Production of Mimetic BesemUances among Buttcrjiies. 331 



in this direction does not seem to be altogether satisfactory 

 as it stands at present. In South Africa, at all events, 

 the vast majority of lizards live only on the ground or 

 among rocks, and personally I have never met with any 

 truly arboreal species. Yet it is among the tree-frequenting 

 butterflies that warning colours and mimicry find their 

 highest development ; nor can I call to mind any un- 

 doubtedly mimetic butterfly which normally settles on 

 rocks or on the ground, with the exception of a few species 

 such as Atcrica gcilene or Papilio echerioides, which only 

 frequent dense forests — localities in which ground lizards 

 are, so far as my own experience goes, conspicuous by 

 their entire absence. Indeed, the habits of the South 

 African Danaines and their many mimics are such as to 

 render it antecedently improbable that they are normally 

 liable to be preyed upon by lizards. It is possible that the 

 conditions are quite different in other tropical countries, 

 but the question does not appear to have been investigated 

 from this point of view, and further observations are to 

 be desired. 



Objections to the View that Birds attack 

 Butterflies. 



It is now about twelve years ago that this question was 

 discussed at some length at a meeting of this Society 

 (Proc. Ent. Soc, 1897, pp. xiii-xxvi) as the outcome of an 

 interesting paper by Dr. Dixey on the subject of mimetic 

 attraction. Judging by the views advanced during this 

 discussion, as well as those published by other observers, it 

 would appear that a considerable majority of entomologists 

 are of opinion that on the whole butterflies suffer but little 

 from the attacks of birds. 



The supporters of these adverse views who are most 

 usually cited in criticisms directed against the theories of 

 mimicry are Scudder (whose experience is confined to 

 N. America), Packard (N. America), Pryer (Japan and 

 Borneo), Piepers (Java) and Skertchley (Borneo). But in 

 regard to Scudder it must be noted that, though he has 

 seen but few cases of birds pursuing butterflies in N. 

 America, yet he readily admits that such occurrences are 

 probably much more frequent in tropical countries (Butterfl. 

 of Eastern U.S., II, p. 1612). Similar negative evidence 

 has more recently been put forward by Sir George Hampson, 

 z 2 



