FAMILY TIIRAUPIDAE 



419 



and found that Golden-hooded Tanagers were in the middle zone of 

 the tree 46% of the time and in the lower zone 37% of the time. By 

 day they move about in small groups, sometimes part of mixed species 

 flocks, and at night, especially out of the nesting season, they gather in 

 larger groups to roost — Skutch (Pac. Coast Avif., no. 31, 1954, p. 202) 

 found a roost in Costa Rica with at least 17, and probably more, birds. 

 No song has been recorded for this species; its usual note is a tssp used 

 to keep members of a pair or family together. Ridgely (1976, p. 318) 

 also notes "a repeated tsit-tsit-tsit-tsit often given in flight as it moves 

 from tree to tree." 



TANGARA LARVATA CENTRALIS (Berlepsch) 



Calospiza larvata centralis Berlepsch, 1912, Verh. V. Intern. Ornith, Kongr., Ber- 

 lin, 1911, p. 1034. (Calovevora, Veragua [Panama].) 



Characters. — Cheeks, except posterior border, deep blue. 



Measurements. — Males (10 from Bocas del Toro and Costa Rica), 

 wing 68.8-72.0 (70.2), tail 43.0-47.9 (45.9), culmen from base 11.0- 

 12.1 (11.4), tarsus 15.2-17.7 (16.4) mm. 



Females (6 from Bocas del Toro and Costa Rica), wing 66.0-69.5 

 (67.4), tail 41.4-44.9 (42.9), culmen from base 11.1-12.6 (11.5), 

 tarsus 15.7-17.1 (16.6) mm. 



Resident. Common in the lowlands of Bocas del Toro and on the 

 Caribbean slope of Veraguas; in the interior it ranges east to the low- 

 lands of northern Code, where I have collected it at El Uracillo. 

 Beyond Panama, it is found on the Caribbean slope from southeastern 

 Nicaragua through eastern Costa Rica. In general it is like T. 1. larvata, 

 found from southern Mexico through Guatemala and Honduras to 

 central Nicaragua; the population of southeastern Nicaragua (at the 

 Rio Escondido) and northeastern Costa Rica (Carillo) is somewhat 

 intermediate, but is nearer centralis. 



When Berlepsch described centralis as new, to replace u francescae" 

 of Sclater, a name that he considered invalid on the grounds that it was 

 merely an emendation of the term fanny of Lafresnaye, he listed a 

 specimen from Calovevora, Veraguas, collected by Arce, as type. Spec- 

 imens from the type locality, which is on the Caribbean slope of Vera- 

 guas, differ as indicated above. Peters (Bull. Mus. Comp. Zool., vol. 

 71, 1931, p. 340) overlooked this difference in his examination of skins 

 from the Almirante region when he recognized franciscae of Sclater 

 and placed centralis as a synonym. 



A male I collected at Almirante, Bocas del Toro, on February 12, 

 1958, was in breeding condition. 



