122 



AFRICAN MIMETIC BUTTERFLIES 



these there were only four instances of attacks on Danainae, and in these two of the species were 

 refused, one, {T. septentrionis) after inspection, and the other {Danaida sp.) after tasting. Only 

 once was a,nAcraea seen to be attacked and eaten. In the Nearctic region forty-six identified 

 species of birds, and many unidentified, were either seen to attack butterflies or were found on 

 dissection to have eaten butterflies, of which some twenty-six species were identified. Here 

 again it is a remarkable fact that on the only occasion when the common Danaine, Anosia plexip- 

 pus^ was seen to be attacked, the insect was immediately dropped again without being eaten. 



The records from the Neotropical region are extremely few, and yet, judging from the 

 extent to which butterfly mimicry has been developed in the South iVmerican species, such 

 attacks must be of frequent occurrence. One instance, however, is very instructive. It is 

 the statement of Poppig, given by Haase in his * Researches on Mimicry ', that ' there is 

 no difficulty in recognizing the favourite perch of a Galbula, for the wings of the largest and 

 most splendid butterflies, whose bodies alone are eaten, cover the ground for some steps 

 around'. It may safely be assumed that such an observation, though proving that 'the 

 largest and most splendid butterflies ' are eaten, does not preclude the destruction of other 

 and smaller species, since the wings of these would be less easily observed. Many of the 

 observed instances of the attacks of birds on butterflies are of special interest. Thus in Africa, 

 Swynnerton records that a White Spotted Flycatcher {Trochocercus alhonotatus) darted out 

 at a Mylothris, but turned back on reaching it without an attempt at capture. Most of the 

 forms of Mylothris serve as models for other species, and we are justified in concluding that 

 the bird in this case recognized a distasteful species. In other instances given by Captain Boyd 

 Alexander, a Bee-Eater which he shot had its mouth ' so stuffed with butterflies that it 

 appeared to me marvellous that it had not choked '. Again, in the Indo-Malayan region, 

 it is recorded by Sir George Hampson (Proc. Ent. Soc, 1897, p. xxxviii), that in the cases which 

 he had seen of birds attacking butterflies the Euploeae and Danainae were attacked as often as 

 any others, ' but usually escaped eventually from the beak of the bird, and flew away none the 

 worse, owing to the toughness of their integuments.' This statement, which was made rather 

 with the intention of discounting the mimetic theory, certainly strengthens it. It is not the 

 habit of animals in a state of nature to allow their prey when once captured to escape, and if 

 the Euploeae and Danainae do thus avoid destruction, it must be on account of qualities 

 which convey unpleasant sensations to their enemies. E. L. Arnold, in ' On the Indian Hills 

 describes the ground beneath the ' watch towers ' of the Green Bee-Eater as being ' thickly 

 strewn with gaity painted shreds of unfortunate butterflies and bees '. Now whilst it is 

 the habit of these birds to eat butterflies they are evidently not always specially adapted 

 to eat distasteful species, since E. H. Aitken (Journ. Bombay Soc, xvi, 1904, p. 156) records 

 that a Bee-Eater ' caught a Danais, but dropped it as soon as it had tasted it, and the Danais 

 flew away little the worse '. Lieut. -Col. Bingham (Trans. Ent. Soc, 1902, p. 362) records 

 that Bee-Eaters habitually attacked many species of butterflies, but that he particularly 

 noticed that they never went for a Danais or a Eupioea, or for P. macareits and P. xenocles, 

 which are mimics of Danais, though two or three species of Danais, four or five of Eupioea, 

 and the two above mentioned mimicking Papilios simply swarmed along the whole road. 



Apart from the direct evidence afforded by the numerous instances in Marshall's paper, 

 there is in the Hope Museum at Oxford a large collection of butterflies, which have been 

 injured by the attacks of insectivorous animals, principally birds and lizards. Very many 

 of these have been seized when at rest, as is shown by the perfect^ symmetrical injuries, 

 taking the form of notches in opposite wings which exactly coincide when the wings are 

 closed. Some eighty-two such specimens are illustrated on Plates IX, X, and XI, in Trans. 

 Ent. Soc, 1902, and the descriptions of the plates give notes on each species, and the probable 



