REPORT ON THE CRINOIDEA. 11 



The centro-dorsal is at first a simple ring, in no way different from the other stem- 

 joints ; but when the basals come to assume a definite shape and the calyx acquires the 

 doubly conical form of the Cystic! phase, the centro-dorsal becomes distinctly wider than 

 the annular stem-joints below it and takes on a pentagonal shape. The basals rest 

 against the sides of the pentagon, and its angles which fit in between them are therefore 

 radial in position, as seen in PI. XIV. figs. 1,8. At this early stage the basals are only 

 in contact with the centro-dorsal by their lower edges ; but it soon begins to increase in 

 diameter and extends itself over the bottom of the calyx in the manner described by 

 Dr. Carpenter. 1 It increases at the same time in vertical depth, and the first cirri make 

 their appearance. These are radial in position, and the portion of the centro-dorsal 

 between every two sockets rapidly enlarges, so that it comes to project beneath each basal 

 plate, and the angles of the centro-dorsal thus become interradial instead of radial. This 

 change is very clearly seen in larvae which have only one or two cirri, so that one part of 

 the centro-dorsal shows the primitive radial symmetry, and another part the acquired 

 interradial symmetry. 



Thus then the centro-dorsal of Comatula, when it first assumes definite form, has a 

 most distinct radial symmetry. Its angles occupy the same position with regard to the 

 basals as do those of the enlarged top stem-joint in Guettardicrinus and Apiocrinus, 

 which are also distinctly radial in situation. I desire to lay particular stress upon this 

 fact, because Wachsmuth and Springer, in support of their assertion that Neocrinoids are 

 built upon the plan of dicyclic Crinoids, have stated that the top stem-joint " is disposed 

 interradially in the Apiocrinidae, Pentacrinidae, and Comatulse, similar to dicyclic Palseo- 

 crinoids." 2 But the ridges and angles of the top stem-joint are radial in every species of 

 Apiocrinus, as is seen with especial clearness in Apiocrinus magnificus. 3 Wachsmuth 

 and Springer 4 say, however, that " the plate in Apiocrinus magnificus is not, as should 

 be supposed from appearances, disposed radially, but interradially, as shown by comparison 

 with species having a pentangular stem. It attained its radial angles accidentally by 

 adapting its form to the basal concavity which is naturally angular." This is a form 

 of teleological argument which is very easily employed but is very difficult to refute. 

 Neither Wachsmuth nor Springer, nor any one else, is acquainted with the post-embryonic 

 development of Apiocrinus, and the changes which may or may not have taken place in 

 the symmetry of its top stem-joint ; though from the positive way in which the 

 American authors write one would imagine that they had watched the whole process of 

 the " accidental " change of symmetry which they describe. If the basal concavity 

 "naturally" has radial angles, it is surely a "natural" and not an "accidental" 

 circumstance that the top stem-joint which occupies this cavity should also have radial 

 angles. This is the case in every species of Apiocrinus, in the single species of 



1 Researches on the Structure, Physiology, and Development of Antedon rosaceus, Phil. Trans., 1866, p. 742. 



2 Revision, pt. iii. p. 299. 



3 See de Loriol, Paleontologie Franchise, Terrain Jurassique, t. xi. pis. 46-49. * Revision, pt. iii. p. 297. 



