BIRDS OF ETHIOPIA AND KENYA COLONY 47 



March and June. Blanford found it at Undel Wells in April, but 

 observed it only at low or moderate elevations, not on the high 

 plateau country. 



The breeding season is from June to August. Riippell states that 

 the nests are built on rocks much like those of Eirundo rustica. 

 Antinori found it nesting during June, July, and August, near 

 Mahal-Uonz, in Shoa. 



It is a rather sad commentary on our knowledge of the habits of 

 this bird that practically nothing has been added in all the years 

 since Sharpe and Wyatt's monograph appeared. Their account (pp. 

 379-380) is still a summation of what is known of the life history of 

 the mosque swallow in Ethiopia. 



HIRUNDO RUFULA EMINI Reichenow 



Eirundo emini Reichenow, Jour, fiir Orn., 1892, p. 215: Bussisi, west shoree of 



Lake Victoria. 

 Specimens collected: 2 males, 20 miles abovee mouth of Thika River, Kenka 



Colony, August 27, 1912. 



These two specimens are the darkest examples of emini I have seen 

 (19 examined) and have slightly narrower bills than any of the others 

 studied. Both birds are molting the remiges and rectrices. 



As mentioned under the discussion of melanocrissa, the present 

 form is very closely related to it, and, as their respective ranges do 

 not overlap, I keep them as races of a single species. 



The range of e7nim is as follows: The Mlanje Plateau, of Nyasa- 

 land, north through Tanganyika Territory and the eastern Belgian 

 Congo, Ruanda, and Urundi through western Uganda and Kenya 

 Colony to the Turkana and Eendile districts to the west and east, 

 respectively, of Lake Rudolf, north to the lake region of Shoa. 

 Lonnberg ''^ recorded Punda Melia (near Fort Hall) as "on the north- 

 eastern frontier of its known distribution." However, six years 

 earlier Neumann ^^ definitely recorded emini as breeding in Malo, 

 southern Shoa, and Erlanger^° recorded it even farther north — at 

 Adis Abeba and the Hakaki River — but Erlanger's birds are prob- 

 ably all really melmiocrissa. In the southern part of its range it 

 appears to be more of a higliland bird than farther north, and con- 

 sequently its distribution in Tanganyika Territory is somewhat 

 patchy and discontinuous. 



It has been recorded from altitudes as great as 8,500 feet on 

 Ruwenzori ^^ and 7,000 feet on Mount Elgon.^^ 



'« Kongl. Svenska Vet.-Akad. Handl., 1911, p. 79. 



" Journ. fur Orn., 1905, p. 202. 



8»lbid., pp. 678-679. 



" Cf. Ogilvle-Grant, Trans. Zool. Soc. London, vol. 19, p. 409, 1910. 



«2 Granvik, Journ. fur Orn., 1923, Sonderheft, p. 120. 



