( 501 ) 



were the !<iiiuc, the limits of division were also the same, whether the delimitation 

 of each subspecies was carried out according to the wiug-characters, or according to 

 the ajii)ai-atus of copulation. This result is most probabl}- owing to the areas to 

 which the various subspecies of the Papilios examined are restricted being mostly 

 islands. 



Hence we can accept it as a general rule applying to most subspecies that the 

 distinguishing characters taken from the wings are associated with distinguishing 

 characters in the genital armature, at least in the »iale sex. There are, however, 

 many exceptions to this rule, in so far as a good number of subspecies characterised 

 by some certain peculiarity in the wings have no peculiarity in the apparatus of 

 copulation. The Chinese specimens of P. mrppclon, so very conspicuously aberrant 

 in pattern, are in the valve and harpe identical with the Indian and .Japanese form 

 of sarjiec/o/t; the Moluccau subs])ecies of P. aristeus, though in the extent of the 

 markings very difi'erent from the Queenslaiidian representative, is in the sexual 

 organs the same ; so that in an arrangement of the forms according to the develop- 

 ment of the apparatus of copulation of the male, P. aristeus aristcm and 7^. aristeiis 

 pannatas, and P. sarpedon sfirpcdon and P. sarpechn semijasciatits, would have to 

 be united. We meet, further, many subspecies which in external characters are 

 conspicuously and almost constantly different, while there is only a slight and very 

 inconstant character of distinction in the genital ajjparatus. We have found it 

 correct in all the species examined that, if in a given species the character of 

 distinction of a certain subspecies lies in one only of two independently varying 

 organs, wings and apparatus of copulation, it is invariably the wing which exhibits 

 the character by which the subspecies is distinguished. This coincides in a remark- 

 able manner with the above observation that seasonal dimorphism does not affect 

 the sexual armature in the case of the North Indian and Japanese Papilios, and 

 renders it certain that the effect of (seasonally or geographically) isolated trans- 

 muting factors is first noticeable in the wing-markings, and that accordingly the 

 transmutation of a sjjecies begins wth a change in the wing-markings. Hence it 

 is also correct to say that in an evolntionistic sense the wiug-markings are less 

 constant than the genital armature, and that conseipiently a difference in the latter 

 constantly met with in the individuals of two forms of Papilio is much more likely 

 to be of specific value than a difference in the wing-pattern, a conclusion which 

 borders closely ujion one of the main objects of our researches, the question as to 

 the classificatory value of the sexual armature, which we will now briefly discuss. 



A. Taxongsiic Value of the Organs of Oopclation. 



As our researches have proved that there is a certain amount of individual varia- 

 tion in the sexual armature, the limits of which can, as we have seen above, only be 

 determined by continued examination of a large material, and as we further have 

 found that there is geographical polymorphism in the organs in tpicstion in which the 

 localised forms are connected by intcrgradations (P. sarpedon sarpedon and adonar- 

 ensis, P. aristeus hermocrates and anlicrates), the premisses of tlie discussion of 

 the question which characters are and which are not of specific value arc precisely 

 the same as in the case of external characters. Referring to what we have said in 

 tlie introduction aliout our means of recognition of specific distinctness, it will be 

 sufficient to recall to mind the fact that the presence of a character in a certain form 

 not met with in other forms is a priori not a proof of the respective form being 



