to limitations of the Milllerian Hypothesis of Mimicry. 133 



butterfly life in South Africa I am convinced that we shall 

 obtain a truer insight into the mimetic relationships of 

 these insects if we take their natural behaviour as our 

 guide, than if we rely too exclusively upon theoretical 

 deductions based on colour patterns alone. 



To take an example. We know nothing whatever 

 about the edibility, or otherwise, of the numerous species 

 of South American ErycinidiB. There appear to be a 

 good many cases of mimicry among them, and a certain 

 number of the species are very brightly coloured. It has 

 therefore been supposed that the family is probably un- 

 palatable as a whole. On the other hand, Wallace tells 

 us (Trans. Ent. Soc. (2), II, 1868, p. 262) that the brightly- 

 coloured species of Erycina have a very quick, jerky, 

 "skipper "-like flight; whereas of the other genera " the 

 great mass of the species" have the curious habit of 

 always settling on the under-sides of leaves with wings 

 outspread.* Now habits such as these are not known 

 to occur among any distasteful butterflies in any part of 

 the world ; for, as we have seen, leisurely movements and 

 a contempt for concealment are the most essential charac- 

 teristics of these insects, for which the display of their 

 warning colours is of the very first importance. In the light 

 of our present knowledge therefore it seems difficult to 

 justify such a far-reaching assumption of distastefulness ; 

 and we can scarce hope for stability in the theoretical 

 edifices which may be raised on so dubious a foundation. 



Again, it has been suggested that the roughly Acr^a- 

 like facies of the giant Pajnlio antimachus of West Africa 

 has probably a Mlillerian significance, apparently without 

 any regard to its habits. My friend, Professor Yngve 

 Sjostedt, of Stockholm, who collected for some time in 

 the Cameroons, has given me a graphic account of the 

 great wariness and tremendous speed of this strange insect, 

 whose flight, he said, could only be compared to that of a 

 swallow. On the other hand, he told me that the lovely blue 

 Pcqnlio zahnoxis, which is often associated with antimachus 

 by systematists, was far more common, having a slow 

 heavy flight and being easily captured. In this case the 

 habits serve to confirm the idea of distastefulness suggested 

 by the appearance of the insect. But are we then to 



* Bates also gives some interesting notes on the remarkable 

 differences of habits which are to be found in this family {pp. cii. 

 (2), V, 1858, pp. 4, 5. 



