318 Dr. Frederick A. Dixey on 



A fall illustration of this position would require a 

 larger series of diagrams than could well be given in this 

 place ; it is hoped^ however, that the series of figures on 

 Plate VII. may suffice to show the more important 

 links in each chain, or at any rate to make the drift of 

 ray statements intelligible. It should be borne in mind 

 that all the forms here spoken of are neotropical. 



1, The "agna" line. The first of these lines of 

 development may be called the agna line, inasmuch as it 

 leads towards a well-marked group of distasteful forms 

 of which Aeria agna, Godm. and Salv., (PL VII., fig. 6) is 

 a good example. Starting from a white Fioris of ordinary 

 aspect like P. phalue, we find that the first step in the 

 growth of this mimetic pattern is furnished by the pro- 

 longation of the diagonal dark bar, which in P. phaloe $ * 

 extends from the costa to the distal end of the cell in 

 the forewing, to meet the dark hind margin as in P. caly- 

 donia $ (fig. l,h). All stages in the development ' of 

 this first feature can indeed be traced by comparing 

 specimens of both sexes of P. calydonia itself. The next 

 step is the extension of the dark hind border of the 

 forewing, already more pronounced in P. calydonia $ 

 than in P. phaloe $ , along the inner margin, as in P. 

 demophile $ (fig. 3, I). Simultaneously with this change 

 the dark border of the hindwing is much broadened 

 (ib., A), and in some specimens of P. demophile $ , as in 

 the one figured, the white ground colour is replaced by 

 yellow. These changes are sufficient strongly to suggest 

 the general aspect of the protected group referred to, 

 and it seems difficult to believe that the appearance of 

 the yellow female of P. demophile has not a significance 

 derived from this fact. 



2. The " atthis " line. But the last-named form, viz., 

 the yellow P. demophile $ , though the final Pierine term 

 in one transitional series, is but an intermediate term in 

 another. This second line of development, starting 

 afresh from P. demopliile $ , passes into an unmistakable 

 mimetic relation with the protected group that centres 

 round such forms as Heliconius atthis (fig. 8) and 

 Tifhorea pavonii. 



Comparing P. viardi $ with P. demopliile J , we find 

 that in the former insect a further stage of divergence 



* See Trans. Ent. Soc. Lond. 180G, pi. III., fig. 2, h. 



