396 



nection with the rest of the stem. Within the vesicle the circu- 

 lation of the globules may be seen very distinctly. On the oppo- 

 site side of the thin région tentacle „Anlagen" appear also (fig. 10 [i) 

 soon afterwards. I could observe the contractions of the coenosarc 

 at intervais of a few minutes, and during (me of its contractions 

 the connection between two hydranth „Anlagen" was broken so that 

 two separate hydranths were formed (fig. 11). The process of for- 

 mation uf the hydranths and the contraction of the coenosarc take 

 place within the perisarc (fig. 11 np.) which is produeed' by the 

 ectoderm of the closing membrane. The conti'actions of the coeno- 

 sarc observed before the breaking of the stem within the perisarc 

 brought positive proofs that the morphological connection between 

 the perisarc and coenosarc at the région of the hydranths forma- 

 tion is broken. 



In the next stages after complète development of the tenta- 

 cles the hydranths break through the perisarc and émerge through 

 the two -apei'tures. 



The development of the hydranth in the center of the stem 

 is, I think, not without importance. It is a proof that the process 

 of formation of the hydranth can take place while the continuity 

 of the stem is not broken. The hydranths which are formed by 

 the pièces eut longitudinally bave a reduced number (9 — 16) of 

 tentacles. 



The hydranths of the pièces eut longitudinally break off after 

 2 to 5 days (autotomy). In some cases the breaking olf of the li- 

 ving hydranths takes place, in other cases the hydranths die and 

 subsequently become broken oflF. In the place of broken hydranths 

 a new head régénérâtes after 3 to 4 days, and the number of its 

 tentacles is still more reduced. I hâve never observed the régéné- 

 ration of the hydranth a third time which often occurs in pièces 

 eut crossvvise. 



Napoli. Zoological >>tation. 



