C 4fi ) 



'/. Mirafi-a africana athi .su1js]i. imv. 



Athi Waiu, Britisli East Africa, ~4. and ,'•">. 1. ISii'.i. Tbis luaguiliceiil ionu is 

 mncli more distinct than forms b and c, and most ornithologists would probably 

 call it a " good s.i)ecies." It difi'crs at a glance from the other forms by its mnch 

 paler and more greyish general apj)oarance. The wing-coverts are not rufous, but 

 greyish-brown with black ante-apical patches. The feathers of the uppcrside are 

 black with wide whitish-brown borders, the head much lighter than in the allied 

 forms, the nape not conspicuously rufons, the superciliary stripe almost white and 

 very conspicuous, the throat white, the size perhaps a little smaller : wing in one 

 male 10], in the females 05 — 06 mm. Under wing-coverts and wing-liuir.g much 

 jialir than in tlie other forms, the under tail-coverts very pale isabelliue, nearly 

 white. " Iris orange golden, feet flesh-colour, upper bill dark grey, lower yellowisli 

 grey." 



A young bird, Nairobe, :'>!. 1. IS'.iii, evidently belongs to this same form. It is 

 mucli more blackish, the crown of the head almost black, the feathers of tlie upper- 

 side and wing-coverts are blackish brown with sharply limited whitish edges. The 

 spots on the cro]i-region are larger tlian in the adult bird, the rectrices, as usual in- 

 yonng larks, are narrower and somewhat pointed. -'Tlie iris was brown, feet light 

 brown."' 



I-"-'. Mirafra fischeri (Rchw.) 



cJ ad. Kitagwetoa in Torn, 0. 4. Is'.i'.i. " Iris dark brown, feet pale brown, 

 upper bill dark, lower pale grey." Specimens of this lark are in the Tring Museum 

 now from Ussambiro, Fadjuli (Emin Pasha coll.), Dar-es-Salaam (collector un- 

 known), Mombasa (Kretschmer coll.), and Samburu, four days from ^Mombasa 

 (Ansorge coll.). 



1311. Mirafra rufocinnamoinea (Salvad.) (an torrida?) 



J ad. L'ampi-ya-Siniba, 11. 1. 1890. " Iris sepia, feet liglit brown. Ujiper bill 

 dark grey, lower ligiit grey."' 



Rharpe has identified Shelley's Mirafrn torrida from Ugogo with Salvadori"s 

 .1/. riifocwrtiimomcii from Aliyssinia, and Professor Reichenow has kindly named the 

 specimen in question as .1/. nifocinnamomea, thus evidently agreeing with Dr. Sliarpc. 

 It is, of course, quite jiossible that both forms cannot be sejiarated. but a series from 

 the various jilaces shoidd be compared. As it is, the rufous foi'in liefore me i> 

 .apparently only a rufons subspecies of .1/. p'schrri, from which it differs only in the 

 very much more rufous colouration everywhere. In the description of the type by 

 Henglin I do not see the cross-markings mentioned which are so characteristic to 

 the uppcrside o^ M. fischer! t'xclieri audits rufous subspecies. The latter is probably 

 the desert form. (( T. P. Z. .S. 1sn2 pi. XVII. (very bad figure), (>//. [I. lirit. .Uus. 

 XIII. p. OlMI.) 



131. Mirafra intercedens Rchw. 



Shot at ( 'amjji-ya-Simba on the same day as tlie jireceding species, and was 

 therefore on the label marked as the female of the larter. This, however, is erro- 

 neous, the two species being jjerfectiy distinct, (.'olour of iris sepia, feet and bill as 

 in the preceding species. 



