The Non-Marine Mollusca of Portuguese East Africa. 199 



between them. The oviduct, which is separate from the male duct through- 

 out its length, may be divided into three parts— a posterior convoluted 

 portion, narrow near the albumen gland, but becoming broader anteriorly ; a 

 median, greatly swollen portion, with translucent glandular walls ; and an 

 anterior portion, which is also broad and glandular, but is of an opaque yellow 

 colour and very brittle in alcohol. The receptaculum seminis or spermatheca 

 is oval and about 1 mm. long, with a short duct. The vagina is rather short, 

 and is darkly pigmented. 



The vas deferens is slender, and is closely convoluted in the posterior 

 part of its course near the albumen gland. It passes forwards close to the 

 oviduct, occupying a slight groove in the wall of the swollen translucent 

 portion. In front of this, beside the opaque portion of the oviduct, it bears 

 a flattened prostate gland, which is composed of a number of irregular 

 tubules, as shown in the figure. After being embedded in the skin between 

 the female and male openings, the vas deferens emerges again into the body- 

 cavity beside the penis-sac and pursues a sinuous course to the posterior 

 end of the penis, this part of it being very long. The penis itself is narrow, 

 except at the end where the vas deferens enters it, where it is slightly swollen. 

 It is nearly as long as the penis-sac, which, however, is a much broader 

 structure, measuring about 2 X -5 mm. Internally the penis-sac bears two 

 longitudinal folds, as well as numerous smaller transverse or circular folds. 

 The penis bears no flagella, such as occur in the genus Segmentina. The 

 penial retractor is inserted where the penis enters the penis-sac. 



The spermatozoa (PI. VIII, fig. 16) have narrow, spirally twisted heads, 

 •0055 mm. in length, and long slender tails, which have the appearance of 

 being furnished with two extremely fine spiral ridges or striae. 



Until more is known about the detailed anatomy of all the species of 

 Planorbinae for which various sectional names have been proposed, I prefer 

 to follow those authors who regard most of them as subgenera of Planorbis 

 and Segmentina rather than as distinct genera. The apertural dentition of 

 Planorbula, however, appears to be merely a passing phase at an early stage 

 of growth, and I think it more logical to retain it in Planorbis than in Segmen- 

 tina, while on conchological and, I believe, anatomical grounds, I may be 

 justified in regarding Hippeutis as an unsegmented subgenus of Segmentina. 



Subgenus gyraulus Agassiz, 1837. 



Planorbis {Gyraulus) costulatus Krs., 1848. 



1848. Planorbis costulatus Krs., Siidafr. Moll., p. 83, pi. 5, fig. 8. D.F. 

 Hab. L. Marques. Makulane ; R. Mitembe, not far from Little 

 Lebombo Hills (Junod). 



Several specimens in perfect agreement with the ordinary Natalian form. 



