132 THE CRINOIDEA CAMERATA OF NORTH AMERICA. 



See our diagram (Fig. 11). So also the plate t of Heteroerinus hellevillensis 

 (Fig. 10), though less symmetrically disposed, rests upon both radials; and 

 not on the plate to the right only, as figured by Bather. 



We now pass to those forms in which the inferradial B' performs the 

 functions of an anal plate. The first step in this direction is shown by 

 Homocrinus (Fig. 14), Botryocrinus, Oncocrinus, and Barycrinus, in which the 

 superradial B has shifted sHghtly to the right, and is connected with the 

 inferradial by an obhque suture ; contrary to the case of Dendrocrimis, 

 in which the two plates are arranged vertically (Fig. 13). The plate t in 

 those forms rests upon R without touching R\ and the plate x, which is 

 comparatively large, rests against both sections of the compound radial. 



Bather explains the evolution that took place in these forms by " the 

 shifting of the radianal," and that " x has sunk still lower into the dorsal 

 cup, and is now on a line with the radials." Nothing of this kind is indicated 

 by the specimens, which clearly show that the radianal throughout the 

 Fistulata retains the same position, whether it constitutes a part of the radial 

 or serves as anal plate. It only changes its outlines so as to conform to the 

 shape of contiguous plates. As the tube became larger, the radials spread 

 out, and the vacant space thus formed was filled by a new plate, x. There 

 was no sinking of the plate t, which never moved from its place above the 

 radials. 



Another stage in the developmental history of the anal area among the 

 Fistulata is presented by Parisocrimis, Atelestocrinus^ Euspirocrinus, and the 

 typical Poteriocrinidse. Bather, in alluding to them, makes the following 

 statement : " In the Poteriocrinites (see our diagram Fig. 2) another change 

 has taken place ; the radianal has passed through a revolution of 90°, and the 

 lowest plate of the ventral sac {t) has sunk down between R and x." If we 

 understand what this means, he assumes that the lower section of the radial 

 moved to the left ; while in fact, as the specimens show, it was the upper 

 section of the plate that moved away, shifting to the right and leaving 

 a space for R and t to meet. Thus it was that x came to rest against R\ 

 but not against R, from which it was separated by the plate t. An increase 

 of width in the anal area became necessary, as the surface for the support of 

 the tube was insufficient to hold it. In Homocrinus and De7idrocrinus, the 

 costals occupy only a comparatively small part of the radial, and a rather 

 large portion of the latter serves as a support for the tube. In the Poterio- 

 crinidae, however, and in Parisocrinus, Euspirocrinus, and Atelestocrinus, in 



