300 Annals of the Carnegie Museum. 



resented an undescribed race of Grallaria varia. With topotypical mate- 

 rial of both these species now before us it is obvious that the form in 

 question is conspecific with G. regulus of Ecuador, instead of with the 

 much larger and paler G. varia of Cayenne. In fact, carmelita is 

 merely a dark race of regulus, differing in the darker, more brownish, 

 less olivaceous color of the upper parts, and in being- darker, more 

 brownish, less ochraceous below, with more dark mottling. The 

 throat and sides of the head are also darker and more uniform. The 

 feet are marked as " bright leaden blue," the bill as " blackish horn, the 

 extreme base below flesh-color." The male (type) measures : wing, 

 100; tail, 38; bill, 21.5; tarsus, 45. Female: wing, 103; tail, 35; bill, 

 21 ; tarsus, 42. 



This fine species was one of the surprises of the trip to the Sierra 

 Nevada. Both birds were taken in a tract of almost • impenetrable 

 forest on the southeast bank of the river below Pueblo Viejo. It is a 

 mass of gigantic boulders, tangled with vines and undergrowth inter- 

 mingled with heavy forest, and traversed by numerous small creeks. 

 The taking of the first specimen was scarcely more than accidental, 

 and then came the search for more. Four half-days were spent in the 

 search, with the result that one more was secured and another was seen 

 which escaped. They seemed to have no special call-note, and could 

 not be '■ whistled up." 



253. Grallaria bangsi Allen. (Plate IV.) 



Grallaria bangsi Allen, Bull. Am. Mus. Nat. Hist., XIII, 1900, 159 (El Libano 

 [type-locality] and San Lorenzo ; orig. descr. ; type in coll. Am. Mus. Nat. 

 Hist.). — Sharpe, Hand-List Birds, III, 1901, 43 (ref. orig. descr.; range). — 

 Dubois, Syri. Avium, II, 1903, 1068 (Santa Marta [region], in range; ref.' 

 orig. descr.). — Braeoukne and Chubb, Birds S. Am., I, 1912, 218 (ref. 

 orig. descr.; range). 



Fourteen specimens : El Libano, San Lorenzo, San Miguel, and 

 Heights of Chirua. 



Mr. Ridgway, in his dismemberment of the genus Grallaria, admit- 

 tedly did not have a good representation of the species before him, 

 and his diagnoses are thus not always satisfactory. Here, for instance, 

 is a species which is exactly intermediate in style of coloration and 

 structural characters between Hypsihemon and Oropezus. To which 

 should it be assigned? To refer it to either one would necessitate a 

 modification of the diagnosis, while the only other alternatives would 



