in Butterflies of the Genus Precis. 35 



marked contrast to the equally striking blue, red-spotted 

 sesamus ; nor is this less remarkable beneath, for the under- 

 side of the former is almost as brightly coloured as its upper- 

 side, whereas in the dry form the sombre greenish black is 

 evidently of a procryptic nature. It is clear that some potent 

 cause must have been at work to produce such an effect. 

 My kind friend Prof. E. B. Poulton, in one of his interesting 

 and instructive letters, has suggested that the colouring of all 

 brilliant butterflies, whose undersides are not of a procryptic 

 character, may indicate that they possess a certain amount of 

 distastefulness — in other words, that they exhibit in a more or 

 less incipient stage the phenomenon of warning or aposematic 

 coloration. Although it may appear presumptuous for me to 

 differ from such an eminent authority on these matters, yet I 

 must admit that the habits and general behaviour of octavia- 

 natalensis in its native haunts raise very strong doubts in my 

 mind as to its inedibility even in a slight degree. Yet had I 

 known this insect from cabinet specimens alone, I should 

 certainly have been strongly inclined to include it in the 

 category of aposematic butterflies, in consideration of its 

 colouring, which is very noticeably brighter and more con- 

 spicuous on both surfaces than in several of the S. -African 

 species known to possess distasteful qualities. It is true 

 that I have never seen the insect attacked by birds (though 

 this would apply to the vast majority of 8. -African butterflies), 

 but I have not unfrequently seen both forms captured by 

 rock-frequenting lizards, particularly a dull brownish species 

 with a bright blue tail — an admirable example, by the way, 

 of colouring to attract attention to a non-vital part. I have 

 also seen these reptiles capture Precis archesia, Oram., 

 1'seudonympha vigilans, Trim., and Durbania amakosa, Trim., 

 and doubtless many other rock-luving butterflies fall a prey 

 to them. Another significant fact, pointing to an absence of 

 inedible qualities, is the eminently procryptic underside 

 colouring of the dry form, which is quite at variance with what 

 we should naturally expect in a distasteful species. Moreover 

 the activity and constant alertness of the insect is in marked 

 contrast to the slow, laboured, or gently sailing, flight of its 

 protected relatives ; and indeed I opine that it would prove 

 more than a match for the average insectivorous bird. 



As we have already seen, o.-natalensis is particularly fond 

 of bare elevated ridges and hill-tops, in which stations insec- 

 tivorous birds and mammals are generally scarce or even 

 absent. Here its chief enemies would only be the rock- 

 lizards and such invertebrates as Asilidaj and Mantidaj. 

 Now it is a generally recognized fact that animals, and more 



3* 



