obtained in Somaliland. 395 



it is met with in small flocks occasionally. In its habits, 

 note, and choice of breeding-places it is quite distinct from 

 Spreo superbus. 



4. Amydrus blythi Hartl. ; Elliot, Field Columb. Mus., 

 Orn. i. no. 2, p. 31. 



Amydrus morio (nee L.) ; Lort Phillips, Ibis, 1896, p. 83. 



No. 46. S juv. Dobar springs, Jan. 24, 1897. lids hazel. 



No. 71. ? ad. Goolis foot-hills, Jan. 28, „ „ „ 



What I take to be the immature male is much duller than 

 the adults, and the little gloss it has is green above and below. 

 In this respect it approaches ^./ra^er of Socotra, but the 

 latter bird has a longer bill and the gloss on the uuderparts 

 is distinctly metallic. 



Dr. Sharpe agrees with me that the specimen which he 

 identified as A. morio in my first Goolis collection must be 

 referred to A. blythi. 



I was much interested to find a flock of these Grakles in 

 the gardens at the Dobar springs, eight miles inland from 

 Berbera, as I had previously noticed them only on the upper 

 ledges of the Goolis, where they were evidently breeding in 

 holes in the face of a clifl". 



5. Heteropsar albicapillus (Blyth) ; Sharpe, P. Z. S. 

 1895, p. 461. 



No. 40. (? ad. The Gooban, Jan. 21, 1897. Iris white. 



No. 41. c? ad. „ 



No. 82. (? . Sheikh, Jan. 30, „ „ „ 



No. 83. Juv. „ „ „ „ „ 



No. 84. Juv. „ „ „ 



The male from the Sheikh plateau, 3000 feet (No. 82), 

 differs from the two other specimens of the same sex in 

 having large spots of black at the end of the long upper tail- 

 coverts. There is a great diff'erence in the breadth of the 

 white stripes on the under surface of the body in the male 

 (No. 40), and the white spotting of the upper surface is 

 rather more distinct than in the other adult males. This 

 individual is probably a young bird in its first full plumage. 

 The two young birds shot on January the 30th differ very 



