REPORT ON THE CRINOIDEA. 21 



altogether misunderstood by Vogt and Yung, who have apparently attempted to work 

 out the anatomy of the type by one method only, that of thin sections, and have almost 

 completely ignored its osteology. Had they devoted a little more attention to the 

 characters of a prepared skeleton of Comatula they would have avoided not a few errors 

 which are calculated to give the student an altogether erroneous conception, not only of 

 Crinoid morphology, but of that of Echinoderms in general. The basal plates are among 

 the earliest calcareous structures which appear in the larva of any Echinoderm, and their 

 relation to the great nerve centre of a Crinoid renders them additionally important 

 morphologically. But no student of Messrs. Vogt and Yung would ever learn of their 

 existence at all. 



The gradual development of the rosette out of the original basal plates of the 

 Pentacrinoid larva was fully described by Dr. Carpenter, 1 who showed that it is 

 " essentially formed at the expense of the secondary or ventral layer of the original 

 basals, the ends of the curved rays being the sole residue of their primary or dorsal 

 layer." Alternating with these spout-like processes, which are radial in position, are five 

 others of a more triangular form, which occupy a somewhat deeper situation within the 

 radial pentagon. The apex of each of them is attached to a suture between two con- 

 tiguous radials, just between the two adjacent apertures of their central canals. Each of 

 these canals receives a branch of the primary basal cord proceeding from the central 

 capsule, that lies on the dorsal side of the interradial process of the rosette ; and when 

 the rosette is in its natural position in the calyx, an opening for the passage of one of 

 these secondary basal cords is visible between every two of the processes of the rosette. 

 This is well seen in Antedon eschrichti (PI. I. fig. 8c). The example of this species 

 which is here represented, has a comparatively simple rosette, which is almost entirely 

 free from any trace of the accessory structures to which I have given the general name of 

 the "basal star," such for instance as is represented in figs. 1-5, c on PI. II. In all these 

 forms, and more especially in Antedon angusticalyx and Antedon insequalis (figs. 4c, 5c), 

 a larger proportion of the embryonic basal has been left unabsorbed than is usually the 

 case in . the European and Arctic Comatulae ; but the peripheral margins of each plate 

 remain, and form, by their union with the corresponding parts of the adjacent plates, the 

 structure which I have called the basal bridge. This is united to each radial along the 

 inner margin of its dorsal face, and partially covers in the two secondary basal cords 

 which are converging; on its single axial canal. It is well shown in Acti?iometra 

 maculata and Actinometra stelligera (PI. V. figs, lc, 5d) and also in the rosette of the 

 latter species disconnected from the radials as seen in fig. 5c ; and it appears, so far as I 

 am aware, to be of pretty constant occurrence in this genus, though absent or at any rate 

 undistinguishable in some species of Antedon (PI. III. fig. 6b). 



United with each angle of the pentagon formed by the five basal bridges is one of the 



1 Phil. Trans., 1866, p. 745. 



