I9I4-] S. Kemp : Onychophora. 483 



Transverse sections through the anterior ends of the two less 

 fully developed spermatophores at the point where the walls are 

 thickening preparatory to the formation of the cap present the 

 very characteristic appearance shown in fig. G In the case of the 

 most fully developed spermatophore only the faintest indications 

 of this structure can be made out, the star-shaped central portion 

 having been obliterated in the further thickening which has taken 

 place. 



It seems th:it the spermatophore of Typhloperipaius differs 

 from that of ftoperifyatus in possessing a chitinous coating through- 

 out its length and in the fact that the spermatozoa are not ar- 

 ranged radially round a central core. The coat is evidently a 

 great deal thinner than that of Peripatus, as typified by " P 

 edwardsii'" and studied by Gaffron', and the three to five swell- 

 ings or chambers described by that author — separate receptacles 

 in which masses of spermatozoa are accommodated— do n(_)t ap- 

 pear to have their counterpart in the Abor genus. The spherical 

 globules found by Gaffron on the surface of the spermatophore do 

 not seem to exist in Typhloperipaius. 



But the most noteworthy feature of the male reproductive 

 system of Typhlopcripatiis is the extremely great length of the 

 unpaired duct. Evans in his account of Eoperipatiis lays stress 

 on this choracter and makes use of it in maintaining the close 

 affinity of the Malayan genus with the neotropical Penpaius, a 

 conclusion also accepted by Bouvier. In Typhloperipaius the com- 

 mon duct is about a^ long as the entire annual, much longer than 

 in Eoperipatiis and. in its proportional development, at least 

 equalling that of any neotropical species. 



The iTiale accessory glands (pi. xxxvi, lig. i, in.a.g.). open by 

 separate openings placed side by side close behind the last pair of 

 legs. They consist of simple tubes, which, however, are much 

 convoluted. They run directly upwards from the openings and 

 lie for the most part near the dorsal aspect of the animal, termin- 

 ating blindly when they have reached the middle of the interspace 

 between the sixteenth and seventeenth pairs,of legs. In structure 

 the glands agree with those of previously described forms ; near 

 the aperture the lumen is lined for a short distance with a thin 

 investment of cliitin. 



In the possession of separate openings to the male accessory 

 glands Tvphloperipatus resembles Peripatus (American species only), 

 the Australasian species {Peripatoides and Oo peripatus) and the 

 S. African Peripatopsis cinctipes'^ (Purcell) and differs from Eoperi- 

 patus and all other species. 



i Gaffron, Zool. Beitrage (Schneider), I, pp. 152-157, pi. xxiii, figs. 62-76 

 (1885). According to Bouvier, Ann. Sci. nat. Zool. (9), II, p. 292 (1905), the 

 specimens used by Gaffron for his study of the genital organs should properly be 

 referred to Peripatus sedgivictii, Bou\ier. 



"^ According to the views expressed by Sedgwick (Quart. Journ. Microsc. 

 Sci., LII, p. 379, 1908) this species belongs to the group " Capo-peripatus " or to 



