Annals of the South African Museum. 



The structure of the embryos also could not be made out, and in 

 particular it was quite impossible to determine the presence or absence 

 of the ventral plate. 



Nevertheless, there is no doubt that this is a species of Clypeoniscus 

 in view of the close agreement of the $ with lianseni. As to specific 

 distinctness, scarcely any character can be found except the (appa- 

 rently) singleness of the marginal folds on the brood-lamellae. Sars 

 doubts the specific distinctness of the two northern species. These 

 two forms were considered as belonging to two species by Giard & 

 Bonnier in conformity with their assumption that each species of host 

 is infested by its own particular species of parasite. This assumption 

 has been proved to have no foundation in fact, or at least to have 

 many exceptions. 



I have instituted a new species for the South African specimens, not 

 in support of the above hypothesis, but in order to indicate the 

 occurrence of the genus on a member of a family different from that 

 on which the northern species are found. 



Length: $ -75mm., $ 2mm.; breadth: ? 1*5 mm. 



Locality : Vasco da Gama Peak N. 71 E., distant 18 miles (off 

 Cape Peninsula). 230 fathoms. $ and 9 attached separately to 

 the ventral surface of the same specimen of Stenetrium dagatna (see 

 supra, p. 399). s.s. "Pieter Faure." 4/5/00. (S.A.M. No. A4167.) 



GEN. ET SP. INCERT. 

 (Text-figs. 1, 2.) 



Attached to the ventral side of the peraeon of a specimen of 

 Ilychthonos capensis (supra, p. 415) were two minute spherical bodies. 

 They are both evidently ? $ , but as no larvae or < <$ were present 

 their systematic position is uncertain. 



They bear some resemblance to Munnoniscus Giard & Bonnier 1895, 

 but this genus possesses no definite fixing apparatus. Oosaccus 

 Eichardson (Bull. U.S. Nat. Mus. no. 54, p. 582, fig. 644) has no 

 attachment cord, but appears to have a kind of suction-disk composed 

 of a raised rim with 3 or 4 valvular flaps within. 



The two specimens in question have the following structure, so far 

 as I have been able to elucidate it. Having only the two specimens, 

 which, moreover, are not exactly alike, I have not resorted to section- 

 ing, but contented myself with mounting them whole in glycerine 



Both are spherical in shape, measuring '5 mm. in diameter. In the 

 one specimen (Fig. 1) there are two little contiguous conical processes, 



