188 RHOPALOCERA AFRICA AUSTRALIA. 



The variations of the under-side in this species are so 

 numerous and so finely graduated into each other that it is 

 impossible to fix upon any variety properly so termed, i.e., a 

 constant form differing from the type. The ground-colour is 

 tinged with ochre, pinkish-red, dark-brown, or purple-grey ; 

 the transverse stripes wanting, indicated by detached dark 

 blotches, or broadly shaded with dark-brown internally (while 

 the ground beyond them is very pale) ; the ocelli very 

 conspicuous in both or one wing to their full number, half 

 wanting, ill-defined, without rings, without black, very 

 indistinct, or barely traceable as whitish or pale dots. 

 The ocellus of upper-side of fore-wing is sometimes com- 

 pounded of three black spots. The outline of the wings 

 also varies much, especially as regards the fore-wing, the 

 hind-margin of which presents every gradation between being 

 almost straight (save for a slight prominency in apical region), 

 and the assumption of an almost falcate form. 



LARVA. Wholly pale-yellowish green, with numerous 

 rows of small black dots : a few short hairs generally scattered 

 and extending over horns of head and anal fork. Three 

 longitudinal streaks, slightly darker than ground-colour, on 

 each side. 



PUPA. Light-green, paler on wing-covers which are 

 marked with three longitudinal black lines. Attached by the 

 tail only to the stem of some plant. 



Larva and Pupa described from the figures on plate VI of Hors field and 

 Moore's Catalogue. 



Boisduval's descriptions in " Faune Ent. de Mad. " agree therewith, but 

 he only mentions four longitudinal streaks in the Larva, which, he states, 

 feeds on grasses, like most of the Satyrida. 



I find it impossible to admit Cyllo BanJcia (or Banksia) of Fabricius as 

 distinct from Leda, though all authors appear hitherto to have so considered 

 it.* The only real difference noted by the founder of the species consists 

 in the nearly obsolete condition of the ocelli of hind-wing, a character 

 admitted to be most variable in Leda, as is proved by the universal acknow- 

 ledgement of Solandra, Fab., as a variety of Linne's species, this insect 

 only differing from Leda in the absence of some of these ocelli. The stria3 

 of the under-side, which Fabricius mentions as characteristic of Bankia, 

 also exist in Leda, Cramer giving them in his fig. (pi. 196). Having 

 arranged under Leda and Bankia the various habitats assigned to them 

 respectively by all the authors to whom 1 have access, I find their range to 

 be equal in extent, and that they are allowed to co-exist in seven distinct 

 regions, viz. : West Africa, Mauritius, Cororaandel, Darjeeling, Bengal, 

 Java, and Australia. The four Cramerian species, by most authors referred 

 to Bankia, might with equal reason have been held variations of Leda, all 

 naving been brought from the East Indies, where the latter is known to 

 abound, and none varying more from Fabricius' description of that species 



* Boisduval, however ("Faune Ent. de Mad.," &c., p. 59), hints at the 

 likelihood of the two species being identical. 



