190n.] FROM EAST AFRICA AXD ZANZIBAR. 385 



most are set almost at right angles to the rhachis. The stomach 

 is large and free, but thin and not laminated nor muscular. The 

 blood-gland is large, and the nervous system is very concentrated. 

 The generative system appears to be as described by Bergh, but 

 the glandula and hasta amatoria are difficult to see and were 

 satisfactorily detected only in one specimen. 



These specimens are, I think, clearly Ehrenberg's Asteronotus 

 JiempricM, from Massaua. He describes it as " sex-pollicaris, 

 oblongus, glaber, vesiculosus, supra fuscus, lineis circulisque niveis 

 sparsis, vesicas dorsuales cingentibus, subtus lateritus, pede flavido, 

 branchiarum apertura lobulis sex stellatim positis priecludenda. 

 .... Branchiarum e dilute laterito sen carneo albicantium 

 fasciculus amplus." 



Prof. Bergh seems inclined to think (S. R. xvii, p. 917) that 

 the real species of this genus ai-e not more than three, hemprichi, 

 mahilla, and ccespitosus. The differences between these three 

 do not seem to me to be clearly defined, and my numerous 

 specimens, which I unhesitatingly refer to one species, present 

 connecting links, especially in colour, which make me think that 

 the three species are merely varieties of one. 



EXPLANATION OF THE PLATES. 

 N.B. — Except in the cases noted, the figures are drawn from the living animal. 



Plate XXXIL 

 Fig. 1. Tliordisa villosa (p. 367), ventral view. 



2. Dorsal view of the same. 



3. Thnrclisa crosslandi (p. 368), ventral view. The marsrin of the mantle is 



inturned here and there, showing the mobile papilla; which cover the 

 dorsal surface. 



4. Trippa monsoni (p. 371), dorsal view, much enlarged. 



5. Salgerda willeyi (p. 372), from a drawing by Dr. Arthur Willej'. 



Plate XXXIII. 

 Fig. 1. Thordisa villosa (p. 367), head and anterior end. The figure shows an 

 extreme elevation of the body, which normally is fiat. 



2. Dorsal papilla of the same species, with flexible pigmented end and spicnle- 



stiifened base. Also a portion of the mantle-edge magnified. 



3. Branchiffi of the same. 



4. Thordisa crosslandi (p. 368). Teeth from the radula : a, upstanding; 



h, laid flat. 



5. Gills and anirs of the same. A ridge {a) connects the higher part of the 



rhachis with the anal papilla (6). 



6. Dissection of the retracted penis of the same, showing the shape and structure 



of the enclosed glans. 



7. The glans penis of the same is slit open, showing it to be hollow and to 



contain a prolongation of the vas deferens which passes to its tip. 



8. The central nervous system of the same in its sheathing of connective tissue. 



Plate XXXIV. 

 Fig. 1. Ealgerda loasinensis* (p. 373), dorsal view. 



2. Diagram of the arrangement of its gill on the rhachis. 



3. Sclerodoris coriacea (p. 383). Pencil drawing from the preseryed specimen. 



4. Degenerate Copepod parasite found in the liver oi Sclerodoris (p. 3,84). 



5. Asteronotus hemprichi (p. 384). 



6. Enlarged view of the gill-opening when the branchiie are as completely 



retracted as is possible. 



7. Kentrodoris rubeseens (p. 374), about half natural size. 



* This name is wrongly spelt " wassinensis " on Platje. 



Proc. Zool. Soc— 1903, Vol. II. No. XXY. 25 



