Mimetic Attraction. 321 



species until it terminates in a Pieris ov Mylothris bearing 

 an intimate mimetic relation with some insect of entirely 

 different affinities. It is further to be observed that, in 

 every instance, the species here considered as tlie model 

 towards which these diverging series tend, does not 

 present an isolated and independent scheme of coloration, 

 but is itself a member of a lai'ger or smaller group of 

 forms, in addition to the Pierine mimic, all of which are 

 endowed with an aspect similar to itself — in other words, 

 that the mimetic associations do not run simply in pairs, 

 but in groups. This latter fact has long been recognised ; 

 and the existence of such mimetic groups has been 

 shown by F. Miiller, Meldola, and Poulton to possess 

 a further significance than that originally detected by 

 Bates. The elaborate work of Haase* contains an 

 attempt to give a systematic account of tlie chief cases 

 of mimetic grouping. But in spite of what has already 

 been written by these and other authors, it may be 

 doubted whether the importance of the principle of 

 mimicry among the factors that have determined tlie 

 facies of the insect fauna in such a region as the neo- 



* " Untersuchungen iiber die Mimicry," Stuttgart, 1893. It may 

 here be mentioned that several of the above-named insects have 

 been noticed by Haase ; who, however, has not attempted to trace 

 in any detail the lines of mimetic assimilation that diverge from the 

 common Pierine stock. He speaks, for instance, of Mylothri.f lorena 



5 and M. mulenka $ , which nndoubtedly belong to the numata group, 

 as having arisen from such forms as P. chmophile 9 , and considers 

 that the transition took place through forms resembling P. viardi 



$ . But a careful examination will I believe show that, as stated 

 above, neither P. dfmophile $ nor P. viardi ^ is in the direct 

 line passing from the unaltered Pieris towards M. jjyrrha and 

 Heliconius numata. Both are, in fact, intermediate terms in the 

 series leading up to an entirely distinct assemblage, that typified 

 by Tithorea imvonii and Heliconius attJiis (Haase's " Bonplandi 

 Tracht"); while P. demophile shows evidence of attraction by the 

 protected agna group, and P. viardi by the dominant form Heli- 

 conius charito)iia. Again, the red streak on the underside of the 

 hindwing in the males of M. lorena, etc., is attributed by Haase to 

 " inheritance from the female,'' but its origin is not traced by him 

 to the primitive basal red common to many Pierine genera. (See 

 Trans. Ent. Soc. Lond. 1894, pp.283-289 ; ibid., 1896, pp. 72, 73.) 

 Pieris leptalina, Bates (P. pisonis, Hew.) is spoken of by Haase as 

 representing the first partial assimilation to certain Ithomias ; 

 rightly, s-o far as the main fact is concerned ; but it may also be 

 noticed that earlitr stages of the same assimilation exist in P. 

 kigahu, and, as shown above, in P. pandoaia. 



