The South African Hirudinea. 313 



pouch region, and the fact that sections reveal loose packets of 

 spermatozoa in this region. Otherwise it must be suggested that 

 spermatozoa are obtained from the individual itself, which is certainly 

 not the case, as the male ducts have direct communication with the 

 exterior, well-developed muscular terminal portions, and further in 

 the specimens sectioned were found to have an abundance of 

 mature spermatozoa in the terminal portions, thus indicating that 

 spermatozoa pass to the exterior. 



Although anatomical details of this form will be published at a 

 later date, it may not be out of place to remark here on the character 

 of the brood pouch. Since Glossiphonia carries its eggs and young 

 attached to the ventral surface, and protected in an external pouch 

 or " marsupium," formed by the curvature or inflection of the 

 thin lateral mai'gins of the body, it might reasonably be suggested 

 that the brood pouch in this new form was the result of the con- 

 crescence of such inflected margins. Sections indisputably show 

 that such is not the case. It is found, for instance, that the nerve 

 cord retains its ventral position as far as the brood pouch, and then 

 is directed upwards to the dorsal side to pass over the hump of the 

 brood sac, afterwards descending on the posterior side of this sac to 

 the ventral surface. The dorsal body wall in the region of the sac is 

 most markedly attenuated. Again this region of the body which, 

 were the marsupium or brood pouch so derived, would possess 

 typical dorsal and ventral musculature, possesses only the dorsal set. 

 Similarly the ventral floor of the brood sac, instead of being pro- 

 vided with both dorsal and ventral musculatures possesses only a 

 ventral set. This seems to indicate that the pouch has been formed 

 as an excavation in the substance of the ventral portion of the body, 

 or as a modification of part of the ventral sinus. Supporting the 

 latter of these suggestions is the fact that the ovary is a very short 

 structure more or less spherical in form, and not of the elongate 

 sac-like character as found in other GlossiphoniidaB. 



II. GNATHOBDELLID.E. 



Numerous members of the Hirudinidee have been obtained which 

 show that there occur a large number of varieties or species bridging 

 the differences between H. scmptcmstriata and H. capensis. Un- 

 fortunately no specimens of the latter species have been accessible. 

 Although all the species including the above-mentioned forms have 



