94 Annals of the South African Museum. 



11. The relation of the external sexual opening to the posterior legs in 



Onychopkora. 



The differences in the position of the external sexual opening in 

 the various Onychoi^hora {cf. figs. 1-5), especially between such 

 forms as, for instance, Peripatopsis balfouri (fig. 4), with the sexual 

 opening between the last pair of legs and Peripatus echuardsi, &c. 

 (tig. 1), with the opening between the penultimate pair, is very 

 striking, and may have originated in two ways. 



One possible explanation is to assume a direct transposition of the 

 genital orifice from one position to the other. There is, however, 

 no evidence to support this view, and against it is the circumstance 

 that the position of the genital opening is always perfectly constant 

 in its relation to the posterior legs within the limits of the same 

 species. Thus, in the American Peripatus, the opening always lies 

 between the penultimate pair, no matter how much the total number 

 of legs varies in the species. 



The second, and, I believe, the only correct explanation, is that 

 given by Korschelt and Heider (1892, p. 716), and quite recently by 

 Bouvier (1900c), who assume that the posterior legs and the post- 

 genital segments of the body have been more or less or even com- 

 pletely reduced, causing the genital opening to acquire thereby an 

 apparently more posterior position. Several strong arguments in 

 favour of this view may be adduced. Thus in Peripatopsis capensis 

 Sedgwick found two pairs of postgenital rudimentary somites, which 

 Korschelt and Heider (p. 716) homologise with the two postgenital 

 segments of the adult Peripatus (also corresponding, according to 

 Kennel, to two pairs of postgenital somites in the embryo). It has 

 further been clearly demonstrated that the posterior (genital) pair of 

 legs in Peripatopsis are subject to great variation as regards their 

 degree of development," both within the limits of a single species 

 and in the different species taken together, and every intermediate 

 stage may be shown to occur, from the perfectly formed, although 

 small-sized, legs of some specimens of P. sedgioicM, Pure, to the 

 conditions sometimes found in P. capensis and moseleyi, where the 

 two posterior legs may be so reduced that one or both of them are 

 completely aborted and no longer distinguishable. From this last 

 condition to an imaginary one in which the legs of the genital seg- 

 ment no longer appear in any of the specimens is but a step, and 

 we would then have a clear case in which a form with the genital 



* Bouvier (1900c) makes tise of this argument in this connection. 



