Development of Ilolothurians, 417 



clistlnguisli a first pair, a second pair, and an unpaired 

 tentacle. The true state of the case is exactly the opposite : 

 in front lies an unpaired tentacle, followed bj the four others, 

 in two pairs, one behind the other. This arrangement does 

 not become perfectly distinct until we take into consideration 

 the fact, which has been hitherto overlooked, that the plane 

 of symmetry of the young- Gucumaria diverges in front 

 towards the left and behind towards the right from the plane 

 of symmetry of the larva. The arrangement of the tentacles 

 which I have just indicated refers strictly only to the plane 

 of symmetry of the young Holothurian. With reference to 

 the plane of symmetry of the larva, on the other hand, the 

 tentacles are asymmetrically arranged, so that three of them 

 belong to the left half of the body of the larva and the two 

 others to the right. 



According to Kowalevsky and Selenka the water-vessels of 

 the first five tentacles arise immediately from the water- 

 vascular ring and alternate with the radial vessels. This 

 statement is absolutely erroneous. The tentacular vessels 

 arise, on the contrary^ from the growing radial vessels. 

 Semen's speculations upon the pliylogeny of Echinoderms, in 

 so far as tliey are based upon the assumption that the primary 

 tentacles in all Holothurians arise from the water-vascular 

 ring, and on their part determine the true radii of the Holo- 

 thurian body, consequently entirely miss the mark. In the 

 case of Gucumaria their correctness is entirely overthrown by 

 the fact that the first five tentacular vessels are by no means 

 disposed in regular radial fashion. Were this the case a ten- 

 tacular vessel would be given off from each of the five radial 

 vessels. This, however, is not the fact. The arrangement 

 of the first five tentacular vessels is neither radial nor bilate- 

 rally symmetrical, but asymmetrical, in that the two tentacles 

 of the tv)o ventral interradii receive their loater-vessels from 

 the median ventral radial vessel, while the tentacle of the 

 median dorsal, as well as that of the left dorsal interradius, is 

 supplied from the left dorsal radial vessel, and lastly the ten- 

 tacle of the right dorsal interradius from the right dorsal radial 

 vessel. The median ventral and the left dorsal radial 

 vessels therefore each give off two tentacular vessels, but the 

 right dorsal radial vessel only one. The points of origin of 

 the two tentacular vessels of the median ventral radial canal 

 are situated exactly opposite one another ; so are also the two 

 tentacular vessels of the left dorsal radial canal. The two 

 latero- ventral radial vessels, on the other hand, give off for 

 the present, so long as only five tentacles altogether are 

 present, no tentacular canals at all, and therefore in this 



