The Bionomics of South African Insects. 485 



is due solely to climatic influence, and that such a change 

 in the former is checked by the necessity for keeping the 

 warning coloration uniform, or are we to consider that the 

 slight change in the former is all that climate can effect, 

 and that in the case of ■pctr;va this slight climatic effect 

 has been enhanced by some other caure — presumably 

 protection ? Personally I incline to the latter view, but in 

 either case it is clear that there are varying grades of 

 protection by distastefulness in the genus." 



An extract from another letter states the same im- 

 portant conclusions as to the severity of the struggle 

 during the dry season. 



" Salishurif, March, 10, 1898. — There are very few butter- 

 flies (South African, at least), exclusive of the admittedly 

 protected species and their mimics, of which the bright 

 colour cannot be explained on the TcracohiS-KaUima 

 basis. The most evident exceptions are ByUict, certain 

 species of Precis, as sesamvs (form natalensis), ccryne, etc., 

 which are practically coloured the same below as above, 

 and Behnois severina and nusentina. The first I will 

 admit has been so far a stumbling-block to me, though I 

 am not yet prepared to accept it as a protected species. 

 Provided that my ideas on seasonal variation in Precis be 

 correct, these would also fall under the same heading as 

 Teraeohts, for like them they only assume the protective 

 under-side colouring during winter, when attacks from birds 

 are no doubt a great deal more to be feared, owing to the 

 almost complete absence of easily-caught prey, such as 

 beetles and other small insects ; the summer forms 

 probably are very little molested by birds, owing to their 

 great agility and alertness, and the profusion of other 

 insects at that period ; they do, however, not unfre- 

 quently fall a prey to the rock-lizards, which stalk them 

 with much astuteness, as I have observed on several 

 occasions. 



" With regard to BeJenois my mind is still open, for it is 

 a very curious genus, containing as it does the above two 

 species which might perhaps from certain considerations be 

 considered protected, and at the same time a species like B. 

 gidica which evidently comes under the Teracolus heading, 

 and lastly B. thyset which, to my mind at least, is clearly a 

 Batesian mimic.'' 



After a consideration of the evidence brought forward 

 above, it will be generally admitted that the struggle for 



