470 NOVITATES ZoOLOaiOAH XXVII. 1920. 



This form badly requires confirmation ! It is in good fresh plumage, while 

 the type of centralis is in very much worn garb. This is peculiar, as both were 

 shot in August ; the latter may be juvenile, but it does not look so ! Another 

 question is, if both could not be the same bird ? B. graueri, however, is much 

 larger, and we do not know that the sexes differ appreciably in size in the genus. 



613. Bessonomis (1 Cossypha) gambagae Hart. = Oenanthe familiaris 



gambagae. 



Bessonornis {Cossypha) gambagae Hartert, Bull. B.O. Clvb, x. p. v. (1899 — Gambaga, Gold Coast 

 Hinterland). 



Type : " $," Gambaga, 27.viii. 1898. Capt. W. Giffard leg. 



I must confess that I am still somewhat in doubt about the genus in which 

 this bird should be placed. I admit that I, following Sharpe's arrangement in 

 the British Museum, was wrong in placing this form in " Bessonornis," which is, 

 apparently, not separable from Cossypha. Reichenow certainly came near the 

 truth when he placed familiaris, galtoni, and falkensteini in Phoenicurus. I 

 would gladly agree to this, if I did not consider that they are still better placed 

 in Oenanthe. Will anyone point out a generic difference between Oenanthe 

 chrysopygia and the disputed familiaris, galtoni, falkensteini, gambagae, omoensis ? 

 This, of course, raises the question of what the differences are between Oenanthe 

 (formerly Saxicola) and Phoenicums ? That they are very unsatisfactory may be 

 gathered from comparing the generic characters described in vol. v. of the Cat. 



B. Brit. Mus., Reichenow's Vog. Afr. iii. (where they are most wrongly placed 

 in two different subfamiUes !), and my Vog. pal. Fauna, i. I stiU believe, however, 

 that they can be kept separate, though they are closely allied, by the following 

 characters : 



In Oenanthe the bill is comparatively larger, tail comparatively shorter. 

 In Phoenicurus the beak is weaker, tail comparatively longer. The artificial 

 distinction of the proportion of the tail and bill used by Seebohm holds good, 

 though it is not a nice one. If the above distinction is accepted, the familiaris 

 group goes into Oenanthe by its bill, while the tail is not reaUy shorter, except 

 by comparison with the bill ! I am certainly of opinion that galtoni, falkensteini. 

 gambagae, and omoensis are subspecies, and that gambagae is not identical with 

 falkensteini, having the under tail-coverts reddish. 



614. Saxicola galtoni omoensis Neum. = Oenanthe familiaris omoensis. 

 Saxicola galtoni omoensis Neumann, Orn, Monatsher. 1904. p. 163 (" Oino-Gebiet "). ' 



Type : ^ ad., Baka in Konta, 28. ii. 1901. Oscar Neumann leg. No. 949. 



It must be said that the two specimens collected by Neumann are in very 

 badly worn plumage, evidently shortly after the breeding season, and that a 

 series of fresh specimens are desirable to better explain the differences of this 

 undoubtedly recognizable form. 



615. Oenanthe leucopyga aegra Hart. = Oenanthe leucopyga aegra. 

 Oenanthe leucopyga aegra Hartert, Nov. Zool. xx. p. 55 (1913 — Algerian Sahara). 



Type: c? ad., Jara Krima, near Ouargla, 10. iii. 1914. E. Hartert and 



C. Hilgert leg. No. 206. 



