HESPEKID.E. 373 



and n-ith more or less of a violaeeous gloss. Fore-wing : on inner mar- 

 gin, from base to about middle, a dull-whitish space, not rising above 

 first median nervule. Hind-wing : hcfore anal angle, hctween first 

 median nervule and siibmedian nermire, an orange-red marlcing, divided 

 hg a transverse hlacJc streak ; below this mark an orange spot on inner 

 margin ; orange cilia as above. 



Head and bodg above much darker and browner than in Forestan 

 ov Pisistratus ; the spots of head as in Forestan, but witli an additional 

 median frontal spot, and all orange-red ; ahdomcn with thin whitish 

 median rings and orange-yellow incision rings. Beneath, palpi, median 

 front of thorax, spots and tufts of legs, and median stripe of abdomen 

 all orange-red. 



$ Like ^, but slightly paler. 



This very liandsonie and distinct species is readily known by its 

 want of any white band on the under side of the hind-wings and by 

 the presence there of the vivid orange-red spots near the anal-angular 

 lobe. In the former respect it is allied to the Malagasy Batch, (Boisd.),^ 

 but this species has the under side of the hind-wings without marking 

 of any kind. Perhaps the nearest known ally of Keithloa is the Soco- 

 tran species named Jucunda by Butler {Froe. Zool. Soc. Lond., 1881, 

 p. 179, pi. xviii. f. 8). This is very like Keithloa on the upper side, 

 but wants the orange discal pubescence in the hind-wings ; and on the 

 under side of the same wings it has, in addition to two orange-red 

 marks near anal angle, some median marks of the same colour, the 

 lowest and largest of which bears a black spot, 



I have the skins of two pupte sent to the Museum by Colonel 

 Bowker from D'Urban in August 1881, from one of which the imago 

 emerged in the following October. They are very like the pupa of 

 Forestan, being covered with a dense white efflorescence, but their 

 colouring beneath this is dark-red instead of pale-greenish. 



Unlike its two South-African congeners, Forestan and Pisisfratus, this 

 butterfly has rather a limited range, and I have not found any record of its 

 occurrence in Tropical Africa ; from its presence, however, at Delagoa Bay, it 

 is very likely to extend into Mozambique. It is numerous on the Natal Coast, 

 where its habits quite agree with those of its congeners just mentioned. I 

 took many specimens about D'Urban in February 1867; and met with the 

 species singly at Port Elizabeth in January, and at King William's Town in 

 February 1878. Colonel Bowker, writing from D'Urban on 24th May 1887, 

 observed : *' Keithloa is very common here now, and I have been noticing the 

 curious behaviour of the sexes. The 9 darts away from a flowering-tree the 

 species frequents and settles on the ground, closely followed by the ^ ; after a 

 little she rises slowly, keeping her wings constantly fluttering, while the ^ 

 circles round her; and when they reach about a yard above the ground, off they 



1 This species was, I believe, erroneously catalogued as South-African in my earlier work 

 (ii. p. 317). The specimen then described was ticketed "Natal" on the authority of Mr. 

 E. L. Layard ; but I have little doubt that it was actually taken in Madagascar, and after- 

 wards by accident was misplaced among the South-African allied species. 



