( ^^0 ) 



Notwitlistandinsi it is manifest that tlic difficulty to obtain hybrids is in many 

 cases to be attributed to the diverse development of the copuhitory organs, the 

 question is of very little importance as long as we take into account only an inter- 

 crossino- between species and species. According to our definition, specifically 

 distinct animals or i>lauts cannot fuse together ; tlie appearance of hybrids has no 

 influence on the evolution of the species which have produced tliem, thongli they 

 may obscure the fact of the specific distinctness uf the parents.* Tlie barrier 

 between species is the degree of diversity already attained, and as this barrier is 

 absolute, the jiresence of any other barrier must be indifferent. The (juestion as to 

 the evolntionistic influence of the peculiar development of tlie coimlatory organs, 

 therefore, does not relate to those forms which have already attained that degree of 

 diversity which renders a fusion with other forms impossible, but to those lower 

 degrees of diversity which are still able to fuse together under favourable circum- 

 stances. Hence we have to inquire whether the variation of the copulatory organs 

 within the limits of a species is such that the difference in tlie organs in question 

 between varieties of the same species can be thought to be a barrier against the 

 sexual intercourse between the varieties. 



The efl'ect of the divergency of the copulatory organs in respect to iireventiun of 



intercrossing can very well be compared with that of geographical isolation, as the 



effect of both the morphological and the geographical factor depends on the extent 



of the mechanical barrier. The Atlantic Ocean is certainly an effective barrier 



between the Nearctic and Falaearctic Regions, although even such a wide sea cannot 



absolutely prevent a migration from one region to the other ; the Ganges Valley 



which separates the North Indian from the South Indian fauuii, and the straits 



between the various islands of the Indo-Australian Archipelago, are barriers which 



render the occurrence of an intercrossing between specimens born on the opposite 



sides of the barrier, though by no means impossible, doubtless very rare ; in fact 



every barrier which isolates to any degree, however small the degree may be, a 



number of specimens A from other specimens B renders the intercourse of the two 



sets of individuals with one another less probable than the copulation of the 



specimens of each set inter sc, and hence is a barrier against intercrossing. If, 



therefore, the varieties of the same species exhibit in both sexes any diversity in the 



the organs of copulation, we can rightly infer that this diversity will act like a 



geographical barrier, isolating the varieties from one another to a certain extent 



according as the anatomical diversity is great or slight. As our researches have 



proved that the premiss of this conclusion is correct as regards the Eastern Papilios, 



namely that there are varieties which differ in the genital armature from other 



varieties of the same species in both sexes, it is manifest that the diversity in the 



genital armature of the PiH)ilios has a great bearing upon the divergent develojinient 



of the varieties. 



We have found that there is individual and geographical variation in the genital 

 armature of the Eastern Pa])ilios. As regards individual variation, it is obvious that 

 specimens which have any marked peculiarity in the jireliensile organs, additional 

 hooks or ridges, in which they deviate from tlie normal, will be at a disadvantage in 



* As besiiles the mechanical barrier against intercrossing there is, in most species with highly 

 developed organs of sense, a psychological barrier active, hybrids must in the state of nature be very rare 

 in comparison with the number of specimens which are not hybrids. Specimens connecting two supposed 

 distinct species have not rarely been treated as liybrids, though the natural conclusion from the regular 

 occuiTence of such intergradations would be that the sujiposcd distinct species form one, di- or polymorphic, 

 species, provided that the reverse has not been prurnl. 



