The Bionomics of South African Insects. 409 



of warning or aposematic coloration that we should know. 

 The behaviour is sufficient, and affords convincing proof. 



Plateau also brings two indirect lines of argument to 

 bear on the question. He points out (§ 3) that the larva 

 is not wholly conspicuous, but requires searcliing for if it 

 is to be obtained in large numbers. This is due to its 

 habit when young of resting along the serrated edges of 

 leaves, exposing only the reddish lateral band sprinkled with 

 black points. This appearance is at a little distance very 

 like the reddish edges of many leaves. When it is older 

 he observes that it stretches itself longitudinally along the 

 branches in tJje inner darker part of the bush at do great 

 height above the ground. When the bush is shaken it 

 falls or lets itself down by a thread ; and as soon as it 

 reaches the ground rolls in a ring. In this position, which 

 it retains for a long time, it resembles the excrement of a 

 bird. Mr. Beddard too points out {loc. cit., pp. 167, 1(38) 

 that these larvas " like other Geometers ... do not begin 

 to feed until evening. I have a quantity of these cater- 

 pillars on some thick-leaved shrubs in my garden ; during 

 the daytime none are visible, but in the evening they 

 commence to crawl about quite actively." I have myself 

 often observed the larva moving and freely exposed by 

 day on the sides and tops of hedges, but after the state- 

 ments of these two naturalists I am quite ready to believe 

 that my observations were exceptional. Indeed on general 

 grounds I must believe that this is so ; for it would not be 

 an advantage but a great disadvantage to the Abraxas to 

 display its aposematic colouring at too great a distance. 

 It has certain enemies, such as the cuckoo, and it would 

 doubtless be dangerous if these were able to see it upon 

 the bush at a great distance. Its colours would be con- 

 spicuous enough to the majority of insect enemies hunting 

 the bush for food ; and the very rough resemblance to 

 leaf-edges, branches, and birds' excrement would never 

 impose upon the eyes which enable their owners to find 

 abundant food in spite of the extraordinarily close likeness 

 to each one of these objects which is attained by so many 

 of the cryptic species which they hunt. 



Plateau's second indirect argument depends on the 

 undoubted fact that the pupa of the Abraxas resembles a 

 wasp. This he regards as an example of (Batesian) 

 mimicry, and believes moreover that it has never been 

 pointed out before. This is very far from being the case. 



