlands I found it common, and it is of regular 

 occurrence in the low country from that point eastward. On the 

 Azuero Peninsula it is common. There is a male in the Havemeyer 

 collection at Yale, taken at Chiriqui Grande (labeled "Chiriquicito 

 Grande," in error) Bocas del Toro, on May 4, 1927, by Austin Paul 

 Smith, the only record for the Caribbean slope. The race ranges 

 beyond Chiriqui into southwestern Costa Rica to Boruca and El 

 General. 



A nest (now in the U. S. National Museum) collected by Dr. F. A. 

 Hartman (Condor, 1957, p. 270) near El Volcan, Chiriqui, March 14, 

 1956, is a small, deep cup of plant downs decorated externally with 

 red and gray-green lichens. It was saddled in a crotch of a little 

 branch in a small tree, about 2^ meters above the ground. The col- 

 lector gave the following dimensions : "outside diameter 48 mm. ; 

 outside depth 45 ; inside diameter 22 ; inside depth 24." The parent 

 was identified, but the nest was empty. 



The numerous specimens examined from throughout the extensive 

 range are remarkably uniform in color, when the variability found 

 in the other races is considered. Those from the eastern side of the 

 Azuero Peninsula are slightly darker on the back, with lessened 

 coppery reflection, which led to my description of a separate race 

 from that area. With more material at hand it seems appropriate to 

 group these as a minor variant of the present form. 



This form of the species as stated is the race of western Panama, 

 found mainly on the Pacific slope. Occasional specimens, even from 

 as far west as western Chiriqui, show a reddish brown tint, variable in 

 amount, in some of the rectrices, occasionally of considerable extent. 

 A similar variability is found in the typical race A. e. edward, in which 

 some individuals have the rectrices very dark so that they suggest 

 niveoventer. On close examination it is seen that these are merely 

 a melanic phase, not to be identified as the western race. Records 

 in literature of the subspecies niveoventer in the Canal Zone are 

 based on an early report by Sclater and Salvin (Proc. Zool. Soc. 

 London, 1864, p. 343) who in describing a collection made by 

 McLeannan near Lion Hill on the Panama Railway list both edward 

 and niveoventer without discussion. Salvin later (idem, 1870, p. 210) 

 in an account of collections by Arce made in Veraguas and Chiriqui 



