106 BULLETIN 153, UNITED STATES NATIONAL MUSEUM 



For the present, in the absence of material, I follow Sclater in 

 considering littoralis as not valid, and place it in the synonymy of 

 peasei. 



The ranges and characters of the three races are as follows : 



1. P. dodsoni dodsoni: Central and southern British Somalilancl, 

 south through Italian Somaliland and the eastern Hawash and Oga- 

 den districts of Ethiopia, westward in extreme southern Gallaland 

 and Shoa and northern Kenya Colony to Lakes Stefanie and Kudolf , 

 south approximately to the Northern Guaso Nyiro River, and south 

 along the coast to the mouth of the Tana River. This race is rather 

 small, the wing length varying from T8 to 84 mm in males and 70 to 

 81 mm in females. It has the rectrices tipped with white when fresh. 



2. P. dodsoni spurius: This form is altitudinally distinct from 

 dodsoni, occurring somewhat higher up as a rule, and inhabits Ennia 

 and southern Arussi-Gallaland, w^est to the southern part of the 

 Shoan lakes district. It is larger than the nominate form, wings 

 measuring 83 to 92 mm in males, 80 to 86 mm in females. In the 

 eastern part of its range (topotypical spurius) it lacks the white tips 

 on the rectrices, but the birds of southern Shoa are intermediate in 

 this respect between spurius and dodsoni, and some have white tips 

 while others do not. 



3. P. dodsoni peasei: Central and southeastern Kenya Colony, from 

 a little north of the Equator to the Tanganyika border. (This is 

 assuming that littordlis is not separable.) This race is larger tlian 

 dodsoni, wings 83 to 92 mm in males, 75 to 86 mm in females, and is 

 darker generally than either dodsoni or spurius, and in fresh plumage, 

 has a slight olive-yellowish wash on the edges of the feathers of the 

 upper parts, and wdth the feathers of the lower breast more heavily 

 striped centrally with dark brown than in either of the other races. 

 Inasmuch as the main character on which littordlis was based is the 

 less mottled breast, it appears that the birds forming this aggregate 

 are merely intermediates between peasei and dodsoni, w^hich sugges- 

 tion is supported both by geography and ecology. From spurius, 

 which it resembles in size, peasei may be told by the greater develop- 

 ment of the white tips on the rectrices and its darker, more heavily 

 mottled breast, but it must be admitted that series are needed to 

 show the differences. Thus, Gyldenstolpe had only a small series 

 (four specimens) and was not able to make out any color differences. 



The present series of dodsoni contains birds all collected within 

 approximately a month, but some of them are in worn plumage, 

 some in fresh feathers, and others are molting, indicating that the 

 breeding season was probably recently over when they were taken 

 (most recently finished in the case of those in worn plumage, least 

 recently in the case of the freshly feathered individuals). This di- 



