92 CATALOGUE OF THE BLASTOIDEA. 



presented in Granatocrinus ellipticus. The calyx-plates are relatively thin and not 

 produced upwards in the interradial areas, so that the radial sinus has no great 

 depth, and there is no room on its sides for any hydrospire-slits. But a hydrospire- 

 sac with strongly plated walls is attached to either edge of the radial sinus, and 

 hangs down freely into the body-cavity. This is well shown in the weathered speci- 

 mens of Granatocrinus ellipticus, which are represented on PI. X. figs. 13, 14, and 

 also in the sections shown in PL XVII. figs. 2-10. Each hydrospire-sac may consist 

 simply of one lamellar tube dilated at its lower end as in Granatocrinus ellipticus, 

 G. campanulatus, G. Derbiensis, G. orbicularis, and Schizoblastus Bofei (PL XVII. 

 figs. 2-7) ; or there may be two tubes to each sac as in G. Norivoodi (fig. 8), three 

 as in Mesoblastus angulatus and M. elongatus (figs. 9, 10), or five as in Schizoblastus 

 Sayi (fig. 1). 



The duplex character of the hydrospire-sacs of Granatocrinus Nonvoodi and their 

 relation to the radials is well shown in the internal views of them which are given on 

 PL VII. figs. 1, 2. The outer plate of each lamellar tube is attached to the edge 

 of the radial and does not appear externally except when the radial is removed, 

 as shown in PL X. figs. 12, 14. But the inner plate comes up to the surface from 

 beneath the lancet-plate which rests upon it, as seen in fig. 14. Between this inner 

 plate, which we have called the hydrospire-plate, as explained above 1 , and the edge of 

 the radial sinus is the cleft-like upper opening of the hydrospire-sac, broken up into 

 a series of pores by the repeated connections of the hydrospire-plate and the radial 

 (PL IV. fig. 4; PL VI. fig. 10; PL VII. fig. 15; PL VIII. fig. 20 ; PL IX. figs. 7, 

 16 ; PL X. figs. 8, 11, 12 ; PL XI. figs. 11-15). 



The adhesion between the hydrospire-plate and the radials is well seen in the 

 weathered fragments of Granatocrinus Derbiensis which are represented on PL XL 

 figs. 11-13. The two last figures show the distal ends of the ambulacra, one (fig. 12) 

 with and the other without the worn lancet-plate in situ. In the latter case, and 

 also in the G. ellipticus shown in PI. X. figs. 12-14, the gap between the two hydro- 

 spire-plates leads down into the general cavity of the calyx, as is evident from the 

 sections represented on PL XVII. figs. 4, G, 7. Fig. 11 on PL XI. shows a similar 

 condition of weathering near the middle of an ambulacrum of G. Derbiensis. The 

 impressions of the side plates are visible on the lancet-plate, as is seen on one side 

 of the better preserved individual figured on PL IX. fig. 7, the plates being in situ on 

 the other side. In G. campanulatus, G. ellipticus, and G. orbicularis there is much 

 the same condition as in G. Derbiensis (PL VIII. figs. 15, 20 ; PL IX. fig. 16). The 

 lancet-plate is wide and the hydrospire-plate scarcely exposed except where the 

 lancet-plate is removed, or near the tip of an ambulacrum, and the same is the case 

 in Cryptoblastus melo (PI. VII. fig. 15). 



1 Antea, p. 49. See also Ann. & Mag. Nat. Hist. 1882, vol. ix. p. 215. 



