ACREDULA 159 



0-naga, Jap. 



ad. (Japan). Differs from Acredula rosea merely in having the black 

 supercilium continued across the lores to the base of the bill. 



Hob. Japan on the main island. 



In habits and nidification it does not differ from A. rosea. It 

 breeds on Fuji, and visits the lower country around Tokio and 

 Yokohama in winter. 



226. SUBSP. ACREDULA IRBII. 



Acredula irbii, Sharpe and Dresser, P.Z.S. 1871, p. 312; Dresser, iii,. 

 p. 73, pi. 105, fig. 1 ; Gadow, Cat. B. Br. Mus. viii. p. 63. 



Mito, Span.; Codibugnolo, Ital. 



Ad. (Spain). Differs from A. rosea in having the back bluish grey,, 

 not black, slightly tinged with rose on the sides, the centre of the 

 crown usually white striped with black, and the cheeks dull white 

 finely striped with blackish grey. Culmen 0'25, wing 2 '32, tail 3 '2, 

 tarsus 0'65 inch. 



Hob. Spain, Southern Italy, and Sicily, and strays into- 

 France. 



In note, habits, and nidification it does not differ from 

 Acredula rosea. It breeds near Gibraltar, placing its nest 15 or 

 16 feet above the ground in the Zarzaparilla, a thorny creeper, 

 a species of Smilax, and its eggs are deposited late in February. 



227. SUBSP. ACREDULA CAUCASICA. 



Acredula caucasica (Lorenz.) Beitr. Orn. Nords. Kauk. p. 60 Nachtrag. 

 (1887) ; Dresser, ix. p. 113, pi. 655, fig. 2. 



Ad. (Kuban). Crown white, forehead marked with reddish brown, 

 supercilium reddish brown above, blackish brown below ; back pale slaty 

 grey, darker on the upper part ; upper tail-coverts grey faintly tinged 

 with rose ; wings blackish, secondaries margined with white ; tail black,, 

 all but the central feathers margined and tipped with white, the outermost 

 almost entirely white ; under parts white faintly marked with grey on 

 the sides of the breast ; bill and feet blackish ; iris brown. Culinen 

 0'3, wing 2-5, tail 3'35, tarsus 0'64 inch. 



Hal). The Caucasus where it is resident and not uncommon. 



In habits it assimilates closely with its allies, and though its 

 nest and eggs are unknown, they will doubtless be found to 

 resemble those of A. caudata. Acredula dorsalis Madarasz, of 

 which I examined the type in Buda-Pest, is not separable from 



