174 BULLETIN" 15 3, UNITED STATES NATIONAL MUSEUM 



races: the typical form, which occupies most of the range, and 

 latifrons which is found in southern Somaliland and Kenya Colony 

 east of the Rift Valley. The latter is characterized by the broad 

 white band on the forehead, and generally darker color of the upper- 

 parts, and smaller size. 



The species is quite variable, and inadequate series have led dif- 

 ferent workers to opposite conclusions, as the following notes indi- 

 cate : Van Someren ^- writes that eight birds from northeastern 

 Africa and Ethiopia are paler on the upperside than nine from 

 Senegal. Zedlitz,''^ on the other hand, notes that birds from the high 

 plateau of Mareb (northern Ethiopia) are darker than birds from 

 Eritrea, but that the former are like examples from Senegal. Van 

 Someren ^- gives the following wing measurements : 



Millimeters 



Northeast Africa (8 specimens) 190-198 



Senegal (9 specimens) 185-195 



Uganda (4 specimens) 180-187 



The series collected by Mearns in Ethiopia (11 specimens) have 

 wings 185.5-198 millimeters long, indicating that no constant dif- 

 ference exists between western and eastern birds in this respect. It 

 is true, however, that Ethiopian examples average slightly larger. 



Not only have various writers contradicted each other with regard 

 to variations, but also with reference to the altitudes at which tectus 

 occurs. Blanford ^* says that it is a lowland form, chiefly confined 

 to the coast, although he did meet with it in the Anseba valley. 

 Erlanger ^^ likewise states that it is a lowland form. Zedlitz °^ re- 

 cords it not only on the coast as in the Dahlak Islands and the Barca 

 district of Eritrea, but also up to 2,000 meters (6,600 feet). In an- 

 other paper ^^ he writes that tectus is common in the highlands up to 

 altitudes of more than 2,000 meters, but also occurs low down near 

 the coast. The onh' conclusion to be made is that the species is not 

 as restricted altitudinally as are many others. 



Claude Grant ^' does not recognize laiifrons^ but he made the mis- 

 take of comparing western birds (typical tectus) with all eastern 

 birds (both tectus and Jatifrons). Naturally, he could not differen- 

 tiate between them. « 



Some specimens have no white on the outer web of the outermost 

 primary as do most, but have the basal area pale brownish gray. 

 Birds in fresh plumage have white tips to the middle upper wing 

 coverts, which wear off and disappear rather rapidly. 



82 Nov. Zool., vol. 29, 1922, p. 16. 

 •3 Journ. f. Ormth., 1910, pp. 311-312. 

 w Gcol. and Zool. Abyss., 1870, p. 430. 

 •sjourn. f. Ornith., 1905, pp. 67-68. 

 ««Ideni, 1914, p. 030. 

 "Ibis, 1915, p. 54. 



