354 Mr. CI. A. K. Marshall on 



Proc. 1897, pp. xx-xxxii, xxxiv-xlvii). The following 

 extracts from a letter indicate the line Mr. Marshall would 

 have taken had he been in England at the time. — E. B. P.] 

 Malvern, Natal ; Oct. 7, 1897. — I am much struck with 

 the large amount of adverse criticism levelled against the 

 theory of even Batesian mimicry. The theory of converg- 

 ence (Miillerian mimicry) might perhaps be considered as 

 debatable, but how any one who has paid any attention 

 to the subject can doubt the reality of Batesian mimicry, 

 I cannot understand, and the attempt to explain it away 

 by climatic causes seems to me weak in the extreme. If 

 the view, advocated by many, that birds cannot be reckoned 

 among the principal enemies of butterflies in their imago 

 state, be true, then I consider that we may practically 

 abandon the whole theory of mimicry as at present applied 

 to the Acrseinm and Danainm of South Africa at all 

 events, for from what I have observed of these insects I 

 am convinced that their warning coloration cannot have 

 reference to either mantises, AsilidR\ or lizards, which are 

 practically the only other enemies that can be taken into 

 account. Moreover, the swift flight of the majority of 

 edible species can only have been developed to enable 

 them to escape from winged enemies, and that this de- 

 velopment is due to Asilidm or dragon-flies is more than 

 I can believe. Certainly the paucity of records of birds 

 eating butterflies is somewhat disconcerting, but this is 

 doubtless due to the fact that not sufficient attention has 

 been paid to the subject, which would entail long and 

 patient observation of the birds themselves, an occupation 

 that the average entomologist is not likely to indulge in 

 when out collecting. Personally I do not suppose I have 

 seen such an occurrence more than perhaps half-a-dozen 

 times ; the birds being the Paradise flycatcher {Terpsi- 

 phone perspicillaia), the bee-eater (Merops apiaster), and 

 two xoWevs, {Coracias spatulata aud iJurystovucs afcr'); but 

 then I admit that I have paid little or no attention to 

 the matter until quite recently. 



The habits of the Tcracoli, especially in their winter 

 forms, have always seemed to me strongly suggestive of 

 their being frequently attacked by birds. With hardly an 

 exception they are fairly swift fliers (especially the 

 " purple -tips"), keeping comparatively close to the ground 

 and dodging well. If struck at gently as they fly by, 

 they dodge and hurry onwards but still continue their 



