Vol. 55.] ECHINOIDEA AND OPHITJEOIDEA . 697 



present originally, but if so they have left no traces of their exist- 

 ence. The arms of the distal pairs are the larger, measuring from 

 12 to 13 mm. in length, and about 1*5 mm. in breadth where 

 widest ; thence they taper to the extremity. The number of 

 joints is about thirty. The smaller proximal arms cannot have 

 much exceeded 5 mm. in length. 



Three plates, which (without asserting any homology) may be 

 compared to the lateral and ventral plates of a recent Ophiuroid, 

 occur in each joint of the arms ; their ends are pointed, but not 

 spined. There are indications of the presence of a fourth or dorsal 

 plate. The existence of vertebral ossicles is doubtful, and though 

 it might be inferred from the general character of the arms, yet 

 there is important negative evidence to the contrary. Thus, in the 

 case of two of the larger arms, which possess well-preserved lateral 

 plates but have lost the ventral plates, the interior is merely filled 

 with matrix presenting no signs of ossicles, nor on clearing out the 

 interior with a brass point could any trace of such structures be 

 brought to light. It would certainly seem unlikely that the con- 

 ditions which allowed of the persistence of the lateral plates should 

 have proved so much less favourable to the vertebral ossicles, which 

 are usually the more massive of the two structures. 



The test measures from 7 to 8 mm. in diameter ; its ventral surface 

 is incomplete, indeed has wholly disappeared, with the exception of 

 a ring of polygonal plates in the region of the ambitus. Fifteen 

 plates enter into the composition of this ring ; five are single and 

 radial, while five are paired and interradial or adradial. Each of the 

 radially-situated plates is the most proximal of a series of three, the 

 second of which is cruciform in outline, and the last much smaller 

 and almost square. The adjacent distal corners of the first radial 

 and its neighbouring adradial are excavated so as to bound a semi- 

 circle, which is completed and converted into a circular aperture by 

 the rounded angle of the cruciform plate, and by an additional 

 element which may be regarded as a second adradial. Through 

 this aperture an arm of the first pair proceeds. The second pair of 

 arms proceeds from a similar but larger aperture which is bounded 

 by the distal angle of the cruciform plate, and by additional plates 

 which are not clearly enough revealed for description. By a careful 

 removal of the matrix the sides of the test have been partly exposed 

 to view. A single polygonal interradial plate, relatively large, lies 

 between the two adradial plates above, and on each side of it 

 another polygonal plate is seen, but its relation to the radial plates 

 is hidden by the arms which overlie it. 



The dorsal surface, owing to its envelopment in the matrix, is con- 

 cealed, but by cleaning out the interior of the test a glimpse of part 

 of it may be obtained from within ; the central region, however, 

 remains covered by the buccal armature. Few details can be made 

 out, but it is clear that the roof of the test was as compactly built 

 as the sides ; it presents itself as a continuous calcareous sheet, in 

 which the existence of five comparatively large radial plates is 

 barely suggested. 



Q. J. G. S. No. 220. 2 z 



