582 Dr. F. A. Dixey's reply to Mr. G. A. K. Marshall 



connection is the difference in character of flight shown 

 by the two genera concerned. It is of course no proof of 

 palatability or the reverse that an insect is active and 

 wary on the wing. Many distasteful species, especially 

 the " dominant " models, possess the characteristically 

 deliberate demeanour first noticed by Bates, but others 

 show resemblances of greater or less degree in this, as in 

 other respects, to the forms that usually occupy central 

 positions in mimetic groups. 



Here ends my survey of Mr. Marshall's criticism of 

 particular instances. I feel justified in maintaining, as a 

 result of this examination, that not only has he failed in 

 each single case to prove his point, but that he has also in 

 many particulars been betrayed into actual error. 



The remaining examples impugned by Mr. Marshall, 

 together with that portion of the concluding section of 

 his paper which has not been dealt with by me, bear 

 reference to certain views and observations for which I am 

 not personally responsible. I shall not presume to enter 

 the lists in defence of champions so well able to take care 

 of themselves as my friends Prof. Poulton and Mr. Neave ; 

 but it may not be out of place to add here a remark with 

 regard to Mr. Marshall's footnote on page 122 of his paper. 

 He there calls attention to some apparent discrepancies in 

 the accounts given by Dr. Longstafif and myself of the 

 scents of certain African butterflies. It is a well-estab- 

 lished fact that scents of opposite character may coexist 

 in the same individual (instances are given in my com- 

 munication quoted by Mr. Marshall from Proc. Ent. Soc. 

 Lond., 1906, pp. ii-vii), and it seems probable that the 

 differences between Dr. Longstaff's records and my own — 

 a very few differences, be it noted, amidst a large body of 

 substantial agreement — may be attributed to a reason of 

 this kind. Certainly my recollection of the strong, dis- 

 agreeable odour of Neptis agatha is vivid to this day. 



Before concluding this paper I have a suggestion to offer 

 to students of the Miillerian problem who may still be in 

 an early stage of their investigations. It is that those 

 who wish to avoid a cramped and narrow view of the 

 mimetic problem should refrain from stating and con- 

 sidering it only in terms of " mimetic pairs " or even 

 "mimetic associations." Tlie real unit of study is the 

 aposeme, in its transitions, its modifications and its 



