1894.] JTR. WALTEK GARSTANG OIT COLPODASPIS PUSILLA. 667 



prolongations, "whicli can be mistaken for the posterior border of 

 the foot when the animal is contracted." His figures unfortunately 

 do not show this point at all well (pi. i. fig. 25), and Forbes and 

 Hanley's figure, though clearer, does not seem to represent the 

 anatomical relations correctly (1. c. pi. UTJ. fig. 4). 



In Pliiline aperta the plantar surface also consists both of foot 

 and mantle : but this part of the mantle does not correspond with 

 the palhal appendage of Colpoclasins, as it contains the viscera 

 and shell. If it be examined, however, from the ventral aspect, the 

 pallial siphon is seen on the left hand, as in Colpodaspis (fig. 2), 

 and, to the right of the siphon, the mantle is seen to be prolonged 

 into a short free membranous border, which overhangs the siphonal 

 gi'oove and even extends slightly behind it. The relations of this 

 slight expansion are such that I think it may be regarded as the 

 rudimentary, or probably vestigial, representative of the pallial 

 appendage of Colpodaspis. 



Radula. — This organ was not described in Sars's original paper, 

 but a figure of it was given (without description) in a later work 

 by Gr. O. Sars '. There is a single admedian series of sickle-shaped 

 denticles on either side, and two series of slender laterals, the 

 formula thus being 2, 1, 0, 1, 2. I was unable to lay open the 

 contracted radula of my specimen, owing to its excessive minute- 

 ness ; but I determined that the number of rows in the radula 

 was from 25 to 30, and isolated individual denticles and half -rows 

 by teasing with needles. Some of these are drawn as figure 3 of 

 my Plate. The admedian denticles of this radula differ from those 

 figured by Sars in presenting a sharp distinction between their 

 terminal and proximal parts. The handle of the sickle shows an 

 angular projection from its inner or concave edge, like the corre- 

 sponding denticle in Colohocephalus costeUatus as figured on the 

 same plate (fig. 16). The lateral denticles also furnish an additional 

 point of resemblance between the radulse of these two types in 

 that their points are slightly bent in a plane at right angles to that 

 of their general surface, so that, when the denticles are mounted 

 _ flat upon a slide, their points are directed upwards towards the 

 observer. 



Shell. — Sars has described the shell so accurately that I have 

 nothing to add to his description ; but my figures, being on a larger 

 scale, represent its form and wonderful delicacy rather better. 



Summary. — On the whole, I think this Plymouth specimen pre- 

 sents features which indicate a slight advance on the organization 

 of those described and figured by Sars. I may mention its greater 

 size (3"125 mm. as compared with 2*5 mm.), the greater differen- 

 tiation of the tentacles, pallial siphon, and admedian denticles, and 

 perhaps some increased extension of the free margin of the shell. 



Affinities. — Sars was not quite certain whether Colpodaspis 

 belonged to the Opisthobranchia at all, and was much impressed by 

 the fact that the foot is attached to the body by a somewhat 



1 ' Bidrag til Kundskaben om Norges Arktiske Fauna. I. Mollusca Eegionis 

 ^rcticpe Noryegiffi ' (Christiania, 1878), plate sii. (Tabula anatomicse) fig. 15. 



