Tlie Bionomics of South African Insects. 427 



of large black spots on a liglit ground which renders this 

 part of the wing prominent in such large numbers of 

 Ethiopian butterflies. 



In the wet phase of. Freds scsamns this area is also 

 remarkably conspicuous, but by a method which is as 

 positive to negative in relation to other distasteful butter- 

 Hies inhabiting the same part of the world, viz. by the 

 appearance of light spots on a black ground, instead of 

 black spots on a light ground. 



Thus it is improbable that this particular element in the 

 conspicuous appearance of the under-side of the wet phase 

 of P. scsamus can be mimetic, and its existence, side by 

 side with a general resemblance in colour and pattern to a 

 large Acr/ect, is evidence that such resemblance is Mlil- 

 lerian or synaposematic rather than Batesian or pseudapo- 

 sematic. This argument is much strengthened by the 

 discussion of the wet phase of the allied F. archcsia (see 

 pp. 428-430). 



Much that has been said of P. scsamns applies with 

 greater force to the closely-allied F. antilope. On compar- 

 ing the under-side of the wet phase, shown in Fig. 4, Plate 

 XIII, with the upper-side of the same specimen, shown in 

 Fig. 8, Plate XII, it is at once seen that the difference in 

 tint of the ground-colour and in conspicuousness of the 

 marginal band on the two wing surfaces is far more pro- 

 nounced than in the species which has been just described. 

 In fact, with an upper-side which is much less conspicuous 

 than sesamns (compare Figs. 3 and \h, Plate XII : it must 

 be remembered that the ground-colour of 3 is merely 

 tawny, while that of Ih is salmon-red), the wet phase of 

 aniilofe combines an under-side which is distinctly more 

 conspicuous than that of tlie corresponding form of the 

 allied larger species (compare Figs. 4 and 1, Plate XIII). 

 The increased conspicuousness is especially clear in the 

 relative size of the spots in the basal black patch and the 

 inclusion in it of a very large piece of the ground-colour of 

 the fore-wing. Here too the increased liglitness of the 

 spots of ground-colour in the black area is often distinct, 

 as it is in Fig. 4, Plate XIII, but in a large proportion of 

 the individuals I have had the opportunity of examining 

 it is only feebly marked. 



In F. antUoiw. there is probably some considerable syn- 

 aposematic approach towards the Acrxa type, but to a less 

 extent than in F. scsamus ; while the conspicuous basal 



