1910.] FROM NORTHERN RHODESIA. IT 



underside of an extreme dry-season specimen is well figured bv 

 Trimen, P. Z. S. 1894, pi. iv. fig. 4. 



I took this form sparingly in the Broken Hill district, and 

 commonly in the Luang wa and Chambezi valleys. To the west of 

 these localities, as we shall see, it is replaced by another race. The 

 typical form of acrita appears to extend up the eastern side of 

 Africa from the Transvaal to a little north of the latitude of 

 Mombasa. 



(2) Acrcea acrita ambigua Trim. P. Z. S. 1891, p. 70. 



This appears to be the western or perhaps central race of the 

 species. My collection contains a long series, some 105 individuals, 

 captured in the Bangweolo, Mweru and Tanganyika districts and 

 less commonly in Katanga. This race is characterised by the 

 breadth of the black apex to the primaries, normally 6-7 mm. 

 wide, and by a number of other characters already pointed out 

 by Mr. Trimen (loc. cit.). Mr. Trimen has kindly allowed me to 

 examine the male mentioned by him (loc. cit. p. 71) as captured 

 by Mr. Selous near the Chobe river, Upper Zambezi. I find that 

 my more northern specimens differ only in the very much slighter 

 development of the pale subapical patch to the primary. This 

 pale area appears in the male to be due to the absence of scales 

 rather than to the presence of white pigment. Wet-season males 

 are rather more heavily marked than the dry ones, which 

 frequently have the spots of the primaries much reduced. 

 Occasional individuals of this latter type have the black apex 

 nearly if not quite as narrow as a heavily tipped specimen of the 

 type race. They can, however, be distinguished at a glance, in 

 these cases, by the great reduction, sometimes disappearance, of 

 the spots on the primary and a general reduction of all the 

 markings of both wings. An extreme of this sort is figured by 

 Thurau, loc. cit. pi. ii. fig. 9, as var. utengidensis. Dry females, 

 except that they are duller coloured, do not differ much from the 

 males. With regard to wet females of this race it is unfortunate 

 that the specimen figured by Trimen is probably somewhat 

 aberrant. I have taken the wet-season phases only in Katanga ; 

 I did not have an opportunity of doing so in the Lake Bangweolo 

 district, etc. These wet Katanga females, four in number, 

 resemble Mr. Trimen's figure in possessing a well-marked white 

 subapical bar, but are larger and the red colour is entirely re- 

 placed by a dusky grey shade. 



This western race of the species appears to extend from the 

 Damara-land localities given by Mr. Trimen across the Upper 

 Zambezi through the south-eastern portion of the Congo State 

 to the districts of Lakes Bangweolo, Mweru, and Tanganyika. 

 There is also one specimen of this race in the British Museum 

 from the Victoria Nyanza, already referred to by Mr. Trimen. 

 The fact that typical acrita, which occurs in the mid-Chambezi 

 valley in N.E. Rhodesia, is replaced by ambigua on the north- 

 eastern shore of Lake Bangweolo is highly remarkable, there being 



Proc. Zool. Soc— 1910. No. II. 2 



