44 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 57 



Echinodermata ^ is considered to have passed through a " Pelmato- 

 zoic " stage in which the animal was attached to some object by a 

 part of its body wall, and in which the mouth and, to a less extent, 

 the other apertures faced upward. This stage is represented by 

 text fig. 2. 



Selecting the characters common to the early stages of all Echin- 

 oderms, a diagrammatic reconstruction of this imaginary phylo- 

 genetic stage gives a marine animal with the longer antero-posterior 

 axis parallel to the sea floor. The mouth was antero-ventral, anus 

 posterior or postero- ventral, the two joined by an uncoiled gut with 

 perhaps a stomachal enlarg-ement in the middle as represented by 



fig-- 3- 



The simplest larval form among recent echinoderms, Auriadaria 

 of the Holothurians, differs from fig. 3 in being bent upon its ven- 

 tral surface so that the mouth lies in the middle of a concavity and 

 the anus on the ventral surface of the lobe back of the concavity. It 

 also shows a decided change in the arrangement of the coil of the 

 alimentary canal and the coelomic cavities, as may be seen by com- 

 paring figs. 3 and 4. 



Fig. 4.- — Diagrammatic reconstruction of the imagined primitive Holo- 

 thurian type. (After Lankester, 1900, fig. 16, p. 18.^) = mouth; As = 

 anus ; Ihc ^= left hydrocctl ; M = hydropore ; g = genital opening. 



I have mentioned the theoretical ancestor of the echinoderm and 

 of one of its classes, Holothurioidea, in order to note that the zoolo- 

 gist has not carried his theoretical line back to the period when the 

 ancestral form was pelagic and had not yet adjusted itself to the 

 conditions of the littoral zone, stages which must have preceded the 

 migration of this organism over the bottom into the deeper water. 

 This still earlier ancestor must have been a free swimming, soft 

 bodied animal. It undoubtedly was more simple than the free 

 swimming Eldonia ludzvigi described in this paper, and I can readily 

 imagine a small bell-shaped body with a simple alimentary canal 

 opening at both ends on the ventral surface — a medusa-like object 



^ See footnote on p. 43. 



