6 



variable in size. Above the fifth range the plates are not dis- 

 posed in ranges nor in any other defined order, and differ so much 

 in size and shape, that one can hardly say how many ranges they 

 would make if disposed in some order. We would estimate, if 

 they were in ranges, there are above the fifth range and below 

 the mouth about seven or eight ranges; or between the basal 

 piece and the mouth about twelve or thirteen ranges. Our 

 specimen is most ventricose on the right side, and there is a base 

 of a young cystidean at the lower side of the mouth and another 

 immediately below one of the arm bases and some smaller bases 

 on other parts of the body. 



Looking at Fig. 3, the mouth may be seen on the upper side 

 of the figure, two arm bases on the right and one on the left, 

 connected by a Y shaped ambulacral furrow, and in the central 

 part, between the mouth and the ambulacral furrow, what has 

 been supposed to be the small anal opening. The mouth is on 

 the margin of the anterior end and appears to have been sur- 

 rounded by six plates, but the cystidean base covers the lower 

 side of it, and hence, possibly, there are seven plates. The anal 

 (?) opening is small and at the summit of a cone which is on 

 the line of two plates. The ambulacral furrow is at the extreme 

 summit and connects the three arm bases by following the sutures 

 between the plates; the plates are denticulated or united by a 

 zigzag line at the bottom of the furrow. Each plate that sup- 

 ports an arm is thickened and prominent and the ambulacral 

 furrow is continued across this plate to the top of the prominence, 

 where the cicatrix shows the place to which the arm was at- 

 tached. The arms are not preserved. On each side near the 

 bottom of the ambulacral furrow there is a row of pores, but a 

 free plate of the same character from the same or a similar 

 species, when examined from below, does not show these- pores in 

 lines, nor can they be distinguished from the other pores that 

 penetrate the plate from all sides. The ambulacral furrow, there- 

 fore, is not homologous with the ambulacral furrows of either 

 crinoids or blastoids. There is no reason to suppose that it was 

 a food groove, was covered with minute plates, or was furnished 

 with pinnules. It appears as a triangular furrow cut only half 

 way through the plates, and where following the suture lines of 

 plates, the plates are more firmly joined than elesewhere by the 

 denticulated edges, but when it enters upon a plate that bears an 



