l6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 76 



C. bradleyi it drops to i to 1.20, and in C. car pent eri still farther to 

 J to 1. 1 7 — although in the latter case the reduction is due to an 

 increase in r. ant. R. at the expense of ant. R. 



The openings at the margin for the dorsal canals in Catillocrinus 

 are round, instead of linear as in Mycocrinus boletus; in the latter 

 they are very similar in form to those upon the radial facets in 

 Synbathocrinus. 



MODIFICATION IN LATER SPECIES 



The three species next following C. tennessceae, up to that of the 

 earlier Keokuk, agree with it in the essential structure of the calyx 

 other than that of the base — the arrangement and relative propor- 

 tions of the radials being about the same within the limits above 

 mentioned. In the progress of the type from there on some remark- 

 able changes took place. The first of these relates to the mechanism of 

 the anal side controlling the attachment of the tube to the cup, and 

 it includes both C. bradleyi of the later Keokuk and C. carpcnteri of 

 the Chester. It is marked externally by the disappearance in the last 

 two species of the raised process on the right posterior radial which 

 is present in the first four, so that the distal face of that radial is at 

 the same level as that of the others. This is accompanied by a singu- 

 lar modification of the anatomy at the base of the tube, which will 

 be explained farther on. 



But while C. bradleyi remained substantially in line with the four 

 preceding species in the distribution and size of the radials, C. car- 

 pcnteri proceeded to depart from all its predecessors by taking on a 

 radial arrangement of its own differing from that of all others in 

 the family. This is effected by an increase in the size of r. post. R. 

 and r. ant. R. and reduction in that of the two larger radials until 

 they occupy no greater part of the circumference of the cup than they 

 do in Mycocrinus. But in addition to this the r. ant. R. is increased 

 much more than its fellow of the pair, so that now, instead of narrow- 

 ing to an apex as in other Catillocrinus, and bearing only a single arm 

 as it does in all other species within the family, it becomes widest 

 above, and supports almost as many arms as the adjoining ant. R. — 

 having 3 to 5, or an average of 23 per cent of the total number 

 of arms, as against 7 per cent in Mycocrinus and 2 per cent in C. 

 tennesseeae. Such a departure from one of the prime characteristics 

 of the family, correlated with the diminution in size, shows a loss 

 of vigor in the type presaging its extinction. 



