226 XF.W YORK STATE MUSEUM 



sutural canals would be inclined toward a point under the middle 

 of a plate suture and the b.vs, would discharge in this direction 

 to reach the common tube or sac of the group. It is quite evident 

 in text figure 3 that the triangles at the plate corners form the 

 related groups in Cliocrinus. Text figure 11 will show that the 

 sutural canals of Palaeoc3\stites are inclined toward a point under 

 the plate corner and that here also the corner groups are the 

 related ones. Such evidence as there is then is markedly in favor 

 of calling the triangular series a natural group and not the parallel 

 series. The primary interradial group just described may be con- 

 sidered as typical also of four others which were simultaneously 

 developed at the orad angles of the other BB. 



Development in complexity. The small portion of the cup of 

 Cliocrinus perforatus shown in text figure i was pos- 

 sessed of more than 230 b.vs. This would indicate that the com- 

 I)lete specimen possessed some 5000 similar structures. P a 1 a e o- 

 c r i n u s s t r i a t n s on the other hand had a cup practically re- 

 duced to but three circlets of five or six plates each and its b.vs. were 

 less than 150 in number. We should therefore expect to find some 

 compensatory arrangement whereby the physiological balance of 

 respiration might be maintained. The arms doubtless came to take 

 a larger share of this function and so relieve the cup surface, but 

 the latter abundantly shows that it also became adapted to carry on 

 the same function to a greater extent than formerly. \\'e have 

 already seen that the oval sutural canals of Cliocrinus per- 

 foratus may be indicative of the movement of the free ends of 

 b.vs. toward the plate centers or to regions of purer water. If the 

 b.vs. could brancli under the theca to give rise to new members, 

 their branching or forking outside of the theca need not be unex- 

 pected. Any b.v. possessing a forked structure would come to hold 

 the arms of the Y in the best functional position which would be in 

 a line crossing the suture. Such a structure could not be easily 

 withdrawn and protection would be first secured by making the 

 arms of the Y lie close to the plate surface and thus changing the 

 Y to a T. The close contact of the arms of the b.vs. with the 

 plate surface might inhibit the formation of epistereom immediately 

 beneath them and stimulate growth between them. Ridges so 

 formed would be an additional means of protection and, however 

 initiated, would be favored by natural selection. 



With the necessity for occasional withdrawal removed, the arms 

 of the b.vs. could extend with plate growth and all available plate 

 area be used for respiratory purposes. The form might thus easily 



