172 
Annals or the Transvaal Museum. 
sheppardi, with Shelley’s description of fratrum, with the result that I think 
there can be no doubt as to the two species being synonymous. Sclater 
in writing of the specimen from Beira does not refer to B. sheppardi ; but 
remarks that Shelley was wrong in stating that the type of fratrum was a 
female, the error having arisen on account of the label not indicating the 
sex, and the specimen not having a black breast band like its congeners 
in South Africa. It is evident that the type was a male, for the dis- 
coverers, after whom the species was named, expressly stated in their 
notes, to which the description was appended by Shelley, that the two 
specimens remitted (the type and cotype) were a male and a female 
respectively. 
Anthus daviesi, spec. nov. 
Most closely allied to A. vaalensis, Shelley, but differing in having a 
longer wing and a shorter and more curved hind claw. In colour it is like 
A. leucophrys on the under surface of the head and body, but like vaalensis 
on the upper parts. A. vaalensis formerly had the longest wing of any 
South African species of the genus, but the present species has an even 
longer wing, in three males from the type locality measuring not less than 
111 mm. as against not more than 106 in vaalensis. Having regard to 
the little variation apparent in wing measurements of specimens of the 
same sex of other members of the genus, this difference of 5 mm. is of 
considerable importance, and having regard to the difference in the hind 
claw, I think its separation as a distinct species is warranted. The type 
is T.M. No. 8666, ex C. G. Davies, Matatiele, East Griqualand, 4th May, 
1911. Wing 111, tail 83, tarsus 27, culmen 16; hind claw, dorsal length 9, 
ventral length 7 *5. 
Cossypha haagneri, Gunning. 
There are good reasons for supposing this speces to be a synonym of 
C. bicolor. There are in the collection at the present time sixteen typical 
specimens of C. bicolor from Ngqeleni or Port St. Johns District, all taken 
by the same collector (H. H. Swinny), from whom the type of C. haagneri was 
obtained ; also another specimen from the same place and collector, which 
was identified by Haagner with C. bicolor, but which might equally well 
be placed with C. haagneri, as it has a yellow eyebrow and black ear- 
coverts ribbed with yellow. The yellow pigmentation of the feathers in 
the three allied species— bicolor, natalensis, and heuglini — is by no means 
always evenly dispersed over certain groups of feathers, but sometimes 
strays into the darker ones adjoining. This is frequently seen in the 
feathers of the hind neck in specimens of bicolor. There is a specimen of 
heuglini in the collection which has half of one eyebrow yellow in place 
of white, and a single yellowish plume in the midst of the black ear-coverts. 
Then, again, if we examine a number of specimens of C. natalensis , it is 
remarkable how much the yellow colouring runs into the slate colour of 
the back, the amount varying considerably in different individuals. 
Finally, the immature birds of all three species have a mottled yellow 
and black appearance. It is easy to conceive, therefore, of yellow having 
predominated in certain parts, either to the exclusion of black or slate 
