( 19' ) 



(2) The folds of the iutersegmental membrane canaot be studietl from the dried- 

 np individuals iu snch a manner that one perceives the minute differences between the 

 folds of the individuals; to this pnrpose it is necessary to compare live specimens, 

 or material preserved iu an adequate fluid. The armature of the valve of the male 

 is during copulation pressed against the intersegmental membrane of the fcinalf, 

 and the spines, hooks, processes, etc., of the valve find a hold on the ridge-like folds 

 of the intersegmental membrane ; this is plainly visible in a cJ and ? of P. memnon 

 in the Tring Museum which are still united to one another. The special armature of 

 the vagina takes hold on tiie internal portion of the ninth segment of the male and 

 on the scaphium. There is sometimes a peculiarly shaped, broad, vertical, chitinons 

 plate underneath the scaphium, for example in the males of the yellow Troidrs 

 (= Ornithoptera?), such as T. helenus, to which a long and strong ventral hook in 

 the/e)»al/' corresponds. 



(3) One would expect that the variation of the harpe in the male is iu the 

 same species accompanied by a corresponding variation in the/em.ale in those parts 

 which during copulation are in contact with the harpe ; hence the variation in the 

 female genital apparatus parallel to the variation in the male harpe must be 

 searched for in the special kind of folding of the iutersegmental membrane, while 

 the variation in the outline of the processes and ridges at the mouth of the vagina 

 can have nothing to do with an adaptation to the special form of the harpe. 



(4) In the only species examined iu which the harpes vary very conspicuously 

 according to locality, in P. alrdnotis, the iuter.segmental fold of the ? on which 

 the harj)c takes a hold is found to be different in the Chinese and the Japanese 

 subspecies. 



(5) Though the individual variation of the vaginal armature is often great 

 (P. aegeiix and memnon), a variation according to locality is not observed, or is 

 slight. 



(6) In species with polymorphic females the vaginal armature is the same in 

 the various /mf/i't'.? (apart from individual variability). 



III.— CONCLUSIONS. 



The demonstration of the kind and the extent of the variation of the genital 

 armature in both sexes of some species of Pa iJilio which is contained in the pre- 

 ceding chapter, notwithstanding the fact that our researches are still incomplete and 

 our notes on the vaginal apparatus even preliminary, enables us to compare the 

 variation of the genital armature with that of other organs and to draw witli 

 safety some general conclusions. If we almost restrict the comparison to the 

 colour, pattern, and shape of the wings, mentioning only the prominent characters 

 of the respective insects, we do so in order to avoid unnecessary detail, and 

 secondly, because the distinguishing ciiaractevs of species, subspecies, and aberra- 

 tions of butterflies as given in the works of Lcj)idopterists refer especially to the 

 wings. 



As the true basis of work in Natural History is the comparison of specimen 

 with specimen of the lowest classificatory entity, the species, a study of the genital 

 armature has to begin with the comparison of the armature of snch individuals 

 as are doubtless sjiecifically the same. To di>terniine the limits of variation of 

 the genital armature of a species is a dUficnlt ta'^k. Though the number of 

 specimens of the commoner species could easily be enlarged without any great 



