The Bionomic!^ of South African Insects. 437 



In the Acr.vinm we find that the least unpalatable species 

 of an unpalatable and conspicuous family have been com- 

 pelled to produce relatively inconspicuous generations in 

 the severe struggle of the dry season : in the Nipnphalinai 

 we find that some of the less palatable species of a com- 

 paratively palatable and inconspicuous family have been 

 compelled to produce strongly conspicuous generations in 

 the wet season when more edible insect food is abundant. 



The interpretation I have here suggested was put 

 forward very cautiously in a note, dated Nov. 1898, to a 

 short paper on Mr. Marshall's results with P. scsamus in 

 the Proc. Ent. Soc. Lond., Oct. 5, 1898, pp. xxv, xxvi. 

 The note points out tliat insects with warning colours 

 are not to be seen in an English winter. " Those such as 

 Coccinellid^e, which exist in the perfect form, hide them- 

 selves. The reason probably is that the amount of palat- 

 able food available is not sufificient to make it safe to rely 

 on unpalatability, accompanied by warning colouring [see 

 also ' Colours of Animals/ London, 1890, pp. 179, 180]. 

 Experiments with hungry animals support this view. It 

 is possible that the conditions are similar in South Africa 

 [it is perhaps unnecessary to state that organic conditions 

 were alone referred to], and that warning colours are more 

 characteristic of the wet than of the dry season, thus 

 affording greater opportunities for mimetic resemblance. 

 If it should hereafter be shown that Precis is to some 

 extent unpalatable, and that its resemblance to an Acrsean 

 type is synaposematic rather than pseudaposematic, the 

 parallelism with our own fauna would be even closer, the 

 conspicuous species which hide and thus adopt procryptic 

 habits being represented by one which gives rise to 

 another brood with markedly procryptic colouring and 

 habits." 



Mr. Marshall in commenting on this note records the 

 following interesting observations on the habits of South 

 African Coleoptera as determined by damp and dryness. 



"Salislnmj, Feb. 12, 1899.— Do you think that the 

 English Coccinclliclm really hide in winter owing to their 

 increased danger from enemies, and not from climatic 

 causes ? I ask the question because in this country 

 Coleoptera are highly susceptible to weather. They 

 appear to be for the most part absolutely dependent on 

 moisture, this being especially noticeable among the ter- 

 restrial forms such as CicindelidcT , Garahida^, Psammodes, 



