104 SOUTH- AFRICAN BUTTERFLIES. 



springing from edge of subcostal nervure ; discoidal cell very short, 

 rather broad, obliquely truncate terminally by the almost continuous 

 disco-cellular nervules. Fore-legs of ^ small, but quite noticeable, with 

 the femur short, slender, and finely hairy, but with the tibia and tarsus 

 densely hairy ; those of $ much larger and longer, and very finely hairy 

 throughout. Middle and hind legs moderately stout, clothed with scales ; 

 tibial spurs short. Abdomen in $ rather elongated, and tufted with 

 hair at the end. 



This genus of numerous closely-related species is characteristic of 

 the Old- World Tropics, extending from the West-African coast to 

 Australia, only a few species occurring in extra-tropical countries, one 

 of them inhabiting Japan. About half of the ninety species recorded 

 are natives of South-Eastern Asia and the Indo- Malayan Islands, while 

 Australia (including the Austro-Malayan Islands) and Africa nearly 

 equally divide the remaining half. Westwood {Gen. Diurn. Lep., ii. 

 p. 393) has pointed out that the genus is divisible into two groups, — 

 one with only the costal nervure of the fore-wings swollen basally, and 

 with the $ badge borne on the submedian nervure of the same wings, — 

 the other with the median and submedian nervures also swollen, and 

 with the badge borne on the subcostal nervure of the hind-wings.'^ All 

 the African species appear to belong to the latter division. 



These butterflies are of medium size and obscure-brown colouring 

 (varied in some Eastern species with fulvous-ochreous), with submarginal 

 ocellated spots more numerous and distinct on the under than on the 

 upper side of the wings. The fore-wings usually bear only two of these 

 ocelli, the lower of which is almost always considerably the larger of 

 the two. The two South- African species that I have seen in life fre- 

 quent wooded spots, M. Safitzci — which is by far the commoner and 

 more widely distributed — preferring shady spots in woods, and M. inr- 

 S])ieua more open localities on the outskirts. I am not aware of the 

 haunts of M. iSimonsii, a singular pale yellow-ochreous form, which 

 appears just to penetrate extra-tropical South Africa, but in all pro- 

 bability it is also a sylvan butterfly. The flight of Safitza and Per- 

 spicuct is extremely weak and low, and interrupted by frequent settling 

 on the ground or on herbage. 



It must have been in error that the little M. Narcissus^ Fab., so 

 abundant in Mauritius, was recorded as a native of Natal, no specimen 

 having occurred in any of the numerous collections from various parts 

 of that colony which I have examined during the past twenty-two 

 years. 



^ Mr. Moore's genera Orsotrkma and Cali/sisme (Lcp. Ceylon, i. pp. 20, 22) seem respec- 

 tively to correspond with these two groups. The Japanese M. Perdiccas, Hewits., links the 

 two divisions, presenting the three swollen nervules and S badge in the fore-wings, and cdso 

 the 6 badge in the hind-wings. 



