Dei'eiopynent of IloJolhurians. 419 



are considerably smaller than the rcniauiln,:^ eight, bjloug not 

 to the rive secondary tentacles, but to the rive primary ones. 



The whole of the tentacular canals arise from the radial 

 vessels by a basal ))ortion, which is at rirst very short an I 

 narrow, but afterwards increases in lenf];th, and which opens 

 by means of a valve into the wider section of the tentacular 

 canal, lying in the tentacle itself. These valves, in spite of 

 their small size, are constructed of two semilunar foil-?, pre- 

 cisely as is already known to be the case in the tentacles of 

 Synajyia. The narrow basal portions of the tentacular canals, 

 as well as the valves at the distal end of these portions, lie 

 internally to the radial ossicles of the pharyngeal ring, which 

 are already present on the eighth day of development. 

 Beyond the valve the expanded section of the tentacular 

 vessel bulges out backwards, forming a short ca3cal process 

 which lies outside the young calcareous ring, and there rests 

 upon the lateral branches of two neighbouring radial ossicles. 

 'Jliis cajcum is the rudiment of the homologue of a tenta- 

 cular ampulla, which Herouard has siiown to exist in the 

 adult animal. No muscle-fibres could be distinguished in th^ 

 wall of the narrow portion of the tentacular vessel, even in 

 the most advanced of the developmental stages examined. In 

 the expanded portion, on the other hand, distinct longitudinal 

 muscle-jibres (and only such), furnished by the cells of the 

 epithelium of the hydrocoele, appear in a single layer as early 

 as the tenth day. Until the tifteenth day the tentacles are 

 simple cylindrical structures with rounded tips, which are 

 beset by the tiny liyaline papillte already noticed by Krohn 

 and Selenka. On the day named the subsequent arborescent 

 shape of the tentacles begins to be ushered in, by the bifur- 

 cation of the tips. On the following days these two branches 

 are soon succeeded by other branches which appear balow the 

 tip. The wdiole of the branches enclose from the beginning 

 a Cffical process of the tentacular vessel. 



Rudiments of the first two feet are already present on the 

 eighth day. At first they each lie concealed in a pit-shaped 

 hollow of the integument, and on emerging from this pit, 

 wdiich then flattens out, have the form of a small hemi- 

 spherical protuberance. During the following days they 

 elongate more and more into cylindrical tubes, and on the 

 eighteenth day a well-developed terminal disk can already be 

 distinguished. The two primary feet receive their water- 

 vessels, as has already been observed by Selenka, from the 

 terminal portion of the median ventral radial vessel, from 

 which they arise exactly opposite one anothei". Nevertheless, 

 by closely observing them from the eighth to the eighteenth 



