PHYLUM ECHJNODERMATA 



437 



to regard those classes as the more ancient which have the radial 

 symmetry less completely developed. Again, the free condition 

 which characterises all existing Echinoderms with the exception 

 of a few Crinoids, is probably less primitive than the attached, 

 since in other phyla the radial symmetry is co-ordinated with, 

 and seems to be developed on account of, a fixed, usually stalked 

 condition. Probably, then, stalked Echinoderms were the pro- 

 genitors of the existing free forms, and these were preceded by 

 primitive free forms with pronounced bilateral symmetry. It 

 appears to be most probable that this ancestral form possessed the 

 most essential features of the dipleinida larva (p. 432) ; i.e., that 

 it was a bilaterally symmetrical form with a pre-oral lobe, simple 

 alimentary canal with mouth on ventral surface and anus at 

 posterior end ; that it had a ccelome, originally developed from the 

 archenteron of the gastrula ; and that it had a band of strong 

 cilia running around the concave ventral surface. Such a 

 dipleurula-like form became converted, it is supposed, into a fixed 

 form, such as that represented by some of the extinct class of the 

 Cystoidea. The fixation must be supposed to have become 

 effected through the medium of the pre-oral lobe, and further 

 changes must have involved the shifting of the mouth to about 

 the middle of the free surface. From this primitive Cystoid, thus 

 regarded as the most primitive of all known Echinoderms, the 

 remaining classes, both fixed and free, might have been derived by 

 some such order of succession as that indicated in the diagram 

 below (Fig. 346). 



Holothuroidea 



Echinoidea Asteroidea °pW<""oidea 



Cystoidea 



Primitive Cystoid 



Oipleurula 



Fig. 340. — Diagram to ilhuitrate the relationships of the classes of the Eehinodcrmata. 



