NATURAL SELECTION 297 



in specific diagnosis on differences in the male, and to a less 

 extent in the female, genitalia. We may mention the studies 

 of the os penis or baculum in mammals (Lonnberg, 191 1 ; 

 von Bittera, 191 8 ; Pocock, 1923) ; the copulatory fins of fishes, 

 e.g. Selachii (Leigh-Sharpe, 1920, 192 1), Gambusia (Geiser, 

 1923) ; the dart and associated structures in Mollusca (Ashford, 

 1885) ; the genitalia in insects (see Boulange, 1924, pp. 359- 

 392) or the copulatory styles in the Planaria (Eggers, 1925). 

 These differences have been recognised not only as very 

 prevalent, but as of particular systematic importance because 

 of the relatively high degree of discontinuity observed, so 

 that species with sharply distinct genitalia may otherwise differ 

 only in trivial and not easily appreciable characters. 



Two main questions arise from the study of the genitalia : 

 (1) What functions do the remarkable modifications of these 

 organs serve ? Do they act as barriers stopping crossing 

 between species ? (2) How have the variations in genitalia, 

 ultimately leading to specific difference, arisen and become 

 established in the species ? 



Both these questions have been dealt with in some detail 

 in Chapter V, and only our conclusions need be summarised. 

 We have rejected the earlier view that the prime function of 

 differences in the genitalia is to isolate species, chiefly because 

 the members of different species do not in any case often try 

 to mate, and because in some pairs of species considerable 

 differences in the genitalia do not prohibit crossing when it is 

 attempted. We are forced to regard specific differences in 

 the genitalia as of essentially the same nature as other 

 apparently useless specific characters. 



As regards the second question, we have also opposed the 

 view that differentiation of the genitalia is necessarily associated 

 with geographical variation. We believe that even in a 

 relatively homogeneous area divergence of species, including 

 divergence of genitalia, is possible and probably, in many 

 groups of insects, quite common. 



Whether divergence of a type leading ultimately to cessation 

 of interbreeding always depends on geographical isolation, or 

 not, we have to explain how the elements in the divergence 

 became established. It is generally agreed that a variety of 

 habitudinal and structural differences between any pair of 

 species contributes to the absence of interbreeding. Even 

 those who maintain that the genitalia are the main agency 



