Todd-Carriker : Birds of Santa Marta Region, Colombia. 491 



and San Miguel, under similar conditions and at about the same alti- 

 tude. It was met with in the tree-tops, in company with Tangara 

 heinei. 



471. Thraupis palmarum atripeimis Todd. 



Tanagra palmarum (not of Wied) Sa'lvin and Godman, Ibis, 1880, 120 (Minca). 



— ScLATER, Cat. Birds Brit. Mus., XI, 1886, 159 (" Santa Marta " and Minca). 

 Tanagra palmarum melanoptera (not of Sclater) Bangs, Proc. Biol. See, 



Washington, XII, 1898, 141 ("Santa Marta"). — Allen, Bull. Am. Mus. 



Nat. Hist., XIII, 1900, i6g (Bonda and Cacagualito). — Clark, Auk, XX, 



1903, 398 ("Santa Marta"; meas. ; crit.). — von Berlepsch, Verh. V. Int. 



Orn.-Kong., 1911, 1053 ("Santa Marta," in range). 



Twelve specimens: Cacagualito, Don Diego, Fundacion, Tierra 

 Nueva, Mamatoco, Dibulla, Santa Marta, and Tucurinca. 



Individual variation in Thraupis palmarum is so considerable as to 

 greatly complicate the definition of the sevefal geographic races into 

 which the species seems to be divided. The matter has been discussed 

 in some detail by Mr. Clark, but apparently without coming to any 

 very definite conclusion. Santa Martan specimens resemble those from 

 Costa Rica in their rather smaller size and and generally darker colora- 

 tion as compared with birds from Guiana, Brazil, and Bolivia, and 

 seem well entitled to separation under the name recently proposed by 

 the writer {Proceedings Biological Society of Washington, XXXV, 

 1922, 92). 



Although several of the records above quoted refer to localities 

 lying at a higher altitude, this tanager seems to prefer the coastal 

 plain, and is seldom seen above 1,000 feet. Its habits are practically 

 the same as those of T. episcopus cana, with which it is often found 

 associated. 



472. Thraupis glaucocolpa (Cabanis). 



Three specimens : Dibulla, Arroya de Arenas, and Fonseca. 



A single specimen of this species was taken at Dibulla on February 

 21, 1914, but although a sharp lookout was kept for it during the few 

 days afterwards spent there, no others were seen. It proved to be 

 fairly common, however, in the valley of the Rio Rancheria, having 

 been observed at Arroya de Arenas, Loma Larga, and Fonseca during 

 the course of the writer's visit to these. parts in July, 1920. The spe- 

 cies is known mainly from the north poast of Venezuela, but appears 

 to be rare in collections. The above specimens agree with a small 



