- 121 — 



of 3 99- 



Wing, total length, tarsus, 



95 mm. 160 mm. 20 mm. 



98 mm. 168 mm, 21 mm. 



93 mm. 160 mm. 20.5 mm. 



All are in full dress and are shot in the same regions and 

 exhibit no differences in the colours of the plumage worth speak- 

 ing of. Yet the measurements of the length and breadth of 

 the culmen vary very much. The visible portion of the bill va- 

 ries between 12 and 13 mm., the breadth at the base between 

 6.8 mm and 8 mm. 



Two young birds, one shot on the 18 th May, the other on 

 the 27th July, are, with respect to colour, as Reichenow has 

 described them (op. cit. p. 435). Both have a wing-measurement 

 of 98 mm., tarsus 21 mm. 



In specimens from the upland regions the measurements 

 for the length of wing range from 90 to 102 mm in adults. In 

 those from the coastal districts the figure is considerably lower 

 and I therefore consider that I am fully justified in separating 

 them, all the more as there are also other differences present. 



Bradornis pallidiis subalaris Sharpe. — Rchw. II. p. 436. 

 1 (5 ad. 29. 4.; 1 Q ad. 27. 4.; 1 S ad. 20. 9. Mombasa. 



Individuals belonging to the tropical coastal regions are in 

 general considerably smaller than the preceding race and, in 

 addition, they have not the clearly marked grey colour on the 

 fore-neck and underparts, but have a pale brownish tint over the 

 more purely white ground. The white colour of the throat is 

 therefore not so pronounced as in the former but shades evenly 

 and almost imperceptibly into the colour of the fore-neck and 

 belly. Further, those from the coast are not dark-grey on the 

 back, but have a much lighter brownish grey upper surface. 

 The respective measurements are: 

 for 2 99 

 wing, total length, tarsus, 

 80 mm. 160 mm. 20 mm. 



82 mm. 150 mm. 21 mm. 



for 1 cf 

 85 mm. 155 mm. 21 mm. 



Ogilvie Grant (Ibis, 1913, p. 637) gives the wiug- 

 lenght as 80—86 mm. . 



I have compared my specimens with those in the Berlin 

 Museum and as none of the coastal specimens even attain the 

 minimum for those from the interior of the country I have come 

 to the conclusion that it would be more correct to separate them. 



