BIRDS OF AVESTERN COLOMBIA. 1 1 65 



however, feel justified in bestowing a name upon this subspecies, 

 having seen but a few skins in not very good condition. 



123. Cercomacra tyrannina rufiventris Lawr. 



\Pyriglena tyrannina Sclater, P. Z. S. 1855, p. 90, pi. xcviii. 

 (1855.— Bogota coll.).] 



Disythammcs (sic) rufiventris Lawrence, Ann. Lj'c. N. H. N.Y. 

 viii. p. 131 (1865. — New Granada, line of Panama Railroad — 

 descr. S juv.) ; cfr. Salvin, Ibis, 1874, p. 316. 



Cercomacra crepera Bangs, Auk, xviii. p. 365 (1901. — Divala, 

 Chiriqui). 



No. 1997. (S ad. Boca de Calima, Rio San Juan, 18.viii.08. — 

 Wing 67; tail 58; bill \U mm. 



No. 1981. § imm. Guineo, R. Calima : 8.viii.08.— Wing 60 ; 

 bill 16 mm. 



No. 2387. 2 ad. Novita : 10.xii.08.— Wing 59; bill 16^ mm. 



No. 2742. $ ad. Pueblo Rico, 5200 ft., 8.ix.08.— Wing 62 ; 

 tail 67 (!) ; bill l^ mm. 



" Iris dark brown, feet blue-grey, bill black, mandible horn- 

 coloured in females." 



These birds strictly belong to the dark western form separated by 

 Mr. Bangs as (J. crepera. Mr, Carriker * has already alluded to 

 the variability of its characters, basing his conclusions upon Costa 

 Rica skins. On comparing the series in the Munich Museum, 

 it becomes at once evident that two fairly marked races can be 

 distinguished. Adult males from Guatemala, Costa Rica, Chiriqui, 

 W. Colombia and N.W. Ecviador are, notwithstanding some 

 individual variation, much darker, more slaty blackish, both on 

 upper and lower parts than specimens from Bogota (topotypes), 

 the upper Orinoco, British Guiana, and North Brazil (Para, Rio 

 Negro). The females also are somewhat darker ferruginous 

 underneath. The single male from Calima is slightly paler 

 grey on the belly, but similar specimens I have seen from 

 S.W, Costa Rica, while several from N.W. Ecuador (San Javier) 

 in intensity of coloration aie practically identical with topo- 

 typical Chiriqui-skins. Mr. Bangs (l. c. p. 366) refers the birds 

 from Loma del Leon (= Lion Hill), Panama, to typical tyrannina. 

 I doubt, however, the correctness of this view, since two males 

 from Monte Oscuro, near Panama City, though not extreme 

 examples of crepera, are certainly much nearer that form than to 

 tyrannina. Therefore, it seems to me that Lawi-ence's term 

 rufiventris, founded upon an immatui-e male from Lion Hill, must 

 take precedence over crepera. 



According to my views, the range of the two subspecies would 

 be as follows : — 



(a) G. tyrannina tyrannina (east of the Andes) : E. Colombia 

 (Bogota, Bucaramanga, etc.) ; Venezuela : Orinoco (Munduapo) 

 and Caura Valley ; British Guiana ; Northern Brazil : Rio Negro 

 down to Manaos ; Obidos, north bank of Amazons ; Para district, 



* Ann. Carnegie Mus. v, 1908, p. 8 ; I.e. vi. 2-4, 1910, p. 613. 



