suggestions as to insect colours and patter iis. oil 



W, J. Kaye if he can remember the colour of the 

 flowers visited by the black, cow-red, and yellow Melin/ea 

 group and its mimics in British Guiana, and he tells 

 me they are either wliite or cream-coloured. Further- 

 more, Mr. Thayer treats this group as though it 

 were uniform throughout tropical South America, dis- 

 regarding the extraordinary changes of colour and 

 pattern undergone by its representative species as we pass 

 from one part of the Neotropical region to another. It 

 is almost inconceivable that the following features, which 

 are characteristic of whole groups in particular areas, can 

 be due to the special flowers of those areas. The barred 

 form of Central America, Colombia, and Venezuela, the 

 black hind-wing of the Guianas, the bright yellow band of 

 Eastern Brazil, the chestnut ground-colour of Ega on the 

 Amazon, the black marked fulvous of the Napo Kiver, 

 passing on into the black forms with fulvous marks which 

 constitute so large and characteristic a group in Ecuador, 

 Peru, and Bolivia. In all these cases, nothing short of 

 actual evidence on the spot can warrant the improbable 

 suggestion that we are dealing with syncryptic groups, 

 changing as the species of flowers are replaced by others 

 in passing from one district to another. 



Moreover, the theory of a syncryptic resemblance to 

 flowers fails to account for certain broad characteristics of 

 the groups in question, which on the other hand receive 

 a ready explanation on the theory of common warning 

 (synaposematic) coloration. These are (1) the predomin- 

 ance of forms belonging to the sub-families Itliomiina}, 

 with the allied Danednx, and Heliconinx, with the allied 

 Acrmno) : (2) the fact that the predominant members of 

 the chief groups in all the other tropical parts of the 

 world are also contributed by the DanainiB and Acrmnx : 

 (8) the flaunting flight, exposure at rest, and general 

 want of alertness exhibited by the species of these sub- 

 families as compared Avith others : (4) the more or less 

 exact similarity of the pattern on the under to that on the 

 upper surface, an arrangement comparatively rare in other 

 Rhopalocera: (5) the experimental evidence of the un- 

 jDalatability of these very sub-families to a large number 

 of the enemies of insects. 



Hence, until positive evidence is obtained on the spot 

 in favour of Mr. Thayer's suggestion of syncryptic resem- 

 blance, I must regard such an interpretation as highly 



