the Crinoidea, Cystldea, and Blastoidea. 151 



the arms. If the ventral perisome of the Criuoid were to be 

 removed (the internal organs remaining undisturbed) the ar- 

 rangement diselosed would be that represented in fig. 9 — a 

 convoluted plate^in the centre with the canals radiating from 

 it. The most striking dilfercnce is the absence of the oeso- 

 phageal ring. According to the organization of Actinocrinus 

 there could be no oesophagus at that point ; and consequently 

 there is no ring. The convoluted plate represents the madre- 

 porie apparatus. The sucking-feet of the starfish most pro- 

 bably represent the respiratory tentacles that border the 

 grooves of the Crinoids, but modified into prehensile and loco- 

 motive organs. Bipinnaria and Actinocrinus agree in having 

 the mouth in one of the interradial areas, and in the absence 

 of an orifice through the peristome at the ambulacral centre. 

 These two characters are embryonic and transitory in the star- 

 fish, but they were permanent in most palaeozoic Crinoids. 



In Codonites stelliformis {Pentremites stelliformis^ Owen and 

 Shumard), figs. 10, 11, the ambulacral centre, c, is completely 

 closed. Five minute 'grooves radiate to the extremities of 

 the five angles of the disk. These grooves are identical with 

 those of Pentremites and Nucleocrimis and were occupied by 

 the ovarian tubes. The ambulacral canals of the true Cri- 

 noids and of the starfishes are represented in a rudimentary 

 condition, in this species, by the hydrospires, which open out 

 to the surface through the ten fissure-like spiracles (.s-). The 

 oro-anal orifice is interradial. C. stelliformis in external form, 

 the interradial position of the mouth, and the closed ambulacral 

 centre resembles Bijnnnaria and Actinocrinus, but differs im- 

 portantly in having its respiratory organs arranged in ten 

 separate tracts, all totally disconnected from each other. It is 

 a lower form than Actinocrinus^ which, in its turn, is lower 

 than Bipinnaria ; and yet all three are constructed on the same 

 general plan. 



C. stelliformis, although much resembling ix, Pentremites, is 

 a true Cystidean. Its affinity to Codaster was first pointed 

 out by Dr. C. A. White, who also suggested that it should be 

 assigned to a distinct group (Bost. Journ. N. H. vol. vii. 

 pp. 486, 487). The main difference between the Cystidea and 

 the Blastoidea is, that in the former the hydrospires do not 

 communicate with the pinnule, whilst in the latter the cavi- 

 ties of the pinnuhe and hydrospires are directly connected by 

 the ambulacral pores. 



The development of the recent Crinoid Antedon rosaceus, 

 as described by Prof. Wyville Thomson (Phil. Trans. 1866), 

 pursues a course that could not possibly result in the produc- 

 tion of such an animal as Actinocrinus. The pseudembryo, 



