196 Dr. F. A. Bather — Studies in Edrioasteroidea. 



The Radials are of the roughly pentagonal shield-shape usual in 

 normal crinoids and in blastoids. Since they have no brachial facet, 

 but are notched for the excavation of the subvective grooves, they 

 more closely resemble the radials of blastoids. The basi-radial and 

 interradial sutures can be traced without excessive difficulty ; the 

 lengths of the latter, in the case of the anterior radial of specimen A , 

 are left anterior 3-5 mm., right anterior 4 mm. 



From the top of the interradial sutures, the shoulders of the radial 

 slope upwards towards the subvective groove, and the sutures 

 bounding them can be traced for some distance. The point at which 

 they meet the flooring-plates of the groove cannot, however, be 

 distinguished, owing to the close union of the various plates in the 

 neighbourhood of the groove. Estimated roughly, for the anterior 

 radial as before, the level of the top of the radial would be about 

 5 mm. from the lip, or distal end, of the groove ; and the distance of 

 the lip from the lower apex of the radial is about 4'1 mm. 



The rhomboidal area above the shoulders of the radials and 

 between the grooves is filled with Interambulacral plates. The 

 precise number and arrangement of these varies in the different areas, 

 and in all except the posterior area their outlines are hard to see 

 (Text-fig. 3). To judge from the folds, three or four small plates 

 have joined in the centre of each area, except the posterior, to form 

 a relatively large plate of somewhat swollen appearance (IB,). This 

 plate rests on the radials, and other smaller plates lie between it 

 and the grooves. 



The posterior area, which is more swollen and a trifle wider than 

 the others, measuring almost exactly 10 ram. from lip to lip of the 

 bounding grooves, is filled with a larger number of plates. Except 

 near the boundaries of the area, these are more loosely united, so 

 that the sutures are readily seen. About the centre of this area 

 in specimen A the plates are rather sharply depressed ; but in B 

 the corresponding depression is very slight, so that it has here proved 

 easier to remove the matrix, and thus to disclose seven plates con- 

 verging to a point, where doubtless was the vent. These seven 

 Circumanal plates are surrounded by about fifteen other Periproctal 

 plates, equally well defined, very diverse in shape but never long and 

 tapering. Outside these again come plates less well defined and 

 therefore less easily counted, but in total number about nine, one 

 being interradial and resting on the radials, and four bounding the 

 sides of the periproctal area and meeting above it in an interradial 

 suture. The twenty-two periproctal plates are but faintly pitted, 

 and each has an equably swollen surface. The nine bounding inter- 

 ambulacrals, however, are coarsely folded on their abanal margins, and 

 the folds meet similar folds from the radials and from the adambu- 

 lacral plates. 



By Adambulacral plates I mean those which lie between the last 

 set of plates and the grooves ; but Avhether they are really distinct 

 from the floor-plates of the grooves is uncertain. To this point we 

 shall recur. 



At the adoral end of the interambulacral area there lies in the 

 posterior interradius, and apparently in the other interradii, a single 



