26 



is without thorns (PI. IV. Fig. 14). hi transverse sections the spines look essentially 

 like those of Astropyga (Comp. Mackintosh. Researches on the structure of the 

 spines of the Diademalidæ. Trans. R. Irish. Acad. XXV. 1870. p. 544. PI. XXXIII. 

 figs. 31 — 35). The central cavity is filled with a very open reticular network; 

 the solid radii may be connected by a few transverse bars. In the flattened end 

 of the spines at the ambitus the cavity is quite filled by a close nieshwork. 

 (PI. V. Fig. 18—19). 



The pedicellariæ are tridcntate, triphyllous and claviform as in Astropyga. 

 The tridentate pedicellariæ (PI. IV. P'ig. 1- PI. V. Fig. 22, 35) have narrow, elongated 

 valves, a little widened towards the end; the edge is .strongly serrate; Ihej' 

 are wide apart, joining only at the end. A little meshwork maj' be found at the 

 lower end of the larger ones. Only one form of tridentate pedicellariæ is found, 

 but Ihey are very ditferent in size, from c. 0,3 mm. to c. 1,5 mm. (length of head). 

 In the larger ones the neck is short, in the small ones rather long. Sometimes 

 also the stalk may be very, short. The valves of the triphyllous pedicellariæ (^Pl. IV. 

 Fig. 24) have only some of the inner holes a little elongated, thus not presenting 

 the beautiful aspect of those of Astropyga. The claviform jiedicellariæ (PI. IV. 

 Fig. 4, 32) are especially large and numerous between the large spines just below 

 the ambitus; on the abactinal side much smaller ones occur. As de Meijere has 

 pointed out there may be found claviform pedicellariæ with the head developed, 

 the valves being simple, with almost straight, finely serrate edges (PI. III. Fig. 11), 

 but undoubtedly of the ophicephalous type; these pedicellariæ alone would prove suffi- 

 ciently that the claviform pedicellariæ of Diadeinatidœ are develoi)ed from ophice- 

 phalous ones. The stalk of the pedicellariæ is as in Diadeina and Astropyga. 



The sphæridiæ are globular, quite smooth; they are placed outside the 

 granulated portion, almost at the ambitus, 3 — 4 to each pore-area ; other- 

 wise they are placed at the lubefeet as in other Diadematids, not in the median line 

 of the anil)ulacral area. In a young specimen of 19 mm. (Challenger, see below) 

 they proceed to the edge of the peristome. In another young specimen of 22 mm. 

 diameter one of the sphæ^ridia» in each series is placed in the granulated portion, 

 the other outside between the large spines. It thus seems that the sphæ^-idiæ 

 disappear from the granulated part as it becomes more densely covered with 

 spines — as if they could not perform their functions in that dense forest 

 of spines. 



The spicules (PI. IV. Fig. 13, 15) are triradiale, very slender, and are more 

 or less distinctly arranged in longitudinal series; in the buccal feet thej^ are more 

 irregular. The sucking disk is small, the rosette slightly developed. In the ab- 

 actinal tubcfcet the spicules are exceedingly fine, more or less irregularly branched; 

 they are found especially at the point of the foot. (In the preliminary description 

 it is wrongly staled that spicules are wanting in the abactinal tubefeet). No 

 spicules are found in the walls of the intestine and the genital organs. 



