128 Mr. G. A. K. Marshall on Diaposematism, with reference, 



universally and saliently denoting the unpalatable groups 

 of butterliies. It is impossible to doubt that in all these 

 groups the advantage of the prolongation of the fbre-wiugs 

 is primarily to increase conspicuousness by that special 

 form of wing, and secondly to increase the area available 

 for warning colours. The character being so general and 

 so fixed in tendency, one cannot be surprised to find some 

 species developing it more than others ; and there seems 

 no need, in such a case as that of the Melinda Danaines, 

 to have recourse to the far-fetched idea of these abundant 

 and distasteful butterflies having to borrow so natural a 

 feature from the rare Fajnlio rex and extremely rare 

 P. mimetictcs. In the genus Acrma no one deems it 

 necessary to account for such great prolongation of the 

 fore- wing as is found in A. 'pcrenna or A. 2)Iiarsalus by 

 assigning it to mimicry of any butterfly of a remote 

 family." 



These then are the cases which have so far been 

 adduced to demonstrate the actual occurrence in nature 

 of Reciprocal Mimicry. To my mind, the facts when 

 critically examined do not lend any valid support to such 

 a hypothesis ; nor even do they appear to justify in any 

 instance the assumption of that mimetic inter-action which 

 I have termed Alternating Mimicry or Resemblance. 



The Scope of Batesicin Mvmic7'y. 



Up to the present time it has generally been considered 

 that mimicry between two species which both possess dis- 

 tasteful qualities cannot possibly be explained on the lines 

 of Bates' theory, but that the association must be Miillerian 

 in character ; in other words, that it can only have been 

 produced by the selective action of experimental tasting 

 by inexperienced animals. Now when I began experi- 

 menting on these subjects in South Africa, some twelve 

 years ago, one of the first things that impressed me was 

 the fact that there was clearly a considerable difference in 

 the degrees of distastefulness, not only between different 

 genera of butterflies, but even within the limits of a single 

 genus, such as Acrxa. Further, the question was compli- 

 cated by the divergence in the likes and dislikes of various 

 insectivorous animals. No doubt all this has long been 

 recognised ; yet in practice, the application of the Miillerian 

 interpretation involves the assumption of a uniform standard 



