ISOLATION 153 



very rare and there is little evidence that differences in the 

 male genitalia are often a very serious barrier between species, 

 except when the structures are extremely different, as between 

 species belonging to different genera or families. In a number of 

 species of Diptera the male genitalia are extremely diverse, and 

 there appear to be no corresponding differences in the female ; 

 sometimes (Lucilia) it is only with great difficulty that the 

 females can be distinguished, if at all. In many Hymenoptera 

 the male genitalia differ greatly, with little or no differentia- 

 tion in the females (Richards, 1927a, p. 262 ; Boulange, I.e.). 

 In Bombus, where the female genitalia do to some extent vary 

 specifically, it is largely groups of species which differ and the 

 structures showing differences come into contact only with 

 part (and that not the most complicated) of the male armature. 

 Further, in some species the male genitalia, though nearly 

 identical, show certain minute but constant differences, too 

 small to have, with any probability, any functional significance 

 (Richards, I.e.). Although there seems to be usually no 

 detailed co-adaptation between the male and female, there are 

 some exceptions. Edwards (1929, p. 40) records a correlation 

 between the length of the male penis and of the female sper- 

 mathecal duct in the flies of the family Blepharoceridae. A 

 similar correlation is sometimes observed in beetles of the 

 family Chrysomelidae (Harnisch, 19 15), but how far this is 

 specific rather than generic requires investigation. 



A more rational explanation would appear to be that 

 differences in instinct — possibly {e.g. in insects) in the nature 

 of the scent produced — are the first stage in the permanent 

 isolation of species ; later, differences in genitalia may arise 

 and may sometimes, incidentally, make the isolation more 

 perfect. In this way it is possible to explain the occurrence 

 of groups (families, genera, etc.) in which the genitalia are 

 scarcely specifically differentiated. All who have studied 

 insect genitalia agree that the value of these structures to the 

 taxonomist varies greatly in different families, in some pro- 

 viding characters of little more than generic value, in others 

 differing very greatly in species otherwise very similar. 



In the preceding paragraph we have advisedly used the 

 phrase ' permanent isolation ' to describe the result of changes 

 in instinct, for temporary isolation may result from geographi- 

 cal barriers. It is a matter of controversy whether some 



