442 



BULLETIN 15 3, UNITED STATES NATIONAL MUSEUM 



As may be sensed from the foregoing discussion, the taxonomic 

 conclusions to be drawn are rendered somewhat uncertain by the 

 extent of nongeographic variation. For the present I follow 

 Sclater's arrangement but hope that workers with more satisfactory 

 material, and especially residents of eastern Africa, who have a 

 chance to study the birds in life, may consider the problem open for 

 investigation and by no means a settled issue. 



The width of the red frontal band varies greatly in the males but 

 is not correlated with locality or wear. Hawash and Dire Daoua 

 specimens are grayer on the back than any others seen. There is a 

 tendency for the birds to be smaller near the Equator and larger to 

 the north, as may be seen from table 78. 



As Lynes ^^ correctly writes, there is no evidence of any form of 

 the melba finch in northern Ethiopia. The species is a bird of com- 

 paratively low altitudes and occurs all around the base of the high 

 plateau regions of northern and north-central Ethiopia, but not 

 high up, Adis Abeba being the highest locality from which it is 

 known. 



Table 78. — Measurements of 15 specimens of Pytilia melba soudanensis 



The species inhabits bushy and scrub country and is usually seen 

 in small groups or singly. At times larger numbers are observed, as, 

 for example, on the Tana Eiver, August 16-19, when Mearns noted 

 150 birds. 



»Ibis, 1926, pp. 399-400. 



