( 277) 
lower Rio Madeira. North of the Amazons, near Mandos, it is represented by 
T. 8s. surinamus, while farther up, in the vicinity of Teffé, and on the upper Rio 
Negro (Marabitanas, Rio Icanna) another closely allied race, 7. s. napensis, takes 
its place. Cf. Nov. Zool. xiii. pp. 357-8.] 
56. Tachyphonus cristatus madeirae n. subsp. 
[ Tanagra cristata Linnaeus, Syst. Nat. xii. 1, p. 317 (1766.—ex Brisson : Cayenne). } 
Tachyphonus cristatus (nec Linnaeus) Pelzeln, Zur Orn, Bras. iii. p. 213 (part. : Engenho do Gama, 
Rio Guaporé ; Borba, r. Madeira). 
T. cristatus subsp. Hellmayr, Nov. Zool, xiv. p, 349 (Humaytha). 
Nos. 209, 249, 329, 386, 483. ddad., Calama, 12, 22. vii, 2, 12, 29. viii, 
1907.—Wing 79—83 ; tail 70—77; bill 124—14 mm. “Iris brown, feet and bill 
black.” 
dad. Nearest to, and agreeing with, 7. cristatus brunneus (Spix), of Eastern 
Brazil (from Para to S. Paulo), in having the crest deep fiery or vermilion-red and 
bordered in front only by a rather narrow, bright buff line, but crest-feathers much 
shorter—scarcely longer than in 7. ¢. cristatus, from Cayenne, etc.—and whole 
throat, except a small blackish chin-spot, deep ochraceous buff. 
? ad. Not different from that of 7. c. cristatus. 
Type in Tring Museum: ¢ ad., Calama, 2. viii. 1907 (W. Hoffmanns coll., 
No. 329). 
Besides the above series, I have before me the adult male from Humaytha 
obtained on Mr. Hoffmanns’ first trip to the Rio Madeira, and several examples, 
among them an adult female, belonging to the Vienna Museum (Natterer coll.). 
Compared with some thirty skins from Para, Pernambuco, Bahia, and Rio de 
Janeiro, the Madeira birds have the crest invariably shorter and generally of a 
deeper fiery red. Even more strongly marked is the difference in the extent of the 
ochraceous gular spot. The Calama and Humaytha specimens have the whole 
throat, with the exception of a very small black chin-spot, deep ochraceous buff, 
while in 7. c. brunneus there is but a broad stripe of a clearer buff along the middle 
of the throat, the sides of the latter, as well as a large patch on the chin, being 
dull black like the remainder of the lower parts. 7. c. eristatus (and the doubt- 
fully distinct 7. c. cristatellus) may be distinguished from the new form by their 
smaller gular spot, and particularly by the much lighter, orange-red crest, bordered 
in front and laterally by a broad, creamy buff margin. 
T. c. madeirae appears to be restricted to the Madeira Valley and its head- 
waters, Guaporé, etc. ; for specimens from the Rio Negro (Barcellos, Marabitanas) 
and Northern Peru (Loretoyacu, Pern) agree in every respect with Bogoté skins 
(so-called cristatellus).* 
57. Eucometis penicillata penicillata (Spix). 
Tanagra penicillata Spix, Av. Bras. ii, p. 36. pl. xlix. fig. 1 (1825,—no locality ; type examined). 
Eucometis penicillata Pelzeln, Zur Orn, Bras, iii. p. 212 (Borba) ; Hellmayr, Nov. Zool. xiv. p. 349 
(Humaytha), 
Nos. 413, 414. d imm., d juv., Calama (island), 17. viii. 1907. ‘Iris greyish 
brown, feet pale greyish yellow, bill black.”—Wing 85, 82; tail 82; bill 15}, 
16 mm. 
* The supposed female of Tachyphonus nattereri Pelz. (lc. p. 328: ¢ ad. Villa Maria, Rio Paraguay) 
was secured at Salto do Girao by Natterer. It is, however, doubtful whether it really belongs to 
T. nattereri, and the question must be left in abeyance until adult males from that locality come to 
hand, i 
