1894.] MANICA, SOUTH-EAST AFBICA. 57 



1885), and Smitli and Kirby (Ebop. Bxot. i. Afr. Lye. pi. ii. pp. 2 & 4, 

 1887) in recognizing the MS. genus Tingra, Boisd., with T. tropi- 

 calis, Boisd., as type, or in taking the same author's MS, species 

 Fentila undularis as the type of Pentila, a genus first defined by 

 Westv\ood (Gren. Dium. Lap. p. 503) in 1851. Although West- 

 wood undoubtedly places P. undularis first on the list of species 

 included under Pentila, it is equally certain, on studying his dia- 

 gnosis of the genus, that the characters he gives are not those 

 presented by undularis, but are (out of the four species he names) 

 solely applicable to the second species, viz. P. a6raa-as,Westw., which 

 should tlierefore be held as the type of Pentila. With P. abraxas, 

 P. tropicalis is unquestionably congeneric, and the MS. genus Tingra 

 should consequently be abandoned, Westwood defines Pentila as 

 having " labial palpi very minute ; " in the fore wings, " upper 

 discocellular arising from the postcostal at about the same distance 

 beyond the second branch as the space between the first and second 

 branches ; it is also about equal in length to the same space and 

 oblique ; middle discocellular short, less obhque ;" and in the hind 

 wings, " lower discocellular nearly transverse and very slender, 

 &c." P. a?)>-a.ras presents these important characters, as well as all 

 the others described by Westwood, whereas P. undidaris has rather 

 long, slender, and porrect palpi ; the upper discocellular nervule of 

 the fore wings so exceedingly short as to be scarcely distinguish- 

 able, and the middle one very short and quite transverse ; and in 

 the hind wings an open discoidal cell, the lower discocellular ner- 

 vule being wanting altogether. The different arrangement of the 

 discocellular neuration of the fore wings gives P. abraa-as a long 

 discoidal cell and P. undularis a short one. Butler (l. c. p. 60) 

 recognizes that P. undularis " differs considerably both in neuration 

 and palpi from the other species associated with it," and also that, 

 if no longer held as type of Pentila, a new genus would have to be 

 founded for it. 



103. Pentila tkopicalis (Boisd.). 



cS . Tingra tropivalis, Boisd. App. V^oy. Deleg. dans I'Afr. Aust. 

 p. 589. n. 46 (1847). 



S . Pentila tropicalis, Hewits, Exot. Butt. iii. pi. 60. fig. 2 

 (1866). 



5 . Tingra tropicalis, Smith & Kirby, Ehop. Exot. i. p. 3, Lycaen. 

 Afr. pi. ii. figs. 9, 10 (1887). 



The examples collected by Mr. Selous (three from the Mineni 

 Valley, one at the Lopodzi Eiver, and three near the Vunduzi 

 Eiver) resemble the variation from Mombasa, named lasti by 

 Messrs. Smith and Kirby {op. cit.Jjjcsixi.Mv. pi. viii. figs. 1-4,1889), 

 in the better development of the upperside fuscous border and 

 discocellular spots in the fore wings of both sexes, but want on 

 the upperside the comnjon discal series of small spots (reproducing 

 that always present on the underside) described and figured in the 

 Mombasa examples. As regards the macular hind-margiual border 

 on the upperside of the hind wings of the male, it is observable 



