a 
276 THE AMERICAN NATURALIST. [VoL. XXXII. 
development, and it is equally certain that in the adult of this same 
species, as well as in many or all other Synaptidæ, they are entirely 
lacking.! In S. vévépara the five primary interradial outgrowths of 
the hydrocoel grow forward and constitute the canals of the five 
primary tentacles; the five secondary radial outgrowths extend for- 
ward, each to a point immediately in front of which a radial nerve 
passes outward in its course to the body wall. From three of the 
radial outgrowths of the hydrocoel branches soon arise, which grow 
forward on either side of a radial nerve and form the basis of 
the accessory tentacles; from the fourth or left-dorsal radial out- 
growth a single branch arises, ventral to the radial nerve, forming an 
accessory tentacle of the left-dorsal interradius ; whereas the fifth or 
mid-ventral outgrowth is never more than a slight protuberance, 
from which no accessory tentacles are normally developed, and which 
usually soon atrophies and disappears. None of the radial out- 
growths of the hydrocoel is prolonged to form a radial canal in the 
body wall. 
From the fact that no radial canals are ever developed in S. vzvipara 
it is evident that we have in this form a more degenerate condition 
of the water vascular system even than in S. digitata. The fact that 
the stone canal in the adult S. včvýģara has an opening directly upon 
the surface of the body, with a madreporite near the body wall 
having openings into the ccelom, would seem to indicate, on the 
other hand, that in this one particular the water vascular system has 
retained its primitive structure. 
When, furthermore, we compare the water vascular system of the 
larva of S. vivipara at the stage with ten tentacles with that in 
Cucumaria at a similar stage, as described by Ludwig,” we find 
strikingly similar conditions. If the Synaptide have been derived, 
as Ludwig has suggested, from an ancestor the tentacles of which 
arose as branches from five radial outgrowths of the hydrocoel, and 
if we should, furthermore, suppose that subsequently by the gradual 
shortening of those outgrowths the five primary tentacles came 
eventually to arise directly from the hydrocoel ring, losing their 
immediate connection with the radial outgrowths, then we wou 
have a complete homology between the conditions which Dr. Clark 
has found in S. vivipara and those in Cucumaria as described by 
1 H. Ludwig m P. Barthels, Zur Anatomie der Synaptiden. Zool. Anz, 
Jahrg. xiv, hid 117-1 1891. 
2 H. Ludwig, Zur Entwickelungsgeschichte der Holothurien. S#tzungsber. k. 
preuss. Akad. Wiss., Nr. 10, pp. 179-192 ; Nr. 32, pp. 603-612. 1891 
