132 Mr. G. A. K. Marshall on Diaj)oscinatisin, with reference 



existence of nauseous qualities has been well established, 

 and this tenacity does not appear to occur apart from those 

 qualities. The great utility of such a faculty to insects 

 which are liable to be experimentally seized and rejected 

 by enemies is so obvious that its very existence may be 

 taken as good circumstantial evidence that Miiller's factor is 

 an objective reality, and not a mere figment of the imagina- 

 tion. But it must not be forgotten that the very efficacy of 

 this characteristic in saving the lives of the insects and 

 enabling them to propagate their kind after experimental 

 attacks (an essential condition for its development) causes 

 it to become a serious check on the production of Miillerian 

 mimicry ; for such mimicry can be brought about only by 

 a progressive destruction, or sexual disablement, by experi- 

 mental tasting, of those individuals of the mimic which 

 do not conform to the mimetic pattern. Wherefore any 

 quality evolved by the species which diminishes this 

 destruction must, so far as it is successful, prevent a 

 mimetic approach. This limitation of Miiller's conception 

 does not appear to have been sufficiently taken into 

 account. 



• Colour as a guide to linpalatdbility. 



Although the existence of displeasing qualities in butter- 

 flies is very generally indicated by bright colours on both 

 surfaces of the wings, such is by no means always the 

 case. Among the Acrxinie, ItJioiwiiniv, and especially 

 the Eitplceini, there are many instances of colouring so 

 sombre that were it associated with the requisite pro- 

 cryptic habits it would serve most effectually as a means 

 of concealment. In such instances it is only the slow 

 heavy flight and generally gregarious habits of the insects 

 which cause them to become conspicuous objects ; and 

 these peculiarities of manner are to be found in every 

 species whose distastefulness has been satisfactorily de- 

 monstrated. On the otiier hand, there is at present no 

 experimental evidence which clearly proves the existence 

 of inedible qualities in butterflies which exhibit a strong 

 active flight and general wariness of manner, together 

 with obviously procryptic habits. We may of course 

 assume the existence of such qualities, but the value and 

 desirability of such an assumption must remain merely 

 a matter of opinion. 



As a result of a good many years' experience of 



