868 Annals of the Carnegie Museum. 



Cat. Birds Brit. Mus., XI, 1886, 75 (Costa Rica [Endres, Carmiol, and Arce]). 

 — Zeledon, An. Mus. Nac. de C. R., I, 1887, 109 (San Jose, Naranjo de 

 Cartago, Navarro, Monte Redondo). — Cherrie, Auk, IX, 1892, 24 (San 

 Jose; descr. nest and eggs). — Underwood, Ibis, 1896, 435 (Miravalles).— 

 Ridgway, Birds N. and Mid. Amer., II, 1902, 25, part (southern Mexico to 

 Chiriqui (?) and Veragua (?); — Costa Rica: San Jose, Naranjo de Cartago, 

 San Juan, Turrialba, etc.). 



Phonasca hirandinacea Cabanis, Jour, fur Orn., i860, 334 (Costa Rica [Frant- 

 zius]). 



Phonasca gnatho Cabanis, Jour, fur Orn., i860, 335 (Costa Rica; coll. Berlin 

 Museum). 



Enphonia gnatho Lawrence, Ann. Lye. N. Y., IX, 1868, 98 (Costa Rica). — 

 Frantzius, Jour, fur Orn., 1869, 297 (Costa Rica). — Salvin and Godman, 

 Biol. Centr.-Am., Aves, I, 1883, 262 (Tempate de Nicoya [Arce]). — Sclater, 

 Cat. Birds Brit. Mus., XI, 1886, 76 (Tempate and Turrialba [Arce]).— 

 Zeledon, An. Mus. Nac. de C. R., I, 1887, 109 (Costa Rica). — Bangs, Proc. 

 Biol. Soc. Wash., XXII, 1909, 37 (E. hirundinacea in Costa Rica equals E. 

 gnatho; crit.). 



U. S. Nat. Museum: Guayabo (Ridgway and Zeledon), Monte Redondo 



(Ridgway), Turrialba (Zeledon), Sabanilla de Alajuela (Alfaro). 

 Bangs Collection: Cartago, Bolson, Coralillo, Sabanilla (Underwood). 

 Carnegie Museum: Bebedero and Miravalles (Carriker). Five skins. 



Mr. Bangs first suggested to me that E. gnatho (Cabanis) was probably 

 the bird known in Costa Rica under the name of E. hirundinacea, and 

 after carefully going over the subject Mr. Todd and I came to the same 

 conclusion, the results of w r hich are given by Mr. Todd below. 



"Four adult males collected at Bebedero and Miravalles (May 2 to June 

 25, 1906), compared with four males of E. hirundinacea from British 

 Honduras and Mexico exhibit the characters of this form to a striking 

 degree. Not only is the bill larger and differently shaped, but the color 

 of the upper part is bluer, less violaceous. The differences between this 

 form and E. hirundinacea, while perfectly obvious to an unpredjudiced 

 eye, are of such a character that a trinomial designation seems preferable, 

 unless, indeed, it can be shown that the two forms occur side by side with- 

 out intergradation. The type locality is not known." (W. E. C. Todd.) 



After examining the series in Mr. Bangs' collection I am positive that 

 true E. hirundinacea does not occur in Costa Rica, and that all birds from 

 that country are referable to this race. 



This species inhabits the plateau region, occasionally straggling 

 down to 2,500 feet on the Caribbean slope, and the whole of the 

 Pacific slope and lowlands north of the Dota Mountains. It is found in 

 company with E. affinis about Miravalles and other places in Guanacaste. 



